Automatically Evaluating Balance: A Machine Learning Approach

TL;DR

This study demonstrates that machine learning models trained on trunk sway data can accurately evaluate balance performance, potentially enabling effective remote assessments outside clinical settings.

Contribution

The paper introduces a novel ML-based method for automatic balance assessment using trunk sway features, achieving high accuracy and closer alignment with PT ratings than self-assessments.

Findings

Support vector machine achieved 82% accuracy in classifying balance ratings.

ML assessments were significantly closer to PT ratings than self-assessments.

Automated evaluation could reduce PT consultation time and improve in-home balance training.

Abstract

Compared to in-clinic balance training, in-home training is not as effective. This is, in part, due to the lack of feedback from physical therapists (PTs). Here, we analyze the feasibility of using trunk sway data and machine learning (ML) techniques to automatically evaluate balance, providing accurate assessments outside of the clinic. We recruited sixteen participants to perform standing balance exercises. For each exercise, we recorded trunk sway data and had a PT rate balance performance on a scale of 1 to 5. The rating scale was adapted from the Functional Independence Measure. From the trunk sway data, we extracted a 61-dimensional feature vector representing performance of each exercise. Given these labeled data, we trained a multi-class support vector machine (SVM) to map trunk sway features to PT ratings. Evaluated in a leave-one-participant-out scheme, the model achieved a classification accuracy of 82%. Compared to participant self-assessment ratings, the SVM outputs were significantly closer to PT ratings. The results of this pilot study suggest that in the absence of PTs, ML techniques can provide accurate assessments during standing balance exercises. Such automated assessments could reduce PT consultation time and increase user compliance outside of the clinic.