A self-attention enhanced deep CNN-LSTM based irregular surface recognition approach for integration into lower limb prosthesis systems to ensure safety through predictive walking

Advancements in instrumentation and control systems for lower limb prostheses have substantially improved mobility for amputees. However, significant challenges persist when users encounter irregular surfaces, as most prosthetic systems lack the capability to dynamically adapt to surface variations....

Full description

Bibliographic Details
Main Authors: Norazian, Subari, Kamarul Hawari, Ghazali, Ji, Yuanfa
Format: Article
Language:English
Published: IEEE 2025
Subjects:
Online Access:http://umpir.ump.edu.my/id/eprint/44911/
http://umpir.ump.edu.my/id/eprint/44911/1/A%20self-attention%20enhanced%20deep%20CNN-LSTM%20based%20irregular%20surface.pdf
_version_ 1848827213430915072
author Norazian, Subari
Kamarul Hawari, Ghazali
Ji, Yuanfa
author_facet Norazian, Subari
Kamarul Hawari, Ghazali
Ji, Yuanfa
author_sort Norazian, Subari
building UMP Institutional Repository
collection Online Access
description Advancements in instrumentation and control systems for lower limb prostheses have substantially improved mobility for amputees. However, significant challenges persist when users encounter irregular surfaces, as most prosthetic systems lack the capability to dynamically adapt to surface variations. This limitation restricts mobility, compromises safety, and diminishes user confidence and security during walking. To address these challenges, integrating inertial measurement units (IMUs) with artificial intelligence (AI) techniques, particularly deep learning (DL) methods, has emerged as a promising solution for surface classification and safety enhancement. This study proposes a self-attention enhanced deep CNNLSTM model to automatically classify walking surfaces as regular or irregular, utilizing IMU acceleration data collected from prosthetic limbs. The model employs the strengths of convolutional and recurrent neural networks combined with a self-attention mechanism to enhance feature representation and improve classification accuracy. Experimental evaluations reveal that the proposed method achieves a classification accuracy of 99.94%, outperforming existing approaches. This result underscores the model’s potential to serve as the basis for AI-driven IMU-based systems, enabling real-time surface recognition and safety alerts in prosthetic devices. By enhancing walking safety and user confidence, this method represents a significant advancement for lower limb prosthesis systems.
first_indexed 2025-11-15T03:57:08Z
format Article
id ump-44911
institution Universiti Malaysia Pahang
institution_category Local University
language English
last_indexed 2025-11-15T03:57:08Z
publishDate 2025
publisher IEEE
recordtype eprints
repository_type Digital Repository
spelling ump-449112025-06-23T07:06:19Z http://umpir.ump.edu.my/id/eprint/44911/ A self-attention enhanced deep CNN-LSTM based irregular surface recognition approach for integration into lower limb prosthesis systems to ensure safety through predictive walking Norazian, Subari Kamarul Hawari, Ghazali Ji, Yuanfa TJ Mechanical engineering and machinery TK Electrical engineering. Electronics Nuclear engineering Advancements in instrumentation and control systems for lower limb prostheses have substantially improved mobility for amputees. However, significant challenges persist when users encounter irregular surfaces, as most prosthetic systems lack the capability to dynamically adapt to surface variations. This limitation restricts mobility, compromises safety, and diminishes user confidence and security during walking. To address these challenges, integrating inertial measurement units (IMUs) with artificial intelligence (AI) techniques, particularly deep learning (DL) methods, has emerged as a promising solution for surface classification and safety enhancement. This study proposes a self-attention enhanced deep CNNLSTM model to automatically classify walking surfaces as regular or irregular, utilizing IMU acceleration data collected from prosthetic limbs. The model employs the strengths of convolutional and recurrent neural networks combined with a self-attention mechanism to enhance feature representation and improve classification accuracy. Experimental evaluations reveal that the proposed method achieves a classification accuracy of 99.94%, outperforming existing approaches. This result underscores the model’s potential to serve as the basis for AI-driven IMU-based systems, enabling real-time surface recognition and safety alerts in prosthetic devices. By enhancing walking safety and user confidence, this method represents a significant advancement for lower limb prosthesis systems. IEEE 2025 Article PeerReviewed pdf en cc_by_4 http://umpir.ump.edu.my/id/eprint/44911/1/A%20self-attention%20enhanced%20deep%20CNN-LSTM%20based%20irregular%20surface.pdf Norazian, Subari and Kamarul Hawari, Ghazali and Ji, Yuanfa (2025) A self-attention enhanced deep CNN-LSTM based irregular surface recognition approach for integration into lower limb prosthesis systems to ensure safety through predictive walking. IEEE Access, 13. pp. 1-15. ISSN 2169-3536. (Published) https://doi.org/10.1109/ACCESS.2025.3567456 https://doi.org/10.1109/ACCESS.2025.3567456
spellingShingle TJ Mechanical engineering and machinery
TK Electrical engineering. Electronics Nuclear engineering
Norazian, Subari
Kamarul Hawari, Ghazali
Ji, Yuanfa
A self-attention enhanced deep CNN-LSTM based irregular surface recognition approach for integration into lower limb prosthesis systems to ensure safety through predictive walking
title A self-attention enhanced deep CNN-LSTM based irregular surface recognition approach for integration into lower limb prosthesis systems to ensure safety through predictive walking
title_full A self-attention enhanced deep CNN-LSTM based irregular surface recognition approach for integration into lower limb prosthesis systems to ensure safety through predictive walking
title_fullStr A self-attention enhanced deep CNN-LSTM based irregular surface recognition approach for integration into lower limb prosthesis systems to ensure safety through predictive walking
title_full_unstemmed A self-attention enhanced deep CNN-LSTM based irregular surface recognition approach for integration into lower limb prosthesis systems to ensure safety through predictive walking
title_short A self-attention enhanced deep CNN-LSTM based irregular surface recognition approach for integration into lower limb prosthesis systems to ensure safety through predictive walking
title_sort self-attention enhanced deep cnn-lstm based irregular surface recognition approach for integration into lower limb prosthesis systems to ensure safety through predictive walking
topic TJ Mechanical engineering and machinery
TK Electrical engineering. Electronics Nuclear engineering
url http://umpir.ump.edu.my/id/eprint/44911/
http://umpir.ump.edu.my/id/eprint/44911/
http://umpir.ump.edu.my/id/eprint/44911/
http://umpir.ump.edu.my/id/eprint/44911/1/A%20self-attention%20enhanced%20deep%20CNN-LSTM%20based%20irregular%20surface.pdf