Research on Intelligent Diagnosis of Motion System Injuries Based on Gait and Kinematic Parameters of Multimodal Data

Dangwei Cheng; Guangping Zhuo; Xinpeng Miao; Pengyuan Ma; Fei Ma

doi:10.62051/ijcsit.v5n1.11

Authors

Dangwei Cheng
Guangping Zhuo
Xinpeng Miao
Pengyuan Ma
Fei Ma

DOI:

https://doi.org/10.62051/ijcsit.v5n1.11

Keywords:

Gait Detection, Transformer Model, Multi-Head Attention Mechanism, Deep Learning

Abstract

This study aims to explore the use of the Transformer model for analyzing multimodal and multidimensional gait data, validating the effectiveness of the multi-head attention mechanism in case summarization and disease diagnosis. A retrospective case series design was adopted, involving 1200 samples collected from the Gait Laboratory of the Second Affiliated Hospital of Shanxi Medical University. The data include gait information from patients with osteoarthritis, anterior cruciate ligament injuries, and meniscus injuries, as well as healthy volunteers. Each sample encompasses foot pressure distribution, pressure intensity distribution, and motion data of the ankle, knee, and hip joints, along with labeled disease categories for model training and testing. Key data of clinical interest were curated into datasets to compare the performance of the Transformer model, based on the multi-head attention mechanism, with conventional convolutional neural networks in training and diagnosing conditions from gait data. The results demonstrate that the Transformer model achieves high diagnostic accuracy in gait analysis, with precision reaching approximately 95%. This study proves the feasibility of AI-based gait diagnosis using the Transformer model, providing valuable diagnostic references for orthopedic clinicians.

Downloads

Download data is not yet available.

References

[1] ZOU B, HAN J, WANG Y, et al. Semi-structural interview-based Chinese multimodal depression corpus towards automatic preliminary screening of depressive disorders [J]. IEEE Transactions on Affective Computing, 2022.

[2] LIU H, ZHANG S, LIANG X, et al. Transformer-based gait recognition for person identification [J]. Journal of Visual Communication and Image Representation, 2022.

[3] SUN J, LIU Z, XU Q, et al. Gait recognition using Transformer networks for enhanced security [J]. IEEE Access, 2023.

[4] WANG X, ZHANG J, WANG S, et al. A novel Transformer-based gait classification method for early detection of Parkinson's disease [J]. IEEE Transactions on Biomedical Engineering, 2023.

[5] LIU Y, ZHOU H, ZHU L, et al. Transformer and deep learning models for gait biometrics recognition [J]. Neurocomputing, 2023.

[6] CHEN X, ZHANG R, ZHANG X, et al. Gait analysis and recognition based on Transformer architecture [J]. Pattern Recognition Letters, 2022.

[7] LI J, LIU Y, LIU Z, et al. Gait recognition using Transformer models for healthcare applications [J]. Journal of Healthcare Engineering, 2023.

[8] MA Y, LI X, JIN D, et al. Transformer-based fusion model for gait-based person authentication [J]. Ieee Transactions on Systems, Man, and Cybernetics: Systems, 2023.

[9] ZHANG J, YU H, LI B, et al. Transformer-based model for human gait recognition in surveillance systems [J]. International Journal of Computer Vision, 2022.

[10] LI X, LIU S, WANG J, et al. Multi-view gait recognition using Transformer networks [J]. Computer Vision and Image Understanding, 2022.

[11] JIANG T, LI B, YU W, et al. A Transformer-based framework for gait recognition with sequence learning [J]. Neurocomputing, 2022.

[12] LIU D, ZHANG L, FANG Y, et al. Gait recognition with Transformer for elderly monitoring system [J]. Journal of Ambient Intelligence and Humanized Computing, 2023.

[13] WANG Z, LIU M, ZHOU W, et al. Deep Transformer networks for gait-based person verification [J]. IEEE Transactions on Information Forensics and Security, 2022.

[14] XU H, LIANG Q, CHEN Y, et al. Gait recognition using a deep learning Transformer model for medical diagnosis [J]. Biomedical Signal Processing and Control, 2022.

[15] WANG J, LI H, LI Y, et al. An end-to-end Transformer model for gait recognition in surveillance systems [J]. Computer Vision and Image Understanding, 2023.

[16] ZHANG X, WANG H, LI Y, et al. Transformer-based deep learning model for gait analysis in rehabilitation medicine [J]. IEEE Transactions on Medical Imaging, 2023.

[17] XU Y, ZHANG D, WANG S, et al. A novel Transformer-based model for human gait classification and analysis [J]. Pattern Recognition, 2023.

[18] sFENG Y, ZHANG J, WANG S, et al. Transformer-based gait recognition model for biometric security [J]. Journal of Biometrics and Biostatistics, 2023.

[19] HU J, LI J, LI Q, et al. Gait pattern recognition based on Transformer models for fall detection [J]. IEEE Journal of Biomedical and Health Informatics, 2022.

[20] YANG Z, ZHAO W, XU C, et al. Transformer-based approach for gait-based human action recognition [J]. Journal of Machine Learning Research, 2023.

[21] ZHANG M, LI Z, WANG X, et al. Transformer and deep neural network-based gait recognition for health monitoring [J]. Journal of Neural Engineering, 2023.

[22] LI B, YU Q, LIU Z, et al. A new Transformer model for gait recognition in multi-task learning scenarios [J]. Cognitive Computation, 2022.

[23] ZHANG Y, LIU T, MA Y, et al. Robust gait recognition with Transformer networks for security applications [J]. Security and Privacy, 2023.

[24] LU X, WANG X, YAN Z, et al. Transformer-based spatiotemporal modeling for gait recognition in video surveillance [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2023.

[25] LI Y, FENG W, LI H, et al. Gait-based human authentication using Transformer models [J]. Journal of Applied Research and Technology, 2023.

[26] LI Z, ZHANG Q, LI J, et al. Transformer-based fusion model for gait analysis in medical applications [J]. Journal of Medical Systems, 2023.

[27] LIU Q, WANG Y, LI L, et al. A Transformer-based system for gait-based recognition of individuals in smart environments [J]. Smart Health, 2023.

[28] ZHAO T, ZHU J, LUO Z, et al. A Transformer-based deep learning method for human gait recognition [J]. Multimedia Tools and Applications, 2022.

[29] ZHAO Z, LI Y, FANG J, et al. Transformer-based model for gait anomaly detection in healthcare systems [J]. IEEE Transactions on Biomedical Engineering, 2023.