A Fusion Network With Stacked Denoise Autoencoder and Meta Learning for Lateral Walking Gait Phase Recognition and Multi-Step-Ahead Prediction

• Published in: IEEE journal of biomedical and health informatics Vol. 29; no. 1; pp. 68 - 80
• Main Authors: Cao, Wujing, Li, Changyu, Yang, Lijun, Yin, Meng, Chen, Chunjie, Worawarit, Kobsiriphat, Thanak, Utakapan, Yang, Yizhuang, Yu, Haoyong, Wu, Xinyu
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.01.2025
• Subjects: Adult; Autoencoder; Exoskeletons; Feature extraction; Female; Gait - physiology; gait phase recognition; Hip; hip exoskeleton; Humans; Lateral walking; Legged locomotion; Long short term memory; Machine Learning; Male; meta learning; multi-step-ahead prediction; Neural Networks, Computer; Signal Processing, Computer-Assisted; stacked denoise autoencoder; Support Vector Machine; Support vector machines; Training; Walking - physiology; Young Adult

Lateral walking gait phase recognition and prediction are the premise of hip exoskeleton application in lateral resistance walk exercise. We presented a fusion network with stacked denoise autoencoder and meta learning (SDA-NN-ML) to recognize gait phase and predict gait percentage from IMU signals....