Validation of a video analysis software package for quantifying movement velocity in resistance exercises

Jou nal o S eng h and Condi ioning Resea ch Publish Ahead o P in DOI: 10.1519/JSC.0000000000000563 ACCEPTED Copy igh Ó Lippinco Williams & Wilkins. All igh s ese ed. VALIDATION OF A VIDEO ANALYSIS SOFTWARE PACKAGE FOR QUANTIFYING MOVEMENT VELOCITY IN RESISTANCE EXERCISES Sho unning head: Valida ion o ideo analysis in bench p ess Bo ja Sañudo 1 , Da id Rueda 1 , Bo ja del Pozo-C uz 1 , Moisés de Hoyo 1 , Luis Ca asco 1 1 Depa men o Physical Educa ion and Spo . Uni e si y o Se ille, Se ille, Spain. Disclaime s: none Sou ce o suppo : Uni e si y o Se ille Co esponding Au ho : Bo ja Sañudo, PhD Physical Educa ion and Spo s Depa men Uni e si y o Se ille Pi o écnia s/n, N-41012 Se ille Spain Tel: ++34 652 387090 Email: bsanco[email p o ec ed] Abs ac : The aim o his s udy was o es ablish he alidi y o a ideo analysis so wa e package in measu ing mean p opulsi e eloci y (MPV) and he maximal eloci y du ing bench p ess. Twen y-one heal hy males (21 ± 1 yea ) wi h weigh aining expe ience we e ec ui ed and he MPV and he maximal eloci y o he concen ic phas e (Vmax) we e compa ed o a linea posi ion ansduce sys em du ing a s anda d bench p ess exe cise. Pa icipan s pe o med a one epe i ion maximum (1RM) es using he supine bench p ess exe cise. The es ing p ocedu es in ol ed he simul aneous assessmen o bench p ess p opulsi e eloci y using wo kinema ic (linea posi ion ansduce and semi-au oma ed acking so wa e) sys ems. High Pea son´ co ela ion coe icien s o MPV and Vmax be ween bo h de ices ( = .473 o .993) we e obse ed. The in aclass co ela ion coe icien s (ICCs) o ba bell eloci y da a and he kinema ic da a ob ained om ideo analysis we e high (>0.79). In addi ion, he low coe icien s o a ia ion indica e ha measu emen s had low a iabili y. Finally, Bland-Al man plo s wi h he limi s o ag eemen o he MPV and Vmax wi h di e en loads showed a nega i e end which indica ed ha he ideo analysis had highe alues han he linea ansduce . In conclusion, his s udy has demons a ed ha he so wa e employed o he ideo analysis was an easy o use and cos -e ec i e ool wi h a e y high deg ee o concu en alidi y. This so wa e can be used o e alua e changes in eloci y o aining load in esis ance aining which may be impo an o he p esc ip ion and moni o ing o aining p og ammes. Keywo ds: aining in ensi y; alidi y; Resis ance T aining; bench p ess ACCEPTED Copy igh Ó Lippinco Williams & Wilkins. All igh s ese ed. ACCEPTED Copy igh Ó Lippinco Williams & Wilkins. All igh s ese ed. INTRODUCTION The con ol o he aining load du ing esis ance aining is one o he main conce ns o inc easing he a hle es’ pe o mance (9) and bo h coaches and a hle es modi y he ype o exe cise, olume and in ensi y o ge he be e esul s. Exe cise in ensi y is gene ally ecognized o be he mos impo an s imulus associa ed o changes in s eng h le els (7). The mos commonly used me hod o assessing he s eng h aining in ensi y is he pe cen age o he one epe i ion maximum (1RM), conside ed o be he mos accu a e me hod o de e mine maximal dynamic s eng h. O he app oaches o de e mine exe cise in ensi y include he pe o mance o a gi en maximal numbe o epe i ions in each se (XRM) o iden i ying a ela i e load using epe i ions pe o med o a igue. Howe e , he e a e some inhe en complica ions associa ed wi h hese me hods, he 1RM assessmen may be e y ime- consuming, i may be associa ed wi h inju y when pe o med inco ec ly o by no ice subjec s and is imp ac ical o la ge g oups (9). These limi a ions encou age he de elopmen o new me hods o objec i ely moni o aining load du ing esis ance exe cise. Many di e en p o ocols and de ices ha e been used o assess muscle pe o mance, including he use o isokine ic dynamome y, linea posi ion ansduce s, accele ome e s o o ce pla o ms, howe e , he cos o such equipmen , i s size, and expe imen al equi emen s limi i s applica ion o lab-based assessmen s (5). Due o he a o emen ioned limi a ions he e is a need o easy- o-use es ing ins umen s o cap u e and e alua e a hle es´ pe o mance a he aining si e, wi h eal- ime eedback o coaches and a hle es (16). The analyses o kinema ic a iables a e becoming inc easingly accep ed o es ima ing he o ce and powe ou pu s wi h exe cise (4). Mo emen eloci y is inc easingly being used o moni o exe cise in ensi y because mo emen eloci y has a di ec ela ionship wi h he amoun o o ce exe ed (13,15). Mo emen eloci y can also be used as a good es ima e o a ela i e load (% 1RM), and is addi ionally a p ecise indica o o he ACCEPTED Copy igh Ó Lippinco Williams & Wilkins. All igh s ese ed. neu omuscula a igue induced by esis ance aining (9). Howe e , he e has been ela i ely li le discussion in he li e a u e on mo emen eloci y (8). To ou knowledge only one p e ious s udy has no ed ha high-speed came as can be a use ul sys em o p o ide a eal- ime eloci y acking (17). Sa o e al. (17) in es ing a sna ch li ound a high co ela ion be ween high speed came a and a wi eless accele ome e , which has p e iously been epo ed o be a alid de ice o calcula ing o ce, eloci y and powe (1,10). The e is a need o s eng h and condi ioning p o essionals o isualize wo kou pe o mances and o ack aining p og ess (2), he e o e he de elopmen o po able, cos -e ec i e equipmen ha allows kinema ic in o ma ion o be isualized would ha e ob ious ad an ages in a ield- es ing si ua ion (5). Any such equipmen , howe e , needs o be alid and eliable. The e o e, in his s udy we es ed he alidi y o eely a ailable so wa e (Kino ea, e sion 0.8.15.) ha allows semi-au oma ed acking o objec s, speci ically we used his so wa e o measu e mean eloci y in he bench p ess exe cise. Thus, he aim o he cu en s udy was o conduc a concu en alidi y analysis o he Kino ea so wa e package in measu ing mean p opulsi e eloci y (MPV) and he maximal eloci y o he concen ic phase (Vmax) as compa ed o a linea posi ion ansduce sys em du ing bench p ess in weigh - ained males. METHODS Expe imen al App oach o he P oblem The in a-session alidi y o a linea posi ion ansduce sys em and a digi izing so wa e p og am was calcula ed when analyzing speed– ime a iables. Twen y-one subjec s pe o med 1RM es ing using he supine bench p ess exe cise. Da a we e collec ed simul aneously wi h a linea posi ion ansduce and a digi al ideo came a, MPV and Vmax we e de e mined o each epe i ion o he bench p ess. The alidi y o a eely a ailable digi izing so wa e p og am in measu ing he bench p ess pe o mance was de e mined by compa ing he MPV measu emen s wi h da a ob ained simul aneously wi h a linea posi ion ansduce . ACCEPTED Copy igh Ó Lippinco Williams & Wilkins. All igh s ese ed. Subjec s Twen y-one heal hy males wi h a mean (± SD) age, heigh and body mass o 21 ± 1 yea , 176.1 ± 4.1 cm and 72.3 ± 7.0 kg, espec i ely we e ec ui ed. The c i e ia o s udy inclusion we e a weigh aining backg ound o a minimum o h ee yea s, display a p ope echnique in bench p ess and no inju ies o condi ions ha would p e en indi iduals om sa ely unde aking he es ing p ocedu es. The pa icipan s we e no i ied abou he po en ial isks in ol ed and p o ided hei w i en in o med consen . P ocedu es we e app o ed by he ins i u ional Human Resea ch E hics Commi ee. P ocedu es Pa icipan s lay supine on a la bench, wi h hei ee es ing on he bench, and hands placed on he ba bell sligh ly wide han shoulde wid h. Posi ions on he bench and g ip wid hs we e measu ed so ha hey could be indi idually ep oduced on e e y li . Pa icipan s pe o med a s anda dized wa m-up consis ing o 5 min o pedaling on a cycle e gome e (Ke le Axiom P2, GmbH & Co.KG, Ense-Pa si , Ge many) a a load o 50 W, and uppe -body join mobiliza ion exe cises. The wa m up was ollowed by one se o 5 epe i ions o bench p ess wi h a ixed load o 20 kg. Pa icipan s we e gi en ins uc ions on p ope li ing echniques and assessmen p ocedu es, and we e also amilia ized wi h he es ing p o ocols and allowed o p ac ice he es s o elimina e any neu ological lea ning e ec s. In addi ion, each subjec was ca e ully ins uc ed o always pe o m each bench p ess in an explosi e manne , exploding he ba o he ches as as as possible. A Smi h machine (Mul ipowe , Technogym, Spain) was used o all es s. The es ing p ocedu es in ol ed he simul aneous assessmen o bench p ess p opulsi e eloci y using wo kinema ic (linea posi ion ansduce and a semi-au oma ed acking so wa e) sys ems. ACCEPTED Copy igh Ó Lippinco Williams & Wilkins. All igh s ese ed. One Repe i ion Maximum Tes ing A 1RM using he supine bench p ess exe cise was used o assess uppe -body s eng h in each pa icipan . Ini ial load was se a 20 kg o all subjec s, and was p og essi ely inc eased in 10 kg inc emen s un il he a ained MPV was lowe han 0.5 m·s −1 . The ea e , load was adjus ed wi h smalle inc emen s (2.5 – 5 kg). This p ocedu e con inued un il he pa icipan was no able o comple e a single epe i ion h ough he ull ange o mo ion ollowing he p ocedu es epo ed elsewhe e (15). Res pe iods o 2 minu es we e allowed be ween ials. Only he bes epe i ion a each load, acco ding o he c i e ia o as es MPV, was conside ed o analysis. Da a Collec ion and Analysis All es ing was pe o med wi h he igh side o he ba bell a ached o one linea posi ion ansduce (T-Fo ce, T-Fo ce Sys em E go ech, Mu cia, Spain). This sys em consis s o a cable- ex ension linea eloci y ansduce in e aced o a pe sonal compu e by means o a 14-bi esolu ion analog- o-digi al da a acquisi ion boa d. Ve ical ins an aneous eloci y was sampled a a equency o 1000 Hz. Simul aneously; a digi al ideo came a (sampling equency o 50Hz) was placed a 1 m heigh a a dis ance o 2 m om he Smi h machine and aligned pe pendicula ly o he sagi al plane. The digi al images we e impo ed in o a eely a ailable digi izing so wa e p og am (Kino ea 0.8.15, www.kino ea.o g ). Based on a semi- au oma ic acking unc ion, he so wa e allows he calcula ion o he ele an kinema ic pa ame e s o e e y epe i ion, p o iding eal ime in o ma ion on sc een. In he cu en s udy, a e he came a se -up was comple ed, a clea e ical and ho izon al e e ence in he plane o mo ion was eco ded. To minimize he e o in he scaling p ocess, he dimensions o he Smi h machine we e eco ded (me e s high x guide ba me e s long). Ins an aneous ba eloci y was calcula ed o each ime in e al as ba displacemen o e change in ime. MPV and Vmax o each mo emen we e ob ained o he concen ic po ion ACCEPTED Copy igh Ó Lippinco Williams & Wilkins. All igh s ese ed. o he mo emen . F om T-Fo ce and ideo da a, eloci y- ime cu es we e c ea ed and Peak alues we e hen ob ained o compa ison. S a is ical analyses Two-dimensional coo dina es esul ing om he digi izing p ocess ( aw da a) we e hen scaled and smoo hed using a Bu e wo h low pass il e wi h cu o 6 - 12 Hz be o e eloci ies we e calcula ed. Pai ed - es compa isons we e used o compa e T-Fo ce and Kino ea esul s. A eg ession analysis was pe o med o es ing he linea ela ionship be ween he so wa e semi-au oma ed acking unc ion and he linea ansduce mean eloci y and mean p opulsi e eloci y alues. The goodness o i and he slope wi h he 95% o con idence in e al we e hen calcula ed. Finally, o es he ag eemen be ween bo h de ices, Bland- Al man plo s we e employed. S a is ical signi icance was assumed a p < 0.05. All analyses we e pe o med using SPSS .18 (SPSS Inc., Chicago, IL). RESULTS Desc ip i e s a is ics o he pa icipan s a e shown in Table 1. Pa icipan s had a BMI o 23.3 ± 2.5 kg/m 2 and displayed a Maximal P opulsi e Powe (PMP) in he concen ic phase o 255.1 ± 74.9 wa s. The analysis o he MPV and Vmax wi h he linea ansduce and he ideo analysis so wa e o all he di e en loads a e epo ed in Table 2. The linea ela ionships o each es wi h bo h de ices a e p esen ed in Table 3, along wi h he slope (95% con idence in e al –CI-), he 95% CIs o he limi s o ag eemen om he Bland-Al man analyses and he y-in e cep alues. The ela ionship o ba bell eloci y da a and he kinema ic da a ob ained om ideo analysis we e high (>0.84) o MPV wi h he excep ion o 20 kg ( 2 = 0.45) and anged om 0.41 (20 Kg) o 0.98 (80 Kg) o Vmax. In addi ion, eg ession plo s wi h he s anda d eg ession line o bes i o MPV and Vmax a e p esen ed in Figu es 1 and 2 espec i ely. ACCEPTED Copy igh Ó Lippinco Williams & Wilkins. All igh s ese ed. Figu e 3 shows Bland-Al man plo s wi h he limi s o ag eemen o he MPV wi h di e en loads. The bias ep esen ing he a e age di e ence o measu es be ween bo h de ices was nega i e which indica e ha he ideo analysis had highe alues han he linea ansduce . The same end can be obse ed o Vmax (Figu e 4). DISCUSSION This s udy was designed o alida e measu es ob ained by kinema ic da a de i ed om ideo analysis in compa ison wi h a linea ansduce . The p ima y inding was ha he analysis om he Kino ea so wa e demons a ed high concu en alidi y when compa ed o he labo a o y-based ins umen a ion. Moni o ing he ba eloci y du ing esis ance exe cise (bench p ess) is essen ial since he aining e ec depends on he eloci y a which loads a e li ed (8). In he cu en s udy, he high co ela ion coe icien s ob ained o MPV and Vmax indica e ha he da a ob ained wi h bo h me hods show a high le el o associa ion. The close ela ionship obse ed be ween he MPV in bo h me hods indica e ha he ideo analysis may also allow us o accu a ely de e mine he eal in ensi y o e o du ing exe cise (9). Simila alues o he cu en s udy we e epo ed by Coms ock e al. (2) when alida ing an accele ome e du ing bench p ess. I is impo an o no e ha lowe co ela ion coe icien s we e ob ained wi h 20 Kg ( = .676 and = .473) and 30 Kg ( = .921 and = .594) o MPV and Vmax espec i ely, bu co ela ion we e be e wi h he hea ie loads es ed, which is consis en wi h he esul s obse ed by C ew he e al. (4). One possible explana ion o hese esul s may be echnique modi ica ions adop ed by pa icipan s wi h low loads (6), his may lead o he use o di e en muscles and/o biomechanics du ing he bench p ess. Despi e he good co ela ion be ween he wo me hods, he 95% CIs o he slope alues om he eg essions and he - es s om he mean compa isons may indica e an o e -p edic ion o he ideo analyses ela i e o he ansduce . In addi ion, cau ion exe cised in in e p e ing he ACCEPTED Copy igh Ó Lippinco Williams & Wilkins. All igh s ese ed. da a as ho izon al mo emen s o he ba away om he e ical ec o , especially du ing ee- weigh exe cises, can modi y he esul s and may dis up he accu acy o eloci y assessmen . Howe e , in he cu en s udy hese issues we e elimina ed by using he Smi h machine. Al hough i is beyond he scope o he cu en s udy, p e ious s udies demons a ed ha a change in ho izon al displacemen o a weigh ba may esul in o e es ima ions o e ical eloci y (3). Howe e , wi h he au o- acking module o he Kino ea so wa e hese de ia ions can be iden i ied and can be minimized. Fu u e s udies a e wa an ed o add ess he alidi y o each kinema ic sys em du ing exe cise pe o med in mul iple planes (4). Ano he in e es ing inding o he p esen s udy is ha high eg ession coe icien s also indica e a high ag eemen be ween bo h de ices (11). Toge he wi h he Bland–Al man analysis i can be con i med he good consis ency, as esul s om his analysis showed ha sys ema ic e o s (mean di e ence be ween de ices) o he kinema ics ou comes assessed we e nea ly ze o and he 95% limi s o ag eemen we e na ow. Howe e i is impo an o no e ha he da a ob ained om he ideo analysis o e es ima ed he eloci y. The e o e, while he wo me hods a e co ela ed wi h each o he , he ag eemen be ween he measu es can be conside ed as modes . The di e en eloci ies o da a collec ion (linea ansduce ; 1000 Hz compa ed o ideo analysis; 50 Hz) may explain hese a ia ions. The e a e some limi a ions when using he so wa e o ideo analysis. Fi s , he e is a need o s anda dize he mo emen as da a a e di ec ionally dependen and, as s a ed ea lie , i is usually ecommended ha he mo emen is pe o med on a Smi h machine. The Smi h machine which elimina e sagi al mo emen o he ba bell du ing he es . This sou ce o e o can also be ela ed o he expe ience o he li e s. Fu he s udy would be necessa y o de e mine he in luence o hese issues on he kinema ic a iables measu ed. Second, i should be no ed ha possible di e ences in he echnical cha ac e is ics o bo h de ices (e.g. sampling a e) may lead o di e en esul s. ACCEPTED Copy igh Ó Lippinco Williams & Wilkins. All igh s ese ed. Table 3. Linea ela ionship be ween he MVP and Vmax alues as measu ed by ideo analysis and he linea ansduce . Load Ou come n Bias (95%CI) slope (95%CI) y-in e cep (95%CI) Goodness o Fi ( 2 ) P alue (slope) MPV (m s -1 ) 21 -0.43 (-0.68 o -0.17) 0.51 (0.24 o 0.77) 0.29 (-0.10 o 0.69) 0.45 0.0008 20 Kg VMAX (m s -1 ) 21 -0.57 (-1.14 o -0.01) 0.41 (0.17 o 0.65) 0.70 (0.18 o 1.23) 0.41 0.0017 MPV (m s -1 ) 21 -0.41 (-0.53 o -0.28) 0.80 (0.64 o 0.97) -0.16 (-0.37 o 0.05) 0.84 <0.0001 30 Kg VMAX (m s -1 ) 21 -0.59 (-1.01 o -0.16) 0.40 (0.14 o 0.66) 0.55 (0.04 o 1.06) 0.35 0.0045 MPV (m s -1 ) 21 -0.30 (-0.41 o -0.20) 0.79 (0.70 o 0.87) -0.09 (-0.18 o -0.008) 0.95 <0.0001 40 Kg VMAX (m s -1 ) 21 -0.42 (-0.67 o -0.16) 0.61 (0.50 o 0.71) 0.17 (0.008 o 0.34) 0.88 <0.0001 MPV (m s -1 ) 20 -0.23 (-0.35 o -0.12) 0.73 (0.68 o 0.78) -0.03 (-0.07 o 0.005) 0.98 <0.0001 50 Kg VMAX (m s -1 ) 20 -0.36 (-0.53 o -0.19) 0.70 (0.62 o 0.79) 0.002 (-0.10 o 0.11) 0.94 <0.0001 MPV (m s -1 ) 19 -0.16 (-0.25 o -0.06) 0.73 (0.68 o 0.77) -0.01 (-0.04 o 0.01) 0.98 <0.0001 60 Kg VMAX (m s -1 ) 19 -0.28 (-0.46 o -0.09) 0.63 (0.53 o 0.74) 0.07 (-0.02 o 0.18) 0.91 <0.0001 MPV (m s -1 ) 8 -0.14 (-0.29 o -0.003) 0.73 (0.65 o 0.81) -0.02 (-0.08 o 0.04) 0.98 <0.0001 70 Kg VMAX (m s -1 ) 8 -0.28 (-0.64 o 0.08) 0.61 (0.43 o 0.80) 0.09 (-0.11 o 0.31) 0.89 <0.0001 MPV (m s -1 ) 4 -0.16 (-0.31 o -0.01) 0.60 (0.46 o 0.73) 0.02 (-0.04 o 0.09) 0.99 0.0046 80 Kg VMAX (m s -1 ) 4 -0.23 (-0.33 o -0.13) 0.81 (-0.53 o 2.15) -0.07 (-1.23 o 1.09) 0.98 0.0821 MPV: mean p opulsi e eloci y; VMAX: maximal eloci y o he concen ic phase; Copy igh Ó Lippinco Williams & Wilkins. All igh s ese ed. ACCEPTED Copy igh Ó Lippinco Williams & Wilkins. All igh s ese ed. ACCEPTED Copy igh Ó Lippinco Williams & Wilkins. All igh s ese ed. ACCEPTED Copy igh Ó Lippinco Williams & Wilkins. All igh s ese ed. ACCEPTED