AIoT-Enabled Rehabilitation Recognition System-Exemplified by Hybrid Lower-Limb Exercises.

Yi-Chun Lai,Hsueh-Chun Lin,Yu-Chiang Lin,Yao-Chiang Kan

doi:10.3390/s21144761

Yi-Chun Lai, Hsueh-Chun Lin + Show 2 more

Open Access

https://doi.org/10.3390/s21144761

Copy DOI

Abstract

Ubiquitous health management (UHM) is vital in the aging society. The UHM services with artificial intelligence of things (AIoT) can assist home-isolated healthcare in tracking rehabilitation exercises for clinical diagnosis. This study combined a personalized rehabilitation recognition (PRR) system with the AIoT for the UHM of lower-limb rehabilitation exercises. The three-tier infrastructure integrated the recognition pattern bank with the sensor, network, and application layers. The wearable sensor collected and uploaded the rehab data to the network layer for AI-based modeling, including the data preprocessing, featuring, machine learning (ML), and evaluation, to build the recognition pattern. We employed the SVM and ANFIS methods in the ML process to evaluate 63 features in the time and frequency domains for multiclass recognition. The Hilbert-Huang transform (HHT) process was applied to derive the frequency-domain features. As a result, the patterns combining the time- and frequency-domain features, such as relative motion angles in y- and z-axis, and the HHT-based frequency and energy, could achieve successful recognition. Finally, the suggestive patterns stored in the AIoT-PRR system enabled the ML models for intelligent computation. The PRR system can incorporate the proposed modeling with the UHM service to track the rehabilitation program in the future.

Highlights

Home rehabilitation care serves many people with chronic disease with body degradation, and it is imperative in an aging society [1]
The modeling refers to the typical physical therapy that requests the patient to do personalized rehabilitation exercises based on the motion guide
In addition to all 3-axis features in both domains, we explored the qualified feature sets based on y- and z-axis variables, such as sum, the sum of absolute, average, and average of absolute in the time domain, and that plus frequency and energy of IMFj (j = 1−3) due to marginal Hilbert spectrum (MHS) area in the frequency domain

Summary

Introduction

Home rehabilitation care serves many people with chronic disease with body degradation, and it is imperative in an aging society [1]. Physicians design physiotherapy programs for patients to take regular rehabilitation exercises to reduce pain and restore physical functions. Programs are designed for stroke patients who have lower-limb hemiplegia, which usually causes abnormal gait due to the affected joints, such as the moment of hip landing, the maximum extension angle when standing, the joint angle when the toe is off the ground, the maximum flexion angle of the stepping, the range of knee joint angle, etc. Correct lower-limb rehabilitation exercises, such as hooking leg and padding toe, can increase the angle range of knee flexion and ankle dorsiflexion of the patient to enhance their walking stability [3,4]. Physicians can remotely track the routine home rehabilitation records of the outpatients to reduce clinics and save medical resources [5,6]

Methods

Results

Conclusion