A machine learning technique that tries to infer probability distribution functions over the possible outputs by using either supervised or unsupervised approach.
Published in Chapter:
Machine Learning Algorithms in Human Gait Analysis
Aditi A. Bhoir (Sardar Patel College of Engineering, India), Tanish A. Mishra (Sardar Patel College of Engineering, India), Jyotindra Narayan (Indian Institute of Technology, Guwahati, India), and Santosha K. Dwivedy (Indian Institute of Technology, Guwahati, India)
Copyright: © 2023
|Pages: 16
DOI: 10.4018/978-1-7998-9220-5.ch053
Abstract
The gait cycle is a study of human locomotion achieved by a combination of efforts made by the nervous system, muscles, and joints. Gait analysis has been extensively studied and applied in recent years for several applications including biometrics, healthcare, sports, and many more. It includes a large number of interrelated parameters, which are difficult to implement because of the high volume of data. The application of machine learning is a potential and promising solution to streamline this process. This work will review the latest developments in gait analysis performed by many researchers, with a primary focus on gait analysis using machine learning techniques. This review will briefly present the success of machine learning in data acquisition, detection of the disorders, and identification of the rehabilitation measures. Furthermore, the implementation of ML algorithms to correct gait abnormalities will be discussed. Finally, the possible opportunities of ML algorithms to improve the assessment of clinical gait analysis will be presented along with the concluding remarks.