Home

Development of a Graphical User Interface for a

Contents

1. 5 4 1 4 conventions 4 1 5 Link between the different conventions 4 2 Programming of the interface 4 2 1 inner e ter reine 4 2 2 The detailed code eese 4 3 Operation of the interface 4 3 1 Structure of the 4 3 2 How to use the B Work ans e een eterne dre tete eer tise eee VENUE PNEU ILE Attachments o de aee HERR RSS den TN DOR eie ere Attachment 1 Enlargements of Figure 2 and Figure 3 representing th UVOKOMM AU sides EM Attachment 2 Comparison between programming environments for Attachment 3 Matlab code for the trajectory generation Development of a graphical user interface for a rehabilitation exoskeleton Contents 49 settings of the GUI of making 1 62 viii List of abbreviations List of abbreviations ALTACRO Automated Locomotion Training using an Actuated Compliant Robotic Orthosis BWS Body Weight Support CP Control Parameter CS Coordinate System DOF Degree Of Freedom GUI Graphical User Interface HHR Heilig Hart Hospital Roeselare HR Heart Rate MACCEPA Mechanically Adjustable Compliance and Controllable Equilibrium Position Actuator MAS Motor Assessment Scale MRMI Modified Rivermead Mobility Index PCI Physiological Cost Index RAGT Rob
2. An error part This part should alert the user if there is some malfunctioning of the device such as communication errors heating of the device etc 2 2 3 The physiotherapist The physiotherapist needs an interface for different purposes must be able to easily and safely control the rehabilitation device So he needs some kinematically and physiologically correct trajectories that can be sent to the device For the user friendliness the trajectories that will be sent to the device must be easily adaptable on some parameters that are important for the therapy such as the step height and step length for a gait rehabilitation robot He must also be able to control the robot on other session parameters such as the treadmill speed and the body weight support level if applicable for the device e Secondly for the objectiveness of the session the physiotherapist needs an objective way for evaluating the patient s performance Ideally this evaluation is made out of two parts The first part is an evaluation of the real time effectuated movements such as a visualization of the difference between the desired and the real effectuated trajectory This part gives an indication about the quality of the effectuated movements The second part gives an indication of how far the patient is advanced in his rehabilitation therapy such as with a bio indicator If it is the user interface which gives a session evaluation the patient s session resu
3. 250 2 Right range fs Ha Is pe Jmm fiso Left range so 180 10 o ECN Right range gt 5 0 7 Fo Left range fom 480 180 2o Right range 100 100 2 zo Freeze C 7180 7 foso 91 00 4000 47 00 5900 70 00 82 00 300 105 00 11600 128 00 13900 15100 162 00 174 00 185 00 197 00 208 00 162 01 174 01 185 01 p aM am ON om o m p am an ir Em ru FE Figure 22 Real time measurements sub tab of the GUI 4 3 1 3 Patient tab The patient tab see further on Figure 25 amp 26 contains the information directly related to the patients These are mainly his personal data and his dimensions This tab contains the possibility to open an existing patient file or to create a new one If an existing patient file is opened all the boxes with the personal data and the dimensions of the patient are automatically filled in if a new file is created the user needs to fill in all the boxes and push the Save patient settings button The file is then added to the patient database with as pathname the surname the name of the patient which makes it easy to find the file back during the next trainings session Development of a graphical user interface for a rehabilitation exoskeleton 48
4. Phase 1 A first double support phase e The right foot is going from the initial heel contact to a flat foot on the ground e The left foot is going from a flat foot to the end of the stance phase with the heel off the ground the toes are still on the ground gt Phase 2 single support phase e The right foot is in the stance phase foot flat on the ground e The left foot is the swing phase gt Phase3 A second double support phase e The right foot is going from a flat foot to the end of the stance phase with the heel off the ground the toes are still on the ground e The left foot is going from the initial heel contact to a flat foot on the ground gt Phase 4 second single support phase e The right foot is in the swing phase The left foot is in the stance phase Development of a graphical user interface for a rehabilitation exoskeleton 22 Trajectory generation LOADING MID TERMINAL INITIAL TERMINAL RESPONSE STANCE STANCE SWING SWING DOUBLE SINGLE DOUBLE SINGLE SUPPORT SUPPORT SUPPORT SUPPORT STANCE SWING 0 80 90 100 0 10 20 30 40 50 60 OF STRIDE Figure 8 The different phases in a cycle 23 3 2 Requirements Beyond the fact that the generated trajectories must be kinematically correct they must comply with the patients dimensions and with the constraints given by the presence of the ground treadmill they must also satisfy requirements coming from the
5. Vrije Universiteit Brussel FACULTY OF ENGINEERING B Brussels Faculty of Engineering Development of a Graphical User Interface for a Rehabilitation Exoskeleton Laura De Rijcke Promoter prof dr ir Dirk Lefeber Co promoter prof dr ir Bram Vanderborght Supervisors Branko Brackx ir Victor Grosu Thesis submitted in partial fulfilment of the requirements for the Master s degree in Master of Science in Electro Mechanical Engineering Academic year 2012 2013 Development of graphical user interface for rehabilitation exoskeleton Acknowledgment Acknowledgment In the first place want to thank my thesis promoter professor dr ir Dirk Lefeber He guided me through my investigation and was always willing to help me would like to thank the thesis president and my co promoter professor dr ir Bram Vanderborght for all the support and advice on my master thesis am very grateful to my supervisors ir Branko Brackx and ir Victor Grosu Their opinion really helped me in making progress Their expertise and helpfulness is really remarkable Finally would thank Doctor Piet Mortel and the other physiotherapists of the department of physical medicine locomotory rehabilitation at the Heilig Hart Hospital of Roeselare They were really helpful for discussing the physiotherapeutic aspects of a rehabilitation therapy Development of a graphical user interface for a rehabilitation
6. At normal walking speed the mean amplitude of the pelvic obliquity is 7 72 peak to peak with a standard deviation of 2 26 amplitude of the sinus changes as a function of the walking speed within a range from 6 to 20 24 e For the pelvic rotation in the transverse plane pelvic rotation during one gait cycle can be approximated by a simple sinus The sinus is starting at its maximum amplitude during the initial contact of the heel This can be visualized in Figure 10 mean amplitude of the sinus is 10 40 peak to peak with a standard deviation of 3 22 as reported by Whittle 25 The sinus is illustrated in Figure 10 The mean amplitude of the sinus is represented by the full line the standard deviation is represented by the dotted lines 110 5 Marker 100 5 f 2 2 B oel di 90 5 5 8 5 8 E 8 80 3 5 20 rs 3 H 5 E E i EN 60 S 2 i t ei 4 x amp Knee Flexion 50 2 A 2 A 2 i i 40 E 30 44 m Pelvic Obliquity 20 gt 0 10 20 30 40 50 60 70 80 90 100 Gait Cycle Figure 9 Vertical movement of the trunk upper curve flexion middle curve and pelvic obliquity bottom curve rotation angles in degree for cycle 24 Development of a graphical user interface for a rehabilitation exoskeleton 24 Trajectory generation Figure 10 Pelvic rotation in d
7. 19 pp 1017 1023 1999 15 M J Bailey et C M Ratcliffe Reliability of Physiological Cost Index measurements in Walking Normal subjects using steady state non steady state and post exercise heart rate recording Physiotherapy vol 81 n 110 pp 618 623 October 1995 16 E Eng Qt GUI Toolkit vol 31 1996 17 B Rempt et D Mertz Qt and PyQt Februari 2003 Online Available http www ibm com developerworks linux library 1 qt Accessed 12 2012 18 A Writing The advantages amp disadvantages of Visual Basic Online Available http www ehow com list 6148486 adavantages disadvantages visual basic html Accessed 12 2012 19 D Eyre Matlab basics and a little beyond august 1998 Online Available http www math utah edu eyre computing matlab intro Accessed 2012 20 R Tabone Matlab Introduction Online Available http www yorku ca jdc Matlab Accessed 12 2012 21 N instruments Advantages of Using LabView in Academic Research Online Available http www ni com white paper 8534 Accessed 12 2012 22 Advantages Disadvantages Online Available http www labviewportal eu en introduction advantages disadvantages Accessed 12 2012 23 J Perry Gait analysis Normal and pathological function SLACK Incorporated 1992 24 S A Gard et D S Childress What determines the vertical displacement of the body during normal walking
8. 4 2 2 The detailed code LabView is a graphical programming environment which is capable of parallel coding For this reason the main program situated in the second while loop is programmed in separated blocks Each block has a dedicated task and is coherent with the structure of the interface as defined in section 2 2 The communication between the different blocks is done with nodes This is a special structure in LabView which replaces the wires to pass the value of a variable from one location to another in the program The use of these property nodes instead of wires makes a large program more readable A disadvantage is that it is more difficult to see the flow of the variables so during the programming this ambiguity has to be taken into account The programming structure logic is illustrated in Figure 18 In this figure the structure is illustrated as a sequential flowchart structure Although the code is in reality programmed in separated code blocks this is done to easily illustrate the dependencies between the different parts of the program A second reason for this is to make the structure more comprehensible for the reader For the same reasons the flowchart goes through when the Booleans are false Only if the Booleans are true extra code blocks the later explained event structures are executed and byways have to be followed in the flowchart In Figure 18 the black arrows correspond to the flow in the second w
9. 60 Ws E Offset Ei Figure 3 Screenshot of the training settings of the Lokomat 8 A screenshot of the patient settings tab with the patient identification settings is displayed in Figure 2 A screenshot of the training settings for the Lokomat is given in Figure 3 The physiotherapist has to Development of a graphical user interface for a rehabilitation exoskeleton 5 Introduction impose the treadmill speed and the unloading percentage later also called the body weight support percentage BWS at the left hand side of the screen The patient coefficient at the left lower side of Figure 3 is a synchronization parameter between the Lokomat robot and the treadmill It has to be changed if the patient s foot remains too long on the treadmill or if his foot slips backward at the end of the stance phase In the middle of the training settings screen the range of motion and the offset of the hip and the knee angles can be changed It is possible to give different values for the left and the right hip and knee angles The last parameter that can be changed here is the guidance force of the robot that acts on the patient By reducing this parameter the patient can walk more freely A larger image of Figure 2 and of Figure 3 is given in attachment 1 1 2 1 2 OptoGait OptoGait is not really a rehabilitation device it is made as a training instrument for athletes OptoGait is a system for optical step detection
10. L feet sin knee i ankle i hip i end end Development of a graphical user interface for a rehabilitation exoskeleton Attachments 70
11. Some possible extra settings 19 2 2 Requirements of a GUI classified by user type 0 nnn nnn nnns 19 2 21 The different sers teret ete te ie ic tee D eii 19 2 2 2 Th r ngin ers ee PG RES RES RI ca OR NR Ede 20 2 2 3 The physiotherapist sii b 20 2 2 4 The Patients ee deo ed ren Be 21 3 Trajectory 22 3 1 sd sehen Mann RES 22 3 2 titan tete url den denn de dal ense deden 23 3 3 Trajectory generation based on previously measured joint 25 3 4 Trajectory generation independent of previously measured joint angles 26 Development of a graphical user interface for a rehabilitation exoskeleton vii 3 5 Influence of the input parameters the generated joint angles 3 6 Conclusion TS 4 Practical implementation realization of a GUI for 4 1 ALTACRO 5 4 1 1 Robot DOP S oss escis ient kn enn nne e 4 1 2 Robot frame definitions essere 4 1 3 Winter
12. gen inv L upper body L upper leg L lower leg L feet the ta hip 1 theta knee 1 theta ankle 1 Step length z gen phasel z mean phasel x gen phasel x gen phasel calculations of phase 2 x inter 2 x phase 1 6 phase 1 6 traj 2 x end L feet length traj 2 x 1 traj 2 x end L feet line toes 2 0 z phase 1 6 end traj 2 x end L feet x phase 1 6 end x inter 2 x phase 1 6 end z phase 1 6 end x toes 2 traj 2 1 feet 2 line toes 2 traj 2 z 2 oe creation of a trajectory for the toes x 3 end traj 3 x end sqrt L feet 2 traj 3 z end 2 x inter 3 x toes 2 end x toes 2 end x 3 end length traj 3 x 1 x 3 eng line toes 3 0 x inter 3 x toes 2 end line toes 2 end x toes 3 traj 3 1 feet 2 line toes 3 traj 3 z 2 x end 4 traj 1 x 1 L feet cos asin Heel off L feet x inter 4 x toes 3 end x toes 3 end x end 4 length traj 4 x 1 x end 4 line toes 4 0 x inter 4 traj 3 x end line toes 3 x toes 4 traj 4 x sqrt 1 feet 2 line toes 4 traj 4 z 2 x inter 1 x toes 4 end x toes 2 1 x toes 4 end length traj 1 x 1 x toes 2 1 spl spline x toes 4 end x toes 2 1 line toes 4 end line toes 2 1 1 line toes l ppval spl x inter 1 x toes l traj 1 1 feet 2 line toes 1 1 1 z 2 x inter 2 x toes l end x toes l end traj
13. x sol z sol calculation of the invers kinematics for the hip knee and ankle joints calculation of the new knee angles with the invers kinematics x and z are the given coordinates of the hip extremity s for i 1 length traj x c2 i traj z i z i 2 traj x i x i 2 L upper leg 2 lower leg 2 2 L upper leg L lower leg end mi in max c2 m max ml m p for i 1 length traj x c2 1 5c2 1 7 s2 i sqrt 1 c2 i 2 theta knee i atan2 s2 i c2 1i end initialisation of zero matrices for the calculation of the angles with the inverse kinematics in the swing phase theta hip zeros l1 length traj x Development of a graphical user interface for a rehabilitation exoskeleton 68 Attachments cl zeros 1 length c2 sl zeros 1 length c2 calculation of the new hip angles if phase phasel for i l length c2 if theta knee i 0 theta knee i theta knee i end cl i 1 L upper leg 2 L lower 1 9 2 2 1 upper leg L lower leg c2 i L up per leg L lower leg c2 i z i traj z i L lower leg s2 i traj x i x i 51 1 1 upper leg 24L lower leg 2 2 L upper leg L lower leg c2 i L low leg s2 i z i traj z i L upper leg L lower leg c2 i traj x i x i theta hip i atan2 s1 i cl i end theta ankle ankle stance elseif phase phase2 for i l length c2 if theta knee i 0 theta knee
14. JPO Journal of Prosthetics and Orthotics vol 13 n 13 pp 64 67 2001 25 M W Whittle et D Levine Three dimensional relationships between the movements of the pelvis and lumbar spine during normal gait Human Movement Science vol 18 n 15 pp 681 692 1999 26 D A Winter Foot trajectory in human gait a precise and multifactorial motor control task Physical Therapy vol 72 n 11 pp 45 53 1992 Development of a graphical user interface for a rehabilitation exoskeleton 57 Sources 27 berg Karsznia et berg Basic gait parameters reference data for normal subjects 10 79 years of age Journal of rehabilitation research and development vol 30 pp 210 210 1993 28 T Oberg A Karsznia et K Oberg Joint angle parameters in gait reference data for normal subjects 10 79 years of age Journal of rehabilitation research and development vol 31 n 13 p 199 1994 29 J L Astephen K J Deluzio G E Caldwell et M J Dunbar Biomechanical changes at the hip knee and ankle joints during gait are associated with knee osteoarthritis severity Journal of Orthopaedic Research vol 26 n 13 pp 332 341 2008 30 The ALTACRO Prototype VUB Online Available altacro vub ac be Accessed 01 06 2013 31 R van Ham B Vanderborght M van Damme B Verrelst et D Lefeber MACCEPA the mechanically adjustable compliance and controllable equilibrium position a
15. Physical human robot interaction The focus of this work is on the design and realization of a Graphical user interface GUI of ALTACRO A GUI is very important for medical robots This has several reasons 3 e First a medical robot is usually a complicated system that may include precise mechanical parts advanced electronic circuits an automatic control system and sophisticated computer software e Second the users of medical robots may not have any expertise in engineering e Third medical robots have special requirements on operation and working environments With the integration of a well designed GUI the handling of the robot is simplified and the requirements are fulfilled in a safe way During the therapy the physiotherapist is the decision maker of the evolution of the session and of the movements that the patient has to accomplish The GUI has to provide him the available choices in a readable way to perform this operation well By making a well designed GUI the possibilities for erroneous user actions would be reduced to a minimum Also the feedback to the patient about their performance and motivating exercises are key aspects of the successful deployment of robotic systems within routine clinical use A GUI can provide this feedback as it facilitates interaction between the stroke patient and the robotic system during the treatment session 4 By training the motoric sensory and cognitive deficits in a playful manner the moti
16. actuators MACCEPA s Development of a graphical user interface for a rehabilitation exoskeleton 33 Practical implementation realization of a GUI for ALTACRO These three different CS s are often used when handling the robot So it is necessary to define these CS s and to define the relations between them 4 1 1 Robot DOF s The structure of ALTACRO is made of an exoskeleton and a support system which are both actuated The robot has 8 DOF s for the exoskeleton structure and 5 DOF s for the support system The DOF s of the exoskeleton are for both legs the hip knee and ankle flexion extension and the ab adduction of the hip The DOF s of the support system are the lateral sagittal and vertical translation of the pelvis the internal external rotation and the pelvic obliquity 30 4 1 2 Robot frame definitions The coordinate systems CS defined with the robot angles will be the most used coordinate systems It is also the most complete set of definitions The CS s are defined as follow Figure 14 and Figure 15 CS Base CS connected to the outside fixed world It is in the middle of the complete robot structure and connected to the ground e CS CS connected to the middle of the complete robot structure It is obtained by a vertical translation of the robot over A e CS CS connected to the middle of the pelvis It is obtained by a horizontal translation from CS over CS CS connected to the left end of the p
17. contact with the ground In phase 4 The heel is following a polynomial generated with respect to the desired step length and step height Development of a graphical user interface for a rehabilitation exoskeleton 26 Trajectory generation The toes go linearly in the vertical 2 direction from the ground to their highest point given as the place of the toes at the initial heel contact Their longitudinal x position is defined by the trajectory of the heel the length of the foot and the vertical position of the toes An example of a generated heel trajectory can be visualized in Figure 11 A toe trajectory is given in Figure 12 Both trajectories are generated for a step length of 50 cm and a step height of 10 cm The heel off height was 5 cm Note that the heel is taken as the rear end at the bottom of the foot The actual ankle joint is 8 to 10 cm above this point The longitudinal x position is changing when the foot is on the ground because the figure is taken as if it was observed from beside the treadmill for the heel Trajectory for the toes T T 9 T T T T T Trajectory 12 T T height cm height cm 0 5 10 15 20 25 30 35 40 45 50 20 30 40 50 60 70 80 x position x position cm Figure 11 Generated trajectory for the heel for a step length Figure 12 Generated trajectory for the toes for a step of 50 cm and a foot clearance of 10 cm length of 50 cm In the two dimensional trajectory gen
18. exoskeleton iii Abstract Abstract Thesis title Development of a Graphical User Interface for a Rehabilitation Exoskeleton Author Laura De Rijcke Master s degreein Science in Electro mechanical Engineering Academic year 2012 2013 Keywords Graphical User Interface Rehabilitation Robot Trajectory Generation Gait impairment has a highly negative influence in the health related quality of life In order to help patients with gait impairment a gait rehabilitation robot was developed at the Vrije Universiteit Brussel namely ALTACRO In this context there is investigated about the development of a graphical user interface for such device A graphical user interface is very important for a rehabilitation device The existence of such interface improves the user friendliness and the safety of the device It facilitates the control over the device and the communication between the user and the robot The first part of the thesis examines existing graphical user interfaces of rehabilitation devices These interfaces are compared with each other and the main options of each interface are listed up Thereafter is investigated on performance indicators These indicators are parameters which give an indication about the performance of the patient during a rehabilitation session Several performance indicators are compared to see which one gives the most accurate results used in combination with a gait rehabilitation device Finally different pro
19. for approximately five years The interview was made in the offices of the HHR on the 26 of November 2012 To have more objective results some other physiotherapists of the department were called during the interview to also have their point of view in the different subjects In what follows a summary of the main results will be made 2 1 1 Important factors for the rehabilitation process In the first part of the interview the most important factors in a rehabilitation session were examined These are the factors which counts the most for the physiotherapist to evaluate the quality of the exercise that the patient carried out The two main factors are e The body weight support BWS quality of the carried gait phases 2 1 1 1 The body weight support BWS The most important factor for a physiotherapist is the percentage of body weight support BWS that the patient is carrying himself In the beginning of the rehabilitation process the patient is not able to support his own weight This factor gives a good indication to the physiotherapist how far the patient is evolved in his rehabilitation process This BWS can be done by a harness or by the stiffness of an exoskeleton robot but it is mandatory for the rehabilitation that it is present The measurement of the BWS is easy if it is done by a harness but becomes much more complicated if it is done by the stiffness of the robot For this reason it was also examined if it should be poss
20. graphical user interface GUI will be detailed In a second section a startup sequence for the GUI will be given 4 3 1 Structure of the GUI The GUI is constituted of one main tab level containing three tabs the technical specifications tab the patient tab and the session tab The first tab the technical specifications tab is further divided in a second tab level This structure together with the main contents of each tab is illustrated in a flowchart in Figure 19 and will be detailed in the next sections Graphical display of Setting of TCPIIP Generation Patient dimensions Patient personal data Patient performance Session real time data control parameters configurations name age indication adjustments Visualization of trajectories En Figure 19 Overview of the structure of the GUI 4 3 1 1 Upper tab level The GUI is made out of 2 tab levels see Figure 20 one main tab level and a sub tab level which is only accessible from the technical specifications main tab The upper main tab level contains Connection open Technical specifications Patient Session STOP Save patient settings Open connection m Error Trajectories TCP IP communication Real time measurements Errors Figure 20 Illustration of the different tab levels in the GUI e 3main tabs Technical specifications This is the main tab for the engineers who are working with ALT
21. in table 7 The value of a large step length is a bit out of range Oberg 28 reported that the maximum hip angle on a normal walking speed is around 47 This value is also in the same range cfr Table 7 as the values found with the trajectory generation method The ankle angle flexion range is around 30 as reported by Astephen 29 The maximum ankle flexion is around 20 to 25 The maximum ankle angle is acceptable for a short step length but has to be improved for higher step lengths The reason that some values are out of range for larger step lengths is that the other parameters such as the amplitude of the sinus are kept constant It was already stated in section 3 2 that these parameters vary with the walking speed As the step length increases for an increasing walking speed the other inputs should also be changed The influence of the other input parameters on the generated joint angles is now explored The step length is always taken 60 cm For each table there is only one parameter changing The values of the other parameters are taken according to table 6 Table 8 Influence of the step height on the joint angles Step height cm Min hip joint angle Max hip joint angle Min knee joint angle knee extension 2 oom ba ie Max knee joint angle knee flexion 2 Min ankle joint angle Max ankle joint angle The first parameter that is examined is the desired step height As can be seen in Table 8 the
22. is a very expensive program to buy for the customers 1 2 3 4 4 LabView LabView is a visual programming environment It is also one of the most used languages for making industrial instrumentation automation programs LabView is especially recommended in real time applications as it provides as well data acquisition tools data analysis tools as data visualization tools LabView works with dataflow programming so each node is activated as soon as data is available at all of its inputs 21 LabView programs can also easily be connected with other programs such as with Microsoft Excel Matlab or Mathematica It is possible to create distributable EXE s and DLL s that runs with the freely available LabView run time engine LabView programs can be executed on multiple operation systems such as Windows Mac OS Linux and RTOS In addition of the previous advantages a LabView tool exists which permits the user to make lightweight web based applications These web based applications serve as GUI which allows the user to control systems from a web browser Another benefit of LabView is the extensive support for accessing instrumentation hardware The provided driver interface saves program development time 22 LabView is linked to National Instruments hardware which makes it easy to interact between the hardware and the software 1 2 3 5 Conclusion Although programs like Qt and VB provide a lot of advantages for the ease of programming a GUI
23. length ankle stance phase2 traj toes z r traj toes x r 0 x right z right coordinates gen inv L upper body L upper leg L lower leg L traj ankle x r x trans obliq right z obliq right L upper leg feet theta hip r theta knee r theta ankle r Step length z obliq right z mean hip x trans obliq right x mean hip x global zeros 11 _ en left x_global 1 x_left 6 x_global 2 x_left 5 x_global 3 x_left 4 x_global 4 x left 3 x_globa 5 x left 3 x_right 3 2 x globa 6 x_left 1 x_global 7 x_global 5 x_global 8 x right 3 x_global 9 x right 4 x_global 10 x right 5 x_global 11 x right 6 z global zeros 11 length z left z global 1 z left 6 z global 2 z left 5 z global 3 z left 4 z global 4 z left 3 z global 5 z left 3 z right 3 2 z_global 6 z global 5 L_upper z_global 7 z global 5 z global 8 z right 3 s z global 9 2z right 4 z global 10 2z right 5 z global 11 2z right 6 figure for i 1 length x left eDlot x lett r z left t zr be x ight a r o plot x global i z global i b axis 10 120 10 120 square pause 0 001 end as control for the trajectories figure plot
24. less feedback 2 1 4 Some possible extra settings As Dr Mortel has a lot of experience with the Lokomat device a lot of time the discussion was referred to the Lokomat Starting from the Lokomat device it was analyzed if some extra settings which are currently not available in the Lokomat would be useful to have in a gait rehabilitation device A first possibility is to exaggerate the movements of the gait during the rehabilitation session For Dr Mortel this would be a very bad thing to do With the gait rehabilitation it is the aim to relearn patients how to walk If an exaggerated movement is applied to the patient during the rehabilitation process the patient would learn to always walk with this exaggerated movement So applying an exaggerated movement would probably have a counter productive effect on the patient A second possibility was the implementation of three dimensional gait kinematics of the hip For Dr Mortel this will be a nice improvement for a gait rehabilitation device By implementing a three dimensional guidance of the hip the patient can relearn a much more natural gait pattern than if the guidance is only in two dimensions In this way it would be interesting to display the desired and the real trajectory imposed to the hip It would also be interesting to display the difference between the left and the right hip displacement especially for patients suffering from hemiplegia Care has to be taken that the imposed t
25. minimum joint angles are independent of this parameter The maximum hip and knee angles increase for an increasing step height The maximum joint angle decreases for an increasing step height Development of a graphical user interface for a rehabilitation exoskeleton 29 Trajectory generation Table 9 Influence of the heel off height on the joint angles Heel off height cm Min hip joint angle hip joint angle 2 Min knee joint angle knee extension Sa a ni Max knee joint angle knee flexion Min ankle joint angle Max ankle joint angle The second parameter that is examined is the heel off height An increase of the heel off angle will decrease the maximum hip joint angle and the maximum ankle joint angle By comparing table 8 and table 9 it is clear that the heel off height has much more importance for the maximum ankle joint angle than the step height On the contrary the step height has much more effect on the maximal hip and knee angles Notable Table 9 is that the minimal hip joint angle is decreasing much for very low heel off heights The maximal hip joint angle is decreasing very slowly for an increasing heel off height Table 10 Influence of the amplitude of the pelvic obliquity sinus on the joint angles Amplitude of the pelvic obliquity sinus Min hip joint angle Max hip joint angle Min knee joint angle knee extension Max knee joint angle knee flexion Min ank
26. of a graphical user interface for a rehabilitation exoskeleton V Abstract Samenvatting Thesis titel Ontwikkeling van een Grafische Gebruikersinterface voor een Rehabilitatie Exoskelet Auteur Laura De Rijcke Master graad in Science in de Ingenieurswetenschappen Werktuigkunde Electrotechniek Academisch jaar 2012 2013 Sleutelwoorden Grafische Gebruikersinterface Rehabilitatie Robot Traject Generatie Een probleem in het stappatroon kan een sterke negatieve invloed hebben op de levenskwaliteit van een persoon Om mensen met een dergelijk probleem te helpen werd aan de Vrije Universiteit Brussel een stap rehabilitatie robot ontwikkeld namelijk ALTACRO In het kader van dit onderzoek wordt in deze thesis de ontwikkeling van een grafische gebruikersinterface voor een rehabilitatie toestel onderzocht Een grafische gebruikersinterface is zeer belangrijk voor een rehabilitatie toestel De aanwezigheid van een dergelijke interface verbetert de gebruiksvriendelijkheid en de veiligheid van een robot Het vergemakkelijkt de controle over het toestel en de communicatie tussen de gebruiker en de robot Het eerste deel van deze master thesis onderzoekt grafische gebruikersinterfaces van bestaande rehabilitatie toestellen Deze interfaces worden vergeleken en de belangrijkste opties van iedere interface worden opgelijst Vervolgens worden er meerdere prestatie indicatoren vergeleken Deze prestatie indicatoren zijn variabelen die een indicatie geven o
27. of life The inability or reduced ability to walk affects a person s possibility to perform activities of daily living induces physical deconditioning and puts strain on his her psychological and psychosocial well being Rehabilitation of gait is essential for promoting recovery and improving the quality of life The physical therapy involved varies significantly depending on the cause and nature of the gait impairment Prevalent causes of gait impairment are neurological disorders and injuries such as stroke spinal cord injury multiple sclerosis and Parkinson s disease The burden of these disorders on the worldwide public health and the related demands for health resources are mostly underestimated and assumed to keep rising in the next years 1 Gait rehabilitation sessions are very exhausting for a physiotherapist They really have to do a lot of efforts to move the legs of the patient during the rehabilitation session In order to facilitate the work of the physiotherapists and to help patients suffering from gait impairment with their rehabilitation a five year concerted research action project ALTACRO Automated Locomotion Training using an Actuated Compliant Robotic Orthosis has been started in 2008 at the VUB Vrije Universiteit Brussel The research is mainly focused on four aspects of robot assisted gait training 2 e Active ankle assistance Balance and load distribution Functional three dimensional gait kinematics e
28. traj ankle x l traj ankle z l r x global 2 z global 2 b title trajectory of the ankle left legend desired trajectory real trajectory Development of a graphical user interface for a rehabilitation exoskeleton 67 Attachments dev sgrt traj ankle x 1 global 2 2 traj ankle 2 1 2 global 2 2 figure plot traj toes x l traj toes z l r x global 1 z global 1 5 title trajectory of the toes left legend desired trajectory real trajectory function traj x traj z x heel y heel traj gen Step length Step heigth Heel off V in contact V heel off numb er of points x 1 0 Step length 1 40 05 y heel Heel off 01 Generation of the ankle trajectory by making a cubic spline between the SHeel off and the initial contact xx 1 0 x heel 2 number of points 1 x heel 2 spl spline x heel V heel off y heel V in contact pp ppval spl xx 1 m i max sreshaping of the Generated trajectory in order to have the desired foot Slift xx 2 0 Step length number of points 1 Step length lin y heel 2 Heel off Step length 0 xx 2 0 Heel off pp pp lin pp pp m lin i Step heigth lin i pp pp lin traj x xx 1 traj z ppl end function theta hip theta knee theta ankle m invers traj x traj 2 x z L upper leg L lower leg L upper body L feet Step length ankle stance phase z toes x toes m p
29. 115 2003 Development of a graphical user interface for a rehabilitation exoskeleton 58 Sources 40 R Hug E Lu Wang et A Mihailidis Development of a portable robot and graphical user interface for haptic rehabilitation exercise chez Biomedical Robotics and Biomechatronics BioRob 2012 4th IEEE RAS amp EMBS International Conference on 2012 41 t GmbH nstruction ManualPablo System Hand arm rehabilitation 2011 42 GUIs and which one to use Online Available http www scriptol com programming gui php Accessed 12 2012 43 G Colombo M Joerg R Schreier et V Dietz Treadmill training of paraplegic patients using a rabotic orthosis Journal of rehabilitation research and development vol 37 n 16 pp 693 700 November December 2000 44 V T Inman H D Eberhart et others The major determinants in normal and pathological gait The Journal of Bone amp Joint Surgery vol 35 n 13 pp 543 558 1953 45 A Leung A Mak et J Evans Biomechanical gait evaluation of the immediate effect of orthotic treatment for flexible flat foot Prosthetics and orthotics international vol 22 n 11 pp 25 34 1998 Development of a graphical user interface for a rehabilitation exoskeleton 59 Attachments Attachments Attachment 1 Enlargements of Figure 2 and Figure 3 representing the settings of the GUI of Lokomat T Michelle David Hocoma 7 18 2008 10 13 31 Setti
30. 2 x end L feet length traj 2 x 1 traj 2 x end L feet line toes 2 0 line toes l end traj 2 x end L feet x toes l end x inter 2 x toes l end line toes 1 x toes 2 traj 2 1 feet 2 line toes 2 traj 2 z 2 traj ankle x l traj 2 x traj 3 x traj 4 x traj 1 x traj ankle z l traj 2 2 traj 3 2 traj 4 2 traj 1 z traj toes x l x toes 2 x toes 3 x toes 4 x toes 1 traj toes z line toes 2 line toes 3 line toes 4 line toes 1 traj ankl traj ankl traj toes x r traj toes z r _r traj 4 x traj 1 x traj 2 x traj 3 x traj 4 2 traj 1 72 traj 2 72 traj 3 zl x toes 4 x toes 1 x toes 2 x toes 3 line toes 4 line toes 1 line toes 2 line toes 3 Il theta hip 1 theta knee 1 theta ankle 1 invers traj ankle x 1 traj ankle 2 1 x trans obliq left z obliq left L upper leg lower leg L upper body L feet Step length ankle stance phase2 traj toes z l traj toes x 1 0 Development of a graphical user interface for a rehabilitation exoskeleton 66 x_left z left coordinates gen inv L upper body L upper leg L lower Attachments eg L feet theta hip 1 theta knee 1 theta ankle 1l Step length z obliq left z mean hip x trans obl iq left x mean hi p theta hip r theta knee r theta ankle r m traj ankle z r L lower leg L invers upper body L feet Step
31. 7 p 15 2010 7 Lokomat enhanced functional locomotion therapy Online Available http www hocoma com products lokomat Accessed 12 2012 8 Lokomat System User Manual 2009 9 Optogait Online Available http www optogait com Accessed 12 2012 10 Noraxon Clinical gait analysis and gait therapy solutions Online Available http www noraxon com products instruments complete gait analysis php Accessed 2012 11 F Collen D Wade G Robb et C Bradshaw The Rivermead mobility index a further development of the Rivermead motor assessment Disability amp Rehabilitation vol 13 n 12 pp 50 54 1991 12 L Johnson et J Selfe Measurement of mobility stroke a comparison of the Modified Rivermead Mobility Index and the Motor Assessment Scale Physiotherapy vol 90 n 13 pp 132 138 September 2004 13 V Hood M Granat D Maxwell et J Hasler A new method of using heart rate to represent energy expenditure the Total Heart Beat Index Archives of physical medicine and rehabilitation vol 83 n 19 pp 1266 1273 2002 Development of a graphical user interface for a rehabilitation exoskeleton 56 Sources 14 M ljzerman G Baardman M van t Hof Boom Hermens et P Veltink Validity and reproducibility of crutch force and heart rate measurements to assess energy expenditure of paraplegic gait Archives of physical medicine and rehabilitation vol 80 n
32. ACRO It gives access to all the inputs and outputs of the robot such as the settings for the trajectories the control parameters the real time sensor readings and the TCP IP communication settings Development of a graphical user interface for a rehabilitation exoskeleton 45 Practical implementation realization of a GUI for ALTACRO gt Patient This tab contains information related to the patient such as his personal data and his dimensions gt Session This tab is planned to contain information about the session itself With a graph that gives an indication about the performance of the patient e g graphical display of the real effectuated trajectory and the asked trajectory and some variables for the session adjustments such as the treadmill speed This session information will be stored in the same file as the patient data The STOP button Stops the robot and the program This button is placed in the main window so that it is accessible from anywhere in the program In this way it is easy to stop the robot in the case that something went wrong save patient settings button This button is accessible from each tab so that you can always easily save the latest changes in the settings for the patient session all the specific session options such as the used trajectories are saved in the patient file so that they can easily be re used for the next session e open connection button This button op
33. GTH OF 50 7 1 424 4 0 0 27 FIGURE 13 THE ALTACRO REHABILITATION DEVICE 30 een 33 FIGURE 14 ROBOT ANGLES FRONT VIEW ee eee eene eene seres sete sans eno 35 FIGURE 15 ROBOT ANGLES SIDE VIEW sien suis oon o ne co Fa a n rne Ve dater ai 36 FIGURE 16 WINTER ANGLE nennen nna sentes antes sisse ntes senses essen senes ens 37 FIGURE 17 MAIN STRUCTURE OF THE PROGRAMMING OF THE INTERFACE 40 FIGURE 18 PROGRAM STRUCTURE THE ORANGE ARROWS CORRESPOND TO ACTIONS WHICH SEND THE EXECUTION TO THE CASE STRUCTURE THE BLACK ARROWS CORRESPOND TO THE FLOW IN THE SECOND WHILE LOOP 44 FIGURE 20 ILLUSTRATION OF THE DIFFERENT TAB LEVELS IN THE nnn nnns 45 FIGURE 19 OVERVIEW OF THE STRUCTURE OF THE 8 45 FIGURE 21 TRAJECTORIES SUB TAB OF THE GUI inner 47 FIGURE 22 TIME MEASUREMENTS SUB TAB OF THE GUI 48 Development of a graphical user interface for a rehabilitation exoskeleton x List of figure
34. L upper leg 40 L upper body 45 L hips 25 Step length 50 Step heigth 10 Heel off 5 V in contact 10000 V heel off 5 percentage swing 40 phasel percentage ic 10 phase2 percentage mid stance 40 phase3 percentage terminal stance 100 percentage swing percentage ic percentage mid stance phase4 amp obl L hips 2 sin 8 180 pi this is the peak to peak amplitude of the Sine imposed on the pelvic obliquity in degree amp forw L hips 2 sin 10 180 pi peak to peak amplitude of the sine for the pelvic rotation amp trunk 3 5 hip neutral factor 2 1 1000 length knee number of points single support percentage swing 100 1 number of points double support 1 2 number of points single support 2 oe Names Traj 1 Trajectory of the foot in the swing phase completely removed from ground Traj 2 Heel is on the ground the toes are going to the ground 3 The foot is completely flat Heel amp toes on the ground 4 Heel is going from the ground to Heel off the toes are staying on the ground oe oe oe oe oe generation of the trajectory during the swing phase traj 1 x traj 1 z x heel z heel traj gen Step length Step heigth Heel off V in contact V heel off numb er of points single support Sgeneration of the trajectory of the heel during the stance phase Generation of the trajectory of the heel during the initial contact Traj 2 p2 percentage mid st
35. LabView will be used for making the GUI The main reason is that a GUI for a rehabilitation device has to provide real time results during the complete training session LabView has proven itself to be robust for programming real time applications LabView is also well designed for making easily connection with the instrumentation hardware which is important when making a real time GUI Matlab is very easy and robust for performing mathematical calculations As Matlab is especially designed for doing calculations it would be easier to use Matlab instead of LabView for eventually complicate calculations so Matlab would be used if some pre post session calculations have to be made It is also easy to move data from LabView to Matlab or from Matlab to LabView so these two programs are a good combination for making a GUI for a gait rehabilitation device Development of a graphical user interface for a rehabilitation exoskeleton 16 Specifications of a GUI 2 Specifications of a GUI 2 1 Contents of a GUI interview with a physiotherapist To understand clearly the needs and the wishes of the physiotherapists which will work a lot with ALTACRO an interview was made with Doctor Piet Mortel Doctor Mortel is the head of the department of physical medicine locomotory rehabilitation at the Heilig Hart Hospital of Roeselare He is specialized in active gait rehabilitation and in his service they have been working with the Lokomat device
36. Line graphics for hips amp knees with as Values of the joints during the displays for the reference the unloaded Lokomat for stance phase patient Smiley s focus on only one leg joint or phase The more the patient supports his own movement the broader the smile 5 Actual performance the performance during the current step for the selected gait characteristics Display for physiotherapist PL Swing phase Possibility to put on a The average of measured forces weighted with gait cycle phase in separate biofeedback units see also paragraph biofeedback system monitor Global Preprogrammed training programs Walking speed calculated as a function ofthe leg lengths Possibility to save additional notes Development of a graphical user interface for a rehabilitation exoskeleton 4 Introduction Michelle Davide Hocoma 7 18 2008 10 13 31 AM Prev Finish Figure 2 Screenshot of the patient settings tab of the Lokomat 8 Time Distance sS Hoc oma Michelle Davide a L t 7 17 2008 10 30 32 AM Training Monter Light Curtain Training t waux settmgs Programs uft Recorder Augmented reedback Patient Guidance Force Selection 9 45 Range of Motion 4 9 L B Right 100 a a a M Pt 100 TERRE 0 Offset 7 a ac El Knee Left H 5 EE H 60 Range
37. Practical implementation realization of a GUI for ALTACRO 4 3 2 How to use the GUI Although the interface is made as clear as possible for the ease of use a correct startup sequence will be given 4 3 2 1 Startup sequence The steps that are now explained for a correct startup of the GUI are summarized in the flowchart in Figure 23 Choose load trajectories for DOF s ofthe hip knee amp ankle for an existing patient the last used trajectories are automatically selected Enter patient personal data name Enter patient dimensions hips width length legs Send trajectories to TCP IP Check change control parameters Send control parameters to TCP IP Figure 23 Flowchart for a correct use of the GUI in yellow the tab names in blue the mandatory actions and in salmon the optional actions Development of a graphical user interface for a rehabilitation exoskeleton 49 Practical implementation realization of a GUI for ALTACRO 1 Turn on the robot 2 Start the program This is done by pushing the RUN button in LabView Figure 24 Connection open 3 Error Figure 24 Starting of the program the arrow is indicating the RUN button in LabView 3 Open the Patient tab Figure 25 amp 26 Choose an existing patient file or add a new patient file a Foran existing file Figure 25 e Push the Search existing patient button Choose the right patient in the
38. able 12 it can be seen that the influence of the amplitude of the sinus on the trunk movement is almost the same as the influence of the amplitude of the sinus on the pelvic obliquity The maximum knee and joint angles are decreasing for increasing amplitude of the sinus The maximum ankle joint angle is increasing for increasing amplitude of the trunk sinus Before sending the generated trajectories to the rehabilitation robot a correction should eventually be made on the ankle angles as they may not exceed 25 This can be improved by changing several parameters The heel off height the step height and or the amplitude of the sinus on the pelvic rotation can be increased or the amplitude of the pelvic obliquity and or the trunk sinus can be decreased This can also be improved by changing the trajectory imposed on the toes The trajectory imposed on the toes can especially be changed during the swing phase At the moment it is assumed that the height is varying linearly between ground and the maximum height during the initial heel contact phase and that the horizontal position is changing according to the height position of the toes and the heel position This linear assumption can be changed into other constraints 3 6 Conclusion The trajectory generation produces good trajectories for the hip and knee joints The generated joint angles are in the same range as the values measured during a human gait cycle The generated joint angles for the ankle
39. al joint is creating the ab adduction see Figure 15 Development of a graphical user interface for a rehabilitation exoskeleton 34 Practical implementation realization of a GUI for ALTACRO Note that the coordinate systems CS until CS are here defined as the coordinate systems of the left leg of ALTACRO The definitions of the CS s of the right leg are analog as indicated by an accent at the end of the name So they are defined CS until CSg 25 z4 z3 Figure 14 Robot angles front view Development of a graphical user interface for a rehabilitation exoskeleton 35 Practical implementation realization of a GUI for ALTACRO 24 24 23 25 Figure 15 Robot angles side view 4 1 3 Winter Conventions The data used to generate reference trajectories of the hip knee and ankle joint angles is based on the data from Winter 26 To be able to use this data it is necessary to firstly recall the used conventions Later on these coordinate systems will be transformed to the correspondent robot angles The Winter data defines only three angles Figure 16 the angle between the vertical the upper body and the upper leg measured between the negative z axis and the upper leg e the angle between the extension of the upper leg and the lower leg the angle between the perpendicular on the lower leg and the foot as illustrated in Figure 16 Development of a graphical user interface for a rehabili
40. always a treadmill so the global performance of a patient will nearly remain the same Therefore these indicators are not so useful in this case and another way to measure the performance has to be found From the previous paragraphs it is clear that a performance indicator for a rehabilitation device as ALTACRO has to be based on biofeedback parameters instead of on a questionnaire This performance indicator may be based on e g velocity measurements combined with HR measurements as with the PCI oxygen consumption rates HR measurements or based on the torque provided from the drive motors as for the L walk In table 5 a comparison is made between the different performance indicators Taking the fact into account that a questionnaire based performance indicator is not suitable to evaluate a performance with a gait rehabilitation device an L walk based method seems to be the most appropriated method in this case Table 5 Comparison of the performance indicators Performance Legend of table 5 indicators 2 5 Very good 22 5 Good 2 2 2 2 2 Mediocre o o S 5 8 E T Very bad 5 8 E S x Specification 8 8 8 is valid m Option not MRMI X 5 X Indirect calorimetry E THBI L walk torque measurement of actuators Development of a graphical user interface for a rehabili
41. ancetpercentage terminal stance percentage mid stance p ercentage terminal stance percentage ic Development of a graphical user interface for a rehabilitation exoskeleton 64 Attachments traj 2 x Step length p2 Step length Step length percentage ic 100 1 1 p2 Step length traj 2 z 0 traj 2 x Generation of the trajectory during the mid stance Traj 3 p3 percentage terminal stance percentage mid stancetpercentage terminal sta ic traj 3 x p2 Step length p3 p2 Step length percentage mid stance 100 1 1 p3 Step length traj 3 z 0 traj 3 x Generation of the trajectory during the terminal stance pre swing Traj 4 p4 percentage mid stancetpercentage terminal stance percentage mid stance p ercentage terminal stance percentage ic traj 4 x p3 Step length p3 Step length percentage terminal stance 100 1 1 0 traj 4 z traj 1 z 1 traj 3 z end traj 1 x 1 traj 3 x end traj 4 x traj 3 x end traj 3 z enda ankle stance ankle 1 1 2 take only the 40 of values of the single support phase out of the Winter data ankle swing ankle 1 2 Generation of a mean hip height a superposed double sine amp up work amp obl 2 amp forw work amp forw 2 x sine pi 4 pi 1 1 5 pi 2 sine sin x sine z neutral sqrt L upper leg L lower 1leg 2 Step length 2 2 hip neutral factor 2 gen left z neutral amp up work z sine z mea
42. as 3 4 The total heartbeats occurring during recovery 13 Heart rate beats min 4 Exercise Recovery Resting heart rate Time s Figure 7 Schematic heart rate plotted against time during exercise 13 An index based on the number of heartbeats provides a measure of a person s total energy consumption and this measure is independent of whether the activity is steady or non steady state By including the recovery period the repayment of the oxygen debt that occurs at the onset of exercise is included The THBI is easy to calculate and requires equipment that is readily available comfortable to wear and noninvasive Repeatability statistics found the to be comparable to oxygen cost indirect calorimetry and better than the PCI The is sensitive to change in workload with a profile similar to that of indirect calorimetry 13 1 2 2 6 L walk The L walk is the system used by the Lokomat device This system measures the torque of the Lokomat s drive motors If the patient actively moves with the programmed gait pattern the power required from the drives is small The less active the patient is or the more awkward the movements he makes e g due to spasticity the greater the drive force required to move his limbs according to the preprogrammed pattern The values displayed on the Lokomat are the average values for the forces measured within the Lokomat drives weighted according to the gait cycle ph
43. ase Therefore the values displayed are strongly correlated with the force or torque values However they are not displayed in Newtons or Newton meters but in biofeedback units Two values are calculated for each driven joint one for the swing phase the other for the stance phase The calculation method is designed in such a Biofeedback is a way for having information on physiological functions by measuring the activity of these functions This can be done with brain waves muscle tone or HR Development of a graphical user interface for a rehabilitation exoskeleton 13 Introduction that therapeutically desirable movements increase the values e g actively swinging the leg forward during the swing phase hip swing actively extending and flexing the knee during the swing phase and actively extending the knee during the stance phase 8 Although this is a relatively efficient way of feedback one has to note that this indicator works best for a guidance force of 100 of the device If the guidance force is less than 80 the changes of the biofeedback values would be much smaller Note that it is better for the rehabilitation if the guidance force is lower A lower guidance force means that the patient is more active 1 2 2 7 Conclusion A problem with the first two performance indicators MRMI and MAS is that they are calculated as a function of the tools that the patient needs to move The tool that the patient will use is
44. ated and time consuming programming It allows users to accurately solve problems produce graphics easily and callback is a pointer to a function which may be executed in the first function For example if you push the close button you want that the close function is executed wherever you are in the program These callbacks are not type safe It is never sure that the processing function calls the callback with the correct arguments 16 Development of a graphical user interface for a rehabilitation exoskeleton 15 Introduction produce code efficiently 19 This makes Matlab an efficient program to do all kind of mathematical operations and for displaying graphics in an easy way The main disadvantage of Matlab is that it uses a large amount of memory so it can become very slow It uses also the maximum CPU time that Windows allows it to have which makes real time applications very complicated 20 Another disadvantage is that Matlab is not so optimal for making GUI s The programming of it is rather complex Each time a control is placed the programmer has to manually compute his location For moving a control the programmer has to recompute all the object positions instead of only drag and drops it as is possible in the other programming environments which are presented here A last disadvantage of Matlab is that it is almost not possible to make a stand alone executable so Matlab always has to be installed Matlab
45. atics the required hip and knee joint angles are calculated Different problems arose during the implementation of this procedure The angles measured by Winter are recorded independently of the patient dimensions and the step length When the desired step length was larger than the one used for the Winter data it was kinematically impossible to follow the desired heel trajectory with the given leg length the heel could not attain the desired trajectory Development of a graphical user interface for a rehabilitation exoskeleton 25 Trajectory generation The Winter data is independent of the step length Discontinuities arose for going from the stance phase to the swing phase as the stance phase was based on the Winter data and the swing phase on the desired step length During the swing phase trajectory was only imposed on the hips on the heel The ankle angle was given by the Winter data Due to this combination the toes were sometimes asked to go through the ground level To solve these problems the procedure was extended as follows The calculation of the hip height is done in the same way as before the Winter data is used to calculate the height of the hip of the leg in the stance phase This hip height is used by calculating the inverse kinematics to calculate the joint angles of the leg in the swing phase trajectory that covers the four phases of the cycle is imposed on the heel This instead of only imposi
46. ck style The larger applications get the more important this advantage of QT becomes 17 It makes QT a very powerful component programming system 1 2 3 2 Visual Basics Visual Basics is an event driven programming language developed by Microsoft to make applications for Microsoft and to make GUI s The fact that it is a programming language made by Microsoft makes it very difficult to export programs made with Visual Basics to other operating systems The biggest advantage of VB is for sure its ease of use for most programmers VB is very straight forward for making GUI s as it has a GUI based development tool with graphical aspects incorporated It also offers an extremely rapid development tool for applications compared with other programs Another advantage of VB is that it is widely used so there are a lot of tutorials examples and online forums that can be consulted in case of problems A disadvantage of VB is that it uses a lot of memory as well disk space for the initial installation as in order to function efficiently after installation This makes that VB is not recommended for making programs that use a lot of processing time 18 1 2 8 5 Matlab Matlab is an object oriented fourth generation programming language developed by MathWorks This programming language is especially used for numerical computing It allows performing numerical calculations in an easy way and the visualization of results without the need for complic
47. cks at the same time which makes LabView capable of parallel coding By the use of multi processing and multi threading hardware of the built in scheduler the code is multiplexed Some special structures are foreseen in LabView in the case that a strict sequential structure is needed for a program 32 LabView subroutines are called virtual instruments VI s and can also be represented by graphical blocks A regular use of these VI blocks can make a large program much more maintainable The use of these VI blocks enables the programmer to reuse the same VI multiple times 4 2 1 The main structure The main structure of the program must meet the next requirements e program has to be initialized the graphs have to be cleaned the variables have to be initialized to their default values and reference control values and trajectories have to be loaded e The program has to keep running until the STOP button is pressed e fa special event occurs like the opening of the TCP IP connection the corresponding code has to be executed and the program has to be restart with the new parameters To satisfy these requirements the effective program is principally made out of a while loop preceded by a sequential structure see figure 17 The sequential structure ensures the initialization of the program The while loop contains a case structure and a second while loop and will run during the entire duration of the ses
48. constituted of a transmitting and a receiving bar Each one contains 96 LEDs communicating on an infrared visible frequency with the same number of LEDs on the opposite bar Once positioned on the floor or on the treadmill the system detects the interruptions of the communication between the bars caused by the patient s movement and calculates the duration between two interruptions and the position of the feet 9 The device is represented in Figure 4 Figure 4 Optogait 9 10 It can be strange for the lecturer that this device is also taken in this state of the art But this is also a device for gait training and it is interesting to take a closer look to the software It gives a good idea how to display the different gait data in a readable way It also gives some ideas of the way to give feedback to the user of the device The possibilities of the software are listed up in Table 2 In Figure 5 a screenshot of the user interface of the Optogait device is shown The different feedback options camera images graphical feedback and numerical values of the performance are represented Development of a graphical user interface for a rehabilitation exoskeleton 6 Introduction Table 2 GUI properties of OptoGait 8 Possibilities Patient settings Saving of personal data and notes Displays for the user Three kinds of feedback Numerical graphical with a video from a webcam see Figure 5 Possibilities of analysis of the
49. ctuator Design and implementation in a biped robot Robotics and Autonomous Systems vol 55 n 110 pp 761 768 October 2007 32 N Instruments NI LabView Improving the Productivity of engineers and scientists Online Available http www ni com labview Accessed 07 05 2013 33 Qt Online Available http qt digia com Product Accessed 11 2012 34 A Gauld GUI Programming with Tkinter Online Available http www freenetpages co uk hp alan gauld tutgui htm Accessed 2012 35 D Cuthbert GUI Programming in Python April 2012 Online Available http wiki python org moin GuiProgramming Accessed 12 2012 36 D A Vogel Software user interface Requirements for medical devices Intertech engineering associates August 2007 37 C Stanger C Anglin W Harwin et D Romilly Devices for assisting manipulation a summary of user task priorities Rehabilitation Engineering IEEE Transactions on vol 2 n 14 pp 256 265 1994 38 J Rose J Medeiros et R Parker ENERGY COST INDEX AS AN ESTIMATE OF ENERGY EXPENDITURE OF CEREBRAL PALSIED CHILDREN DURING ASSISTED Developmental Medicine amp Child Neurology vol 27 n 14 pp 485 490 1985 39 S Jezernik G Colombo T Keller H Frueh et M Morari Robotic orthosis Lokomat a rehabilitation and research tool Neuromodulation Technology at the Neural Interface vol 6 n 12 pp 108
50. data after the test Speed of video reproduction can be changed Global Pre defined tests or tests created by the user Generates automatically a report after saving of the session data Development of a graphical user interface for a rehabilitation exoskeleton 7 IntroductionintroductionAbstract Smith John RUN TREADMILL 41 KMH 04 08 2041 17 00 59 Parameter 241 0 147 4 M 10 MB 34 4 121 29 uy 92 9 1 7 ________ 7 Bar N 70 71 BarN Figure 5 Feedback system of Optogait with above the camera images the middle the graphical feedback and on the bottom the numerical values 9 Development of a graphical user interface for a rehabilitation exoskeleton 8 Introduction 1 2 1 3 Pablo 5 Pablo differs from the previous rehabilitation devices as it is a hand and arm rehabilitation device and not a gait rehabilitation device However this device is considered in this state of the art as it has a lot of similarities with a gait rehabilitation device so it could also be interesting to study the possibilities of the software of a device such as this one DN EE Figure 6 Pablo 5 The main advantage of the software of Pablo is that there are different therapy games with different difficulties levels included in the software T
51. e Lokomat was built by a Swiss company named Hocoma 7 It is at the moment one of the most used and most performing gait rehabilitation robots In Belgium there are only two hospitals that use them currently the Heilig Hart hospital at Roeselare and the Jessa hospital at Herk de stad Hasselt The Lokomat system is an electrically driven gait orthosis consisting of a hip support and two gait orthoses The gait orthoses are each equipped with a hip and a knee joint drive The Lokomat system is Development of a graphical user interface for a rehabilitation exoskeleton 2 Introduction mounted via parallelogram on a swivel door It is operated in combination with a treadmill and body weight support system The Lokomat system is controlled via a PC 8 See Figure 1 for an image of the Lokomat gait rehabilitation device The different settings and possibilities of the Lokomat software are listed in Table 1 They are extracted from the user manual of the Lokomat robot 8 Figure 1 Photo of the Lokomat device 8 Table 1 GUI settings and possibilities of the Lokomat 7 Possibilities Patient settings i Parameters 19 Sliders for be set the physiotherapist 7 Hipangles Saving of patient data Guidance force of the robot on the patient by reducing this parameter the patient can walk more freely Development of a graphical user interface for a rehabilitation exoskeleton 3 Introduction Possible
52. e enne nnn nnne sse annees sena serene 28 TABLE 7 EXTREMES OF THE GENERATED JOINT ANGLES enne nennen nnne nnne 29 TABLE 8 INFLUENCE OF THE STEP HEIGHT ON THE JOINT ANGLES eeeeeeeeeeen eene enne rini 29 TABLE 9 INFLUENCE OF THE HEEL OFF HEIGHT ON THE JOINT 5 102 0 0020 00 020 000000 600 000000 30 TABLE 10 INFLUENCE OF THE AMPLITUDE OF THE PELVIC OBLIQUITY SINUS ON THE JOINT ANGLES 30 TABLE 11 INFLUENCE OF THE AMPLITUDE OF THE PELVIC ROTATION SINUS ON THE JOINT 30 TABLE 12 INFLUENCE OF THE AMPLITUDE OF THE TRUNK SINUS ON THE JOINT 5 31 TABLE 13 MACCEPA ANGLE DEFINITIONS 4 2 1 1 annees arent esa senses senes nan 38 TABLE 14 LINK BETWEEN THE ANGLE DEFINITIONS 1 1 1 nennen enne nennen 39 TABLE 15 COMPARISON BETWEEN DIFFERENT PROGRAMMING ENVIRONMENTS eene enne 62 Development of a graphical user interface for a rehabilitation exoskeleton xii Introduction 1 Introduction 1 1 Motivation impairment has a major impact health related quality
53. ebruikersinterface voor ALTACRO te maken Deze interface focust zich voornamelijk de ingenieursnoden Er wordt eveneens een traject generatie gedeelte in deze interface geimplementeerd Development of a graphical user interface for a rehabilitation exoskeleton vi Contents Contents AcknowledgmeTit emi iacu a te eda tira iii Abstract cete sie A einst uiui MEM NIMM HIM EMI i LOIRE tuer fce iv aum Samenvatting PE vii List Of abbreviations te e List of fIEUFES oco eoram tan t taa chit e e ann ereen RR lu dee eter eh x List au a Mrd xii T a dele UCT ON eos Een ig hte Re fec eua eoo eed at ee ote rue eal age ood 1 1 14 MOtIVatiOn ee 1 1 2 State Of TEA IC RED UE EO 2 1 2 1 Existing GUI s of rehabilitation devices 2 1 2 2 Performance i ss 11 1 2 3 Programming environments 1 2 4 1 0000 15 2 Specificationsof GUI eee eo OO ec EPOR D IMS 17 2 1 Contents of a GUI interview with a physiotherapist 4 17 2 1 1 Important factors for the rehabilitation process 17 2 1 2 Data to display and or save 18 2 1 3 Patient feedback methods issues 18 2 1 4
54. egree for cycle 25 3 3 Trajectory generation based on previously measured joint angles David Winter 26 investigated two dimensional human gait trajectories For this he collected data about the two dimensional gait cycle He measured the hip angle in the sagittal plane the knee and the ankle angles in a normal human gait cycle In a first approach to generate trajectories for a gait rehabilitation robot it seemed straightforward to start from this data The angles measured by Winter are recorded independently of the step height and the step length The idea was to start from this Winter data to make trajectories in two dimensions and expand further to three dimensional trajectories by implementing also the pelvic obliquity the pelvic rotation and the movement of the trunk The method in two dimensions is as follows e dimensions the pelvis is reduced to a single point e The Winter data is used to calculate the hip height during the stance phase e As the pelvis is reduced to a single point and one leg is in the stance phase when the other leg is in the swing phase the calculated hip trajectory is also used for the leg in the swing phase e trajectory is designed for the heel of the foot in the swing phase For this a polynomial is generated with as constraints the required step height and the step length generated heel trajectory and the calculated hip trajectory are used as inputs By using inverse kinem
55. elvis This CS is obtained by a vertical translation over A 3 By setting for example a positive translation to the left side and a negative translation to the right side of the pelvic the pelvic obliquity is obtained CS CS connected to the cylindrical joint at the left end of the pelvis This CS is obtained by a horizontal translation over from CS3 By combination of a horizontal translation on the left and on the right side of the pelvis the obliquity in the transverse plane is created CS CS connected to the left cylindrical hip joint The CS is obtained by a rotation of around It represents the ab adduction of the left leg CSg CS connected to the left upper leg link It is obtained by a rotation of 66 around ys It represents the rotation between the upper leg and the upper body CS CS connected to the left lower leg link It is obtained by a rotation of around yg and represents the rotation between the upper leg and the lower leg CSg CS connected to the left foot link It is obtained by a rotation of around ya It represents the rotation between the foot and the lower leg s Keep attention the robot DOF are not exactly the same as the physiological DOF so the robot left right end of the pelvis is not the same as the physiological In ALTACRO there s an extra bar between the end of the pelvis and the hip joint with a cylindrical joint connected to the bar This extra cylindric
56. ens the TCP IP connection If the connection is opened the led connection open lights up e error line displays the errors 4 3 1 2 Technical tab The Technical specifications main tab contains a second tab line with 4 sub tabs e Trajectories Figure 21 This tab is for the joint trajectories The user physiotherapist or engineer can generate or load the reference trajectories for the hip knee and ankle joints To generate trajectories different gait parameters such as the step height and length and the amplitudes of the hips can be chosen If the save patient settings button is pushed after the selection of the trajectories the program associates these trajectories with the previously opened patient file see the section startup sequence In this way there is no need to redo this step each time a new session is started with the same patient It is also possible to simulate and visualize the imposed trajectories in this tab before sending them to the real robot For this the user first has to choose a simulation speed If the user does not select a simulation speed a default medium speed is selected The simulation speed is defined as the time delay in ms the program waits before plotting a new step of the simulation so the higher the chosen value the slower the simulation The simulation speed is completely independent of the treadmill speed The simulation is started by pushing the Simulate button If
57. ent seine GNU Figure 27 Header of the GUI with the TCP IP connection buttons highlighted 5 Open the Technical specifications tab a Open the Real time measurements sub tab Check and or change the control parameters for the different nodes Push the send control parameters b Open the Trajectories sub tab Development of a graphical user interface for a rehabilitation exoskeleton 52 Practical implementation realization of a for ALTACRO Note the next 5 steps should not be executed if an existing patient file was chosen The trajectories associated to the last run of this patient file are automatically loaded in the program The trajectories are not sent automatically to the robot as the patient file is chosen before the TCP IP connection is opened Choose the desired trajectories for the hip knee and ankle joints Push the Load button to load the chosen trajectories in the program Choose the simulation speed Push the Simulation button to verify the behavior of the chosen trajectories Push the Save patient settings button to associate this trajectories to the patient file e Push the Send trajectories to TCP button to send the trajectories to the robot c Openthe Real time measurements sub tab Choose the nodes to inspect Choose the real time data to display on the graphs Push the Save save to default or save under a new name button to save the
58. eration the hip is represented by a fixed single point The hip position is horizontally taken in the middle of the step length The hip height is calculated as a function of the step length and the length of the leg When the hip is taken as a fixed point the heel is the furthest from the hip during the initial contact The hip height at the initial contact can be calculated as 2 Step length Lipper leg Liower leg With the hip height and the trajectories for the heel and the ankle as inputs the hip ankle angles are calculated by using inverse kinematics The exact kinematic formulations can be found in Attachment 3 in the Matlab function invers In this two dimensional trajectory generation the desired trajectories can perfectly be followed There are no more problems of discontinuities or toes which pass through the ground This trajectory generation tool now has to be extended to a three dimensional generation tool in order to have physiologically more compatible trajectories Development of a graphical user interface for a rehabilitation exoskeleton 27 Trajectory generation In the three dimensional trajectory generation the pelvis is represented by three points one for the middle of the pelvis one for the left extremity and one for the right extremity of the pelvis On both pelvis extremities a trajectory is imposed in accordance with the physiological constraints A double frequent s
59. ference in some consecutive sessions as their walking speed would remain nearly the same The heart rate must also achieve a steady state In unimpaired subjects this state occurs when the cardiovascular system has adapted to the new physiologic demands which occurs about the third minute of exercise In a population with gait impairments the effort of walking may be significantly higher and not be considered submaximal Analysis was made with children with cerebral palsy and it was discovered that in 9 of their subjects the heart rate continued to rise during the walking trials If a nonsteady state heart rate is used as an alternative the repeatability may be compromised 13 1 2 2 5 Total Heart Beat Index THBI To represent more accurately the total energy consumed during an activity Total Heart Beat Index THBI investigates heart rate behavior throughout a period of exercise by finding the total number of heartbeats that occurred during the period This is done with continuous heart rate monitoring which is Development of a graphical user interface for a rehabilitation exoskeleton 12 Introduction now readily available with the development of portable heart rate monitors This is illustrated in Figure 7 where e Area 1 The extra heartbeats required during the exercise e Areas 1 2 The total number of heartbeats during the exercise including the basal level e Area 3 The extra heartbeats that occur during the recovery phase e Are
60. g to the chosen node so that the interface is not overloaded with data It is possible to display the data of maximum 2 different nodes at the same time There are 3 graphs in this tab so that the user can display three different RTS of the same node at the same time If the user chooses to display the data of two different nodes the RTS of both Development of a graphical user interface for a rehabilitation exoskeleton 47 Practical implementation realization of a GUI for ALTACRO nodes are displayed the same graph so that it is easy for the therapist to compare the values of both nodes The program automatically sets ranges for each graph This range is defined by the type of the node that is selected a rotational or a translational and by the RTS that is selected This range can also be changed manually It is possible to freeze the graph on every moment Once the graph is frozen it is also possible to zoom on the graph Errors In this tab error codes are logged and explained more in detail I iiec Technical specifications Patient Session STOP Save patient settings Open connection Error Trajectories TCP IP communication Real time measurements Errors Desired Node left 5 Real tim arameters Desired Node right n me Desired nodes to display Nodes Angle rad m Control parameters position m 1 Parameter Left range number Value Control parameters 4 Value 180 180
61. generation 3 Trajectory generation A robot has many different joints trajectory has to be imposed on these joints to let the robot move For a rotational joint for example the trajectory contains the successive angles that the joint must occupy The generation of the trajectories is very important for rehabilitation robot The desired robot trajectories must be synchronized with the treadmill speed be kinematically compatible with the patients dimensions have kinematically correct gait pattern be similar to a human gait cycle be safe for the patient This patient safety is mandatory especially with exoskeleton robots as the patient is in a certain way imprisoned in the robot and he cannot escape if the robot is executing wrong movements So the gait rehabilitation device must follow trajectories that are physiologically compatible with the human gait cycle If the trajectories are not physiologically compatible the robot may hurt the patient This should absolutely be avoided In this chapter a method will be developed for generating kinematically correct and physiologically compatible trajectories for a gait rehabilitation robot 3 1 Gait phases Before starting the actual trajectory generation the different phases in a gait cycle must be examined For the purpose of trajectory generation the gait cycle is divided in four phases according to Perry 23 These different phases are visualized in Figure 8
62. gramming environments are explored to see which is the most suitable for programming a graphical user interface for a rehabilitation robot In the second part of the thesis the requirements for a graphical user interface are explored The interface is needed for three different types of users engineers physiotherapists and patients Each of these user groups has specific requirements for the contents of the interface The requirements of the different user groups are investigated and listed up For the groups of the physiotherapists and the patients this is done with the cooperation of a physiotherapist In the third part of this master thesis a trajectory generation tool is developed The trajectories are generated based on several user inputs specific for a gait rehabilitation therapy such as the step length the step height and the amplitude of the hip movement These generated trajectories should also be kinematically and physiologically correct In a next step of the thesis the influence of the different gait parameters on the generated joint angles is explored In the final part of this master thesis the gathered knowledge is used to make a graphical user interface for ALTACRO The interface focuses mainly on the needs of the engineers A trajectory generation tool is developed for this interface Development of a graphical user interface for a rehabilitation exoskeleton iv Abstract R sum Titre du m moire D veloppement d une Inte
63. hese games ensure a maximum attention and motivation of the patient The different levels of each game give the possibility to adjust the game to the patient s ability As there are different games available the patient will not be so fast tired from his therapy as if there was only one single game available The different GUI possibilities of Pablo are listed in table 3 Table 3 GUI properties of Pablo 10 Possibilities Patient settings Patient identification Saving of patient data Possibility to export the report in pdf or excel Displays for the user Eight different interactive therapy games results of the game are based on the chosen control settings force or movement Different difficulties in the game Analysis possibilities Report is made with graphics of the elbow wrist and shoulder movements and with the different grip forces Development of a graphical user interface for a rehabilitation exoskeleton 9 Global 1 2 1 4 Introduction Possibility of calculating a force control index compare the patient force with the force of a healthy person Comparison between the right and the left extremity is possible Measure and evaluate the cylindrical grip force and the pincers grip Calculate the movement of the shoulder elbow and wrist measurements are saved with date time and length of the therapy Preprogrammed movements which has to be performed Summary of the existing devices The possibi
64. hile loop The orange arrows correspond to actions which will sent the execution flow of the program to the case structure in the first while loop To speed up the execution of the code event structures are used An event structure is a block of code that is preceded by a boolean The appended code block is only executed if this boolean has the desired value true or false In this way only the part of the code that is needed for the action that the user is asking for is executed Only the code which is susceptible to be used several times in the program such as the code which manage the variables for the desired nodes is read in every repetition of the second while loop The most important code blocks programmed in the second while loop are e A block that sends data control parameters or trajectories over the TCP IP connection It first checks if the connection is open and it gives an error if this is not the case If the connection is open it sends the data over the TCP IP and it checks if an acknowledgment of the arrival of the data is received It checks also whether the same quantity of trajectories has arrived as were sent e block that reads all the data coming over the TCP IP connection This block is active as long as the TCP IP connection is open Development of a graphical user interface for a rehabilitation exoskeleton 41 Practical implementation realization of a GUI for ALTACRO block that handles the representa
65. i theta knee i end 1 i 1 L upper leg 24 L lower leg 2 2 L upper leg L lower leg c2 i L up per leg L lower leg c2 i z i traj z i L lower leg s2 i traj x i x i 51 1 1 upper leg 24L lower leg 2 2 L upper leg L lower leg c2 i L low er leg s2 i z i traj z i L upper leg L lower leg c2 i traj x i x i theta hip i atan2 s1 i cl i c kah i x toes i x i L upper leg sin theta hip i lower leg sin theta hip i theta knee i L feet S kah i z i z toes i L upper leg cos theta hip i lower leg cos theta hip i theta knee i L feet theta kah i atan2 s 1 kah i theta ankle i theta knee i theta hip i theta kah i end end end function x z coordinates gen inv L upper body L upper leg L lower leg L feet hip knee ankle Step length z extremity hip z mean hip x extremity hip x mean hip x zeros 5 length hip z zeros 5 length hip for i l length hip x l i x mean hip i 2 1 1 1 Development of graphical user interface for rehabilitation exoskeleton 69 xtremity hip i 1 1 upper leg sin hip i i L lower leg sin hip i knee i i L feet cos knee i ankle i hip i mean hip i L_upper_body mean hip i extremity hip i 3 i L upper leg cos hip i 4 i L lower leg cos hip i knee i 5 i
66. ible to replace the BWS by another parameter with a kind of a performance indicator for example A lot of performance indicators are developed until now such as the modified Rivermead mobility index MRMI 11 or the Motor Assessment Scale MAS 12 but none of them are really indicated to be used with a device as ALTACRO or Lokomat The reasons that they are not indicated is that their calculations are related to the tools that the patient needs to walk such as crutches for example so you will always get similar results for the indicator as the device remains the same Another possibility could be to use a kind of bio indicator A bio indicator is an indicator that takes for example the effort the force or the energy consumed by the patient into account to evaluate the performance More information about bio indicators is given in paragraph 1 2 2 performance indicators These bio indicators give a good view on the performance of the patient and some of these bio indicators are easier to measure than the BWS Development of a graphical user interface for a rehabilitation exoskeleton 17 Specifications of a GUI 2 1 1 2 The quality of the carried phases A second element that is necessary to see during the session is the quality of the performed gait phases With the currently available devices such as the Lokomat this is mainly evaluated by the therapist who looks at the movements of the patient Unfortunately this is not a very ob
67. inus with as standard amplitude 4 5 cm is imposed on the middle pelvis point for the vertical movement of the trunk The amplitude is chosen as the mean displacement amplitude for mean walking speed see section 3 2 This amplitude can be changed by the physiotherapist The offset from the ground zero position is the same height as the hip height calculated for the two dimensional trajectory generation For both extremity points this double frequent sinus is combined with a superposed sinus This superposed sinus has a standard peak to peak amplitude of 8 in the z direction for the generation of the pelvic obliquity in the frontal plane A simple sinus with standard peak to peak amplitude of 10 is imposed in the x direction for the transverse obliquity The amplitudes of the different sinuses are chosen as the mean values during normal walking as given in the physiological requirements in section 3 2 They are taken as reference values but can be changed manually by the physiotherapist in order to adapt the gait pattern for different therapeutic sessions and for different walking speeds The angles of the two dimensional hip joint the knee joint and the ankle joint are calculated by using inverse kinematics The inputs of this inverse kinematics are the pelvis extremity trajectories and the heel and ankle trajectories The exact kinematic formulas can be found in Attachment 3 3 5 Influence of the input parameters on the generated joint ang
68. jective indicator The quality can also be evaluated by the software by giving the difference between the imposed trajectory and the really performed trajectory or with the L Walk function of the Lokomat see the paragraph 1 2 2 6 performance indicators This quality measurement is important because during the therapy the aim is to reduce the BWS but this may not be done at the expense of the quality of the gait trajectory All the other parameters that can eventually be displayed during a session such as the hip ankle and knee angles and torques may be interesting as they give the physiotherapist the ability to see if the patient is doing his exercises well but they are not mandatory for Doctor Mortel 2 1 2 Data to display and or save To be able to evaluate the progress of a patient some data of the session have to be saved These data are personal data of the patient his name weight length leg dimensions eventually an asymmetry of the legs The percentage of BWS that the patient carried The velocity at which the patient walked The mean velocity of treadmill walking during a session is 2km hour The time of walking may eventually be displayed Normally this is a normalized time of 45 minutes for each session In one week a patient may have two or three sessions For the patient s first sessions the walking time may be less than 45 minutes if the patient has not yet the endurance to complete an entire sessi
69. joint are often too high They can be improved by changing the values for the heel off height the step length and or height and or the amplitude of the sinus The step length and the step height are the parameters with the most influence on the generated hip and knee joint angles Note that the hip and the knee joint angles are both increasing for an increasing step length and step height On the contrary the hip joint angles are decreasing and the knee joint angles are increasing for an increasing heel off height So the hip and the knee joint angles are not always correlated in the same way for the different gait parameters The step length the heel off height and the amplitude of the pelvic obliquity and the trunk sinus are the parameters with the most influence on the ankle joint angle Especially the step length has a major role for the calculation of the ankle joint angle Development of a graphical user interface for a rehabilitation exoskeleton 31 Trajectory generation Before sending the data to a rehabilitation robot the generated trajectories have to be visualized in order to check the correct behavior of the complete trajectory If the angles are not in the physiologically compatible ranges the physiotherapist has to change the input parameters Development of a graphical user interface for a rehabilitation exoskeleton 32 Practical implementation realization of a GUI for ALTACRO 4 Practical implementation realization of a GUI f
70. le joint angle Max ankle joint angle The influence of the amplitude of the sinus of the pelvic obliquity is opposite to the influence of the step height As can be seen in Table 10 the maximal hip and knee angles are decreasing for increasing obliquity amplitude The maximal ankle joint angle is increasing for increasing obliquity amplitude Table 11 Influence of the amplitude of the pelvic rotation sinus on the joint angles Amplitude of the pelvic rotation sinus Min hip joint angle Max hip joint angle Min knee joint angle knee extension 2 remit Max knee joint angle knee flexion Min ankle joint angle TO EHE KA Max ankle joint angle The influence of the amplitude of the sinus of the pelvic rotation on the joint angles is very small For increasing amplitude of the pelvic rotation sinus the maximum hip and ankle joint angles are slightly Development of a graphical user interface for a rehabilitation exoskeleton 30 Trajectory generation decreasing The maximum knee joint angle is approximately remaining the same with increasing rotation amplitude Table 12 Influence of the amplitude of the trunk sinus on the joint angles Amplitude ofthe trunk sinus cm Min hip joint angle Max hip joint angle 2 Min knee joint angle knee extension 2 Sn Max knee joint angle knee flexion Min ankle joint angle Max ankle joint angle Lastly from t
71. les To verify the behavior of the previously developed trajectory generation tool a stick model is generated in Matlab Firstly the reference values for the parameters as found in section 3 2 are put in the program This is done for three different step lengths 47 cm 60 cm and 78 cm These step lengths are chosen as the smallest mean and largest step length in a normal human gait cycle of people between 10 and 79 years based on the results of berg 27 The remaining input parameters of the simulation are given in Table 6 Table 6 Inputs of the generated trajectories Amplitude of the pelvic obliquity sinus Amplitude of the pelvic rotation sinus Amplitude of the trunk sinus 4 5 cm Length upper leg Length lower leg Length feet Width hips Step height Heel off height B 10 5 ____________ Development graphical user interface for rehabilitation exoskeleton 28 Trajectory generation The extremes of the generated joint angles are given in Table 7 Table 7 Extremes of the generated joint angles Step length cm EE y Min hip joint angle Max hip joint angle 2 Min knee joint angle knee extension 2 Loc mro Max knee joint angle knee flexion 2 Min ankle joint angle 2 Max ankle joint angle Oberg 28 reported that the knee angles are maximum 68 during a normal gait cycle This value is relatively close to the ones found for a short and a medium step length with the trajectory generation
72. lities of the previous explained devices and the differences between the different devices are summarized in Table 4 An X means that this option is present a means that this option is not available Only the most important parameters and differences are listed in order to have a clear overview Table 4 Comparison between the options of the different GUI s Rehabilitation device Patient tab Patient identification Automatic saving of patient data Possibility of data transfer to a pdf file a text file or an excel file Way to adapt parameters Feedback methods List of numerical values of parameters Graphical feedback graphic Other Training method Preprogrammed programs Manual set program Development of a graphical user interface for a rehabilitation exoskeleton Lokomat Pablo OptoGait Tyromotion systems x x sliders sliders x ETE smiley Or x x thermometer X X video taken by a webcam 10 Introduction By examining Table 4 one can see that these GUI s have a lot of common properties From this comparison a list of essential options for a useful GUI can be extracted e Patient identification name date of birth leg parameters height weight results of previous sessions The possibility to export the results to a printable file pdf excel e Not only a graphical feedback but also a more funny readable kind of feedback a game or icon or vide
73. lts would be independent of the present therapist Asa third point for the user friendliness it should be possible to save the patient data together with the session results In this way it is easy for the physiotherapist to see the evolution of the patient s rehabilitation process session after session Also the used session parameters such as the step length the step height and the amplitude of the hip movement for a gait rehabilitation robot may be saved in the same file Development of a graphical user interface for a rehabilitation exoskeleton 20 Specifications of a GUI 2 2 4 The patient For an optimal rehabilitation of the patient he must execute his exercises as actively as possible It has been reported by Br tsch 6 that the feedback of a GUI can be as positive for the evolution of rehabilitation therapy as the feedback of a physiotherapist By combining the feedback of a physiotherapist with that of a GUI the evolution of the rehabilitation therapy is even better than by using only one of both 6 Another reason for the importance of the user interface for the patient is that if the exercise is done as a game the attentiveness and the motivation of the patient can be improved If the patient is paying more attention to and motivation for his exercises he will put more energy in the execution of them and have a better result Development of a graphical user interface for a rehabilitation exoskeleton 21 Trajectory
74. n hip z gen left z obliq left generator sine 1 2 obliq left z obliq left z mean hip the left foot starts with the IC phase 2 and the hip at his furthest point in front the right foot starts with phase4 z forw pi 2 2 pi 1 1 5 2 pi x forw amp forw work sin z forw x trans obliq left Step length 2 x forw percent percentage ictpercentage mid stance 100 move factor percent l x trans obliq right x trans obliq left move 1 x trans obliq left 1 move factor z obliq right z obliq left move factor 1 end z obliq left 1 move factor x mean hip x sine x sine Step length 2 calculations of phase 1 theta ankle ankle stance ankle stance percent 100 percentage swing 100 move factor percent l 2 gen phasel z oblig left gen phasel x trans obliq left 2 gen phasel z gen phasel move factortl end 2 gen phasel 1 move factor Development of a graphical user interface for a rehabilitation exoskeleton 65 Attachments z mean phasel z mean hip move factortl end 2 mean hip l move factor x gen phasel x gen phasel move factor l end x gen phasel l move factor theta hip 1 theta knee 1 theta ankle 1 ml invers traj 1 x traj 1 z x gen phasel z gen phasel L upper leg L lower leg L upper body L feet Step length ankle swing phasel traj 4 x traj 4 0 x phase 1 z phase 1
75. new control parameters Development of a graphical user interface for a rehabilitation exoskeleton 53 Conclusions 5 Conclusions In this thesis the development of graphical user interface for rehabilitation exoskeleton was investigated For this in the first parts of this thesis the requirements for a graphical user interface were investigated It was found that the GUI must fulfill the needs of the engineers the physiotherapists and the patients For the engineers the GUI must contain options to check and change the control parameters of the robot check the data coming from the robot generate trajectories for the robot and manage the communication These generated trajectories represent the sequence of desired positions of the joints of the robot The requirements of the physiotherapists and the patient have been developed in cooperation with physiotherapists The must enable the physiotherapist to easily and safely control the robot For this the physiotherapist must be able to generate physiologically and kinematically correct trajectories based on several therapy specific parameters such as the step height and length The GUI must also contain options to save patient data and session data Lastly it is advisable that the GUI contains a bio indicator which gives an indication of the patient s performance For the patient the GUI must give feedback about his session results In the second part of the thesis a trajectory generation
76. ng a trajectory on the heel during the swing phase as before Another trajectory is imposed on the toes This trajectory covers also the complete gait cycle With these improvements the discontinuities were eliminated Also the problem of the toes which passes through the ground level was solved The problem of following the desired heel trajectories for larger steps remained Due to this problem it was decided to discard the Winter data and to generate all the trajectories completely based on the physiological and kinematical constraints 3 4 Trajectory generation independent of previously measured joint angles First a two dimensional gait trajectory is generated As before a trajectory is generated for the heel and for the toes These trajectories are generated according to the four different phases in a gait cycle explained in section 3 1 The constraints for the heel and the toes in each of these phases are In phase 1 The heel is in contact with the ground toes go from their highest point given as the place of the toes at the initial heel contact to zero ground level This by taking the position of the heel and the length of the foot into account In phase 2 Both the heel and the toes are in contact with the ground In phase 3 heel is going from the ground level to the heel off position The heel off height is defined as a variable and can be changed by the physiotherapist The toes remain in
77. ngs Lastname First Patient Lokomat Notes Last name First Miche Weight Height Note uUi eri Figure 28 Screenshot of the patient settings tab of the Lokomat 8 Development of a graphical user interface for a rehabilitation exoskeleton 60 Attachments Michelle Davide Hocoma 7 17 2008 10 30 32 AM Hip Right 45 Range 45 O0 0 Knee Left 60 Range of Motion 60 Fl B Offset 0 Figure 29 Screenshot of the training settings of the Lokomat 8 Development of a graphical user interface for a rehabilitation exoskeleton 61 Attachments Attachment 2 Comparison between programming environments for making GUI s Table 15 Comparison between different programming environments at 33 17 16 Advantages Disadvantages GUI programming toolkit High portability High size of the program QtSDK High flexibility Requires vast amounts of disk space for compilation Openness Mechanism called signals slots much easier to control communications between objects in a flexible and maintainable manner than it is with a fragile callback style 17 Full API Fast language of interfaces Availableforfree i Several programming languages possible Allows 3D animations OtCreator visual development interface Used on some mobile Good documentation Java C
78. nt can be attached in the exoskeleton when he is sitting in the wheelchair and the rehabilitation device will help him to stand up This can be difficult to implement for ALTACRO but is useful to take into consideration when developing a new gait rehabilitation device Development of a graphical user interface for a rehabilitation exoskeleton 55 Sources 7 Sources 1 W Organization Neurological Disorders public health challenges 2006 2 P Beyl Design and control of a knee exoskeleton powered by pleated pneumatic artificial muscles for robot assisted gait rehabilitation 2012 3 B Fei W S Ng S Chauhan et C K Kwoh The safety issues of medical robotics Reliability engineering amp system safety vol 73 pp 183 192 2001 4 S Kemna P Culmer A Jackson S Makower J Gallagher Holt F Cnossen J Cozens M Levesley et B Bhakta Developing a user interface for the iPAM stroke rehabilitation system chez Rehabilitation Robotics 2009 ICORR 2009 IEEE International Conference on 2009 5 Pablo system Online Available http www tyromotion com fr produits pablo apercu html Accessed 12 2012 6 K Br tsch T Schuler A Koenig L Zimmerli S M rillat L L nenburger R Riener Jancke et A Meyer Heim Research Influence of virtual reality soccer game on walking performance in robotic assisted gait training for children Journal of NeuroEngineering and Rehabilitation vol
79. o Need to have preprogrammed training programs that can be followed These demands for a GUI will be integrated when a GUI for the rehabilitation robot ALTACRO will be made in chapter 5 1 2 2 Performance indicators To give feedback to the patient it is useful to have an indicator of his global performance during the rehabilitation session It is also useful for the physiotherapists and for the objectiveness of the session evaluation to have an indicator about the patient s performance There already exist multiple kinds of such indicators In the next paragraph the utility of these indicators sometimes also called bio indicators will be studied 1 2 21 Modified Rivermead Mobility Index The Modified Rivermead Mobility Index MRMI is a measure of disability related to bodily mobility It demonstrates the patient s ability to move his own body It does not measure the effective use of a wheelchair or the mobility when aided by someone else It was developed for patients who had suffered a head injury or stroke at the Rivermead Rehabilitation Centre in Oxford England 11 The mobility of the person is evaluated by 15 mobility related questions as for example Can the patient move from bed to chair and back without any help or Can the patient sit on the edge of the bed without holding on for 10 seconds Points are attributed to the responses of the questions with a score from 0 to 5 for each question The global mobility of
80. omplete for the desktop or the Web Rather complex to use but JavaFX Swing JavaFX language of interfaces Not especially fast Several development tools see resources Portability Google GWT framework for the Web converts Java to JavaScript Access to OpenGL with jogl and Swing Canvas object Development of a graphical user interface for a rehabilitation exoskeleton 62 Attachments WPF Language of interface XAML Rather slow in execution ES Easy programming with Visual Studio Tcl language toolkit 34 35 Ease of use Tcl is a dynamic language and therefore Tk Tkinter is slow Portable Fewer built in widgets Runs on any system where Python is installed Lightweight Restricted to the C programming language FLUID editor with a visual interface GTK WxWidget More difficult to program Development of graphical user interface for rehabilitation exoskeleton 63 Attachments Attachment 3 Matlab code for the trajectory generation all the dimensions will be given in cm in this program initialisation of the program ear all ose all lc import of the Winter angles for hip knee amp ankle nkle importdata Winter NormalSpeed Ankle Angle Periodic 1000p mat A initialisation of the patient session dimensions L feet 25 L lower leg 45
81. on The other data such as graphics with the data of the joints the torques in the joints or the data of the gait are not very important to be saved They are only subject of interest during the session itself 2 1 3 Patient feedback methods A last important point is that it would be nice to have some direct feedback methods for the patient This is because although there is always a physiotherapist in the environment of the patient this physiotherapist could not always be attentive to all the movements of the patient In this way it would be useful if there is some signal if the patient is deviating a lot from the desired gait trajectory This signal can be a visual signal such as a led which starts blinking or an auditory feedback for example the device says the patient that he has to lift his left ankle a little bit more By doing so care must be taken to not overuse this kind of feedback If a led is blinking almost all the time to say that the movement is not executed well enough the patient would become dispirited and after a time the physiotherapist would not be aware anymore of the led The same happens if the auditory feedback is exorbitant both the patient as the physiotherapist would be tired to hear the Development of a graphical user interface for a rehabilitation exoskeleton 18 Specifications of a GUI device all the time So by using immediate visual or auditory feedback an equilibrium must be found between too much and too
82. or ALTACRO In this chapter the previously found specifications for a graphical user interface for a rehabilitation robot will be implemented GUI for ALTACRO will be implemented In this work the interface will especially be concentrated on the needs of the engineers who are working with the device In a future work this interface can be extended with a rehabilitation session part with performance feedback 4 1 ALTACRO conventions ALTACRO Automated Locomotion Training using an Actuated Compliant Robotic Orthosis is a project started in 2008 at the VUB in order to help patients suffering from gait impairment A picture of the rehabilitation device is given in Figure 13 For the project a rehabilitation device was made The research was mainly focused on four aspects of robot assisted gait training 2 e Active ankle assistance e Balance and load distribution e Functional three dimensional gait kinematics e Physical human robot interaction Figure 13 The ALTACRO rehabilitation device 30 Before handling the robot the different degree s of freedom DOF of ALTACRO are explained and some conventions of the reference frame are to be made Three angle definitions are used with ALTACRO e The angle definitions in the coordinate systems CS linked to the joints and the DOF s of the actual robot also defined further on as robot angles e The angle definitions as used for the Winter data e The angle definitions inherent to the
83. ory generation buttons Figure 21 Trajectories sub tab of the GUI TCP IP communication The settings for the TCP IP communication are gathered in this tab The user physiotherapist doesn t need to go to this tab as the settings remains constant for each session Only if something changes in the communication protocol e g if another computer port is used an engineer needs this tab Real time measurements Figure 22 This tab contains the returned real time signals RTS as well as the control parameters to tune the robot The control parameters are the parameters that manage the low level controllers of the robot When passing with the mouse over the control parameters the names of the control parameters as well as their units are displayed The control parameters may be saved in two different ways If the user employs the save to default button the next time you run the program this new control parameters will be used automatically independently of the chosen patient If the save under a new name button is used the user needs to choose a new path name and a back up file is made The next time the program is used the old values will be loaded as default values The real time signals are the signals that are collected coming from the robot such as the speed and the position angular or translational of each joint The data as well the control parameters as the real time signals are displayed accordin
84. otic Assisted Gait Training RTS Real Time Signal THBI Total Heart Beat Index VB Visual Basics VUB Vrije Universiteit Brussel Development of a graphical user interface for a rehabilitation exoskeleton List of figures List of figures FIGURE 1 PHOTO OF THE LOKOMAT DEVICE 81 nennen nennen nnne enne 3 FIGURE 2 SCREENSHOT OF THE PATIENT SETTINGS TAB OF THE LOKOMAT 81 5 FIGURE 3 SCREENSHOT OF THE TRAINING SETTINGS OF THE LOKOMAT 81 5 FIGURE OPTOGAIT 9 10 ce rere td c e Eh ere eee AE tira 6 FIGURE 5 FEEDBACK SYSTEM OF OPTOGAIT WITH ABOVE THE CAMERA IMAGES IN THE MIDDLE THE GRAPHICAL FEEDBACK AND ON THE BOTTOM THE NUMERICAL VALUES 9 nnne nnn nnne nennen 8 FIGURE 6 PABLO b i iioc te dau 9 FIGURE 7 SCHEMATIC HEART RATE PLOTTED AGAINST TIME DURING EXERCISE 13 13 FIGURE 8 THE DIFFERENT PHASES IN A GAIT CYCLE 23 nennen enne enne nennen 23 FIGURE 9 VERTICAL MOVEMENT OF THE TRUNK UPPER CURVE KNEE FLEXION MIDDLE CURVE AND PELVIC OBLIQUITY BOTTOM CURVE ROTATION ANGLES IN DEGREE FOR ONE GAIT CYCLE 24 24 FIGURE 10 PELVIC ROTATION IN DEGREE FOR ONE GAIT CYCLE 25 25 FIGURE 11 GENERATED TRAJECTORY FOR THE HEEL FOR A STEP LENGTH OF 50 CM AND A FOOT CLEARANCE OF 10 CM 27 FIGURE 12 GENERATED TRAJECTORY FOR THE TOES FOR A STEP LEN
85. out that the social interaction between the therapist and the participant undoubtedly plays a crucial role especially for patients Thus the use of VR during rehabilitation therapy should not replace the physical therapist but rather provide an additional means of enhancing training efficiency 6 If the exercise is done in a playful manner care has to be taken that the game does not become tiresome after having played it a lot Attention has to be paid to the fact that the patients will use the revalidation device for a long time So the patient interface has to encourage the patient to evolve during the complete revalidation period without becoming tired of it A GUI is also important for the engineers who are developing the robot and who are working with it They need the GUI to control the robot and to check the data coming from the robot This is needed as well during the development phase as during the debugging phase and the maintenance of the device 1 2 State of the art 1 2 1 Existing GUI s of rehabilitation devices Firstly a look is taken on graphical user interfaces GUI s of other commercial available gait rehabilitation robots At the moment little is written over GUI s of gait rehabilitation robots so it is useful to also take a closer look to GUI s of other devices such as GUI s of arm rehabilitation devices Many of the concepts of these GUI s can be compared to those of gait rehabilitation devices 1 2 1 1 Lokomat Th
86. parameters from TCP IP Read choosen node value Read choosen parameters Display on chart true Load the appropriate range values for the graph Load the correct axes labels Load and display the corresponding control parameter false Open TCP IP connection _ Start writing all the incoming data in a binary file true Practical implementation realization of a GUI for ALTACRO Load patient personal data Load patient dimensionts Load trajectories assigned to that patient Update patient database Remove patient file Save patient data personal data dimensions attribuated trajectories false Read patient dimension Choose trajectories for hip knee amp ankle if not done by loading a patient file v Load choosen trajectories create submatrices ofthe trajectories that are used for faster simulation purposes Calculate kinematics Generate amp display a stickmodel true thatfollows the generated trajectories Figure 18 Program structure The orange arrows correspond to actions which send the execution to the case structure The black arrows correspond to the flow in the second while loop 44 Development of a graphical user interface for a rehabilitation exoskeleton Practical implementation realization of a GUI for ALTACRO 4 3 Operation of the interface In this part the functioning of the interface will be explained In a first section the structure of the
87. patient database Push the Select button All the previously added settings are automatically loaded Development of a graphical user interface for a rehabilitation exoskeleton 50 Practical implementation realization of a GUI for ALTACRO Connection open eron D Search existing patient button Patient database ytton Figure 25 Patient tab of the GUI buttons to select an existing patient file b For a new patient file Figure 26 Enter the patient data and patient dimensions into the boxes Note that there is the possibility to give different lengths for the left and the right side of the patient This is for in the rare case that the patient does not have symmetric legs and is done by toggling the switch of Symmetric legs This switch is automatically put on symmetric legs during the creation of a new patient file Push the Save patient settings button Development of a graphical user interface for a rehabilitation exoskeleton 51 Practical implementation realization of a GUI for ALTACRO Connection open rer ______ ane meae Save patient settings button Patient data Patient dimensions Figure 26 Patient tab of the GUI buttons to select a new patient file 4 Open the TCP IP connection Figure 27 a Push the Open connection button b Verify that the Connection open led is lit Connection open Sv pat
88. physiological compatibility of the trajectory and from the wishes of the physiotherapist during the session These requirements are gt Requirements from the physiotherapist To impose a desired step length e To impose a desired step height also called the foot clearance To choose the walking speed of the patient The walking speed will influence the ratio between the stance time to the swing time This ratio decreases with an increase in walking speed 2 Requirements for the compatibility between the robot movements and the patient s physiology e For the vertical movement of the trunk 24 The amplitude of the displacement is related to the walking speed from 3 cm peak to peak for slow walking 0 8 m sec up to 8 cm peak to peak for fast walking 2 2 m sec For a mean walking speed 1 4 m sec the peak to peak amplitude of the displacement is typically 4 to 5 cm The displacement can be approximated by a double frequent sinus for one walking cycle This double frequent sinus is a simple sinus which executes two sinus cycles during one gait cycle The sinus starts at its mean value during the initial contact of the heel as is visualized in Figure 9 Development of a graphical user interface for a rehabilitation exoskeleton 23 Trajectory generation For the pelvic obliquity in the frontal plane 25 The pelvic obliquity during one gait cycle can be approximated by two superimposed sinus as illustrated in Figure 9
89. rajectory is physiologically correct If this is not the case a three dimensional hip guidance can engender more problems than before A third possibility is the implementation of active ankle assistance This seemed also to be a good improvement for a gait rehabilitation device For the current rehabilitation processes the ankle is mostly not taken into account At the end of the rehabilitation this is often solved by implementing an orthosis It should be a nice improvement is this active ankle assistance is combined with the gait rehabilitation 2 2 Requirements of a GUI classified by user type This chapter describes the required options of a graphical user interface GUI of a rehabilitation robot Firstly the different users namely the engineers physiotherapists and the patients are examined Then the specific needs and requirements of each type of users are developed Out of these needs the different options that need to be implemented in the GUI are extracted 2 2 1 The different users As stated before there are three different kind of people that uses the GUI e The engineers who develops the robot The physiotherapist who supervises the patient e The patient who has to use the device for his her rehabilitation These three kinds of users all have specific needs They use the GUI for different things so it is possible that different GUI s need to be made or that a single GUI has different tabs and or screens to fulfill in everybody
90. rface Graphique pour un Exosquelette de R habilitation Auteur Laura De Rijcke Master degr en Ing nieur civil lectrom canique Ann e acad mique 2012 2013 Mots cl s Interface Graphique Robot de R habilitation G n ration de Trajectoire Une anomalie de la d marche peut avoir une forte influence n gative sur la qualit de vie d une personne Afin d aider des patients avec une anomalie de la d marche un robot de r habilitation ALTACRO a t d velopp la Vrije Universiteit Brussel Dans ce contexte ce m moire traite du d veloppement d une interface graphique pour un appareil de r habilitation Une interface graphique est tr s importante pour un appareil de r habilitation La pr sence d une interface de ce type am liore la convivialit et la s curit de l appareil Elle facilite le contr le de l appareil et la communication entre le robot et l utilisateur La premi re partie de ce m moire enqu te sur les interfaces graphiques d appareils de r habilitation existants Ces interfaces sont compar es et les options principales sont extraites de chaque interface Ensuite plusieurs indicateurs de performance sont compar s Ces indicateurs sont des variables qui donnent une indication de la performance du patient durant la session de r habilitation Ils sont compar s afin de voir quel indicateur donne le r sultat le plus pr cis en combinaison avec un appareil de r habilitation Plusieurs environnemen
91. s FIGURE 23 FLOWCHART FOR A CORRECT USE OF THE GUI IN YELLOW THE TAB NAMES IN BLUE THE MANDATORY ACTIONS AND IN SALMON THE OPTIONAL ACTIONS nennen nennen trier inns ass as dass asses eis 49 FIGURE 24 STARTING OF THE PROGRAM THE ARROW IS INDICATING THE BUTTON IN LABVIEW ee 50 FIGURE 25 PATIENT TAB OF THE GUI BUTTONS TO SELECT AN EXISTING PATIENT FILE ss 51 FIGURE 26 PATIENT TAB OF THE GUI BUTTONS TO SELECT A NEW PATIENT FILE 52 FIGURE 27 HEADER OF THE GUI WITH THE TCP IP CONNECTION BUTTONS HIGHLIGHTED eene 52 FIGURE 28 SCREENSHOT OF THE PATIENT SETTINGS TAB OF THE LOKOMAT 81 60 FIGURE 29 SCREENSHOT OF THE TRAINING SETTINGS OF THE LOKOMAT 8 cce 61 Development of a graphical user interface for a rehabilitation exoskeleton xi List of tables List of tables TABLE 1 GUI SETTINGS AND POSSIBILITIES OF THE LOKOMAT 7 120002044060 000000000000 3 TABLE 2 GUI PROPERTIES OF OPTOGAIT 8 nennen 7 TABLE 3 GUI PROPERTIES OF PABLO 10 rr id ceo eo neo cn cere a ara ee ee race aid da 9 TABLE 4 COMPARISON BETWEEN THE OPTIONS OF THE DIFFERENT GUI S sss enhn 10 TABLE 5 COMPARISON OF THE PERFORMANCE INDICATORS sans siens r nennen enn 14 TABLE 6 INPUTS OF THE GENERATED TRAJECTORIES ee eeeeee eee een
92. s needs Development of a graphical user interface for a rehabilitation exoskeleton 19 Specifications of a GUI 2 2 2 The engineers These are the people who will develop the rehabilitation device They will mainly need an interface to test the behavior of the rehabilitation device This is as well needed during the elaboration of the device as later on for maintenance of the device They need a user interface which represents all the technical data in a readable way This technical data consists of multiple parts e communication part This part is handling the settings which are needed to establish a correct communication between the device and the user interface By making this part clear and easily accessible eventual communication problems can be quickly detected Atrajectory part This part is handling and or generating the trajectory s which are imposed on the joints of the rehabilitation device If the trajectories are generated by the interface they must be kinematically correct and physiologically possible e A technical input part This part is handling the technical inputs which are sent to the device These are mainly the control parameters e g the variables and the gains used in the control loops of the device A feedback part This part of the interface should represent the data which is coming from the device in a readable way These data are for example the current and the torque in the different joints of the device
93. sion The second while loop is the actual program and the case structure handles the special events Development of a graphical user interface for a rehabilitation exoskeleton 39 Practical implementation realization of a GUI for ALTACRO 1st while loop 2nd while loop Stop Figure 17 Main structure of the programming of the interface More in detail the sequential structure sets all the patient variables dimensions personal data to zero cleans the graphs loads the last saved control parameters for the robot controllers and initializes the lists with the available trajectories for the ankle hip and knee nodes With the aim of facilitating and accelerating the use of the program for the engineers during the test phase of the robot values of fictive standard patient are loaded during the startup of the program In this way the engineers can immediately test new trajectories or new control parameters without the need of selecting firstly some patient dimensions After passing this initialization phase the program is entering a first while loop This while loop runs non stop during the execution of the program It can only be stopped by pushing the stop button This first while loop contains a case structure and a second while loop See Figure 17 The case structure contains actions which influence and set many different values that are used in different places in the program I
94. t gait parameters step length step height and the amplitude of the hip movements are dependent off the walking speed but the exact correlation factor has not yet been developed This factor can be implemented in the trajectory development tool to increase the user friendliness of the device and to increase the robustness of the physiological compatibility of the trajectory generation tool It is possible that the option to change the different input parameters manually may still be available as the gait cycle also changes dependent on the individual In a future work investigation can be done for the further development of a graphical user interface for the physiotherapist and for the patient For the physiotherapist the effectuated trajectories may be visualized and compared to the desired trajectories For the patient a feedback method may be developed as stated before In a future work a procedure for going from a sitting position to a standing position and back may be developed This will facilitate the start up and the end of a rehabilitation session At the moment in the beginning of the session the patient has to be put in a standing position in the rehabilitation robot This is difficult for the physiotherapist if the patient is in the beginning of his rehabilitation process especially when the patient is heavy By implementing a stand up and a sit down procedure a wheelchair can be used to move the patient to and from the therapy The patie
95. t is used to load new global parameters such as a file with new control parameters a new trajectory file and a new patient file or to open the TCP IP connection In this way it is also sure that the GUI program restarts the main program loop with the new values The second while loop which is nested in the first one contains the main code of the GUI program It runs continuously and can only be stopped by the STOP button or by one of the actions of the previously explained case structure If the STOP button was pressed the complete program would be stopped if one of the described actions of the case structure was asked this action is executed and the 2 while loop restarts Development of a graphical user interface for a rehabilitation exoskeleton 40 Practical implementation realization of a GUI for ALTACRO So to summarize the overall structure Figure 17 The second while loop and the case structure nested in an overall first while loop The second while loop is the main part of the program When event handled by the case structure occurs the second while loop is left and the program goes back to the first while loop Here it can enter the case structure The code of the case structure is executed and the program goes again to the second while loop Is the STOP button is pressed the program leaves both while loops and the complete program is stopped
96. tation exoskeleton 14 Introduction 1 2 3 Programming environments The programming language environment that is chosen to make the graphical user interface GUI will have a major importance Not only for the ease of programming the GUI but also for the possibilities and the lay out of the final GUI It is useful to examine first different programming environments before choosing one Otherwise it is possible that the programming task has to be restart several times due to the fact of touching the limits of the programming environment Hereafter the most used meaningful environments for making a GUI are exposed Other possibilities than the ones explained can be find in attachment 2 with their pro s and contra s 1 2 3 1 QT Qt is an object oriented GUI toolkit which allows programmers to choose between the Motif and the Windows look and feel It implements its own widgets user interface elements Almost all classes and member functions in Qt are documented The documentation is available in HTML postscript text and as manual pages 16 Qt has a high portability and flexibility It is also very fast and is available for free Another advantage of QT is its mechanism with signals and slots which eliminates the need for callbacks and provides type safe way to send arguments of any type These concepts make it much easier to control the communication between objects in a flexible and maintainable manner than it is with a fragile callba
97. tation exoskeleton 36 Practical implementation realization of a GUI for ALTACRO Figure 16 Winter angle definitions 4 1 4 MACCEPA conventions To ensure the variable stiffness of the actuators of the robot the joints are equipped with MACCEPA actuators Mechanically Adjustable Compliance and Controllable Equilibrium Position Actuators 31 A MACCEPA is a compliant actuator made of 3 bodies namely two links and a lever arm which are pivoting around a common rotation axis as can be seen in Table 13 The angle p between one link and the lever arm is set by a traditional position controlled actuator A spring is attached between the lever arm and the other link The angle between these two bodies is called When a is zero the actuator is in its equilibrium position When a differs from zero the force due to the elongation of the spring will generate a torque that tends to line up the second link with the lever arm 31 As in each joint the relative placements of the MACCEPA actuator and of the motor are a little bit different the MACCEPA angles are defined as follows see also Table 13 Development of a graphical user interface for rehabilitation exoskeleton 37 Practical implementation realization of a GUI for ALTACRO e forthe hip joint gt a angle between the lever arm and the upper leg gt angle between the prolongation of the upper body and the lever arm e for the knee joint gt o angle between
98. the lever arm and the lower leg gt Q angle between the prolongation of the upper leg and the lever arm e forthe ankle joint gt a angle between the lever arm and the lower leg gt Q angle between the perpendicular of the foot and the lever arm Table 13 MACCEPA angle definitions Hip Knee Ej 2 Ankle lever arm Development of a graphical user interface for a rehabilitation exoskeleton 38 Practical implementation realization of a GUI for ALTACRO 4 1 5 Link between the different conventions In order to work with the different coordinate systems and angle definitions they should be linked one to each other This is done in table 14 The MACCEPA angles are called B corresponding to the previous section the Winter angles are indicated with a subscript w and the robot angles have a subscript r Table 14 Link between the angle definitions MACCEPA WINTER WINTER robotangles Robot angles MACCEPA Hip B gt 07 Ankle 0zw B 5 ne 4 2 Programming of the interface The programming of the interface is done in LabView as justified in section 1 2 3 LabView is a graphical dataflow programming language The code is made out of graphical blocks which are connected one to another with wires As soon as all the input data given by the wires is available block can be executed This may be the case for multiple blo
99. the patient if evaluated by his obtained total score Note that these questions are based on the tools needed for moving 1 2 22 Motor Assessment Scale MAS The Motor Assessment Scale MAS is a measure of motor impairment and mobility in stroke patients The MAS uses a seven point ordinal scale to measure five mobility related activities that are similar to the MRMI activities rolling from supine to side lying rising from supine to a sitting position balanced sitting standing up from a sitting position and walking Three additional items measure the impairment and function of the upper limb 12 The way to evaluate the performance of the patient with the MAS is nearly the same as with the MRMI Although the specific questions and the way to attribute the points are a little bit different the principle of the method is to give points based on the result of questions These questions are as for the MRMI based on the ability of the patient to move with or without external tools Development of a graphical user interface for a rehabilitation exoskeleton 11 Introduction 1 2 23 Indirect Calorimetry Indirect Calorimetry is based on the assumption that all energy releasing reactions in the body ultimately depend on oxygen uptake The most common method of measuring oxygen uptake during exercise is by open loop spirometry when exhaled air is sampled and analyzed for its oxygen content Recent developments have led to lightweight portable teleme
100. the user pushes one of the view buttons Top view Front view or Side view the corresponding view is also projected during the simulation of the trajectory By pushing the Show only projection button the 3D stick model is not visible anymore and only the chosen projections are displayed during the simulation Development of a graphical user interface for a rehabilitation exoskeleton 46 Practical implementation realization of a GUI for ALTACRO Connection open Technical specifications Patient Session STOP Save patient settings Open connection Error Selection of the base trajectories Trajectories TCP IP communication Real time ulation Visualization of the peed button trajectories Hip trajectory hip csv 30 Surface hip2 csv Simulation speed hip3 csv winter 1000 points csv 0 50 100 150 Knee trajectory knee csv knee2 csv knee3 csv winter 1000 points csv Step length cm 1 Step height 140 Offset hips 7 100 Amplitude pelvic obliquity y 0 Amplitude transverse obliquity yi Ankle trajectory ankle csv ankle2 csv ankle3 csv winter 1000 points csv Show only the projection view Front view Side view Amplitude trunk movement cm go readmill speed m s 24 View buttons oad choosen trajectories oa Traject
101. tion of the real time signals It checks the selected parameters and nodes and displays the desired data It also changes the range of the graphs and the displayed units as a function of the chosen parameters and the type of the chosen node if it is a translational joint or a rotational joint e block that is handling the information related to the patient data This block saves the entered patient data in a new file or fills the variables for the patient data with the data out of an existing file e A block that handles the simulation of the trajectory with a stick model It loads the angles for the different nodes as well as the patient dimensions and takes them as input From this data it generates a stick model such that the movement of the robot during the execution of the trajectories can be visualized Development of a graphical user interface for a rehabilitation exoskeleton 42 Practical implementation realization of a GUI for ALTACRO Initialisation of the start values default values of variables empty graphes Load the patient amp trajectory databases Load a fictive standard patient gt easier for testing trajectories checking technical parameters gt Insertion of new control parameters CP false a Add time label and save true choosen standard file Sending of ha false false Give a 0 value for the real parameters RTP Read real time
102. tool was developed The generated trajectories must satisfy kinematically imposed constraints as well as physiologically imposed constraints The trajectories were firstly developed in two dimensions and were then extended to three dimensions In two dimensions it was tried to base these generated trajectories on the data recorded by Winter 26 This method was not feasible if the step length and the step height must be possible to change The generated trajectories as well in two dimensions as in three dimensions must finally be completely based on the physiological constraints and the gait parameters given by the user inputs The generated trajectories have then been evaluated by comparing them to measured data By using right combinations for the input parameter values the trajectory generation is working well In the next part of the thesis a GUI is implemented for a gait rehabilitation robot ALTACRO It was chosen to focus this interface on the needs of the engineers The interface was programmed in LabView and the trajectory generation was programmed in Matlab By programming the interface in LabView the GUI could easily and quickly interact with the robot In this way the data coming from the robot joints could be viewed almost in real time The interface satisfied to all the engineers needs it could change and check the control parameters gather and visualize the joint data in an organized way manage the communication and control the trajector
103. tric devices that are capable of performing breath by breath oxygen and carbon dioxide analysis Oxygen uptake is expressed either in unit time Vo ml kg min or in unit distance oxygen cost in ml kg m which can be considered as a measure of metabolic efficiency 13 The problem with Vo measurements is that they are cumbersome to conduct the instrumentation is expensive for a routine laboratory and the measurements require trained personnel 14 Also for the patient the session would become less comfortable which may affect his obtained results 1 2 2 4 Physiological Cost Index PCI The Physiological Cost Index PCI is calculated in beats per meter by a combination of the heart rate HR of the patient and his walking speed It is calculated as Working HR Resting HR PCI EK Walking speed meters minute PCI can be used e measure changes in locomotor efficiency over time measure changes as a result of the use of different orthotic or prosthetic devices e as an indicator of the handicap when compared with matched normative data 15 The calculation of this indicator can easily be made by a rehabilitation device as it needs only the walking speed of the patient and the HR of the patient which can easily be measured A negative point for this method is that the method works best at the preferred walking speed of the user Since the patients are relearning to walk there would not be an important dif
104. ts de programmation sont explor s afin de voir lequel est le plus appropri pour r aliser une interface graphique pour un robot de r habilitation Dans la deuxi me partie de ce m moire les exigences pour une interface graphique sont examin es L interface est n cessaire pour trois types d utilisateurs les ing nieurs les kin sith rapeutes et les patients Chaque groupe d utilisateurs a des exigences particuli res concernant le contenu de l interface Les exigences de chaque groupe sont examin es et num r s Pour les exigences des kin sith rapeutes et des patients cette analyse a t faite avec l aide de kin sith rapeutes Dans la troisieme partie de ce m moire un outil de g n ration de trajectoire est d velopp Ces trajectoires sont g n r es en fonction de plusieurs donn es d entr e sp cifique pour la r habilitation de la d marche comme la longueur du pas la hauteur du pas et l amplitude du mouvement de la hanche Ces trajectoires g n r es doivent tre cin matiquement et physiologiquement correctes L influence des diff rents param tres de la d marche sur les trajectoires g n r s est ensuite explor e Dans la derni re partie de ce m moire les connaissances acquises sont appliqu es afin de r aliser une interface graphique pour ALTACRO Cette interface est ax e sur les besoins des ing nieurs Un programme pour g n rer des trajectoires est galement impl ment dans cette interface Development
105. vation and Development of a graphical user interface for a rehabilitation exoskeleton 1 Introduction attentiveness of the patient which is thus encouraged facilitates the repetitive exercising The focus on the exercise is also reinforced by the software and patients relearn lost or partially lost abilities 5 Care has to be taken that the devices which are implemented to motivate the patient do not distract him Some pediatric rehabilitation centers that use robotic assisted gait training RAGT try to boost the motivation of patients by showing DVDs or by playing music Such strategies may distract the patients especially children from the actual therapy causing them to become less active in the Lokomat 6 It is clear that the GUI must provide a readable way of feedback for the patient This feedback has to be given in an optimistic way to encourage the patient to do his exercises correctly It can be done in different ways e g graphs of the performance results of a game a smiley if the exercise was been done well etc Another possibility is linking the exercise to a virtual reality VR environment 6 VR techniques make it possible to directly interlink the patients motor performances during the gait training with actions in a virtual world such as in a computer game If the VR games are adequately adapted to children s needs it provides motivation and yet keeps the focus on the actual gait training It should be pointed
106. ver de prestatie van de pati nt gedurende de rehabilitatie sessie De vergelijking wordt gebruikt om te zien welke indicator de meest accurate resultaten geeft indien het gebruikt wordt in combinatie met een stap rehabilitatie toestel Ten slotte worden verschillende programmeer omgevingen vergeleken om te zien welke het meest geschikt is om een grafische gebruikersinterface voor een rehabilitatierobot te maken In het tweede deel van deze thesis worden de eisen waar een grafische gebruikersinterface aan moet voldoen onderzocht De interface is nodig voor drie verschillende gebruikersgroepen de ingenieurs de kinesisten en de pati nten Deze gebruikersgroepen hebben elk specifieke eisen betrekkende de inhoud van de gebruikersinterface De noden van iedere gebruikersgroep worden onderzocht en opgelijst Voor de eisen van de kinesisten en van de pati nten werd dit gedaan in samenwerking met kinesisten In het derde deel van deze thesis wordt een traject generatie programma ontwikkeld Deze trajecten worden gegenereerd op basis van meerdere inputs specifiek aan een stap rehabilitatie therapie zoals de stap lengte en hoogte en de amplitude van de heupbeweging Deze gegenereerde trajecten moeten kinematisch en fysiologisch juist zijn In het vervolg van deze thesis wordt de invloed van de verschillende stapparameters op de gegenereerde trajecten onderzocht In het laatste deel van deze master thesis wordt de verzamelde kennis gebruikt om een grafische g
107. y generation The goal of this thesis was to investigate the development of a graphical user interface for a rehabilitation exoskeleton This has been done by exploring the requirements of a GUI specific for a rehabilitation device The GUI has been extended with a trajectory generation tool to generate physiologically safe and kinematically correct trajectories Further on the GUI has been build with a focus on the engineers needs The GUI satisfied the stated requirements and the trajectory tool worked well Development of a graphical user interface for a rehabilitation exoskeleton 54 Future work 6 Future work In a future work a feedback method may be developed for the patient This will improve his attentiveness during the therapy and will give him immediate feedback on his session results An interactive game can be developed for this purpose A bio indicator may also be developed to calculate the global performance of the patient during a gait rehabilitation process The importance of the movement of each joint in a gait cycle for the calculation of this bio indicator can be investigated In this way the global performance indicator can be calculated based on the torque needed in the actuator of each joint For the trajectory generation it may be useful to investigate on the correlation between the different user inputs step length step height walking speed amplitudes of the hip movements It has already been stated that the differen

Download Pdf Manuals

Related Search

Related Contents

12-Channel Thermocouple Datalogger Model TM500    Pressure Calibrator  Epson 36-714 User's Manual  be.ez LArobe 13" MacBook Kakee/ Lime  LMS0192A-2 - topwaydisplay.com  TruBlue Data Loggers - Manual  Notice d`utilisation - la bouillotte magique  取扱説明書 - SoftBank SELECTION  Edição 2096 – 03/05/2015 - Paróquia Porciúncula de Sant`ana  

Copyright © All rights reserved. Failed to retrieve file