Step Length Classification Using Decision Tree Based on IMU Sensors and Body Height

Step Length Classification Using Decision Tree Based on IMU Sensors and Body Height

The ability to walk independently contributes significantly to the protective health benefits and complications reduction in stroke rehabilitation. Taskoriented training involves the repetition of functional activities to improve walking ability and is considered a critical strategy. Human gait anal...

Full description

Bibliographic Details
Published in:Proceedings - 2024 12th Electrical Power, Electronics, Communications, Controls and Informatics Seminar, EECCIS 2024
Main Author: Zaeni I.A.E.; Lestari D.; Mustika S.N.; Rif'a Anzani D.; Osman M.K.; Ahmad K.A.
Format: Conference paper
Language:English
Published: Institute of Electrical and Electronics Engineers Inc. 2024
Online Access:https://www.scopus.com/inward/record.uri?eid=2-s2.0-85217999543&doi=10.1109%2fEECCIS62037.2024.10840151&partnerID=40&md5=a28944d0b44beb72880a98090feef4f6
id 2-s2.0-85217999543
spelling 2-s2.0-85217999543
Zaeni I.A.E.; Lestari D.; Mustika S.N.; Rif'a Anzani D.; Osman M.K.; Ahmad K.A.
Step Length Classification Using Decision Tree Based on IMU Sensors and Body Height
2024
Proceedings - 2024 12th Electrical Power, Electronics, Communications, Controls and Informatics Seminar, EECCIS 2024


10.1109/EECCIS62037.2024.10840151
https://www.scopus.com/inward/record.uri?eid=2-s2.0-85217999543&doi=10.1109%2fEECCIS62037.2024.10840151&partnerID=40&md5=a28944d0b44beb72880a98090feef4f6
The ability to walk independently contributes significantly to the protective health benefits and complications reduction in stroke rehabilitation. Taskoriented training involves the repetition of functional activities to improve walking ability and is considered a critical strategy. Human gait analysis, especially step length measurement, plays an important role in monitoring mobility and rehabilitation progress, though practical application in everyday life can seldom be achieved by means of conventional methods. This paper presents the confirmation that step length classification is achievable using Inertial Measurement Unit sensors. Feature extraction, filtering, and preprocessing were done on the data from the accelerometer to enhance the accuracy of the classification model. A decision tree algorithm was used for its simplicity and interpretability to classify the step lengths into short (22 cm), medium (33 cm), and long (44 cm). Three confusion matrices are presented, showing the performance of three different models in classifying data into classes 0, 1, and 2. For example, the first model is most accurate for class 0, at an accuracy of 34.88%, but has difficulties classifying classes 1 and 2 at an accuracy of 12.40% and 15.50%, with considerable misclassifications each. Then, the second model yields higher results for both classes 1 and 2 with accuracy of 19.38% each, though it still shows high misclassifications. The third model is the most accurate and balanced; class 0 is classified at 24.07%, class 1 at 25.93%, and class 2 at 27.16%, with low misclassification rates. On the whole, this model is outstanding since it offers the most reliable classification results of all the models. © 2024 IEEE.
Institute of Electrical and Electronics Engineers Inc.

English
Conference paper
author Zaeni I.A.E.; Lestari D.; Mustika S.N.; Rif'a Anzani D.; Osman M.K.; Ahmad K.A.
spellingShingle Zaeni I.A.E.; Lestari D.; Mustika S.N.; Rif'a Anzani D.; Osman M.K.; Ahmad K.A.
Step Length Classification Using Decision Tree Based on IMU Sensors and Body Height
author_facet Zaeni I.A.E.; Lestari D.; Mustika S.N.; Rif'a Anzani D.; Osman M.K.; Ahmad K.A.
author_sort Zaeni I.A.E.; Lestari D.; Mustika S.N.; Rif'a Anzani D.; Osman M.K.; Ahmad K.A.
title Step Length Classification Using Decision Tree Based on IMU Sensors and Body Height
title_short Step Length Classification Using Decision Tree Based on IMU Sensors and Body Height
title_full Step Length Classification Using Decision Tree Based on IMU Sensors and Body Height
title_fullStr Step Length Classification Using Decision Tree Based on IMU Sensors and Body Height
title_full_unstemmed Step Length Classification Using Decision Tree Based on IMU Sensors and Body Height
title_sort Step Length Classification Using Decision Tree Based on IMU Sensors and Body Height
publishDate 2024
container_title Proceedings - 2024 12th Electrical Power, Electronics, Communications, Controls and Informatics Seminar, EECCIS 2024
container_volume
container_issue
doi_str_mv 10.1109/EECCIS62037.2024.10840151
url https://www.scopus.com/inward/record.uri?eid=2-s2.0-85217999543&doi=10.1109%2fEECCIS62037.2024.10840151&partnerID=40&md5=a28944d0b44beb72880a98090feef4f6
description The ability to walk independently contributes significantly to the protective health benefits and complications reduction in stroke rehabilitation. Taskoriented training involves the repetition of functional activities to improve walking ability and is considered a critical strategy. Human gait analysis, especially step length measurement, plays an important role in monitoring mobility and rehabilitation progress, though practical application in everyday life can seldom be achieved by means of conventional methods. This paper presents the confirmation that step length classification is achievable using Inertial Measurement Unit sensors. Feature extraction, filtering, and preprocessing were done on the data from the accelerometer to enhance the accuracy of the classification model. A decision tree algorithm was used for its simplicity and interpretability to classify the step lengths into short (22 cm), medium (33 cm), and long (44 cm). Three confusion matrices are presented, showing the performance of three different models in classifying data into classes 0, 1, and 2. For example, the first model is most accurate for class 0, at an accuracy of 34.88%, but has difficulties classifying classes 1 and 2 at an accuracy of 12.40% and 15.50%, with considerable misclassifications each. Then, the second model yields higher results for both classes 1 and 2 with accuracy of 19.38% each, though it still shows high misclassifications. The third model is the most accurate and balanced; class 0 is classified at 24.07%, class 1 at 25.93%, and class 2 at 27.16%, with low misclassification rates. On the whole, this model is outstanding since it offers the most reliable classification results of all the models. © 2024 IEEE.
publisher Institute of Electrical and Electronics Engineers Inc.
issn
language English
format Conference paper
accesstype
record_format scopus
collection Scopus
_version_ 1825722578886459392

Similar Items

  • Personal bankruptcy prediction using decision tree model
    by: Syed Nor S.H.; Ismail S.; Yap B.W.
    Published: (2019)
  • Evaluation of support vector machine and decision tree for emotion recognition of malay folklores
    by: Saad M.M.; Jamil N.; Hamzah R.
    Published: (2018)
  • Navigation System in Dynamic Indoor Environment Using Deep Reinforcement Learning with Actor-Critic
    by: Sendari S.; Muladi; Ardiyansyah F.; Zaeni I.A.E.; Widyaningtyas T.; Wibowo D.A.; Yuztar M.; Setumin S.
    Published: (2024)
  • A century of trends in adult human height
    by: Bentham J.; Di Cesare M.; Stevens G.A.; Zhou B.; Bixby H.; Cowan M.; Fortunato L.; Bennett J.E.; Danaei G.; Hajifathalian K.; Lu Y.; Riley L.M.; Laxmaiah A.; Kontis V.; Paciorek C.J.; Riboli E.; Ezzati M.; Abdeen Z.A.; Hamid Z.A.; Abu-Rmeileh N.M.; Acosta-Cazares B.; Adams R.; Aekplakorn W.; Aguilar-Salinas C.A.; Agyemang C.; Ahmadvand A.; Ahrens W.; Alhazzaa H.M.; Al-Othman A.R.; Raddadi R.A.; Ali M.M.; Alkerwi A.; Alvarez-Pedrerol M.; Aly E.; Amouyel P.; Amuzu A.; Andersen L.B.; Anderssen S.A.; Anjana R.M.; Aounallah-Skhiri H.; Ariansen I.; Aris T.; Arlappa N.; Arveiler D.; Assah F.K.; Avdicová M.; Azizi F.; Babu B.V.; Bahijri S.; Balakrishna N.; Bandosz P.; Banegas J.R.; Barbagallo C.M.; Barceló A.; Barkat A.; Barros M.V.; Bata I.; Batieha A.M.; Batista R.L.; Baur L.A.; Beaglehole R.; Romdhane H.B.; Benet M.; Bernabe-Ortiz A.; Bernotiene G.; Bettiol H.; Bhagyalaxmi A.; Bharadwaj S.; Bhargava S.K.; Bhatti Z.; Bhutta Z.A.; Bi H.; Bi Y.; Bjerregaard P.; Bjertness E.; Bjertness M.B.; Björkelund C.; Blokstra A.; Bo S.; Bobak M.; Boddy L.M.; Boehm B.O.; Boeing H.; Boissonnet C.P.; Bongard V.; Bovet P.; Braeckman L.; Bragt M.C.E.; Brajkovich I.; Branca F.; Breckenkamp J.; Brenner H.; Brewster L.M.; Brian G.R.; Bruno G.; Bueno-De-mesquita H.B.; Bugge A.; Burns C.; de León A.C.; Cacciottolo J.; Cama T.; Cameron C.; Camolas J.; Can G.; Cândido A.P.C.; Capuano V.; Cardoso V.C.; Carlsson A.C.; Carvalho M.J.; Casanueva F.F.; Casas J.-P.; Caserta C.A.; Chamukuttan S.; Chan A.W.; Chan Q.; Chaturvedi H.K.; Chaturvedi N.; Chen C.-J.; Chen F.; Chen H.; Chen S.; Chen Z.; Cheng C.-Y.; Chetrit A.; Chiolero A.; Chiou S.-T.; Chirita-Emandi A.; Cho B.; Cho Y.; Christensen K.; Chudek J.; Cifkova R.; Claessens F.; Clays E.; Concin H.; Cooper C.; Cooper R.; Coppinger T.C.; Costanzo S.; Cottel D.; Cowell C.; Craig C.L.; Crujeiras A.B.; D’Arrigo G.; D’Orsi E.; Dallongeville J.; Damasceno A.; Damsgaard C.T.; Dankner R.; Dauchet L.; De Backer G.; De Bacquer D.; de Gaetano G.; De Henauw S.; De Smedt D.; Deepa M.; Deev A.D.; Dehghan A.; Delisle H.; Delpeuch F.; Deschamps V.; Dhana K.; Di Castelnuovo A.F.; Dias-Da-costa J.S.; Diaz A.; Djalalinia S.; Do H.T.P.; Dobson A.J.; Donfrancesco C.; Donoso S.P.; Döring A.; Doua K.; Drygas W.; Dzerve V.; Egbagbe E.E.; Eggertsen R.; Ekelund U.; El Ati J.; Elliott P.; Engle-Stone R.; Erasmus R.T.; Erem C.; Eriksen L.; Escobedo-De la Peña J.; Evans A.; Faeh D.; Fall C.H.; Farzadfar F.; Felix-Redondo F.J.; Ferguson T.S.; Fernández-Bergés D.; Ferrante D.; Ferrari M.; Ferreccio C.; Ferrieres J.; Finn J.D.; Fischer K.; Flores E.M.; Föger B.; Foo L.H.; Forslund A.-S.; Forsner M.; Fortmann S.P.; Fouad H.M.; Francis D.K.; Franco M.D.C.; Franco O.H.; Frontera G.; Fuchs F.D.; Fuchs S.C.; Fujita Y.; Furusawa T.; Gaciong Z.; Gafencu M.; Gareta D.; Garnett S.P.; Gaspoz J.-M.; Gasull M.; Gates L.; Geleijnse J.M.; Ghasemian A.; Giampaoli S.; Gianfagna F.; Giovannelli J.; Giwercman A.; Goldsmith R.A.; Gonçalves H.; Gross M.G.; Rivas J.P.G.; Gorbea M.B.; Gottrand F.; Graff-Iversen S.; Grafnetter D.; Grajda A.; Grammatikopoulou M.G.; Gregor R.D.; Grodzicki T.; Grøntved A.; Gruden G.; Grujic V.; Gu D.; Gualdi-Russo E.; Guan O.P.; Gudnason V.; Guerrero R.; Guessous I.; Guimaraes A.L.; Gulliford M.C.; Gunnlaugsdottir J.; Gunter M.; Guo X.; Guo Y.; Gupta P.C.; Gureje O.; Gurzkowska B.; Gutierrez L.; Gutzwiller F.; Halkjær J.; Hambleton I.R.; Hardy R.; Kumar R.H.; Hata J.; Hayes A.J.; He J.; Hendriks M.E.; Cadena L.H.; Herrala S.; Heshmat R.; Hihtaniemi I.T.; Ho S.Y.; Ho S.C.; Hobbs M.; Hofman A.; Hormiga C.M.; Horta B.L.; Houti L.; Howitt C.; Htay T.T.; Htet A.S.; Htike M.M.T.; Hu Y.; Husseini A.; Huu C.N.; Huybrechts I.; Hwalla N.; Iacoviello L.; Iannone A.G.; Ibrahim M.M.; Ikeda N.; Ikram M.A.; Irazola V.E.; Islam M.; Ivkovic V.; Iwasaki M.; Jackson R.T.; Jacobs J.M.; Jafar T.; Jamil K.M.; Jamrozik K.; Janszky I.; Jasienska G.; Jelakovic B.; Jiang C.Q.; Joffres M.; Johansson M.; Jonas J.B.; Jørgensen T.; Joshi P.; Juolevi A.; Jurak G.; Jureša V.; Kaaks R.; Kafatos A.; Kalter-Leibovici O.; Kapantais E.; Kasaeian A.; Katz J.; Kaur P.; Kavousi M.; Keil U.; Boker L.K.; Keinänen-Kiukaanniemi S.; Kelishadi R.; Kemper H.C.G.; Kengne A.P.; Kersting M.; Key T.; Khader Y.S.; Khalili D.; Khang Y.-H.; Khaw K.-T.H.; Khouw I.M.S.L.; Kiechl S.; Killewo J.; Kim J.; Klimont J.; Klumbiene J.; Koirala B.; Kolle E.; Kolsteren P.; Korrovits P.; Koskinen S.; Kouda K.; Koziel S.; Kratzer W.; Krokstad S.; Kromhout D.; Kruger H.S.; Kubinova R.; Kujala U.M.; Kula K.; Kulaga Z.; Kumar R.K.; Kurjata P.; Kusuma Y.S.; Kuulasmaa K.; Kyobutungi C.; Laamiri F.Z.; Laatikainen T.; Lachat C.; Laid Y.; Lam T.H.; Landrove O.; Lanska V.; Lappas G.; Larijani B.; Laugsand L.E.; Bao K.L.N.; Le T.D.; Leclercq C.; Lee J.; Lee J.; Lehtimäki T.; Lekhraj R.; León-Muñoz L.M.; Li Y.; Lilly C.L.; Lim W.-Y.; Lima-Costa M.F.; Lin H.-H.; Lin X.; Linneberg A.; Lissner L.; Litwin M.; Liu J.; Lorbeer R.; Lotufo P.A.; Lozano J.E.; Luksiene D.; Lundqvist A.; Lunet N.; Lytsy P.; Ma G.; Ma J.; Machado-Coelho G.L.L.; Machi S.; Maggi S.; Magliano D.J.; Maire B.; Makdisse M.; Malekzadeh R.; Malhotra R.; Rao K.M.; Malyutina S.; Manios Y.; Mann J.I.; Manzato E.; Margozzini P.; Markey O.; Marques-Vidal P.; Marrugat J.; Martin-Prevel Y.; Martorell R.; Masoodi S.R.; Mathiesen E.B.; Matsha T.E.; Mazur A.; Mbanya J.C.N.; McFarlane S.R.; McGarvey S.T.; McKee M.; McLachlan S.; McLean R.M.; McNulty B.A.; Yusof S.M.; Mediene-Benchekor S.; Meirhaeghe A.; Meisinger C.; Menezes A.M.B.; Mensink G.B.M.; Meshram I.I.; Metspalu A.; Mi J.; Michaelsen K.F.; Mikkel K.; Miller J.C.; Miquel J.F.; Miranda J.J.; Mišigoj-Durakovic M.; Mohamed M.K.; Mohammad K.; Mohammadifard N.; Mohan V.; Yusoff M.F.M.; Molbo D.; Møller N.C.; Molnár D.; Mondo C.K.; Monterrubio E.A.; Monyeki K.D.K.; Moreira L.B.; Morejon A.; Moreno L.A.; Morgan K.; Mortensen E.L.; Moschonis G.; Mossakowska M.; Mostafa A.; Mota J.; Motlagh M.E.; Motta J.; Mu T.T.; Muiesan M.L.; Müller-Nurasyid M.; Murphy N.; Mursu J.; Murtagh E.M.; Musa K.I.; Musil V.; Nagel G.; Nakamura H.; Námešná J.; Nang E.E.K.; Nangia V.B.; Nankap M.; Narake S.; Navarrete-Muñoz E.M.; Neal W.A.; Nenko I.; Neovius M.; Nervi F.; Neuhauser H.K.; Nguyen N.D.; Nguyen Q.N.; Nieto-Martínez R.E.; Ning G.; Ninomiya T.; Nishtar S.; Noale M.; Norat T.; Noto D.; Nsour M.A.; O’Reilly D.; Oh K.; Olayan I.H.; Olinto M.T.A.; Oltarzewski M.; Omar M.A.; Onat A.; Ordunez P.; Ortiz A.P.; Osler M.; Osmond C.; Ostojic S.M.; Otero J.A.; Overvad K.; Owusu-Dabo E.; Paccaud F.M.; Padez C.; Pahomova E.; Pajak A.; Palli D.; Palloni A.; Palmieri L.; Panda-Jonas S.; Panza F.; Parnell W.R.; Parsaeian M.; Pecin I.; Pednekar M.S.; Peeters P.H.; Peixoto S.V.; Peltonen M.; Pereira A.C.; Pérez C.M.; Peters A.; Petkeviciene J.; Peykari N.; Pham S.T.; Pigeot I.; Pikhart H.; Pilav A.; Pilotto L.; Pistelli F.; Pitakaka F.; Piwonska A.; Plans-Rubió P.; Poh B.K.; Porta M.; Portegies M.L.P.; Poulimeneas D.; Pradeepa R.; Prashant M.; Price J.F.; Puiu M.; Punab M.; Qasrawi R.F.; Qorbani M.; Bao T.Q.; Radic I.; Radisauskas R.; Rahman M.; Raitakari O.; Raj M.; Rao S.R.; Ramachandran A.; Ramke J.; Ramos R.; Rampal S.; Rasmussen F.; Redon J.; Reganit P.F.M.; Ribeiro R.; Rigo F.; de Wit T.F.R.; Ritti-Dias R.M.; Rivera J.A.; Robinson S.M.; Robitaille C.; Rodríguez-Artalejo F.; Rodriguez-Perez M.C.; Rodríguez-Villamizar L.A.; Rojas-Martinez R.; Rojroongwasinkul N.; Romaguera D.; Ronkainen K.; Rosengren A.; Rouse I.; Rubinstein A.; Rühli F.J.; Rui O.; Ruiz-Betancourt B.S.; Horimoto A.R.V.R.; Rutkowski M.; Sabanayagam C.; Sachdev H.S.; Saidi O.; Salanave B.; Martinez E.S.; Salomaa V.; Salonen J.T.; Salvetti M.; Sánchez-Abanto J.; Sandjaja; Sans S.; Santos D.A.; Santos O.; Dos Santos R.N.; Santos R.; Saramies J.L.; Sardinha L.B.; Sarrafzadegan N.; Saum K.-U.; Savva S.C.; Scazufca M.; Rosario A.S.; Schargrodsky H.; Schienkiewitz A.; Schmidt I.M.; Schneider I.J.; Schultsz C.; Schutte A.E.; Sein A.A.; Sen A.; Senbanjo I.O.; Sepanlou S.G.; Shalnova S.A.; Sharma S.K.; Shaw J.E.; Shibuya K.; Shin D.W.; Shin Y.; Shiri R.; Siantar R.; Sibai A.M.; Silva A.M.; Silva D.A.S.; Simon M.; Simons J.; Simons L.A.; Sjostrom M.; Slowikowska-Hilczer J.; Slusarczyk P.; Smeeth L.; Smith M.C.; Snijder M.B.; So H.-K.; Sobngwi E.; Söderberg S.; Soekatri M.Y.E.; Solfrizzi V.; Sonestedt E.; Song Y.; Sørensen T.I.A.; Soric M.; Jérome C.S.; Soumare A.; Staessen J.A.; Starc G.; Stathopoulou M.G.; Staub K.; Stavreski B.; Steene-Johannessen J.; Stehle P.; Stein A.D.; Stergiou G.S.; Stessman J.; Stieber J.; Stöckl D.; Stocks T.; Stokwiszewski J.; Stratton G.; Stronks K.; Strufaldi M.W.; Sun C.-A.; Sundström J.; Sung Y.-T.; Sunyer J.; Suriyawongpaisal P.; Swinburn B.A.; Sy R.G.; Szponar L.; Tai E.S.; Tammesoo M.-L.; Tamosiunas A.; Tang L.; Tang X.; Tanser F.; Tao Y.; Tarawneh M.R.; Tarp J.; Tarqui-Mamani C.B.; Taylor A.; Tchibindat F.; Theobald H.; Thijs L.; Thuesen B.H.; Tjonneland A.; Tolonen H.K.; Tolstrup J.S.; Topbas M.; Topór-Madry R.; Torrent M.; Toselli S.; Traissac P.; Trichopoulou A.; Trichopoulos D.; Trinh O.T.H.; Trivedi A.; Tshepo L.; Tulloch-Reid M.K.; Tuomainen T.-P.; Tuomilehto J.; Turley M.L.; Tynelius P.; Tzotzas T.; Tzourio C.; Ueda P.; Ukoli F.A.M.; Ulmer H.; Unal B.; Uusitalo H.M.T.; Valdivia G.; Vale S.; Valvi D.; van der Schouw Y.T.; Van Herck K.; Van Minh H.; van Rossem L.; van Valkengoed I.G.M.; Vanderschueren D.; Vanuzzo D.; Vatten L.; Vega T.; Velasquez-Melendez G.; Veronesi G.; Verschuren W.M.M.; Verstraeten R.; Victora C.G.; Viegi G.; Viet L.; Viikari-Juntura E.; Vineis P.; Vioque J.; Virtanen J.K.; Visvikis-Siest S.; Viswanathan B.; Vollenweider P.; Voutilainen S.; Vrdoljak A.; Vrijheid M.; Wade A.N.; Wagner A.; Walton J.; Mohamud W.N.W.; Wang M.-D.; Wang Q.; Wang Y.X.; Wannamethee S.G.; Wareham N.; Weerasekera D.; Whincup P.H.; Widhalm K.; Widyahening I.S.; Wiecek A.; Wijga A.H.; Wilks R.J.; Willeit J.; Wilsgaard T.; Wojtyniak B.; Wong J.E.; Wong T.Y.; Woo J.; Woodward M.; Wu F.C.; Wu J.; Wu S.L.; Xu H.; Xu L.; Yamborisut U.; Yan W.; Yang X.; Yardim N.; Ye X.; Yiallouros P.K.; Yoshihara A.; You Q.S.; Younger-Coleman N.O.; Yusoff A.F.; Zainuddin A.A.; Zambon S.; Zdrojewski T.; Zeng Y.; Zhao D.; Zhao W.; Zheng Y.; Zhou M.; Zhu D.; Zimmermann E.; Cisneros J.Z.
    Published: (2016)
  • GIS in management of landscape trees: Determining, monitoring and implementing safety measures on potentially hazardous trees
    by: Isa M.M.; Othman N.; Ghazali R.
    Published: (2012)