Home

Binded Manual - The University of Kansas

1. 8733 9000 9500 10000 10500 11000 11500 12000 12500 13000 13500 14000 14500 15000 15500 16000 16500 17000 17500 18000 18733 Figure 44 Extracted Signal 36 NeoSuck RT CNL U of Kansas 3 2 STI Math Fundamentals 3 2 1 What is STI STI The Spatiotemporal Index Mathematically STI is the cumulative sum of the standard deviations SD for a set of trajectories STI shows the stability of a sequence of 13 movements 3 2 2 Data selection The purpose of the non nutritive suck spatiotemporal index NNS STI is to compare the suck pressure control profiles among NNS bursts for comparative studies within and among infants This comparison is based on a set of selected bursts with fixed peak number For example for any individual comparison user make if five peaks are selected for the first burst then another five peaks must be selected for the second burst and always five peaks for all the other bursts in this comparison Figure 45 shows the first five peaks from eight bursts Selecting the same number of peaks from each burst is necessary to make different sets of trajectories comparable 25 FF f 20 Hif 1 4 i Py PA a Ur f LI J 10 Uf bh Y T E y f nent hn rtf Figure 45 Original data plot 3 2 4 Data normalization As shown in Figure 45 the five NNS peaks from different bursts have different lengths
2. Figure 10 Choose new file for Burst Extraciton The Sampling frequency is a parameter used for analysis hereafter Please input the desired sampling frequency to analyze your data file Sampling Frequency Hz 100 00 2 3 2 Step 2 Start analyzing Once the file for analysis is selected click this button for analysis Once this button 1s clicked you are not allowed to go back to choose other files Meanwhile the status bar red Stopped Figure 11 will turn into green Running Figure 12 Figure 11 Status bar when Burst Extract is not running Figure 12 Statur bar when Burst Extract is running 2 3 3 Step 3 Coarse tuning Find the number of bursts There are 3 tabs in this step Tab 1 starting from the top Original Signal 11 NeoSuck RT CNL U of Kansas This tab displays the original signal Figure 13 as read from the file you selected as shown in Figure 9 Original Signal 30 a I ih l i il HH wlll I i E y Ul il nT mh ili EE WN yl I W l il i i lil it 1 I I 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 Time La Amplitude N Peak Detect HPF Original Signal Figure 13 Original signal Tab 2 High pass filter HPF for advanced user This tab allows the user to set a desired High pass filter by choosing filter order and cut off frequency to filter the original signal The filtered signal with the low frequen
3. Patient Configuration Reports User Help Followup Dates Month Followup Date Pt No Pt Medical Record No 123456 Pt First Name oe Cans Pt Last Name ob Birth Date 02 24 200 BirthOrder Ethnicity Race Gender Patient Type Discharge Date Record Position Username Location 6 12 2006 3 10 PM Figure 6 Entering new patient s information In Patient gt Patient Info there are some other buttons like View Modify Delete etc Clicking on those buttons will allow user to make changes for an individual patient in similar window as shown in Figure 6 NICU Database CNL U of Kansas 2 3 Add and modify medical information for existing patient After setup and creation of the fundamental data file outlined in section 2 2 for an individual patient the Patient Test Info table will allow the user to view the patients information database The User can add or modify the patient s database by choosing Modify icon This will allow the user to add new information to the patient s database as shown in Figure 7 Note that there are 7 pages tabs located along the top panel to allow the user to modify or add new information into the patient database These are shown on the following panels Patient Configuration Reports User Help Pt No Trial No Trial Date Pt First Name Pt Last Name Pt Medical Record No Status Birth Date BirthOrder Ethnicity Race Gender Patient Type Discharge D
4. 3 5 5 5 0 F E H o 3 m 0 04 l l l I LOAD CELL N Z 0 4 O 5 0 6 10 0 7 T T T Amplitude cm H20 N 5 5 04 a 2 2 w D 2 oO 10 0 z 0 5 5 gt 0 a Pause s 0 04 I l l f 0 0 2 0 4 0 oO 8 0 10 0 Time s 2 5 8 V NeoSuck RT E Real Time l 0 4 Period s Block Peak Count Date Created Figure 5 Blank window for recording and analysis 2 1 Record new data by NeoSuck RT To Record data for a baby go to File gt New then a Save As file dialog will pop up as shown in Figure 6 Input the file name as dat file which will be used for future analysis and click Save button NeoSuck RT ma NeoSuck RT C Program Files Neuro Logic NeoSuck RTWata test TestingA 061206 dat File Settings ols Help CNL U of Kansas MOTOR PULSE fv Save As LYDT Imm My Recent Documents 1 LOAD CELL Places SUCK PRESSURE c Time s 1 Save in D test EMG OOS F u e ex EB Testing _061206 dat My Network File name Baby_A_01 0204 X Save as type Data files dat Y NeoSuck RT Beep On Off Real Time Observations Cancel 1 0 4 0 5 0 6 Amplitude cm H20 0 5 Pause s Period s Blocks Sampled 0 Figure 6 Recording new data Date Created 6 15 2006 3 49
5. How can trajectories of different lengths be compared The next step required for STI calculation is to normalize the trajectories to the same length in order to match valley to valley 37 9000O NeoSuck RT CNL U of Kansas and peak to peak across different trials Figure 46 shows the eight normalized trajectories based on the original data in Figure 45 Important note Due to the mathematical requirements of normalization the sample number after normalization has to be larger than the maximum non normalized sample number For example in Figure 45 the maximum non normalized data length is around 8500 samples In Figure 46 after normalization all the data are lengthened to 10000 samples 3 2 5 k or hes LL jak ji 2 L J n y j f Yi i f af 1 j i i l L F P Ef t r f ah ma mm fy ays Pf phy af ree 1 5 if 4 l l ft l f y t j i J k i We Fi AUT bet if At oat ek EY to pau th if y pL cy l if J J LL ut in th ashi N TE AMA UP ARAN HARA 7 if OM ip fo Y ir fy wR LPP TA ty j N Hy Bh l Y l l T cA UL oO fi Th Ep 5 v i a A S oe AAAA i y ib N i f A A A Py oe D f ca 1 T i h p n l l J J fd j i j j PEI u P 0 5 4 j 5 ed J J J ne ha f f f J 3 l i j f th L Z J 1 ea z maa p fe g A S hbo f Ri aan i l ri bee at a 4 ide 4 ago Se SS B 1 A 5 i O 1000 2000 3000 A000 5000 Sooo
6. K69 854796 L Doe 2 26 2006 First of twins Hispanic or Latino American Indi K70 852365 J Doe 2 28 2006 First of twins Hispanic or Latino American Indi K71 753962 Doe 37172006 Singleton Hispanic or Latino American Indi K72 158742 S Doe 3 24 2006 Singleton Hispanic or Latino American Indi K73 253698 X Doe 3 5 2006 First of twins Unknown Black or Afric K74 852741 Z Doe A 2NN06 Singleton is i i American Indi K75 896542 M Doe Patient Info Search Saf Asian K76 158752 Doe American Indi K77 857963 P Doe American Indi K78 956327 Doe Q American Indi K79 785596 R Doe si American Indi K80 785962 T Doe American Indi Medical gt Record No Modify Mody C Last Name C First Name Cancel Record Position 1 78 Username Location kumc 6 14 2006 3 43 PM Figure 8 Search in Patient Info 14 NICU Database CNL U of Kansas The Sort button will also allow the user to sort the patients database by different options such as Patient Number Patient Medical Record First Name Last Name Date of Birth Birth Order Ethnicity Race or Gender refer Figure 9 Pt No Pt Medical Record No Pt First Name Pt Last Name BirthD ate Birth Order Ethnicity Race a 123456 Joe Doe 1424 2006 Singleton Hispanic or Latino American Indi 234567 A Doe 2 24 2006 Singleton Hispanic or Latino American Indi 456785 B Doe 1422 2006 Singleton Hispanic or Latino American Indi 458796 C Doe 17272006 Second of Twin Not Hispanic or Latin Native Hawai 456789
7. Math Fundamentals 1 1 What is STI STI the spatiotemporal index Mathematically STI is the cumulative sum of the standard deviations SD for a set of trajectories STI shows the stability of a sequence of movements 1 2 Data selection The purpose of non nutritive suck spatiotemporal index NNS STT is to compare the suck pressure control profiles among NNS bursts for comparative studies within and among preterm babies This comparison is based on a set of selected bursts with fixed peak number For example for any individual comparison you make if 5 peaks are selected for the 1 burst then another 5 peaks must be for the 2 burst and always 5 peaks for all the other bursts in this comparison Figure 1 shows the first 5 peaks from eight bursts which are called eight trajectories Selecting the same number of peaks from every burst is necessary to make different sets of trajectories comparable 30O 25 gt r Lf AO A e a l 20 Hii l Aiie l A 15 me Jy Lp r A ERREA ty 10 oO 1000 2000 3000 4000 5000 S6000 ZOOO 3000 9000 Figure 1 Original data plot 1 3 Data normalization As shown in Figure 1 the five NNS peaks from different bursts have different length How could we compare among trajectories with different length In this section we need to normalize different trajectories to the same length in order to match valley to valley and peak to peak of
8. NeoSuck RT CNL U of Kansas User manual There are six panels in this NeoSuck RT Labview version they are l 2 gt 6 DAQ Data Acquisition BurstExtract Burst Extraction for more accurate peak detection NNS Param includes peak detection burst detection histogram generation mean and standard deviation calculation NNS FFT power spectrum generation for entire data or selected part of data NNS STI STI analysis Oxygen Important note All blue texts are inputs that user needs to tune data analysis All purple texts are outputs that user needs to keep in mind for data processing thereafter or indicators for generated file path and name NeoSuck RT CNL U of Kansas 2 1 Basic of Labview exe While running an executable file of Labview it is necessary to make sure that Labview is running which is indicated by the status bar under the tool bar Figure 3 exemplifies that Labview is running with solid arrow and red stop sign gt 2 n Figure 3 Status bar Running Figure 4 shows what the status bar looks when the red Aborted button is clicked The red Aborted sign is utilized when ending the program abruptly This abort button is not typically used when stopping the program as it will not save any files or data Aborted button should only be used when the program has errors To resume the program please click the arrow button to re activate this program before running anything else on Figu
9. Normalized suck pressure 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 Standard Deviation STl 47 602 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 Relative sample number Figure 3 STI Calculations for NNS skilled preterm infant Figure 3 is the STI analysis result for a preterm infant with good suck pressure patterning with an STI 47 60 Figure 4 shows an example of poor NNS patterning sampled from a preterm infant with respiratory distress syndrome In this case the poor suck patterning results in an STI increase to 93 01 Suck pressure cmH 0 Normalized suck pressure NNS STI CNL U of Kansas Original Difference Signals 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 Original sample number Linear Time and Amplitude Normalized Signals 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 Standard Deviation STI 93 0103 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 Relative sarnple number Figure 4 STI Calculations for NNS impaired preterm infant NNS STI CNL U of Kansas 2 Part Il User manual The NNS STI is programmed using LabVIEW The plots in LabVIEW are hard to export to other file If a pretty plot like Figure 3 or Figure 4 is desired a MATLAB code NormAndOrigPlot m needs to be called for STI calculation based on LabVIEW data file Important note All blue are inputs that user needs to tune to do the STI data selection All purple text are outputs that user needs t
10. Periventricular leukomalacia Yes No CT of head C Yes No Modify _Delete_ 3 page 11 Save Cancel NICU Database CNL U of Kansas Patient Configuration Reports User Help MRI of head pH probe C Yes No C Yes Modify GER documented on pH probe Yes No No _Delete_ VP Shunt C Yer No ee __ Add Caffeine Delete Medications Mode of Feeding 5 NG PO Percent of feeding j 25 oral by voulme Past Medication __ Add Days of oral feedings 2 Medications Delete prior to study Breast Milk C Yes No Number of emesis day ir fo pres aes E 2 last 72 hours FET C Yes Hodify a G No _Delete_ Scintiscan Yes GER documented on Blood Culture Sepsis Date scintiscan Yes C Yes Modify FEN Delayed gastric emptying No _Delete_ on scintiscan Yes Save Cancel Patient Configuration Reports User Help EGhiciiea aaive Oxymetry swallow results P ass No Aspiration meningitis Modify re C Yes Fe Indications for Oxymetry Cough or choke witl i i Xi feeding swallow G No _Delete_ At the time of study ROP stage at last exam fis Neurologic Exam IN l Normal ROP zone at last exam fy Gag reflex IN l Normal State Control during study Eyes Closed normal 1 Resp absent d Modify movements ABR Date
11. Prococol gt lt ProtocolDataSet gt 18 NeoFlex RT CNL U of Kansas 3 References Barlow SM Finan DS amp Rowland SG 1992 Mechanically evoked perioral reflexes in infants Brain Research 599 158 160 Barlow SM Finan D Bradford PT amp Andreatta R 1993 Transitional properties of the mechanically evoked perioral reflex from infancy through adulthood Brain Research 623 181 188 Barlow SM Dusick A Finan DS Coltart S amp Biswas A 2001 Mechanically evoked perioral reflexes in premature and term human infants Brain Research 899 251 254 Barlow SM Estep M Vantipalli R amp Finan DS 2005 Mechanosensory modulation of the trigeminofacial system during non nutritive suck in premature infants Society for Neuroscience 753 12 19 NICU oromotor database R Vantipalli J Wang amp S M Barlow CNL Technical Research Report 2006 4 1 16 University of Kansas Lawrence Kansas USA Reprint available at www ku edu cnl NICU Database CNL U of Kansas Table of contents t NICU Database Setup cciccccacesusesdeondemedrsentedsvandeensiesuseadsarndenatenmesteddamdineddauteadianddeetanneseddiaes 3 DZ Urmand onee E E OA 4 2 1 Overview of men bar OPuOns ices ou ssivaetendeteteahee A Modiualem entices 5 22 Add new Patient m NICU database vssscsscuestiwciecess n 9 2 3 Add and modify medical information for existing patient cccccssseeeeeeeeees 10 2 4 Other funcuons 1 NICU data ase erer a tea
12. are sampled without a perturbating stimulus 5 on The trigger level in suck pressure is set to 5 cmH20 The average signal calculated by this method just includes the blocks with Motor Impulse triggered 10 off The trigger level in suck pressure is set to 10 cmH20 The average signal calculated by this method doesn t include the blocks with Motor Impulse triggered In other words the blocks contributed to this average signal represent an experimental control condition in which the EMG profiles associated with suck are sampled without a perturbating stimulus 10 on The trigger level in suck pressure is set to 10 cmH20 The average signal calculated by this method just includes the blocks with Motor Impulse triggered 5 on off The average signal showed by this method is calculated by the signal from second method 5 on minus signal from first method 5 off It displays the net response of EMG signal after the baby s lip is tapped by the probe 10 on off The average signal shown by this method is calculated by the signal from fourth method 10 on minus signal from third method 10 off It displays the net response of EMG signal after the baby s lip is tapped by the probe 15 NeoFlex RT CNL U of Kansas Signal Average C Program Files Neuro Logic NeoF lex RI Wata t49_101405_reflex dat File Settings Motor Pulse EMG OOS F uv nl Trigget I 1 1 1 I I 6 al 1 j I I 1 LYDT
13. different trials Figure 2 shows the eight normalized trajectories based on the original data in Figure 1 NNS STI CNL U of Kansas Important note Due to the mathematical requirements of the normalization the sample number after normalization has to be larger than the maximum non normalized sample number For example in Figure 1 the maximum non normalized data length is around 8500 while after normalization all the data are lengthened to 10000 L i i L L i i O 1000 2000 3000 A000 5000 Sooo SOOO S000 9000O 1 OOOO Figure 2 Normalized Data plot 1 4 SD and STI calculation Suppose we are given a population x1 xy Of values The arithmetic mean of this population is defined as Eq 1 _ I MK Fit hy x x SS l A 7 1 2 In our case as shown in Figure 2 there are eight trajectories thus an N 8 We will calculate the SD based on eight normalized pressure values for each sample point 10000 SD values will be calculated for the normalized data Figure 3 shows the original data normalized data and standard deviation The STI value represents the sum of standard deviation In order to keep STI simple and neat we will add SD every 100 samples which is named as STI step instead of adding all 10000 SDs NNS STI CNL U of Kansas Original Difference Signals Suck pressure cmH 0 Original sample number Linear Time and Amplitude Normalized Signals
14. preference changing data acquisition configuration selecting data blocks by users saving averaged signal for user selected data blocks and etc Structure Plot of NeoFlex RT _ Data File frmMain csO frmNew csO btnDataFileBrowse_Click New frmNew cs Protocol File frmMain csOfrmNew csO btnProtocolFileBrowse_Click Open frmMain csO mnuFileOpen_Click Open an existing dat file File Page Setup frmMain csO mnuFilePageSetup_Click Print frmMain csO mnuFilePrint_Click Exit frmMain vbO mnuFileExit_Click DAQ pam DAS es 8 AT amp 2 AO calibration settings for DAQ board ee a ane TR Sa Settings Graph eg ae oF Y Axis scale setting for 8 AI plots Analysis T SettingsAnalysis is os Trigger setting for AO Protocol L frmProtocal Sj Neokilex Page Setup emMain cs mE p Print Blocks Window irmBiocksAnalysis cs cs Exit Y Axs saile apn a EP D Settings p Graph f nBlockSettingsGrach os setting for 8 AI plots Save g sf Page Setup Print Signal Average R 8 j ansignaiverage csl Exit View Window ceme ee e i Arjtfact Window Size Artifact frmSigAvgSettingsArtifact Cs Pre Trigger Demean Window Settings amna anh aa Y Axis scale setting Senes j Graph imSoAveosetinosarannes ior srar pios Single Graph ae i Cursor Mode Tools p Protocol Editor fm ProtocolEditor Sy All Help tema frmAbou
15. therefore user could identify if the whole recorded data are satisfied or not In menu Setting gt Graph user could define the upper and lower data display limit of each graph to accommodate individual user preference Or user could simply check Auto Scale box to make all graphs automatically scaled Figure 10 12 NeoFlex RT CNL U of Kansas lt 2 Graph Settings Graph Settings Motor Pulse L YDT Load Cell suck Pressure Y Asis Y Axis Y Anis Y Asis Mir H blin 0929 mm Min eee MH Mir 299 cmH2o Max 4g Max agoq mm Max 1126 Max 720 cmH20 iw Auto Scale W Auto Scale W Auto Scale W Auto Scale EMG 005 A EMG OOS L EMG OOF EMG OOI L Y Anis Y ANis Y Anis Y Asis klin 400 0 UY Mini Foo uy Min 500 0 uy Mir 500 0 uy Max 4o00 uv Max Boog ou Max Boog ou Max Boog uM iw Auto Scale W Auto Scale W Auto Scale W Auto Scale 2 Axis Period 50 a Figure 10 Graph setting window 2 2 2 Screen out bad blocks In this Blocks Analysis window Figure 11 the whole data set shown in the last window NeoFlex window is divided into small windows with 150ms length for each window User could press Previous Next First and Last buttons to view the whole data set When encountering a bad data block that is not desired in final signal average analysis user could check Tag as bad block to remove it from signal averaging that will show in the third window Signal Ave
16. 0 10 0 20 0 0 16000 17000 18000 19000 20000 W_ _ AE lt s M el 1000 2000 3000 4000 5000 6000 7000 s000 9000 10000 11000 12000 13000 14000 15000 Figure 16 Burst extracted signal for NNS Param and STI This format is necessary because it is a more accurate representation of the data which is needed for the peak detection analysis and for future applications Click the Stop and Save button to automatically save the extracted file to the directory Figure 17 shows the file path and name for this data Saved file path and name for Burst Extraction qE Program Files NeoSuck Data BabySuck_BurstE traction t t Figure 17 File path and name Notice e If user needs to re select another file after clicking the Step 2 Start Analyzing a Click Finish Burst Selection in Step 3 b Click the DO NOT SAVE button This could stop current trial and enable user to start a new one e If Burst extract was not stopped as explained with the previous method and the program crashed a Click the Aborted button in the status bar stop sign as in Figure 3 b Resume NeoSuck exe by clicking the arrow button as in Figure 4 Then another trial for Burst Extract could be started 15 NeoSuck RT CNL U of Kansas 2 4 NNS Param This panel has the function of peak detection burst detection and histogram generation 2 4 1 Load file This panel should load the file that w
17. Bound hoo o s Intervals kooo Intervals Intervals jo Y Axis Y Axis Y Axis Min 0 Min an Min 0 Max 4 Max 30 Max a V Auto Scale MV Auto Scale IV Auto Scale VW coe Figure 16 Graph and Histogram setting dialogue The Suck Pr Analysis setting Figure 17 defines parameters for peak selection and suck pressure threshold value for recording etc Settings Suck Pressure Analysis Derivative Peak Settings Threshold 0 3 Peak Valid Range 200 Peak Period 400 Suck Pressure Threshold 2 0 cm H20 Lowpass Cutoff Frequency fg Hz Skip Samples Display m oms Figure 17 Suck Pr Analysis setting 17 NeoSuck RT CNL U of Kansas 3 References Estep M Barlow S M Stumm S Fees M Finan D Seibel L Poore M Cannon S 2005 Non nutritive Suck Burst Parametrics in Preterm Infants Society for Pediatric Research 2141 Seibel L Barlow S M Vantipalli R Finan D Urish M Carlson J 2005 Spectral Dynamics of Non Nutritive Suck in Preterm Infants Society for Pediatric Research 2139 Stumm S Barlow S M Vantipalli R Finan D Estep M Seibel L Urish M amp Fees M 2005 Amplitude Burst Dynamics of the Non Nutritive Suck in Preterm Infants Society for Pediatric Research 2632 Stumm S Barlow SM Estep M Lee J Cannon S Gagnon K Carlson J Finan D 2008 The relation between respiratory distress syndrome and the fine structure of the non nu
18. D Doe 1 25 2006 Second of Twin Not Hispanic or Latin White 456123 e Doe 1 26 2006 First of twins Unknown More than on 159753 F Doe 2 1 2006 First of twins Hispanic or Latino White 852963 G Doe 1424 2006 First of twins Not Hispanic or Latin Native Hawai 456258 H Doe 1425 2006 First of twins Unknown Asian 478596 K Doe 2 3 2006 Singleton Hispanic or Latino American Indi 854796 L Doe 2 26 2006 First of twins Hispanic or Latino American Indi 852365 J Doe 2 28 2006 First of twins Hispanic or Latino American Indi 753962 Doe 3 1 2006 Singleton Hispanic or Latino American Indi 158742 S Doe 3 24 2006 Singleton Hispanic or Latino American Indi 253698 X Doe 3 5 2006 First of twins Unknown Black or Afric 852741 Z Doe 3 6 2006 Singleton Hispanic or Latino American Indi 896542 M Doe 3 28 2006 Singleton Hispanic or Latino Asian 158752 Doe 4 1 2006 Singleton Hispanic or Latino American Indi 857963 P Doe American Indi 956327 Doe Q fm Sort Patient Info Sees ino American Indi 785596 R Doe ino American Indi 785962 T Doe Latin American Indi v b Sort By Modify Refresh Sort Order ASC Cancel Record Position 1 78 Username Location kumc 6 14 2006 3 43 PM Figure 9 Sort Patient Info window is displayed 15 NICU Database CNL U of Kansas 3 Reference NIH R01 DC 03311 05 SM Barlow Principal Investigator Sensorimotor Control and Development of the Human Orofacial System 2 120 000 total costs 01 15 02 12 31 06
19. For example 17 NeoFlex RT CNL U of Kansas our default protocols are The trigger level set in suck pressure is 5 cmH20 which will be repeated 30 times After that the trigger level will automatically rise to 10 cmH20 and the experiment will be repeated for another 30 times under this trigger level User could define new protocol in this window to change the trigger level and repetition time Protocol File Editor Untitled File Protocol SuckPr Repetition Add Modify Delete Figure 14 Protocol Editor dialog All the protocols will be saved in Protocol file pf which is an XML file that contains test protocol a sequence of test conditions with repetition count for running the experiment A protocol file can be created or modified using the Protocol Editor as we detailed earlier User selects the protocol file to be used for the experiment during a File New operation With correct NeoFlex RT installation user can locate the protocol file under directory C Program Files Neuro Logic NeoFlex RT Protocol Here is a sample file lt xml version 1 0 standalone yes gt lt ProtocolDataSet gt lt Prococol gt lt SuckPr gt 5 lt SuckPr gt lt Repetition gt 30 lt Repetition gt lt PrOtLocol gt lt PPrOCOCO gt lt SuckPr gt 10 lt SuckPr gt lt Repetition gt 30 lt Repetition gt lt Protocol gt lt Provcocol gt lt SuckPr gt 15 lt SuckPr gt lt Repetition gt 30 lt Repetition gt lt
20. Normally it works with BNC 2090 interface Figure 3 To work with the BNC 2110 Figure 4 please make the following change Figure 4 BNC 2110 For BNC 2110 To work with BNC 2110 the terminal configuration has to be in differential mode Terminal configuration can be set to differential mode by modifying the parameter value of terminalConfig to 1 under lt systemConstants gt section in the file NeoSuckRT exe config located in C Program Files Neuro Logic NeoSuck RT Here is the line from NeoSuckRT exe config that shows this parameter value lt add key terminalConfig value 1 gt terminalConfig can have the following values NeoSuck RT CNL U of Kansas value 1 for Differential value 2 for Referenced single ended value 3 for Nonreferenced single ended By default the file NeoSuckRT exe config contains value 3 for terminalConfig This setting 1s used for the BNC 2090 signal interface Here is the line from NeoSuckRT exe config that shows this parameter value lt add key terminalConfig value 3 gt NeoSuck RT CNL U of Kansas 2 User manual When opening the NeoSuck RT a blank window is shown as follows Figure 5 User could apply this set of software and hardware to record new data or analysis of existing data ES NeoSuck RT fe fe x Fie Settings Tools Help MOTOR PULSE fv EMG OOS F uv 10 0 10 0 oO 5 5 0 4 5 5 04 2 z 10 0
21. ONG SUT cal ula ON uiis eiA T E 2r Zed OXY GEN PULSE priii E 30 3 APPENDIX koan E T ren tee Try eran err re 31 3 1 UR STAC Tiaa aaa A E tas iatee Ue tencan atlas 31 NeoSuck RT CNL U of Kansas Sls Wal deS Durst EXPACHON UO soxceesioics asatece EEN eee See 3 3 1 2 Problems with the Original Suck Signal Figure 35 icccccccccccsssssccccccccccseesscccceeeassessseceeeeeaaaes 31 R OE U A cea Gas ahs sae See See 3 DLA SOMMMONETIOW GO We PEATE 11 ieiet ei E A e AN A 32 3 2 STEMATH FUNDAMENTALS criari n A E E A 37 Sed Whati SST seai a a e A AE A oeepeneuc daae 37 P MDAA EC ON e EEEE EA E EAE AO AE AAN 37 Iar DGG normal aN OT airaa a A E T RE 37 La PDanna STE CO CUI GI ONiseiririoiiire e t E A EAT LE 38 REFERENCE siin e a 41 NeoSuck RT CNL U of Kansas 1 NeoSuck RT Setup NeoSuck RT real time 1s used to record and analyze data associated with the non nutritive suck NNS including signals related to Suck Pressure and Oxygen which are recorded from 2 channels of NI DAQ 6052E PCI multifunction board installed in the acquisition PC with 100Hz sampling rate on 16 bits vertical resolution We currently apply this software and hardware system to test and re analyze the suck pressure and signals from other channels for premature and term baby The analysis tools in NeoSuck RT include data acquisition burst extraction peak detection histogram analysis FFT analysis STI analysis etc 1 1 Hardware The Hardware for this testing equip
22. Rhythmical C Yes Tongue N1 Rate of 1 sec C Yes CNL U of Kansas Save Cancel NEONATAL OROMOTOR ASSESSMENT SCALE Tongue Dysfunctional r Protrusion with down and Yes out move Tongue Dysfunctional Absence of movement C Yes Save Cancel 7 page Figure 7 Page Tabs 1 through 7 Permit the user to enter or modify patient s clinical information 13 NICU Database CNL U of Kansas 2 4 Other functions in NICU database In Patient Info table by clicking Patient Patient Info selecting Search button will allow the user to search a patient in the database by Patient Number Medical Record Number Last Name or First Name refer Figure 8 Pt No Pt Medical Record No Pt First Name Pt Last Name BirthD ate Birth Order Ethnicity Race a K59 123456 Joe Doe 1424 2006 Singleton Hispanic or Latino American Indi K60 234567 A Doe 2 24 2006 Singleton Hispanic or Latino American Indi K61 456785 B Doe 1422 2006 Singleton Hispanic or Latino American Indi K62 458796 C Doe 1 2 2006 Second of Twin Not Hispanic or Latin Native Hawai K63 456789 D Doe 1425 2006 Second of Twin Not Hispanic or Latin White K64 456123 e Doe 1426 2006 First of twins Unknown More than on K65 159753 F Doe 2 1 2006 First of twins Hispanic or Latino White K66 852963 G Doe 1424 2006 First of twins Not Hispanic or Latin Native Hawai K67 456258 H Doe 142572006 First of twins Unknown Asian K68 478596 K Doe 2 3 2006 Singleton Hispanic or Latino American Indi
23. Yes No Delete ABR Result Pass Oxymetry Swallow study Dae C Yes Modify No Delete Save Cancel 5 page 12 Patient Configuration Reports User NEONATAL OROMOTOR ASSESSMENT SCALE Jaw N1 Continuous degree jaw depression Jaw N1 Rhymth excursions Jaw N1 Spontaneous jaw excursions Jaw N1 Movements of l sec Jaw N1 Sufficient closure for expression Jaw Disorganized Inconsistent depression Jaw Disorganized Arrythmic Jaw Disorganized Difficult initiation Jaw Disorganized Immature suck Patient Configuration Reports User NEONATAL OROMOTOR ASSESSMENT SCALE Tongue N1 E fficientSuck Tongue Disorganized protrusion Tongue Disorganized Arnhythmic Tongue Disorganized Unable to sustain 2 min Tongue Disorganized incoord suck swallow breathe Tongue Dysfunctional flattened or flaccid NICU Database Help Help C Yes C Yes C Yes C Yes C Yes Tongue Dysfunctional Retracted Tongue Dysfunctional Asymmetrical lateral deviation NEONATAL OROMOTOR ASSESSMENT SCALE Jaw Dysfunction wide excursion interrupting seal C Yes Jaw Dysfunction Minimal excursions clenching C Yes Jaw Dysfunction Lat deviation C Yes Jaw Dysfunction Absence of move C Yes Jaw Dysfunction Lack of rate change NNS to NS Yes Tongue N1 Cupped C Yes ely retr nl Tongue N1 Ext elv retr n C Yes Tongue N1
24. ZOOO 83ooo 9000 1 OOOO Figure 46 Normalized Data plot 3 2 4 SD and STI calculation Suppose we are given a population x xy of values The arithmetic mean of this population is defined as Eq 1 _ 1X Ny X Xy x x _ _ 1 T r 1 and the standard deviation SD of this population 1s defined as Eq 2 In our case as shown in Figure 46 there are eight trajectories thus an N 8 We will calculate the SD based on eight normalized pressure values for each sample point 10000 SD values will be calculated for the normalized data Figure 47 shows the original data normalized data and standard deviation 38 NeoSuck RT CNL U of Kansas The STI value represents the sum of standard deviations In order to keep STI simple and neat we will add SD every 100 samples which is named as STI step instead of adding all 10000 SDs Original Difference Signals Suck pressure cmH 0 Original sample number Linear Time and Amplitude Normalized Signals Normalized suck pressure lt o 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 Standard Deviation STl 47 602 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 Relative sample number Figure 47 STI Calculations for NNS skilled preterm infant Figure 47 is the STI analysis result for a preterm infant with good suck pressure patterning with an STI 47 60 39 NeoSuck RT CNL U of Kansas Original Difference Signals a
25. are sampled including the output of the DAC intraluminal nipple pressure and jaw kinematics or other desired channel 1 e oxygen saturation pressure 1 1 Hardware The Hardware for NTrainer includes gt A PC with Windows XP and 1 GB RAM gt NI DAQ 6052E PCI multifunction I O card gt Position Servo controller BioCom Electronics LLC gt Bridge amplifier s 2 channels gt Pneumatic pressure cylinder actuator gt Linear Motor gt Pressure transducer The hardware listed above is shown in Figure Most of the hardware linear motor pneumatic pressure actuator power amplifier pressure transducer etc is concealed in a shielded cabinet within a wheeled cart which 1s position cribside in the NICU A detailed view of the linear motor actuator and air cylinder with pressure lines pressure transducer and Luer fittings 1s given in Figure 2 CNL U of Kansas NTrainer RT RT hardware set up mer NTrai 1 Figure NTrainer RT CNL U of Kansas Figure 2 Linear Motor pneumatic actuator assembly 1 2 Software NTrainer RT is coded by C 5 448 coding lines This software performs the function of controlling the start and stop of the synthesized patterned orofacial somatosensory stimulation generating the impulse with NNS pattern and updating an integrated database of patient parameters including patient ID birth age age at test and a cumulative history of stimulation parameters stimulus train ch
26. becomes solid i E S L h Program will automatically stop when the desired Repeat is reached STOP Before Repeat is Finished 6 User may stop the program at any time by clicking this button Non nutritive suck spatiotemporal index in preterm infants NNS STI J Wang amp S M Barlow CNL Technical Research Report 2006 1 1 12 University of Kansas Lawrence Kansas USA Reprint available at www ku edu cnl NNS STI CNL U of Kansas Table of Contents 1 Part I Background Math Fundamentals cccccccccccccccccceessseeccceeecaeaeeseesceseesueseseeeees 3 1 1 W a E e N E S 3 Le SALAS elecho naan T E E O 3 1 3 Data OL IAIN ALON senn E E 3 1 4 SP ane S LT Calc lation menrs ae a aeudansienazlas shediga ieee tieeibealecntreetieatest 4 2 Part WU Ser manud eosa E Caccuaatiacenaceedeccaadtetonanctacreseentiacenncd 7 ZA SEPO Data prepara OM cceeieecceest tanec cei eliinteeieri ccna amtoteneieeeaenscad 7 22 DIC Op AP ea CCC OMe de see ii ascireaaisterhdenes a eaneeeuseea ome aet cued 7 2 3 oP ZB UES t CeCe 1G Mic ote cele erase sented etuiaaaenctet tans aanshee netuyas Maadatamncnceeeeeiselies 8 2 4 Step 3 Peak shoulder detection and deletion cccccccccccccsssseeeeeececeaeeeeeeeeeeeeeeaaenees 8 2 9 Step 4 select D rst and peaks TOK SDi co vesdasardtebiledaiecentcesneiaendecctetems 8 2 6 Step 50D ands TE Calculations sia a E E 10 3 REENE asia ripe eee ee A 12 NNS STI CNL U of Kansas 1 Part Background
27. by both of them 20 NTrainer RT CNL U of Kansas 2 3 Review an NTrainer data file using NTrainer RT Select Patient gt Open A Patient List will appear and the user can select the patient s file to review Figure 17 CE NTrainer RT Patient Settings Tools Help STIMULUS PULSE fv 26 5 Patient List Select a Patient PatientId First Name Last Name BithDate No of Visits Cumulative Cycles Good Sucker 5 5 2005 8 534 1 4 2006 0 1 26 2006 0 Patient Info Patient ID Cancel Patient First Name Age at Test Patient Last Name DOB OE mm dd ypyy N Trainer Cycles for Current Visit Cumulative N Trainer Cycles for All Visits 156 User Name Lawrence Figure 17 Main window program of Patient List 21 NTrainer RT CNL U of Kansas S NTrainer RT Patient Settings Tools Help STIMULUS PULSE fv SUCK PRESSURE em H20 l 37 5 Time fs T 37 5 Time s Visit Info Visit H Patient ID Test Date 678 206 mm dd yyyy Patient First Name Age at Test 158 days Patient Last Name DOB ia 2006 ram dd yyyy Patient Info N Trainer Cycles for Current Visit 90 Cumulative N Trainer Cycles for All Visits 156 User Name Lawrence Figure 18 Main window of a patient s NTrainer data file After a patient file is selected all of the patient s personal parameters and the corresponding NTrainer data file will be s
28. database Selection of New option will allow the user to insert the patient s information such as Patient ID Patient First Name Patient Last Name and DOB see Figure 7 10 NTrainer RT CNL U of Kansas S NTrainer RT Elsie Tools Help DAQ E ecw STIMULUS PULSE Graph Change User l i 4 5 SUCK PRESSURE em H20 Time fs 5 Time s Patient Info Visit Info Visit H a Patient ID Test Date Sa ram dd pyyy Patient First Name Age at Test 1 Se days Patient Last Name poe Grenadin N Trainer Cycles for Current Visit Cumulative N Trainer Cycles for All Visits User Name Topeka Figure 8 Main program window with the Settings menu option list displayed The Settings menu Figure 8 provides user access to DAQ Stimulus Pulse Graph and Change User User may adjust the parameters within each submenu s dialog box 11 NTrainer RT CNL U of Kansas The DAQ submenu allows the user to select the Device Type and map analog to digital channel ADC assignments for Stimulus Pulse Suck Pressure and Jaw as shown in Figure 9 NTrainer RT STIMULUS PULSE fv DAQ Settings DAG Device PCI B052E Al jao Sample Rate f ooo samples s Buffer Size 5 Channels Stimulus Pulse Suck Pressure Jaw Al Channel facho xl Al Channe fachi xl Al Channe acH2 xl Physical Unit z W Physical Unit lz cm H20 Physical Unit Eoo u Voltage z Y Voltage fos
29. ey Neonate Sensorimotor Oral Development Database CNL University of Kansas 1 RO1 DCO3311 02 Patient Information MPR Patient Configuration Reports User Help Pt No ga ETELE Pt Last Name BirthD ate Birth Order Ethnicity Race a ee ee eee ee Record Position Username Location CAPS 67672006 3 39 PH Figure 1 Main program window of the NICU Database user interface NICU Database CNL U of Kansas 2 1 Overview of menu bar options Selecting the Patient allows the user with access to Patient Info Patient Test Info submenus see Figure 2 These two submenus enable user to view add and modify pertinent Patient information in two tables with different functionality ey Neonate Sensorimotor Oral Development Database CNL University of Kansas 1 RO1 DCO3311 02 Patient Information EES AGA Configuration Reports User Help X Patient Info Patient Test Info Record No Pt First Name Pt Last Name BirthD ate Birth Order Ethnicity Race Exit Ctrl x ee ee ee ee ee ee Record Position Username Location 6 6 2006 3 46 PH Figure 2 Main program window with the Patient menu options list displayed NICU Database CNL U of Kansas The submenu designated as Configuration contains Status Info Birth Order Info Ethnicity Info Race Info Delivery Type Patient Type Feeding Mode Info Medication Info ROP Stage Info Oxymetry Swallow Results Info Oxymetry Swallow Indications Info Neurologic Exam Inf
30. in Figure 3 If the distance from a peak to its shoulder is less than Sample number from peak to its shoulder the shoulder will be deleted For example Sampling Frequency 3000H2 in the first plot in Figure 29 the input sample distance is set at 300 which shows the shoulder as a peak In the second plot the distance is increased to 900 and all the shoulders are deleted Important Note higher sampling frequency of the data leads to more peak shoulders For our current analyzed data with 100Hz sampling frequency peak shoulders are very rare 24 NeoSuck RT CNL U of Kansas FERES 40 0 30 0 20 0 10 0 aH 0 0 207387 4 225000 0 27500 0 30000 0 21106 0 29000 0 27500 0 30000 1 Figure 29 Peaks with Peak Shoulder left and Peaks with Peak Shoulder Deleted right 2 6 6 Step 3d Select Burst and Peaks for SD Up to now all the peaks without shoulders and bursts should both be recognized exactly as what user want them to be Important SD processing is based on a set of bursts with the same peak number In order to proceed with this step user must decide how many peaks and how many bursts user want to compare across this file Because NNS STI is based on the cumulative sum of the standard deviations across bursts the more peaks and the more bursts user choose the higher the STI value will be In order to make comparisons across files NNS STI calculations must be based on the same peak and burst number
31. l Y Voltage z Y A D Min is A D Min o ES A D Min jo A D Max 5 x A D Max fo 8 Orst v A D Max fo ES M Patient Info Patient ID Cancel Patient First Name Patient Last Name DOB EEE iddia N Trainer Cycles for Current Visit Cumulative N Trainer Cycles for All Visits User Name Lawrence Figure 9 Main program window with the DAQ Settings submenu option list displayed 12 NTrainer RT CNL U of Kansas The Stimulus Pulse submenu contains scaling options for Pulse Rate Hz Pulse Train length number of pulses Pulse Amplitude Volts Pause Period seconds and Train Rate Hz as shown below Figure 10 r A 15 EG STIMULUS PULSE fv SUCK PRESSURE cm H20 Stimulus Pulse Attributes l 5 M Stimulus Pulse Attributes Time s 7 Pulse Rate Hz Pulse Train 6 amp 6 Pulse Amplitude 06 y Pause Period 2 Train Rate 0 1875 Hz Time s oms M Patient Info Visit Visit H Patient ID Test Date mmdd yyyy Patient First Name Age at Test days Patient Last Name DOE i iddi N Trainer Cycles for Current Visit Cumulative N Trainer Cycles for All Visits User Name Lawrence Figure 10 Main window program of Stimulus Pulse Attributes 13 NTrainer RT CNL U of Kansas The Graph Settings submenu provides user access to Y axis and X axis display settings for Stimulus Pulse Intraluminal Suck Pressure and Jaw as shown in Fi
32. may be the result of noise 2 3 1 Step 1 Loading file e If you want to burst extract the data you just collected DAQ choose select file from data acquisition which will automatically load the file you created in the DAQ panel as shown in Figure 9 Select File From Data Acquisiton O Select new file Load File For Burst Extraction T OH Program Files NeoSuck Data Baby Suck daka txt ja Figure 9 Load file from DAQ for Burst Extraction However if you want to analyze another file you should click Select new file and then the Browse button to choose the desired file e Burst Extract without collecting data in first panel a pop up File Browse window Figure 10 will enable user to choose new files 10 NeoSuck RT CNL U of Kansas Look in GEE v gt fa 4 IE BabySuck_200_2000_PowerSpectrum 2 terry_NormData i 4 5 BabySuck_BurstExtraction E terry_OrgData My Recent 2 BabySuck_BurstPauseHist E terry_PeakAmpHist Documents IE BabySuck_data f terry_PeakPeriodHist BabySuck_NormData E terry_PeakSummary BabySuck_OrgData terry_Summary BabySuck_PeakAmpHist BabySuck_PeakPeriodHist BabySuck_PeakSummary BabySuck_PowerSpectrum BabySuck_Summary terry_167_2000_PowerSpectrum terry_BurstExtraction terry_BurstPauseHist E E 14 Tw aT My Documents 8 MAMAMA terry_data My Computer GJ Fiene x D My Network Files of type AI Files v
33. mm EMG OOS L uv ACH1 ACH5 l LOAD CELL N EMG OOI R uv 20 ACH2 ACHE 1 1 ner ler oe en see I I 1 15 es ml 5 1 1 I 1 1 SUCK PRESSURE cm H20 EMG OOI L uv 20 ACH3 ACH I l f l 0 20 40 60 80 100 120 140 0 20 40 60 80 100 120 140 ms ms Cursors Time Motor Pulse LVDT LoadCell Suck Pressu EMG OOSR EMG OOSL EMG OOIR EMG OOIL Cursor Blue 35 8 0 0 5 134 0 279 1 037 9 061 13 44 16 337 15 881 Cursor Red 113 0 0 0 6 222 0 276 17 722 10 525 12 975 22 552 20 483 Difference 47 2 0 0 0 088 0 003 16 685 1 464 0 465 6 215 4 602 Save File Name t49_101405_reflex dat N 66 Bad Blocks List 4 Figure 12 Signal average window The other function to assist user to look through the average signal in more details is the blue and red cursor User could drag or move them by up down and right left keys on keyboard the blue and red vertical lines in any graph and the cursor table in left lower corner will show the exact cursor position and the distance between two cursors in which user are interested 16 2 3 Other settings NeoFlex RT CNL U of Kansas The settings in this part shouldn t be changed once they are set at the beginning of data recording 2 3 1 DAQ board set up From the main NeoFlex window data acquisition criterion could be set in Setting gt DAQ as shown in F
34. of each burst as marked by green circle Figure 39 What does Peak Detector do a Locate burst ending points Green points and burst starting points White points Fine tuning Manually finding missed peaks and relocating the burst starting point Each burst detected in Coarse tuning is displayed individually in Figure 6 to enable the user to identify if the starting point is correctly recognized If the starting point is not correctly 33 NeoSuck RT CNL U of Kansas recognized the user could utilize the Shifting right or Shifting left button to tune until desired bursts are found For example one of the bursts detected in Coarse tuning is shown in Figure 41 after Shift right is pressed the first peak is included in current burst as displayed in Figure 42 Plot 0 AN Burst Graph Original Data Tam 30 Amplitude MN m mo h m on I l l 1 l l 1 l 1 l l 1 l 1 l l 1 l 1 l l 1 l 300 350 400 450 500 550 600 650 700 750 800 850 9300 950 1000 1050 1100 1150 1200 1250 1300 1350 I 1 1 i 0 50 100 150 200 250 Time Figure 41 Identified burst by Coarse Tuning whose first peak are undetected Pto B Burst Graph Original Data 2515 Amplitude Nm o I 1 1 I I l l 1 1 1 l I l l l 1 l 1 1 l l l l 350 400 450 500 550 600 650 700 750 800 850 900 950 1000 1050 1100 1150 1200 1250 1300 1350 1400 I I 1 1 0 s50 100 150 200 250 300 Time Figure 42 Fi
35. of peak values Cursor Drag 4 gt Define J 8 The switch button allows the user to choose the cursor movement mode Drag allows the user to click and drag the threshold line Define allows the user to input a specific threshold value Peaks found indicates the number of peaks identified based on the user s previously defined peak width and threshold 2 4 3 Burst detection and Histogram generation Burst Detection for Histogram Generation 8 f Distance between bursts Jo aries ok biras E Peak Amplitude Histogram Peak Amplitude Bins A 10 Peak Period Histogram Peak Period Bins Jio _ Burst Pause Histogram Burst Pause Bins 5 10 20 60 mS FS i 2 ae 2 10 30 1 2 H oO 20 ea 5 t I 0 a le 1 0 I 1 es ee 1 5 15 25 0 6 0 8 1 ee 1 4 1 6 3 4 5 6 7 8 7 Amplitude cmH20 7 a Peak Period s Burst Pause s Peak Amplitude Mean 12 93 Peak Amplitude SD 13 85 Peak Period SD 10 10 Peak Period Mean 0 69 Burst Pause Mean 4 73 Burst Pause SD 1 25 Burst pause histogram Saved file path and name 4E Program Files NeoSuck Data BabySuck_BurstPauseHist txt Save histogram Peak amplitude histogram Peak period histogram 4E Program Files NeoSuck Data BabySuck_PeakAmpHist txt 4 C Program Files NeoSuck Data BabySuck_PeakPeriodHist txt Figure 19 Burst extraction and histogram processing 17 NeoSuck RT CNL U of Kansas e Burst detection Distance between bursts the user def
36. 0 gt Coulbourn waveform discriminator comparator gt Bridge Amplifier gt AC PreAmplifiers up to 4 channels gt DVRT gt Force servo actuator Neuro Logic LLC amp BioCommunication Electronics LLC gt Sensors pressure sensor and EMG signal sensors All the hardware listed above is set up according to Figure 1 Baby s suck pressure and EMG signals sampled proximal to the SOOTHIE silicone pacifier nipple will be input to NeoFlex and then recorded NeoFlex RT CNL U of Kansas Load Cells Ai2 DVRT Actuator by LoadCell Microstrain Linear Motor For Force N Luer Probe 10KHz H2W Tech Entran gf LVDT Ai1 DVRT Input To Controller Current F l PAGU Seinen mpulse AA ts 501 Motor Pulse Controller LoadCell AiO Input to geController DAC1 Ref j 4v Ca er and DA PF10 TTL Coulbourn Suck Pressure Comparator Ai3 N Bridge Suck A re Amplifier Suck Pressure Ai3 D EMG LPF 50Hz aula ict Ai4 7 EMG Ai4 Grass AC EMG Ai4 7 PreAmplifiers 4 GND is connected to baby leg Dec 1 2005 by Joan Figure 1 Hardware setup for NeoFlex RT NeoFlex RT CNL U of Kansas 1 2 Software This software is coded by C 13 195 coding lines according to the structure showed in Figure 2 It performs functions like recording creating new data analyzing existing data plotting all 8 channels data to users
37. 16
38. 2141 5 Seibel L Barlow S M Vantipalli R Finan D Urish M Carlson J 2005 Spectral Dynamics of Non Nutritive Suck in Preterm Infants Society for Pediatric Research 2139 6 Stumm S Barlow S M Vantipalli R Finan D Estep M Seibel L Urish M amp Fees M 2005 Amplitude Burst Dynamics of the Non Nutritive Suck in Preterm Infants Society for Pediatric Research 2632 7 Zimmerman E Barlow S M Seibel L Poore M Stumm S Estep M Chu S Fees M Urish M Gagnon K Cannon S Carlson J 2006 Pacifier stiffness alters the dynamics of the suck central pattern generator Society Pediatric Research 5571 3593 8 Stumm S Barlow S Estep M Lee J Cannon S Gagnon K Carlson J Finan D 2008 The relation between respiratory distress syndrome and the fine structure of the non nutritive suck in preterm infants J Neonatal Nursing in press 9 Estep M Barlow SM Vantipalli R Lee j Finan D 2008 Non nutritive Suck Burst Parametrics in Preterm Infants with RDS and oral feeding complications J Neonate Nursing accepted 10 Zimmerman E Barlow SM Lee J 2008 Pacifier stiffness alters the dynamics of the suck central pattern generator J Neonatal Nursing in press 11 Popescu EA Popescu M Wang J Barlow SM Gustafson KM 2007 Non nutritive sucking recorded in utero via fetal magnetography Physiological Measurement in review 41 Somatosensory entrainment of
39. 3500 4000 4500 5000 5500 6000 6500 7o00 7500 Nurm Amplitude l l l l l 0 1000 2000 3000 4000 5000 6000 7000 S000 S000 10000 Num Figure 10 Plots of Original Data and Normalized data Remember we said in Part I SD and STI calculation that in order to keep STI simple and neat we will just add 100 of 10000 SD values as in Figure 11 INPUT SD Step is for picking one SD value for every SD Step samples to add to STI value Normally this value does need any tuning if the normalized sample points are 10000 10 NNS STI CNL U of Kansas SD Waveform Graph Plot 0 aa l l l l l l l l l l l l 0 1000 2000 3000 4000 5000 6000 7000 e000 9000 10000 11000 Number of samples Figure 11 SD Plot with 100 Sample Points OUTPUT STI Value showed current STI value after the last burst selection Normalized Data File Path and Original Data File Path set the path and file name for data file in Figure 10 Those data files are used for pretty processing in MATLAB as well as keeping a data file record for your STI processing STOP and SAVE DATA Figure 12 STOP button The STOP and SAVE DATA button as in Figure 12 is for stopping the whole program and saving data 11 NNS STI CNL U of Kansas 3 References 1 Smith A and Zelaznik H N 2004 Development of functional synergies for speech motor coordination in childhood and adolescence Dev Psyc
40. 39 39 08 39 42 08 39 45 08 39 48 08 39 51 08 39 54 08 39 57 08 40 00 08 40 03 08 40 06 08 40 09 08 40 12 08 40 15 08 40 18 08 40 21 08 40 24 08 40 27 08 40 30 o8 40 33 Ili in A a lh wl i il WI Mil Figure 6 Suck Pressure channel e Time Elaps shows the time for DAQ excluding pause time Selected file path and name for Data Acquisition Figure 7 shown below shows the final path and file name for the acquired data It displays the final path and name as user sets before data acquisition File path to save data amp Input baby name Selected file path and name for Data Acquisition C Program Files NeoSuck Data BabySuck_data txt Figure 7 Generated file path and name for acquired data NeoSuck RT CNL U of Kansas 2 2 2 DAQ Setting This tab Figure 8 for advanced users enables user to set e Channels to collect data e Minimum and maximum of collected data e Sampling rate e Scale factors for each channel Illustrates to the user e Data length in sample number e DAQ time between pauses Devi ail Aao C r 6 2130 ar a Figure 8 DAQ Setting Do not change parameters in this panel if you are not familiar with the theory of DAQ DSP NeoSuck RT CNL U of Kansas 2 3 Burst Extraction This panel is used for locating the start and end points of each burst Peak removal is also a feature of this panel and is used to remove peaks between bursts that
41. 52 PM Then user can press the Start to begin the recording or Pause button to halt it NeoSuck RT CNL U of Kansas 2 2 Analyze existing data by NeoSuck RT Based on the data collected described in section 2 1 Record new data by NeoSuck user can systematically apply analysis tools in NeoSuck RT to investigate more detailed properties of the human infant NNS 2 2 1 Basic analysis Click File gt Open an open file dialogue will pop up Figure 7 to enable the user to designate the name and path of the target data file ma NeoSuck RT C Program Files Neuro Logic NeoSuck RTWata test Baby_A_010206 dat MOTOR PULSE EMG OOS F u Look in gt test a e ey EJ 1 i Baby_A_010206 dat 2 l TestingA_061206 dat My Recent Documents Desktop SUAE My Documents 33 My Computer My Network File name Testing4_061206 dat y f z Places SUCK PRESSURE e Files of type Data files dat pro 0 4 Time s Observations NeoSuck RT Beep On Off V Real Time Blocks Sampled 0 amp Date Created 6 15 2006 3 50 58 PM Figure 7 Opening an existing data file Figure 8 shows the initial NeoSuck RT data analysis window The three histogram windows on the right display the histogram plots for NNS Amplitude Pause Periods and NNS Burst Periods for each block that the user has selected If the user is only interested in the analysis 10 NeoSuck RT CN
42. A D Max i ew A D Max fio y EMG 005 R EMG OOS L EMG OOI R EMG OOI L l Channel ACH4 l Channel ACHS5 Al Channel ACHE v Al Channel ACH v Physical Unit ho W Physical Unit hoo o W Physical Unit ho W Physical Unit ho W Voltage Ze Y Voltage jac i y Voltage 20 Y Voltage Zz Y A D Min mm y A D Min lio 0C t 4N A D Min mw j A D Min joo y A D Max woo PM A D Max Aon A D Max Dy A D Max my Block Sample Rate 3000 Sample Time 30 No of Samples 90000 Down Sample 100 Cancel Figure 15 DAQ board settings The Graph and Histogram settings Figure 16 include the dialog windows for lower and upper limit selection according to user s preference 16 NeoSuck RT CNL U of Kansas Settings Graph Graph Settings Motor Pulse LYDT Load Cell Suck Pressure Y Axis Y Axis Y Axis Y Axis Min Y Min 8 929 mm Min 1 033 N Min 1 0 cmH20 Max 40 Max 8 929 mm Max 4 033 N Max 5o 0 cm H20 V Auto Scale MV Auto Scale MV Auto Scale MV Auto Scale EMG 00S R EMG 005 L EMG OOI R EMG OOI L Y Axis Y Axis Y Axis Y Axis Min 400 0 uv Min 500 0 uv Min 500 0 uv Min 500 0 uv Max 400 0 u Max 500 0 uv Max 500 0 uv Max 500 0 Y V Auto Scale IV Auto Scale M Auto Scale MV Auto Scale coe Settings Histogram Amplitude Histogram m Pause Histogram Burst Period Histogram Lower Bound bo cm H20 Lower Bound s Lower Bound bo s Upper Bound 1250 cm H20 Upper Bound 0 0 s Upper
43. C Program Files NeoSuck Data BabySuck_PeakSummary txt Figure 20 Summary 18 NeoSuck RT CNL U of Kansas 2 5 NNS FFT This panel analyzes the frequency property of loaded data 2 5 1 Load file User could choose any file for FFT analysis as shown in Figure 21 but usually we do FFT analysis for original signal Load File for FFT C Program Files NeoSuck Data BabySuck_data txt ie l Figure 21 Loaded for NNS FFT Sampling frequency is a parameter used for analysis hereafter Please input the corresponding sampling frequency for your analyzed data file Sampling frequency Hz 100 2 5 2 FFT analysis User could choose to generate FFT for the entire file or for part of the data e FFT for entire data switch the button in Figure 22 to All the spectrum will give you the frequency description Figure 23 of the whole data set Select All Range Figure 22 Switch button for data selection 19 NeoSuck RT CNL U of Kansas 16 00 14 00 12 00 10 00 4 00 6 00 4 00 2 00 0 00 l l l l l l O20 1 00 2 00 3 00 4 00 5 00 6 00 7 00 o 00 9 00 Frequency Hz aawl e amp Figure 23 Power spectrum for selected data e FFT for part of data when part of data is desired for FFT switch the button to Range as shown in Figure 2 The blue and yellow cursors in plot shown in Figure 4 define the starting point and ending point for part of data Cursor Control u
44. Hz TN T ET ysical Channe pm r a J Dev 1 ao0 iidel ians E Repeat Sample period Desired sample Minimum Yalue E J 100 oai 7550 Z55000 Sampling Rate Hz J 10 00 A fioo oo Maximum Yalue gt Current cycle Final sample J 1000 7 110 00 lt l 100 755000 l Figure 1 Parameter panel 2 1 Impulse Generator Parameters Yellow Frame Impulse selector Hz o J2 INPUT The User determines the pulse train rate by selecting 2 4 or 8 Hz Impulse Tab GRAPH PANEL depicts the shape of an individual impulse train waveform that is chosen from Impulse selector Figure 2 Impulse generator waveform Poo B 0 6 0 4 Amplitude v fam o N o M 1 1 5 te 0 6 1 1 1 l l 1 l 1 i i 1 1 i l 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500 7000 7500 8000 Time He Figure 2 Individual impulse train Repeat 100 INPUT The User determines the number of pulse train repetitions for one run Sample period 7550 OUTPUT Display of the length of an individual impulse train Example The sampling rate is 1000Hz the length of current impulse is also 7550 1000 7 55seconds Desired sample 755000 OUTPUT Displays how many samples would need to be generated for the User defined Repeat Example To generate 100 pulse trains the program needs to generate 755000 samples based on 1000 Hz sampling rate 755000 samples 755 seconds Curren
45. L U of Kansas of data from 8 channels displayed in the main window then these three histogram windows can be closed by the user EA NeoSuck RT C Program Files Neuro Logic NeoSuck RTWata K6_033004 K6_033004 dat MOTOR PULSE fv EMG 00S R u 200 0 0 0 ACHO ACH4 200 04 LYDT mm Cumulative Amplitude Histogram EJ ACHS T a z Observations LOAD CELL N 62 5 75 0 le cm H20 50 0 87 6 LOOsO JI2Z S 125 0 Amplitud ACH2 SUCK PRESSURE em H20 20 04 m 5 a 10 04 E l i l 0 0 10 0 20 0 30 Time s Block NeoSuek RT Block No Real Time First Block 3 6 Peak Count 29 Date Created 3 30 2004 1 53 32 PM Figure 8 Open NeoSuck RT window for data analysis The User can also call these windows by clicking desired window in Tools gt Histogram as shown in Figure 9 Furthermore the displayed histograms can also be exported to txt file by clicking Tools gt Generate Histogram Files for future analysis Those files will be saved under C Program Files Neuro Logic NeoSuck RT Data 11 NeoSuck RT CNL U of Kansas NeoSuck RT C Program Files Neuro Logic NeoSuck RTWata K6_033004 K6_033004 dat File Settings Tools Help Select Baseline Amplitude Bad Block List Pause Period EMG OOS R u Suck FFT Burst Period Create ASCII Blocks arae Generate Histogram Files o Generate Amp Period ASCII File lt 5 Con
46. NS STI CNL U of Kansas Normally INPUT Start to Ist Peak doesn t need any tuning It is defined to set the beginning point of the current burst After you input everything a selected burst with desired peak number will be plotted in Selected peaks in current burst as shown in Figure 8 Selected peaks in current burst Plot 0 a i l l l l l l l l 7e000 79000 agg 6 1000 a 2000 83000 84000 85000 S6000 Number of samples Figure 8 Current Data Selection OUPUT Selected length for comparison must be less than INPUT Number of points after normalization in step 5 in order to satisfy the mathematical requirement of normalization Important note e INPUT Select burst will be changed every time a new burst is selected while INPUT Select how many peaks you want to compare will keep unchanged for this whole STI processing e OUPUT Selected length for comparison lt INPUT Number of points after normalization in step 5 Figure 9 Next button for starting next burst selection NNS STI CNL U of Kansas 2 6 Step 5 SD and STI calculation Again INPUT Number of points after normalization the default is 10000 has to be always larger than OUPUT Selected length for comparison in Step 4 Figure 10 shows an example of accumulated five trajectories from LabVIEW Original data Poto B z0 l l l l l l l l l l l l l l I g 500 1000 1500 2000 2500 3000
47. NeoFlex RT data file and whole data analySIs cc eeeeeeceeeeeeeees 11 22 2 SEEEN OUl DAG DIOCKS secaia ni a Siete bean 13 2 2 3 Signal average and save the output file cc csssesseeecceeeceeeeeeseceeeeeenans 14 ZO Other Setna Senare thoes hae eee asia Nuance a cea ce nee eulens 17 23a DAO Dodd SEU UD cscet cis esss eas decee esas aad ee eee ee eee ee 17 Des POCO WC iota fe sae oh a E A E E 17 I REEI eane a extn See ae aes ata Rees tana en eS 19 NeoFlex RT CNL U of Kansas 1 NeoFlex RT Setup NeoFlex RT real time is used to record and on line or post process the data for Motor Pulse LVDT Load Cell Suck Pressure EMG OOS R EMG OOS L EMG OOI R EMG OOI L which are recorded from 8 channels of NI DAQ board The purpose of this software is to select the best user defined data blocks named as signal average txt which will be used for further analysis by other commercial software for example Minitab MATLAB etc Suck Pressure is the signal used to trigger the impulse in Motor Pulse thus the baby will receive a probe tap on the left upper lip from an actuator controlled by its own suck pressure The triggering levels in suck pressure are typically set to 5 cmH20 30 repetitions and 10 cmH 20 30 repetitions in a protocol file which sets the rules of how the data are recorded 1 1 Hardware The Hardware for this testing equipment includes gt A PC with Windows XP and 1 GB RAM gt NI DAQ 6052E and BNC 211
48. Principal Investigator Program Director Last First Middle BARLOW Steven Michael APPENDIX B Communication Neuroscience Laboratory University of Kansas November 2007 Data acquisition analysis and real time stimulus control software developed under NIH RO1 DC03311 06 Sensorimotor Control of the Human Orofacial System for use with premature infants in the neonatal intensive care unit to study the development of ororhythmic motor behavior entrain the suck central pattern generator and patient data base Neosuck RT Visual Basic Neosuck RT LabVIEW NTrainer NTrainer adult LabVIEW NNS STI LabVIEW NeoFlex RT NICU Oromotor Database Automatic analysis of non nutritive suck in time and frequency domain among preterm infants NeoSuck RT R Vantipalli J Wang amp S M Barlow CNL Technical Research Report 2006 5 1 18 University of Kansas Lawrence Kansas USA Reprint available at www ku edu cnl NeoSuck RT CNL U of Kansas Table of contents UNGCOSUCK Rh SCID ivcstisnsensawesustesduanseeeduwstteddilentendauescdtesdiandinndussddlediNendentsuescdesdhandimeduvsedesives 3 1 1 IVI VY A a nc ees eats atcha eee ceae E E eens eae ok 3 1 2 SOTVA ruisciy a east E elena uae ean eee endee 4 1 3 PROSE atin ns Al aU Oe Seams aetcnassieveanteucteud at aaca dane skudsad eetidiasulare 5 R E MANU A E S E A ac te erste E NENANA IAE E ce ea enh as 8 2 1 Record new data by INCOSUCK RV scrccsccrtccisssvetaustnticeeisdv
49. S Suck cmH20 m n 1 1 1 j 1 1 1 1 j 1 1 0 5 10 ky 4 es Ss tn tn a ef aa op te aa alge alan tye eal Shes at als ale in ales ae ies alltel ay et Time S P Figure 35 Oxygen Pulse panel 30 NeoSuck RT CNL U of Kansas 3 Appendix 3 1 Burst Extract 3 1 1 What does Burst Extraction do Locate the beginning and ending points of each burst align the bursts on the same level and flatten the signal between bursts The purpose of Burst Extraction is to enable the Peak Detector to recognize all the desired peaks by a single threshold value 3 1 2 Problems with the Original Suck Signal Figure 36 e Drifting the signal baseline increases over time which is the result of sensor temperature increase etc e Some tiny first peaks as illustrated in yellow circle would fall below the noisy peaks as illustrated in blue circle and therefore probably would not be recognized as a peak Signal Burst start Starting points and ending points on original signal Busrt end 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 24000 26000 28000 30028 0 If aje A BE threshv 500 67 581 1 gt Foy snr 2 Figure 36 Original Suck Signal 3 1 3 Goal Result In order to remove the low frequency signal we need to align all recognized bursts on the same horizontal level Then we want to straighten the signal between bursts and drag them down to a level which is low en
50. There are 3 inputs for user to define gt Select burst this number represents which burst within the file will be chosen for imputation into the STI calculation gt of peaks to compare this number represents the number of peaks that will be chosen for STI calculation and must remain constant across a file gt Start to first peak this number is defined to set the beginning point of the current point and is the number of samples between the beginning point and the first peak within the burst This value normally doesn t need any tuning After user input everything a selected burst with desired peak number will be plotted in Selected peaks in current burst as shown in Figure 30 253 NeoSuck RT CNL U of Kansas Selected peaks in current burst Plot 0 aa 27 l l l l l l I l 78000 79000 s0000 81000 a2000 83000 84000 85000 86000 Number of samples Figure 30 Current Data Selection Important Click on the output tab of Step 3d Note that the Selected length for comparison value must be less than Number of points after normalization in Step 4 in order to satisfy the mathematical requirement of normalization Up to now the first trajectory selection is finished Clicking Select this burst for STI calculation as in Figure 31 will count the current burst into STI calculation This Step 3d must be repeated for each burst desired for inclusion in current STI calculation As user continue to select th
51. and set new starting points for the bursts We are tracking the missed peaks by applying the Zero crossing algorithm Below in Figure 15 the upper plot shows the individual bursts identified in the previous step Select any burst you want to see in detail The Shift right button adds one more peak in the front by detecting the next zero crossing point and moves this burst to the right The Shift left button performs the same algorithm but deletes un desired peak in the front and moves the burst to the left It allows the user to exclude any noisy and unwanted peaks Meanwhile the user can track which burst in the whole signal that is being processed by following the blue cursor in the lower plot see Figure 15 Burst Graph Original Data I I I I 1 I I 1 1 1 I I 1 I I I gt I 0 100 200 300 400 500 600 700 600 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 Sample Number Burst Graph HPF Data Burst Graph Original Data 1000 2000 3000 4000 sooo 6000 7000 8000 9000 10000 11000 12000 13000 14000 15000 16000 17000 18000 19000 20000 ase AF lt PIY ey s Figure 15 Burst start re recognition After the desired bursts are identified Figure 16 shows a new plot which 14 NeoSuck RT CNL U of Kansas e Removed all the noisy signals between bursts e Removed the low frequency noise e g temperature increase of the sensor etc Extracted burst for peak recognition 40 0 30 0 20 0 10 0 0
52. annel based on user selected good data blocks 2 2 1 Open a NeoFlex data file and whole data analysis To open a NeoFlex data file just go to File and click Open as typically done in any Microsoft office software In the open file dialogue Figure 8 select the dat file which will be analyzed in NeoFlex RT Please notice in order to open dat file correctly the ph xml and prm xml file which are generated along with dat file in recording data should also be in the same directory Look in ewa xi E smb3 09 05 dat j E smb3 09 05b dat My Recent E t49_101405_reflex dat Documents E test dat My Documen ts 3 My Computer My Network File name P00 Places Files of type Data Files dat v Cancel Figure 8 Open file dialog After opening the data file the whole data will be showed in NeoFlex RT window similar to Figure 9 11 NeoFlex RT NeoFlex RT C Program Files Neuro Logic NeoF lex RI Wata t49_101405_reflex dat CNL U of Kansas File Settings Yiew Tools Help Motor Pulse LYDT mm LOAD CELL N SUCK PRESSURE cm H20 s Date Created 10 14 2005 11 03 20 AM Figure 9 NeoFlex RT window for entire data analysis EMG OOS R uv EMG OOS L u EMG OOI R uv EMG ODI L uv 42 42 5 s 1 43 43 5 44 44 5 45 NeoFlex RT A Real Time V The scroll bar in right lower corner enables user to look the data for 8 channels from beginning to end
53. aracteristics amplitude frequency pulse width pulse rate pulse length train rate and number of pulses generated per session NTrainer RT CNL U of Kansas 1 3 Program Installation Uninstall any previous installed versions of NTrainer RT using Control Panel gt Add Remove Programs Install NTrainer RT by running setup exe from the installation package Note Normally it works with BNC 2090 Figure 3 To work with BNC 2110 Figure 4 BNC terminal box please make the following change Figure 4 BNC 2010 For BNC 2110 To work with BNC 2110 terminal configuration has to be in differential mode The Terminal configuration can be set to differential mode by modifying the parameter value of NTrainer RT CNL U of Kansas defaultAiTerminalConfig to 1 under lt systemConstants gt section in the file NTrainer exe config located in C Program Files Neuro Logic NTrainer RT Here is the line from NTrainer exe config that shows this parameter value lt add key defaultAiTerminalConfig value 1 gt defaultAiTerminalConfig can have the following values value 1 for Differential value 2 for Referenced single ended value 3 for Nonreferenced single ended By default the file NTrainer exe config contains value 3 for defaultAiTerminalConfig This setting is used to work with BNC 2090 Here is the line from NTrainer exe config that shows this parameter value lt add key defaultAiTerminalConf
54. as generated from the Burst Extract panel e Load file right after Burst Extract this panel will automatically load the file you saved as _BurstExtraction in last panel as shown in Figure 18 Load file For Suck Analysis G Select file from Burst Extraction 7 C Program Files NeoSuck Data BabySuck_BurstExtraction txt i I Select new file Figure 18 Loaded for NNS Param If you want to analyze another file click Select new file and then browse to choose any other file e NNS Param for existing Burst Extract file a pop up File Browse window will enable user to choose new files Sampling frequency is a parameter used for the analysis Enter the corresponding sampling frequency for the file you wish to analyze Sampling frequency Hz 5 100 When performing NNS Param analysis for data files other than _BurstExtraction file generated by the previous tab be sure to select the right Suck pressure channel for your data Suck pressure channel z 0 Data length shows how much data in the current file Data length 13240 16 NeoSuck RT CNL U of Kansas 2 4 2 Peak Detection Suck Pressure Peak detector width 20 Peaks found 19 EBL l lt gt Ee om Q Figure 2 Peak detector Peak Detector width this parameter is defined by the user to adjust the peak width described in the number of samples Cursor selecting the Drag or Define cursor option allows the user to define the threshold
55. ate Delivery Type Pacifier Size 06 14 2006 TE TES M gt Initial UEZIN 2o First of twins Bo H Unknown ls More than one race F E Group1 TERM p ic x Section M xl Medium GA at Birth days Days of Life at Init Study GA at Study days Birth Weight gm Birth Length cm Birth Head Circumference cm Study Weight gm Study Length cm Study Head Circumference cm Apnea in last 72 hours Polysomnogram C Yes No Oxypneumocardiogram C Yes No Record Position 1 154 Save Cancel Username Location kumc 6 14 2006 1 page 10 3 39 PM NICU Database Patient Configuration Reports User Help CNL U of Kansas Comments Patient Configuration Reports User Help Cancel Save Intubated Date Intubated fat DRAE Delete Nasal CPAP CPAP Date No of Days Add Modify Delete Nasal Cannula Cannula Date LOIDE d Modify Delete Oxygen Days on Oxygen Days Since Extub Days Since Nasal CPAP Days Since Nasal Cannula po Seizures treated with anticonvulsants EEG C Yes No Intraventricular Hemorrhage Yes C No Modify _Delete Grade of IVH i i Cranial Ultrasound Yes C No Add Modify Delete Date 01 29 2004 Ventricular dilatation C Yes No Porencephalic cyst C Yes No
56. cy removed is much easier for burst locating in the next tab Tab 3 Peak Detect This tab identifies bursts based on the high passed signal Figure 14 The yellow points mark the peaks the white points mark the start of a burst and the green points represent the end of a burst e There are 3 inputs for the user to define Peak detector width this parameter is used to determine the peak detection algorithm Generally it defines the peak width in sample number Decreasing it will detect skinnier peaks and vice versa Distance between burst this controls the distance between bursts in sample number Last peak to burst end defines the sample number between the end of the burst green point and the previous peak yellow point e There is one active cursor for the user to define the threshold of the peak value 12 NeoSuck RT CNL U of Kansas Cursor Drag D Define 5 8 When the switch button is engaged in the drag mode it enables the user to drag the cursor in the vertical direction to define the peak threshold value When the switch button is engaged in the define mode to the right the user 1s able to define the threshold by input numbers It may be difficult to detect the tiny peaks While as long as the burst number is correct there should be no concern that peaks are missed e There are 2 outputs for user to check Peaks found this is not a critical number it gives the user a number to ro
57. e Je Peaks found 224 amp Peak Detection Plots Peak Values Figure 28 Block Suck Pressure Plot in LabVIEW There are 2 inputs for user to define gt Peak detector width this is the parameter used for the peak detection algorithm Generally it defines the peak width in sample number Decreasing it will detect skinnier peaks and vice versa 23 NeoSuck RT CNL U of Kansas gt Channel this enables user to choose which column in the selected file is desired for STI analysis There is 1 active cursor for user to define the threshold of peak value Cursor Drag D Define 5 8 This switch button enables user either to drag the cursor to define the threshold or to define the threshold by input numbers There are 2 outputs for user to check gt Data length this shows the number of samples in current data gt Peaks found this shows the identified peak number based on user defined peak width and threshold 2 6 4 Step 3b Burst Detection Bursts are recognized by setting the distance from the last peak position of the first burst to the first peak position of the next burst If the Recognized burst number is incorrect user must adjust the Distance between bursts until the correct burst number is displayed 2 6 5 Step 3c Peak Shoulder Detection and Deletion This step need only be performed if user s data contains peak shoulders A peak with a peak shoulder looks like the second peak shown
58. e recording window Figure 7 User could click Start button to start recording Meanwhile the Stop will be enabled and user could click Stop button to stop The protocol window showed in left lower corner window is the current recording protocol for user s reference NeoFlex RT CNL U of Kansas After recording the user could apply the methodologies introduced in 2 2 to analyze reflex data in more detail File Settings Yiew Tools Help Motor Pulse v EMG OOS R u ACHO on ACH4 on i E 0 ig E l 1 a al oOo 1 DF ha i ef i i 1 DE eT i ml atz El 1i EMG OOS L u LYDT mm ACH1 n 1 ACH5 on 1 LOAD CELL N EMG O0I R uv ACH2 a i ACHE on i 0 I I 1 l P 1 1 E I SUCK PRESSURE cm H20 EMG OOI L u ACH3 on ACH on I I 0 0 5 1 1 5 2 245 3 a5 4 4 5 5 0 0 5 1 fled 2 Ag a hel 4 45 5 vse gt ae Date Created 6 8 2006 4 14 12 PM Figure 7 Start recording window 10 NeoFlex RT CNL U of Kansas 2 2 Analyze existing data by NeoFlex RT After opening an existing data file dat for NeoFlex analysis another two windows will automatically pop up These include a blocks analysis window for screening out bad blocks and a signal average window for saving averaged signal according users criterion In other words these two windows perform as the next two steps that lead users to find out the averaged signal for each ch
59. e remaining bursts monitor user s selection in the Selected peaks in current burst window Figure 30 If at any point a burst and its peaks is not represented accurately modifications can be made immediately by adjusting values in Steps 3a through 3c as required before hitting Select this burst for STI calculation Select this burst for STI calculation Figure 31 Next button for starting next burst selection Important Note Select burst will be changed every time a new burst is selected while INPUT Select how many peaks user want to compare will remain unchanged for STI processing 26 NeoSuck RT CNL U of Kansas 2 6 7 Step 4 SD and STI calculation Again Number of points after normalization default 10000 samples must always be larger than Selected length for comparison in Step 3d Figure 32 shows an example of five accumulated trajectories Original data Plot 0 aa l l l l l l l l l I i l l l l 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500 7o00 7500 Num Linear time and amplitude norm Plot 0 Ba 2 5 Amplitude Ea Lv l l l l l l l l l l l 0 1000 2000 S000 4000 5000 6000 7000 8000 9000 10000 Slum Figure 32 Plots of Original Data and Normalized data SD Step represents a constant sample number used as the step length in SD calculation for the STI value Normally this value does not need any tuning if the normalized sam
60. eduseathaneiauateeidveduees 8 2 2 Analyze existing data by NEOSUCK R Porense ea 10 22l BASIC ANALY SIS eirese a a ai 10 2 2 2 Fourier transform of suck pressure eeessesssseeeeesssssssseererssssssseceressssssseeeeees 12 223 Other analysis Tool Sac ou ieesinhahahnuseauieiaieleee ex eae eeteeec ene 15 23 OHer UL oie acct cec gc ee a lee cetera esa eet tas lec etna ta cea icee teulen 16 T REC EMC CS dace ce testes N ate ease e ase tence ae east arses 18 NeoSuck RT CNL U of Kansas 1 NeoSuck RT Setup NeoSuck RT real time 1s used to record and analyze data associated with the non nutritive suck NNS including signals related to Suck Pressure EMG OOS R EMG OOS L EMG OOI R EMG OOI L which are recorded from 5 channels of NI DAQ 6052E PCI multifunction board installed in the acquisition PC with 3 kHz sampling rate on 16 bits vertical resolution We currently apply this software and hardware system to test and re analyze the suck pressure and signals from other channels for premature and term baby The analysis tools in NeoSuck RT include histogram analysis FFT analysis bad block screening digital signal processing generate output ASCII files for time and frequency domain etc 1 1 Hardware The Hardware for this testing equipment includes gt A CPU with 1 GB RAM running Windows XP gt NI DAQ 6052E and BNC 2110 or BNC 2090 gt Bridge Amplifier gt AC PreAmplifiers 4 gt Sensors pressure sensor and EMG Ag AgCl senso
61. ee i i Figure 6 Block Suck Pressure Plot in LabVIEW 2 3 Step 2 Burst detection Bursts are recognized by setting the distance from the last peak position of the first burst to the first peak position of the next burst User needs to change the INPUT Sample number for distance between bursts until the OUTPUT Number of recognized bursts in current block shows the right burst number if it doesn t display the correct burst number 2 4 Step 3 Peak shoulder detection and deletion Peak with peak shoulder looks like the second peak shown in Figure 7 If the distance from peak to its should is less than INPUT Sample number from peak to its shoulder the shoulder will be deleted For example the first plot in Figure 7 which the input sample distance set as 300 shows the shoulder as a peak While in the second plot the distance is increased to 900 all the shoulders are deleted a 40 0 30 0 20 0 10 0 Dig 0 0 2078 7 4 29000 0 27500 0 30000 0 21106 0 29000 0 2 7500 0 30000 Figure 7 Peaks with Peak Shoulder left and Peaks with Peak Shoulder Deleted right 2 5 Step 4 Select burst and peaks for SD Up to now all the peaks without shoulder and bursts should both be recognized exactly as what user wants them to be Remember our SD processing is based on a set of bursts with same peak number Now it s time to INPUT Select burst and define INPUT Select how many peaks you want to compare N
62. ession NTrainer RT isle Settings Tools Help Open Ctrl O Test Ctrl T STIMULUS PULSE fv Page Setup Print Ctrl P Exit Ctrl Xx 5 Time fs 5 Time fs Patient Info Visit Info Visit m Test Date mm dd yppyy Age at Test days Patient ID Patient First Name Patient Last Name DOB a E N Trainer Cycles for Current Visit Cumulative N Trainer Cycles for All Visits User Name Topeka Figure 6 Main program window with the Patient Info menu option list displayed This menu is divided into three function blocks The first block contains three options The first option for creating a new data filename the second option for opening an existing file for data analysis and the third option for opening an existing file for NTrainer RT stimulation The second block contains page setup and printing options The final block contains the Exit command NTrainer RT CNL U of Kansas NTrainer RT Jex STIMULUS PULSE fY SUCK PRESSURE cm H20 New Patient Details Patient Info Patient ID Patient First Name fa Patient Last Name DOB 0170172008 mm dd pyyy coe Patient Info Patient ID Test Date mmdd yyyy Patient First Name Age at Test days Patient Last Name DOB T ddia N Trainer Cycles for Current Visit Cumulative N Trainer Cycles for All Visits User Name Lawrence Figure 7 Main program window of creating a new patient
63. gure 11 CE NTrainer RT STIMULUS PULSE fv SUCK PRESSURE em H20 Graph Settings Graph Settings Stimulus Pulse Suck Pressure MY Axis Y Axis Min Yy Min 333 333 cm H20 Max 5o Y Max 333 333 cm H20 Max IV Auto Scale V Auto Scale V Auto Scale x Axis Period M Patient Info Visit H Test Date mmdd yyyy Age at Test days DOB een TEE N Trainer Cycles for Current Visit Patient ID Patient First Name Patient Last Name Cumulative N Trainer Cycles for All Visits User Name Lawrence Figure 11 Main program window with the Graph Settings menu option list displayed 14 NTrainer RT CNL U of Kansas The Change User submenu allows the user to choose the data collection sites 1 e NICU hospital test locations as shown in Figure 12 NTrainer RT ex STIMULUS PULSE fv Change User Select a User Time fs 5 Time s Patient Info E Visit Info Visit H Patient ID ai Test Date La rom dd yyyy Patient First Name E i Age at Test days Patient Last Name DOB ennn ridd N Trainer Cycles for Current Visit Cumulative N Trainer Cycles for All Visits User Name Topeka Figure 12 Main program window of users 15 NTrainer RT CNL U of Kansas 2 2 Record new data using NTrainer RT 2 2 1 How to create a new file Select Patient gt New and fill in t
64. he response field for data file name and click OK A window message will appear on the main window screen to inform the user that the patient s information has been added into the database as shown below Figure 13 El NTrainer RT Sag STIMULUS PULSE fV SUCK PRESSURE em H20 New Patient Details Patient Info Patient ID Patient First Name Patient Last Name DOB 01 26 2006 mm dd yyyy Patient Info Patient ID Patient First Name Patient Last Name DOE T E N Trainer Cycles for Current Visit Test Date ram dd pyyy Age at Test days Cumulative N Trainer Cycles for All Visits User Name Lawrence 16 NTrainer RT CNL U of Kansas NTrainer RT _l J Patient details added to database succesfully Figure 13 Main program window of New Patient Details 2 2 2 How to provide NTrainer RT neurotherapeutic oral stimulation After entering patient s information into the NTrainer RT database refer 2 2 1 the user can provide neurotherapeutic oral stimulation by selecting Patient gt Test A Patient List window will appear on the main program window for the user to select the patient Highlight the patient s name you wish to test and click OK 17 NTrainer RT CNL U of Kansas a NTrainer RT Patient Settings Tools Help STIMULUS PULSE fy Patient Id FirstName Last Name Birth Date Good Sucker 6 5 2005 Doe 1 4 2006 Doe 1 26 2006 m Patie
65. hobiol 45 22 33 2 Smith A Johnson M McGillem C and Goffman L 2000 On the assessment of stability and patterning of speech movements Journal of Speech Language and Hearing Research 43 277 286 3 Poore M Barlow SM Wang J Lee J 2007 Respiratory Distress Syndrome history predicts suck spatiotemporal index in preterm infants Developmental Medicine amp Child Neurology in revision 4 Poore M Zimmerman E Barlow SM Wang J Gu F 2007 NTrainer therapy increases suck spatiotemporal stability in preterm infants Acta Paediatrica in review 12 Electrophysiological recording of the trigeminofacial reflex during NNS in human preterm infants NeoFlex RT R Vantipalli J Wang amp S M Barlow CNL Technical Research Report 2006 2 1 19 University of Kansas Lawrence Kansas USA Reprint available at www ku edu cnl NeoFlex RT CNL U of Kansas Table of contents I SINGCORIEK KRI SOU Dx advsssemaavssusieacuamdenndustnedieddilesentauosndtesdiamstion dusted ddan antec ncdesdhandimeduvsadadddens 3 1 1 PO AM e E E E E eee 3 1 2 SONATE aE E E eat oie se etiam eee enna te nteiaaiaaseaieeet 5 1 3 Pro ora lista aO Mii s cscentcitndest Mpemidissmerstucentiudestpetidisimer seutiadsveate disci 6 2 NGC i A earn sea cosas uate thee E E ue tence tata ee 8 2 1 Record new data Dy NCOPICK KI sercicinvatsvacasntunied vadsia casted aekaaata E 9 22 Analyze existing data by NEOPIEX RI oporno iinan a a a 11 2 2 1 Open a
66. hown as exemplified in Figure 18 Note that the user can select a previously completed NTrainer session file by simply choosing it at the Visit icon 22 NTrainer RT CNL U of Kansas 3 References Finan DS amp Barlow SM 1998 Mechanosensory modulation of non nutritive sucking in human infants Journal Early Human Development 52 2 181 197 Barlow SM Finan DS amp Park S Y 2004 Central pattern generation and sensorimotor entrainment of respiratory and orofacial systems In B Maassen W Hulstijn R Kent H F M Peters P H M M van Lieshout Eds Speech Motor Control in Normal and Disordered Speech Oxford University Press 211 224 Barlow SM amp Estep M 2006 Central pattern generation and the motor infrastructure for suck respiration and speech J Communicative Disorders 39 366 380 Barlow SM amp Finan DS 2006 A new therapeutic method for entraining the suck central pattern generator CPG in the premature infant Society Pediatric Research 3153 Barlow SM Finan DS Seibel L Chu S Poore M Zimmerman E Urish M Estep M 2006 Translational neuroscience using patterned somatosensory stimulation to entrain oromotor activity in premature infants 5th International Conference on Speech Motor Control June 7 10 Nijmegen The Netherlands Barlow SM Zimmerman E Chu S Lee J 2007 Patterns for the premature brain Synthetic orocutaneous stimulation entrains preterm infants with feeding difficulties to s
67. hows an example which stores FFT for baby BabySuck this FFT describe the data start from sample 200 with data length 2000 21 NeoSuck RT CNL U of Kansas File path and name for Power spectrum Figure 26 Save FFT file for part of data pp Zi NeoSuck RT CNL U of Kansas 2 6 NNSSTI 2 6 1 Step 1 Load File NNS STI will automatically load the file user created in Burst Extract as shown in Figure Select file From Burst Extraction q Program Files WeoSuckiDatalBabySuck BurstExtraction ket B Select new file Figure 27 Loaded for STI If user wants to analyze another file however user may also click Select new file and then the Browse button to choose any other file A pop up File Browse window will enable user to choose new files 2 6 2 Step 2 Start Analyzing Press Step 2 Start Analyzing button to enable NNS STI analysis Notice once the start button is clicked the program doesn t allow user to reselect other file If user want to resume other files instead of current one user must press the Stop and DON T Save STI Analysis button at the bottom of the STI tab page 2 6 3 Step 3a Peak Detection Peak detector width 2 t20 Channel gfo Data length in sample 130000 1000 0 2000 0 3000 0 4000 0 5000 0 6000 0 7000 0 8000 0 9000 0 10000 0 11000 0 12000 0 13000 0 14000 0 15000 0 16000 0 17000 0 17997 0 0 0 pope rid Weesp 221 0 178 Cursor Dra defin
68. ig value 3 gt NTrainer RT CNL U of Kansas 2 User Manual To run NTrainer RT double click on the program icon The main program should appear with a menu bar along the top three blank waveform panels along the upper half of page patient s information on the bottom left side and patient s visit information displayed along the bottom right margin As shown below Figure 5 the menu bar has four active options including Patient Settings Tools and Help NTrainer RT Patient Settings Tools Help STIMULUS PULSE fv Time s 5 Time s Patient Info Visit Info Visit Test Date mm dd yyyy Age at Test days Patient ID Patient First Name Patient Last Name ni DOB aT AAAA N Trainer Cycles for Current Visit Cumulative N Trainer Cycles for All Visits User Name Topeka Figure 5 Main program window of the NTrainer RT user interface The bottom of the program window appears with the Patient s information and Visit information The Patient s information panel includes the Patient s ID First Name Last Name and Date of Birth DOB The Visit Information panel includes the total number of Ntrainer 8 NTrainer RT CNL U of Kansas Visits Test Date Age at Test NTrainer Cycles for Current Visit and Cumulative NTrainer Cycles for All Visits 2 1 Overview of menu bar options As shown in Figure 6 selecting Patient allows the user to choose a New file name for the recording s
69. igure 13 User could define the minimum and maximum value calibration rate and channel name for all 8 input channels and 2 output channels User could also read and change the sampling rate for input and output channels to accommodate individual user applications DAQ Settings DAG Device Sample Rate Down Sample Buffer Size Channels Motor Pulse Al Channel Voltage A D Min A D Max EMG 005 A Al Channel Voltage A D Min A D Max 2 3 2 Physical Unit Physical Unit PCI 6052E 5000 samples s ho samples 1 samples LYDT facHa o o Al Channel z Y Physical Unit A Y Voltage 5 V A D Min 5 V A D Max EMG OOS L ACH4 v Al Channel ho uV Physical Unit a Y Voltage Rom A D Min Hoa A D Max Protocol file Load Cell ac it Al Channel mm Physical Unit 0 56 Y Voltage ie A D Min z Y A D Max EMG OOI R ACH5 v Al Channel 00 ul Physical Unit 2 Vv Voltage Ro A D Min j A D Max Suck Pressure ACH2 v Al Channel o5 N Physical Unit 0 444 Vv Voltage Hoe A D Min z Y A D Max EMG OO L ACHE v Al Channel ho u Physical Unit ea V Voltage ioy A D Min Hoy A D Max Figure 13 DAQ setting window acha xl jo cmH20 bs Gomi ie 8 jach zl ho w Pik Ro ie Cancel In NeoFlex window click Tools gt Protocol Editor to activate protocol editor window as shown in Figure 14 In this window user could change data recording protocols
70. ile format User must do this before exiting Neosuck 28 NeoSuck RT CNL U of Kansas otherwise the images will have to be recreated by re running STI or created in another program such as MATLAB by opening the Original Data and Normalized Data files 29 NeoSuck RT CNL U of Kansas 2 7 Oxygen Pulse This panel simply displayed the Oxygen and Pulse rate channel It also calculates the minimum maximum and mean value for each of them The suck channel displayed at the bottom is for user to follow the changes in suck signal when Oxygen and Pulse rate change DAQ Burst Extract NNS Param NNS FFT STI Oxygen Pulse File Path org l Size 0 samples Sampling Frequency Hz for Current Data J H100 Select Oxygen Channel f Oxygen Psat opi 1 J on m f on 1 Oxygen 100 amp 1 l l 1 l 1 l l l l 1 1 l l l l I l l l i 1 l l l 1 l l l 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 8 amp 5 0 9 100 105 110 115 120 125 130 135 140 145 150 155 160 165 170 Time 5 r Select Pulse Channel Oxygen maximum 9 Oxygen mininum 0 Oxygen mean 0 00 F Pulse Rate J 1 y aa f Pulse 100bpm 5 2 meer 1 l l i I l l l i 1 l l l l l l l 0 5 10 15 2 23 3 35 4 4 w 5 60 65 70 75 30 wW 9 100 105 110 115 120 125 130 135 140 145 150 155 160 165 170 Time 5 E Select Suck Channel ao Pulse rate maximum J0 Pulse rate mininum 0 Pulse rate mean 0 00 Suck S
71. ines how many samples are required between bursts The user must compare the of bursts recognized with that of the original suck pressure signal If the number of bursts recognized does not match what is found with visual inspection of the original suck pressure signal the value of Peak detector width Distance between bursts and the cursor threshold position should be adjusted until the desired number of bursts are detected e Histogram generation For each histogram the user must define the Bin number in order to justify the desired number of bins in the histograms Bin number change this value to reflect the number of desired bins in the histogram Mean and Standard Deviation are output values generated regarding the histograms Save histogram click this button to save the histogram value to the directory where you loaded the file 2 4 4 Save summary files e Summary this file includes miscellaneous information such as peak number burst number peak number minute burst number minute date and time as well as any comments entered by the user as shown in Figure 20 e Peak Summary this file contains a table that records the peak index peak amplitude burst location etc Comments of peaks Please input comments here Saved file path and name for summary it a p C Program Files NeoSuck Data BabySuck_Summary txt of bursts Save Si T 13 Saved file path and name for peak summary ao umar p
72. ion Pt No MELAEDEE Pt First Name Pt LastName Status Pt Medical Record No Birth Date Birth Order Ethnicity Race ee ee ee ee ee ee Record Position Username Location CAPS 67672006 4 04 PH Figure 4 Main program window with the Reports menu options list displayed NICU Database CNL U of Kansas The User submenu allows users to change the sites of data collections refer Figure 5 pes Neonate Sensorimotor Oral Development Database CNL University of Kansas 1 RO1 DCO3311 02 Patient Information PR Patient Configuration Reports MISSA Help Change User Pt No GRETEL ALALELE Pt Last Name BithDate Birth Order Ethnicity Race von am woay otto sours sat ater Record Position Username Location CAPS 6 12 2006 3 00 PH Figure 5 Main program window with the User menu option list displayed NICU Database CNL U of Kansas 2 2 Add new Patient in NICU database In order to add new patient information choose the Patient gt Patient Info and click the Add button Figure 6 This permits the user to enter Patient information such as Patient Number Patient Medical Record First Name Last Name Date of Birth Birth Order Ethnicity Race Gender Patient Group and Patient s Discharge Date Selecting Save will update the patient s information into the NICU Database a Neonate Sensorimotor Oral Development Database CNL University of Kansas 1 RO1 DCO3311 02 Patient Info Details
73. l N m m z 2 2 oO wo 2 a P oOo o 0 1000 2000 3000 4000 5000 6000 700 8000 9000 10000 Original sample number Linear Time and Amplitude Normalized Signals 2 wo wo 2 x Q W oa a i w Z o me 7D 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 Standard Deviation STI 93 0103 2 1 5 aP 0 5 0 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 Relative sample number Figure 48 STI Calculations for NNS impaired preterm infant Figure 48 shows an example of poor NNS patterning sampled from a preterm infant with respiratory distress syndrome In this case the poor suck patterning results in an STI increase to 93 01 40 NeoSuck RT CNL U of Kansas 4 Reference 1 Smith A and Zelaznik H N 2004 Development of functional synergies for speech motor coordination in childhood and adolescence Dev Psychobiol 45 22 33 2 Smith A Johnson M McGillem C and Goffman L 2000 On the assessment of stability and patterning of speech movements Journal of Speech Language and Hearing Research 43 277 286 3 Poore M Barlow SM Wang J Lee J 2006 Respiratory Distress Syndrome history predicts suck spatiotemporal index in preterm infants Developmental Medicine amp Child Neurology in revision 4 Estep M Barlow S M Stumm S Fees M Finan D Seibel L Poore M Cannon S 2005 Non nutritive Suck Burst Parametrics in Preterm Infants Society for Pediatric Research
74. ment includes gt A CPU with 1 GB RAM running Windows XP gt NI DAQ 6052E and BNC 2090 gt AC PreAmplifiers gt Sensors pressure sensor gt Nellcor Tyco N 600 Pulse Oximeter The hardware listed above is configured up according to Figure 1 and baby s suck pressure sampled from the Soothie silicon nipple receiver is input to NeoSuck RT for real time digitization signal processing and waveform display Bridge Amplifier K Suck Pressure Ai3_ _ eee Computer and DAQ LPF 50Hz ransaucer EMG Z lt N EMG J Grass EMG p Ai4 7 AC EMG Ai4 7 EMG pi PreAmplifiers Sensor GND is connected to baby leg 7 i Dec 1 2005 by Joan NeoSuck RT CNL U of Kansas Figure 1 this is a temporary figure 1 2 Software This software is coded by Labview 8 2 The function of this software includes Real time data acquisition Burst extraction Peak detection Histogram processing Mean and Standard deviation calculation FFT analysis for entire data or for selected data STI analysis Exported all results to txt file which could be used for further analysis in other software etc 1 3 Installation Instructions l Uninstall any previous installed versions of NeoSuck RT using Control Panel gt Add Remove Programs PA Install NeoSuck RT by running setup exe from the installation package Figure 2 BNC 2090 Note Normally it works with BNC 2090 interface Figure 2 2
75. nagls 10 Duh SSID LE LOONES UO iced case tue E E Oona Os 10 Beds SPE P LZ STOEL QI GULY ZIM saaicrle erie Sa caine sas A A TEAC 1l 2 3 3 Step 3 Coarse tuning Find the number of DUPSts ccccccccceseesccccccecceceseesseecceeeeaaasseseeseesaaees Il 2 3 4 Step 4 Fine tuning Burst Start Re recognition ccccccccseecccccceeeseeeceeseeseeeeaaeeeeeeeeaaaeseseeaaees 14 2 4 INNS ARAN secs or anata E EA OIT O sa ancenmeueceausa tence O placentae Fencesas 16 DM TiO FUL Oe EEEN E N Penns isia seme OOE ON ia basal Raat Laue OOE 16 Ds CUR Dele ON aihena a a A E N ETA ET E AE ETNEN I7 2 4 3 Burst detection and Histogram BCNCLALONL ccccccccccssscccccccccecesesssseceeeeeeaesesseeeeeeeaaaasseseeeeeeaaaes 17 ZAG DOVES UNII VCS Gatien wees tasec aceasta too ak ota ate ea EA OOA 18 2 5 ANS EE Tirini a a E TR emraaT Or een dane an Orr er omer 19 I IO Uschi EAE SE E eat lect Nea Gav Ne ce eG eR Tees eats 19 ee TE GINS S ores she stoa se ea ase neces AG ets ee aa a 19 ee HOV Cresta ie eee 21 2 6 N SS E A EE E ceecmeeateatans 23 POE SPL LOUL T UC Aea EE ENE E 23 202 Sep 2 SUE ANONY AINO i e EEE O E a avis eee ee T E NTON 23 20S ICP IAS FAK DECO ee aE NEE E EENE NE OEE EE ERA 23 ZO SIC IDE DUIS Delet Oerein E a E A E E O AAS 24 2 6 5 Step 3c Peak Shoulder Detection and Deletion ccccccccccccssseeeescccceeeenesssseceeeeeeaaeesssseeeseeeaas 24 20 0 Step od Select burstand Peaks fOr SD os cirioss ddssss decane ie a e A AA TEER 23 ZOL SCP FS
76. nt Info Patient ID Cancel Patient First Name Patient Last Name DOE trendiest N Trainer Cycles for Current Visit Age at Test Cumulative N Trainer Cycles for All Visits User Name Lawrence Figure 14 Main program window with the Patient List submenu option list displayed The user may Stop and Resume the patterned orosensory stimulus delivery by simply clicking on the window shown in Figure 15 The pacifier receiver of the NTrainer unit is instrumented with a wireless mouse Figure 16 which allows the user to remotely Start Stop and Resume stimulus delivery while attending to the infant In this way the user does not need to interrupt direct intervention or handling of the infant since program control can easily be directed from the wireless mouse 18 NTrainer RT CNL U of Kansas S NTrainer RT STIMULUS PULSE fv SUCK PRESSURE em H20 l 42 5 Time fs 42 5 Time s M Patient Info Visit Info Visit H 2 Test Date 6 7822006 romn dd pyyy Age at Test 159 days Patient ID Patient First Name Patient Last Name DOB EENT mmn dd pyyy N Trainer Cycles for Current Visit 49 Cumulative N Trainer Cycles for All Visits 138 User Name Lawrence Figure 15 Example of an NTrainer RT main program 19 NTrainer RT CNL U of Kansas Figure 16 Remote control for NTrainer RT Wireless Mouse and regular mouse User could control NTrainer
77. o Gag Reflex Info State Control Info and Pacifier Size Info as shown in Figure 3 Clicking on any of these submenus will enable a new table to popup which allows the user to change or update the patient s information EE Neonate Sensorimotor Oral Development Database CNL University of Kansas 1 RO1 DCO3311 02 Patient Test Information e i ieee Reports User Help ao x Status Info Birth Order Info PI Ethnicity Info Name Pt LastName Status Pt Medical Record No Birth Date Birth Order _ Ethnicity Race Race Info Delivery Type Patient Type Feeding Mode Info Medication Info ROP Stage Info Oxymetry Swallow Results Info Oxymetry Swallow Indications Info Neurologic Exam Info Gag Reflex Info State Control Info Pacifier Size Info oe ee ee ee ee ee Record Position Username Location CAPS 6 6 2006 3 56 PM Figure 3 Main program window with the Configuration menu options list displayed NICU Database CNL U of Kansas The Reports menu allows the user to generate a report of the cumulative distribution of patient race and ethnicity in separate files as shown in Figure 4 This is useful for completion of NIH minority race inclusion reports for annual research progress reports a Neonate Sensorimotor Oral Development Database CNL University of Kansas 1 RO1 DCO3311 02 Patient Test Information MPR Patient Configuration KESS User Help Inclusion Enrollment gt Report Part B Ethnicity Select
78. o keep in mind for data processing thereafter 2 1 Step 0 Data preparation The input suck pressure files for NNS STI are txt file generated from NeoSuck RT Before running NNS STI suck pressure in blocks should be ready In NeoSuck RT go to Tools and then click Create ASCII Blocks 2 2 Step 1 Peak detection Once the NNS STI is opened the Choose file dialog will pop up for your first block selection _ Choose file to read Look in gt Data v QO 2 ega k19 te T18 My Recent W resultsForT 18K 19 doc Documents Z Sample_022704_B1 txt aunt Sample_022704_B2 txt Sample_022704_B3 txt Desktop Sample_022704_B4 txt Sample_022704_B5 txt Sample_022704_B6 txt Sample_022704_B7 txt Sample_022704_B8 txt Sample_022704_B9 txt Sample_022704_B10 txt My Documents ys E VLE KUE KUE LE AVE OE BL ATE My Computer My Network Files of type All Files C E x Figure 5 Choose File Dialog After the block is chosen this block of suck pressure is showed in Block Suck Pressure plot The red line cursor need to be tuned the red line could be dragged up and down in order to select the right peaks we want All recognized peaks are marked by yellow dot NNS STI CNL U of Kansas Block Suck Pressure 50 0 0 0 5000 0 10000 0 15000 0 20000 0 25000 0 30000 0 35000 0 40000 0 45000 0 50000 0 55000 0 60000 0 65000 0 70000 0 75000 0 80000 0 85000 0 89999 frenp aos
79. omain plot of the currently displayed block in the main NeoSuck RT window Figure 10 The first and second graphs are both for current block while the third graph named as composite FFT shows the mean FFT for all the blocks that have been analyzed by the FFT subroutine User could either click Update FFT Plot Save FFT button or go to Tools gt Generate Composite in this window to periodically update the composite FFT E NeoSuck RT C Program Files Neuro Logic NeoSuck RTWata K6_033004 K6_033004 dat pro FFT Suck Pressure Block 5 MOTOF Settings Tools SUCK PRESSURE em H20 ACHO Suck Pr cm H20 Time s 3 0 FFT SUCK PRESSURE cm H20 ACH1 A ika Amplitude em H20 i i 0 0 5 0 10 0 15 0 Frequency Hz FFT Suck Pr File K6_033004_B5 dft ACH2 FFT Suck Pressure Update FFT Plot Save FFT Amplitude 0 895 Frequency 0 046 Hz 0 05 1 SUCK PRES COMPOSITE FFT SUCK PRESSURE ACH3 Amplitude cm H20 0 0 5 0 10 0 15 0 z0 0 Frequency Hz NeoSuck RT Real Time v Composite for blocks 3 5 Composite FFT Suck Pr File K6_033004_BC dft Block 5 6 Peak Count 16 6 Date Created 3 30 2004 1 53 32 PM Figure 10 Suck pressure FFT analysis In the setting menu user could set the filter parameters cutoff frequencies for Low Pass filter and High Pass filter to plot the desired frequency property Figure 11 In addition the g
80. ommercial graphics and statistical software Open Look in K6_033004 ce EB a PEZETA nid i k6_033004 EFI pro My Recent Documents Desktop My Documents My Computer a i l F ae etwork File name K6_033004 dat aces Files of type Data files dat pro v Cancel Figure 14 Create ASCII blocks for dat file 3 The Generate Histogram Files and Generate Amp Period ASCH Files will generate txt files for future analysis 15 NeoSuck RT CNL U of Kansas 2 3 Other setups Some other setups in Setting of the NeoSuck RT menu bar enable the user to set and modify analysis criteria for data acquisition graphs histogram plots and FFT analysis Figure 15 is the DAQ setting dialog where the user can adjust the minimum and maximum acquisition limits as well as calibration parameters Once the DAQ parameters are set sampled data files will assume the format established in the DAQ set up Thus these setup parameters should not be modified unless the specific intent is to change file structure and sampling rates DAG Device Al Channels Motor Pulse LYDT Load Cell Suck Pressure Al Channel ACHO tsi Al Channel ACH l Al Channel lach l Al Channel ACH3 l Physical Unit 1 O w Physical Unit 5 mm Physical Unit 05 N Physical Unit 10 emH20 Voltage y Voltage 0 56 y Voltage 0 484 y Voltage bs Yy A D Min 5 y A D Min m y A D Min 1 tC A D Min e A D Max 5 y A D Max a
81. ough to expose tiny peaks the yellow circle in Figure 1 Our goal is to change Figure 36 to Figure 37 Jl NeoSuck RT CNL U of Kansas Signal LN Extracted burst For peak recognition burst begin bre 45 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 22000 24000 26000 28000 29998 4 gt WE 22 thresh si 7 6 J amp Oy t t be Figure 37 Extracted Suck Signal 3 1 4 Solution How do we realize it Coarse tuning High pass filter and Peak detection Amplitude N a Rainer ee 9000 9500 1 0000 1 o500 11 doo al 500 1 2000 1 2500 1 3000 1 3500 ra 4000 1 4500 1 S000 1 5500 1 6000 1 6500 1 7000 1 7500 1 8000 E 8733 Ly oa Figure 38 Original Signal High Pass Filter 4 32 NeoSuck RT CNL U of Kansas Signal LW Peaks a4 Burst start a Starting point and ending point recognization Burstend i 20 0 8733 0 9500 0 10000 0 10500 0 11000 0 11500 0 12000 0 12500 0 13000 0 13500 0 14000 0 14500 0 15000 0 15500 0 16000 0 16500 0 17000 0 17500 0 18000 0 18733 0 e Ll g m Figure 39 High Pass Filtered Signal after Peak Detector What does High Pass Filter do a Delete low freq signal which corrects drifting For signals with obvious drifting as shown in Figure 40 this is a necessary procedure Figure 40 Drifted Signal b Centralize whole waveform to zero change peak value c Add pseudo peaks at the end
82. ple points are 10000 27 NeoSuck RT CNL U of Kansas SD Waveform Graph Plot 0 Bo l l l l l l l l l l l 0 1000 2000 3000 S000 5000 6000 7000 e000 S000 10000 11000 Number of samples Figure 33 SD Plot with 100 Sample Points Type of Interpolation represents the algorithm used for normalization in STI calculation This does not normally need any tuning as the default is set to Linear STI Value displays the current STI value after the last burst selection Click either the STOP and Save STI Analysis button to save results or STOP and DON T Save STI Analysis to simply stop STI Processing see Figure 34 After one of the STOP buttons has been pressed the final STI value will be displayed under STI Value STOP and Save STI Analysis STOP and DON T Save STI Analysis Figure 34 STOP buttons Save file path and name for Original Data and Save file path and name for Normalized Data set the path and file name for the data used in STI processing Those data files may be used for processing in MATLAB as well as keeping a data file record for user s STI processing Important Note If hard copy displays of the data used for STI processing are desired for example as those represented in Figure 6 user may right click on the desired window and click Copy data to paste into another document such as Microsoft Word or Export simplified image to save as another picture f
83. rage 13 NeoFlex RT CNL U of Kansas Blocks Analysis C Program Files Neuro Logic NeoF lex RTWata t49_101405_reflex dat File Settings Motor Pulse EMG OOS R uv ACHO ACH4 EMG OOS L uv ACH ACH5 ACH2 ACHE 6 6 100 ra m 5 5 a 5 4 a 50 I 1 1 I 1 1 I 1 0 q 1 1 I 1 I I I D 20 40 60 80 100 120 140 0 20 40 60 80 100 120 140 ms ms Block Block No 6 v V Tag as bad block Saye ie First Last Block No 6 195 Suck Pressure 5 Motor On True Figure 11 Block analysis window 2 2 3 Signal average and save the output file After screening out all the bad blocks the average signal showed in this window Figure 12 1s calculated based on the entire user defined good data in window Blocks Analysis Then user could save the averaged data as txt file by go to File gt Save User could also select to see and save different average signal based on different calculation method This function is enabled by user at selection box in the lower right corner of this window Six different signal averaging methods are listed below 1 5 off The trigger level in suck pressure is set to 5 cmH20 The average signal calculated by this method doesn t include the blocks with Motor Impulse triggered In other words the blocks contributed to this average signal represent 14 NeoFlex RT CNL U of Kansas an experimental control condition in which the EMG profiles associated with suck
84. raph setting Figure 12 enables the user to define the minimum and maximum values for both x axis and y axis for each of the three graphs shown in this window 13 NeoSuck RT CNL U of Kansas FT Suck Pressure Block 5 FFT SUCK PRESSURE em H20 Figure 11 Filter setting in FFT analysis SUCK PRESSURE em H20 ma a R r wae Settings Graph Figure 12 Graph setting in FFT analysis 14 NeoSuck RT CNL U of Kansas 2 2 3 Other analysis tools There are some other analysis tools embedded in NeoSuck RT to either display graphs or export txt files which could be used for extended analyses using other commercial software like Minitab LabVIEW MATLAB MS Excel etc 1 Tools gt Bad Block List Figure 13 enables the user to delete bad blocks from the entire data set and thus generate clean analysis data for both exported data and Fourier transform analysis Bad Block List Add Remove Bad Blocks Block List Bad Block List o Add o Add Add All gt gt gt Remove Remove lt lt lt RemoveAll lt lt RemoveAll All cues Moe wr Figure 13 Bad Block List 2 Tools gt Create ASCII Blocks Figure 14 performs the function of exporting recorded dat file to txt file block by block The dat is a binary file which can be read by Visual Basic The purpose of transforming data into txt file is to make the recorded data readable by a7 party c
85. re 4 Status bar Stopping NeoSuck RT CNL U of Kansas 2 2 DAQ Data Acquisition Please make sure the status bar is running Figure 3 before Data Acquisition begins 2 2 1 Start DAQ e File path to save data enables user to choose file path where the acquired data will be saved The default path is C Program Files NeoSuck Data this is also the default directory where NeoSuck is installed e Input baby name this composes part of the file name For example if the baby name is BabySuck then the file that stores the data will be BabySuck_data txt e After labeling files press Start Data Acquisition this button enables data acquisition File path to save data 4 C Program Files NeoSuck Data l April 2007 Sun Mon Tue Wed Thu Fri Sat Start Data Acquisition Pause Stop Data Acquisition 5 1 2a 5 SF Time Elaps ea HO a et a Input baby name gt 00 00 00 0 15 16 17 18 19 20 21 __ 22 23 24 EJ 2 27 28 29 30 _ Today 4 25 2007 DAQ Settings Start DAQ Figure 5 Stat DAQ tab e Pause Stop Data Acquisition click this button whenever you want to pause or stop DAQ When you want to resume DAQ after pause just click the Start Data Acquisition button again If user pauses DAT for a while and re starts it again there will be a data break in the plot of collected data as shown in Figure 6 However this break will be eliminated in the data file i AY l l M 39 3308 39 36 08
86. rs The hardware listed above is configured up according to Figure 1 and baby s suck pressure and EMG signal sampled from the Soothie silicon nipple receiver is input to NeoSuck RT for real time digitization signal processing and waveform display NeoSuck RT CNL U of Kansas Suck I Ai3 Bridge Amplifier K Suck Pressure Ai3_ _ a Computer and DAQ LPF 50Hz ransducer EMG EMG pa aye Ai4 7 Ai4 7 EMG Ai4 7 PreAmpliiers Sensor GND is connected to baby leg Dec 1 2005 by Joan Figure 1 Hardware configuration for NeoSuck RT 1 2 Software This software is coded by Visual Basic Net 17 265 coding lines according to the structure shown in Figure 2 The function of this software includes real time data acquisition histogram processing FFT analysis ASCII file generation and export desired data to txt file NeoSuck RT CNL U of Kansas Structure Plot of NeoSuck New frmMain vb mnuFileNew Name a new dat file Open frmMain vb mnuFileOpen Open an existing dat file File Page Setup frmMain vb mnuFilePageSetup l frmPrintSelection vb Print frmMain vb mnuPrintAmplitudeHist etc ummy 1 7 Exit frmMain vb mnuFilePageSetup DAQ fmSettingsDAa vo 8 AI calibration settings for DAQ board Graph IrmSettingsGraph vb Y Axis scale setting for 8 AI plots Settings a ad Histgram E A scale setting for Histogram of Amplitude Pause and Burs
87. rst Peak is Identified after Fine Tuning This adding deleting peaks is performed by applying the zero crossing method the zero crossing is the instantaneous point at which the signal is zero What is the purpose of the extracted signal 34 NeoSuck RT CNL U of Kansas After user find out the desired starting and ending points this program will convert the original signal Figure 43 to signal in Figure 44 Important Notice This process does change some of the peak values in order to align all the bursts on the same level But this change removes the low frequency signal and is good for peak amplitude histogram calculation This process does not change the shape of the bursts peak and burst locations so it is good for STI analysis peak period histogram as well as burst pause histogram Signal Burst start Starting points and ending points on original signal Busrt end 6000 8000 10000 12000 14000 16000 18000 20000 22000 24000 26000 28000 30028 If 3j e threshy 500 67 581 1 gt PUE try fey i amp Figure 43 Original Signal 4000 Signal LN Extracted burst for peak recognition burst begin bre 20000 22000 24000 26000 28000 29998 4000 6000 8000 10000 12000 14000 16000 18000 J3 NeoSuck RT CNL U of Kansas Signal LN Extracted burst For peak recognition burst begin bre 45 0 40 0 35 05 30 0 25 0 20 0 15 0 10 0 5 0 IHi
88. ser could choose how to define the starting and ending points User could Drag the blue cursor to define the starting point of desired data or input start point to define it after switch to Define as shown in Figure 24 The Dragging cursor position in purple shows the resulted cursor position by dragging the blue cursor of samples in range defines the length ending point of desired data which is shown as the yellow cursor 20 NeoSuck RT CNL U of Kansas Suck Pressure 2 30 0 20 0 10 0 0 0 aln a 192 7 500 0 1000 0 1500 0 2000 0 2500 0 3000 0 3500 0 4000 0 4500 0 5000 0 5500 0 6000 0 6500 0 7000 0 7500 0 8000 0 8500 0 9000 0 9500 0 10000 0 10500 0 10956 1 PASP EJE 200 0 581 1 Q gt FY rry gt Egrrtend 2200 0 00 Cursor Control Drag amp D Define Jo Dragging cursor position of samples in range 200 z 2000 Figure 24 Range data selection 2 5 3 Save file To save the FFT file simply click Save power spectrum button Important note The saved file names for FFT are different for the entire data FFT and part data FFT e FFT for the entire data is named as _PowerSpectrum txt as shown in Figure 25 File path and name for Power spectrum Program Files NeoSuck Data BabySuck PowerSpectrum txt Figure 25 Saved FFT file for the whole data e FFT for part of the data is named as_ starting point _ data length _PowerSpectrum txt Figure 26 s
89. ses entieee tattle ees 14 I RECIO oe laste oi ccuehase sae E id ta Vannou tue sda easeoaomemtantaabetentne 16 NICU Database CNL U of Kansas 1 NICU Database Setup NICU Database is coded by Visual Basic 6 0 and SQL 20 469 coding lines It defines a series of tables that describe patient information like Patient background Patient experiment trial records Birth order Race Feeding mode Medication description ROP stage Oxymetry history State control Ethnicity Patient type Delivery type Neurological exam Gag reflex Pacifier size etc Please follow the steps shown below to install NICU database l Uninstall any previous installed versions of NICU Database using Control Panel gt Add Remove Programs 2 Install NICU Database by running setup exe from the installation package a Copy nicu mdb MS Access database file into folder C Program Files NICU Database Note Take a backup of MS Access database file nicu mdb periodically Nicu mdb contains all the patient data NICU Database CNL U of Kansas 2 User manual To run NICU Database double click on the program icon The main program window should appear with a menu bar along the top As shown in Figure 1 the menu bar has five active options including Patient Configuration Reports User and Help There are seven options at the bottom of the window program including View Add Modify Delete Search Sort and Refresh as shown in Figure 1
90. suck in preterm infants NTrainer R Vantipalli J Wang amp S M Barlow CNL Technical Research Report 2006 3 1 23 University of Kansas Lawrence Kansas USA Reprint available at www ku edu cnl NTrainer RT CNL U of Kansas Table of Contents Ee AND aI Cr RI SC UDe T 3 1 1 IVI VY A 2 este cieee ac A cee ameae nesta cere EE 3 1 2 SOVA O tui sciy a eatin easiest E elena vee ean eee tiaienaeee ies 5 1 3 Prora Lins fa AMOM casas ieeancnitindsgt O 6 2 NISC IVA ATM aad tessa ecstatic he ats E 8 2 1 Overview Ob Ment Dar opos sisse a E 9 22 lt Recordnew data usine NTramer RT siscccsecsssorticieteascmissriaeseecasnesiaviivieakemieruieses 16 22 How tO Create a new TiC xsicssedavunertecettnctagundedtaanidahedasaeadteceitactagetaesveaaiiededewanestexe 16 2 2 2 How to provide NTrainer RT neurotherapeutic oral stimulation 17 23 Review an NTrainer data file using NTrainer RT cccceseeeseeeeeeeeeeeeeeeeees 21 S E E ol E ee ee ee PERE E ene Pee eR 23 NTrainer RT CNL U of Kansas 1 NTrainer RT Setup NTrainer RT real time was created for simulation of suck pressure of non nutritive suck NNS which in turn is used to train the sucking ability of premature babies with no functional suck or at risk for developmental disabilities A biphasic square wave pulse pattern with a pulse frequency similar to NNS is written onto channel DACO of the NI 6052E PCI DAQ board At the same time up to three analog channels
91. t 2 NeoFlex RT User manual CNL U of Kansas When opening the NeoFlex RT a blank window appears as shown in Figure 5 User could apply this set of software and hardware to record new data or analysis existing data EH NeoFlex RT ata ee EE Motor Pulse VJ a ar O LYDT ram pa R 7 a y A a A a A U A aaa A N LOAD CELL N pa p SUCK PRESSURE em H20 EMG OOS R ux Ee AG O0S L fu a 5 E ee A A EMG OOI R u Pe Ree a AE ee ee eee en cn a a EMG OOI L u Figure 5 Blank window for recording NeoFlex RT CNL U of Kansas 2 1 Record new data by NeoFlex AT User has to input a new file name in order to Record new data Go to File gt New and File New dialog Figure 6 will pop up to enable user to input new file name and choose the desired protocol file After the data file and protocol file are selected user could click OK to Start to record EMG OOS R uv st 5 5 4 5 5 4 lt a Data File Name smb3 09 05b dat Path Ic Program Files Neuro Logic NeoFlex RT Data EMG OOS L uv Protocol File Name SMB_M ar03_ 2005 pt Browse Path Ic Program Files Neuro Logic NeoFlex RT Protocol EMG OOI R uv Ok Cancel z T Q 57 D 5 az a s fs RT Real Time Date Created Figure 6 File New dialog An active Start button and a disabled Stop button will appear on th
92. t see d e Nov 17 2005 by Joan Figure 2 Coding structure plot for NeoFlex RT NeoFlex RT CNL U of Kansas 1 3 Program Installation l Be sure to install the NI DAQ 6 9 3 driver 2 Uninstall any previous installed versions of NeoF lex RT using Control Panel gt Add Remove Programs on Install NeoFlex RT by running setup exe from the installation package Note Normally it works with BNC 2090 Figure 3 To work properly with the BNC 2110 Figure 4 please make the following change Figure 4 BNC 2110 NeoFlex RT CNL U of Kansas For BNC 2110 To work with BNC 2110 terminal configuration has to be in differential mode Terminal configuration can be set to differential mode by modifying the parameter value of defaultAiTerminalConfig to 1 under lt systemConstants gt section in the file NeoFlex exe config located in C Program Files Neuro Logic NeoFlex RT Here is the line from NeoFlex exe config that shows this parameter value lt add key defaultAiTerminalConfig value 1 gt defaultAiTerminalConfig can have the following values value 1 for Differential value 2 for Referenced single ended value 3 for Nonreferenced single ended By default the file NeoFlex exe config contains value 3 for defaultAiTerminalConfig This setting is used to work with BNC 2090 Here is the line from NeoFlex exe config that shows this parameter value lt add key defaultAiTerminalConfig value 3 g
93. t Period Suck Pr Analysis fm SattingsPeakvb Peak selection settings for suck pressure plot NeoSuck frmMain vb imMain vo Select Baseline Amplitude mAmpitudetist vb Histogram plot of cumulative amplitude Histogram Pause Period rmBurstPauseHist vb Histogram plot of cumulative bust a Burst Period frmDerivative vo Histogram plot of cumulative pause Tools ee ok ee Bad Block List AA EO A ee Bad block list selection dialog h e amen Soma _ _ _ HPF Filter frmSuckFFTSettingsFilter v ae a Settings ay e o ES frmSuckFFTSettingsGraph vb es SuckFFT ea aaa Graphlfrms O e a j settings Tools Generate Composite Create ASCII Blocks frmMain vb mnuToolsCreateASCII Create data file in txt for every block frmMain vb mnuToolsGenHistFiles Create data file in txt for histogram of Generate Histogram Files amplitude bust and pause Generate Amp Period ASCII File frmMain vb mnuToolsGenAmpPeriodFile Create data file in txt for Convert Old Format to New Format frmMain vb mnuToolsConvertOldToNew Create txt file for Help nko i m mme aw Nov 17 2005 by Joan Figure 2 Coding structure plot for NeoSuck RT 1 3 Program installation l Uninstall any previous installed versions of NeoSuck RT using Control Panel gt Add Remove Programs 2 Install NeoSuck RT by running setup exe from the installation package NeoSuck RT CNL U of Kansas Figure 3 BNC 2090 Note
94. t cycle Current sample 3 1 000 OUTPUT Display of current status within the run After completion of the run Current cycle Repeat and Current sample Desired sample 2 2 Output Channel Parameters Red Frame displays the channel number channel ID of the output port 2 3 Input Channel Parameters Green Frame the input channel should be connected to the output channel in order to display the real time signal The Advanced User may adjust the input channel number and data range for a preferred display of the real time output signal Figure 3 Input channel Voltage Channel MAWA 0 6 Amplitude v eh te o o E M O MM as 1 1 1 j i 0 6 1 1 29 30 646 PM 1 30 00 646 PM 8 15 2007 8 15 2007 Time Figure 3 Real time display of the output signal from the input channel 30s frame Final signal tab GRAPH PANEL provides the User with an overview of the final signal once the program is stopped Figure 4 Saved signal Poo B 0 4 Dzs 0 0 2 Amplitude 0 4 0 6 0 8 l l l 1 l 1 l l 1 l l 1 0 50000 300000 350000 400000 450000 500000 550000 600000 650000 700000 750000 00000 HARI aj P Figure 4 Final signal after the program is stopped l 1 l l 100000 150000 200000 250000 3 Operate Program e After the all the parameters have been set up clicking the arrow button under the menu bar to make the program running when the program running the arrow button
95. tritive suck in preterm infants J Neonatal Nursing in press Zimmerman E Barlow SM Lee J 2008 Pacifier stiffness alters the dynamics of the suck central pattern generator J Neonatal Nursing in press Estep M Barlow SM Vantipalli R Lee J Finan D 2008 Non nutritive suck burst parametrics in preterm infants with RDS and oral feeding complications J Neonatal Nursing accepted Popescu EA Popescu M Wang J Barlow SM Gustafson KM 2007 Non nutritive sucking recorded in utero via fetal magnetography Physiological Measurement in review 18 NeoSuck RT LabView Communication Neuroscience Laboratories University of Kansas Lawrence Kansas USA June 2007 NeoSuck RT CNL U of Kansas Table of Contents 1 NEOSUCK RESETUP sisi casesceseussitciardiavecuatitiveniicteteiiirinnniicnlnaiitiediinnniinnints 4 1 1 FARON ARE aoa A E A Daan areas tener usta oe 4 12 SOPIVAN E a aru ap ict tata aaa ane ni santo asia ai oak nia otal un Soamale ine aiee 5 1 3 INSTAL TATIONMJINS PRUCTIONS irnia i insets bashad am Ona N A 5 2 USER IWAN GAD iraeiccs eeckittecsstaatescea ara poa e aoe aa aa sunssoessensvusetereustsnsseceas 6 2l BASI COF LABVIEW TSE oirre Or T T A E dansaasgunaetetes 7 Zee IDA DATA ACOUISETION heuir er a a o a a a a hase 8 GA SODA O EE E T E a Bae cence Dae dtcee i alc ig Balen ecees tan Dalene ier ten 8 22a PAO SCL aca ees E E E RA 9 2 3 BORS TEX TRAC HON arra ae A te ool atstuasutlaune eueantewats geleeamusaune
96. uck Journal of Perinatology in review Poore M Zimmerman E Barlow SM Wang J Gu F 2007 NTrainer therapy increases suck spatiotemporal stability in preterm infants Acta Paediatrica in review 23 NTrainer Adult Version Communication Neuroscience Laboratories University of Kansas Lawrence Kansas USA August 2007 NTrainer Adult Version performs the following functions e Generate three different impulse trains at 2 Hz 4 Hz and 8Hz e Program produces a User defined number of repetitions 1 Create File Name The default file path is set to C Program Files NTrainer_adult data File path to save data T Oo Program FilesiNTrainer_adultidata oa User needs to input a subject ID which composes the data file name Subject ID The file name used to store data will be created as subject ID _ impulse selector _ Repeat _data txt For example if the subject ID is John impulse selector is set as 2Hz Repeat is 100 then the file name will be John_2Hz_100_data txt The final file path and name will also be shown at the bottom of this program in purple File path and name Program Files NTrainer_adult data John_ 2 Hz_100_data txt 2 Set Parameters User may input pulse parameters in blue fields to determine stimuli characteristics Figure 1 8 Impulse Generator Parameters Output Channel Parameters Input Channel Parameters Physical Channel impulse selector
97. ughly judge if all the desired peaks are detected If there are tiny peaks at the beginning of the burst that have been missed the program will pick them in the next step Recognized burst check the number of bursts with the raw signal If they don t match one another tune the value of the peak detector width distance between burst and cursor position for threshold until the desired bursts number is displayed _ Peak detector width Distance between bursts Last peak to burst ending Peaks Found Recognized bursts JA J200 gso ize i3 0 0 1000 0 2000 0 3000 0 4000 0 5000 0 6000 0 7000 0 8000 0 9000 0 10000 0 11000 0 12000 0 13000 0 14000 0 15000 0 16000 0 17000 0 18000 0 Me A urs location 284 5 0 0 Cursor Drag Define i8 rio vty th a J Finish Burst Selection Figure 14 Burst finding on High passed signal Peak Detect HPF Original Signal After desired bursts are identified press the Finish Burst Selection button to stop this step and start Step 4 Fine tuning Burst Start Re recognition 13 NeoSuck RT CNL U of Kansas Finish Burst Selection Notice this button must be pushed before the next step can be started 2 3 4 Step 4 Fine tuning Burst Start Re recognition In Step 3 burst location is roughly identified although sometimes tiny peaks at the beginning of bursts may be missed Therefore this step is used to select the missing peaks
98. vert Old Format to New Format 5 0 0 Ty fon 200 0 LYDT mm EMG OOS L uv EES 50 04 eS w a 5 0 0 a a 1 5 50 0 LOAD CELL N EMG OOI R uv 50 0 O 15 gt N w 5 5 0 0 1 o 1 lt lt 0 05 i i i 50 0 i i i SUCK PRESSURE em H20 EMG OOI L u 20 0 100 0 ia m q z 1 10 0 F 0 0 0 0 1 1 I l l I I 0 0 10 0 20 0 30 0 0 0 10 0 20 6 30 0 Time s Time s Block NeoSuck RT Block No V Real Time i 3 z Previous First Last Block 3 6 Peak Count 29 Date Created 3 30 2004 1 53 32 PM Figure 9 NeoSuck window without Histogram analysis Similar to the recording process NeoSuck RT analyzes data in successive blocks 30 seconds for each block User can scroll among different blocks by clicking the Previous and Next buttons in the lower part of the window as shown in Figure 9 NeoSuck RT is mainly for suck pressure analysis which is shown as the last graph in the first column NeoSuck RT can automatically recognize the peak in suck pressure which is marked by small diagonal red lines in suck pressure plot 2 2 2 Fourier transform of suck pressure NeoSuck RT can be used to calculate the FFT of suck pressure In order to get the frequency property of suck pressure the user can click Tools gt Suck FFT to call up the FFT Suck Pressure window In this analysis window the user will find the time domain and 12 NeoSuck RT CNL U of Kansas frequency d

Binded Manual - The University of Kansas

Contents

Download Pdf Manuals

Related Search

Related Contents