Computer Vision and Machine Learning-Based Gait Pattern Recognition for Flat Fall Prediction.
convolutional neural network
fall recognition
gait
k nearest neighbor
long short-time memory
machine learning
pattern recognition
support vector machine
Journal
Sensors (Basel, Switzerland)
ISSN: 1424-8220
Titre abrégé: Sensors (Basel)
Pays: Switzerland
ID NLM: 101204366
Informations de publication
Date de publication:
19 Oct 2022
19 Oct 2022
Historique:
received:
21
09
2022
revised:
12
10
2022
accepted:
14
10
2022
entrez:
27
10
2022
pubmed:
28
10
2022
medline:
29
10
2022
Statut:
epublish
Résumé
Gait recognition has been applied in the prediction of the probability of elderly flat ground fall, functional evaluation during rehabilitation, and the training of patients with lower extremity motor dysfunction. Gait distinguishing between seemingly similar kinematic patterns associated with different pathological entities is a challenge for the clinician. How to realize automatic identification and judgment of abnormal gait is a significant challenge in clinical practice. The long-term goal of our study is to develop a gait recognition computer vision system using artificial intelligence (AI) and machine learning (ML) computing. This study aims to find an optimal ML algorithm using computer vision techniques and measure variables from lower limbs to classify gait patterns in healthy people. The purpose of this study is to determine the feasibility of computer vision and machine learning (ML) computing in discriminating different gait patterns associated with flat-ground falls. We used the Kinect The SVM and KNN had a higher accuracy than CNN and LSTM. The SVM (94.9 ± 3.36%) had the highest accuracy in the classification of gait patterns, followed by KNN (94.0 ± 4.22%). The accuracy of CNN was 87.6 ± 7.50% and that of LSTM 83.6 ± 5.35%. This study revealed that the proposed AI machine learning (ML) techniques can be used to design gait biometric systems and machine vision for gait pattern recognition. Potentially, this method can be used to remotely evaluate elderly patients and help clinicians make decisions regarding disposition, follow-up, and treatment.
Sections du résumé
BACKGROUND
BACKGROUND
Gait recognition has been applied in the prediction of the probability of elderly flat ground fall, functional evaluation during rehabilitation, and the training of patients with lower extremity motor dysfunction. Gait distinguishing between seemingly similar kinematic patterns associated with different pathological entities is a challenge for the clinician. How to realize automatic identification and judgment of abnormal gait is a significant challenge in clinical practice. The long-term goal of our study is to develop a gait recognition computer vision system using artificial intelligence (AI) and machine learning (ML) computing. This study aims to find an optimal ML algorithm using computer vision techniques and measure variables from lower limbs to classify gait patterns in healthy people. The purpose of this study is to determine the feasibility of computer vision and machine learning (ML) computing in discriminating different gait patterns associated with flat-ground falls.
METHODS
METHODS
We used the Kinect
RESULTS
RESULTS
The SVM and KNN had a higher accuracy than CNN and LSTM. The SVM (94.9 ± 3.36%) had the highest accuracy in the classification of gait patterns, followed by KNN (94.0 ± 4.22%). The accuracy of CNN was 87.6 ± 7.50% and that of LSTM 83.6 ± 5.35%.
CONCLUSIONS
CONCLUSIONS
This study revealed that the proposed AI machine learning (ML) techniques can be used to design gait biometric systems and machine vision for gait pattern recognition. Potentially, this method can be used to remotely evaluate elderly patients and help clinicians make decisions regarding disposition, follow-up, and treatment.
Identifiants
pubmed: 36298311
pii: s22207960
doi: 10.3390/s22207960
pmc: PMC9612353
pii:
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Références
J UOEH. 2014 Mar 1;36(1):41-8
pubmed: 24633184
Sensors (Basel). 2021 Jul 29;21(15):
pubmed: 34372371
Comput Methods Programs Biomed. 2014 Dec;117(3):489-501
pubmed: 25308505
J Neuroeng Rehabil. 2019 Feb 1;16(1):22
pubmed: 30709363
Sci Rep. 2021 Oct 25;11(1):20976
pubmed: 34697377
Clin Biomech (Bristol, Avon). 2014 May;29(5):518-22
pubmed: 24768223
Sensors (Basel). 2020 Sep 08;20(18):
pubmed: 32911651
Comput Methods Programs Biomed. 2020 Dec;197:105721
pubmed: 32882593
IEEE Trans Inf Technol Biomed. 2009 Sep;13(5):687-702
pubmed: 19447724
Sensors (Basel). 2016 Dec 10;16(12):
pubmed: 27973406
Artif Intell Med. 2019 Mar;94:54-66
pubmed: 30871683
Front Neurol. 2021 May 21;12:666458
pubmed: 34093413
Comput Biol Med. 2019 Mar;106:33-39
pubmed: 30665140
Biomed Eng Online. 2021 Jun 22;20(1):62
pubmed: 34158070
IEEE J Transl Eng Health Med. 2020 May 28;8:2100609
pubmed: 32537265
Sensors (Basel). 2019 Apr 06;19(7):
pubmed: 30959877
Sensors (Basel). 2021 Aug 26;21(17):
pubmed: 34502640
Sensors (Basel). 2021 Apr 19;21(8):
pubmed: 33921769
Med Biol Eng Comput. 2008 Jun;46(6):563-73
pubmed: 18347832
Sci Rep. 2020 Mar 10;10(1):4426
pubmed: 32157168
Front Robot AI. 2022 Jan 03;8:749274
pubmed: 35047564
Sensors (Basel). 2022 Aug 21;22(16):
pubmed: 36016043
Int J Stroke. 2018 Aug;13(6):612-632
pubmed: 29786478
IEEE J Biomed Health Inform. 2022 Jan;26(1):229-242
pubmed: 34181559
N Engl J Med. 1988 Dec 29;319(26):1701-7
pubmed: 3205267
J Med Eng Technol. 2020 Nov;44(8):441-467
pubmed: 33078988
Sensors (Basel). 2022 Jan 21;22(3):
pubmed: 35161570
Phys Med Biol. 2020 Apr 22;65(8):085013
pubmed: 32084652
J Electromyogr Kinesiol. 2001 Feb;11(1):19-30
pubmed: 11166605
Digit Biomark. 2022 Jan 13;6(1):9-18
pubmed: 35224426
Med Biol Eng Comput. 2018 Sep;56(9):1553-1564
pubmed: 29435705
J Neurol. 2022 Jul 11;:
pubmed: 35817988
J Am Acad Orthop Surg. 2007 Feb;15(2):107-17
pubmed: 17277257
Postgrad Med. 1993 May 1;93(6):185-90
pubmed: 8483823
Sensors (Basel). 2021 Aug 12;21(16):
pubmed: 34450879
Biomed Res Int. 2015;2015:528971
pubmed: 25705672
J Safety Res. 2016 Sep;58:99-103
pubmed: 27620939
Sensors (Basel). 2022 May 25;22(11):
pubmed: 35684613
Phys Ther. 2011 Dec;91(12):1740-51
pubmed: 22003158
J Am Geriatr Soc. 2018 Apr;66(4):693-698
pubmed: 29512120
Sensors (Basel). 2017 Feb 28;17(3):
pubmed: 28264503
Sensors (Basel). 2022 Jun 29;22(13):
pubmed: 35808426
Gait Posture. 2013 Jun;38(2):165-9
pubmed: 23755883
J Stroke Cerebrovasc Dis. 2020 Sep;29(9):105035
pubmed: 32807447
IEEE Trans Neural Syst Rehabil Eng. 2017 Dec;25(12):2336-2346
pubmed: 28792901
Sensors (Basel). 2022 May 12;22(10):
pubmed: 35632109
J Biomech. 2019 Apr 18;87:150-156
pubmed: 30876735
Health Informatics J. 2021 Oct-Dec;27(4):14604582211055650
pubmed: 34989252
J Stroke Cerebrovasc Dis. 2021 Oct;30(10):106011
pubmed: 34325274
J Aging Phys Act. 2018 Oct 1;26(4):577-582
pubmed: 29345526
Biomed Eng Online. 2011 Nov 10;10:99
pubmed: 22074315
PLoS One. 2018 Mar 8;13(3):e0192345
pubmed: 29518090
MMWR Morb Mortal Wkly Rep. 2020 Jul 10;69(27):875-881
pubmed: 32644982
IEEE J Biomed Health Inform. 2017 Jan;21(1):4-21
pubmed: 28055930
Sensors (Basel). 2022 Jun 02;22(11):
pubmed: 35684863
Sensors (Basel). 2022 Apr 20;22(9):
pubmed: 35590844
Stroke. 2010 Aug;41(8):1709-14
pubmed: 20576947
IEEE Trans Med Imaging. 2016 May;35(5):1285-98
pubmed: 26886976
Neurol Neurochir Pol. 2021;55(6):513-524
pubmed: 34817060
Sensors (Basel). 2022 Jul 22;22(15):
pubmed: 35897985
Comput Biol Med. 2019 Sep;112:103366
pubmed: 31386972
Gait Posture. 2022 Jan;91:137-148
pubmed: 34695721
Environ Health Prev Med. 2010 Nov;15(6):386-91
pubmed: 21432571
Insights Imaging. 2018 Aug;9(4):611-629
pubmed: 29934920
J Gerontol A Biol Sci Med Sci. 2015 May;70(5):608-15
pubmed: 25568095
BMC Bioinformatics. 2007 May 22;8 Suppl 4:S2
pubmed: 17570145
J Big Data. 2021;8(1):53
pubmed: 33816053
J Public Health Manag Pract. 2019 Mar/Apr;25(2):E17-E24
pubmed: 29757813
Sensors (Basel). 2016 Jan 06;16(1):
pubmed: 26751449
Sensors (Basel). 2016 Dec 22;17(1):
pubmed: 28025484