Predicting fall risk in older adults: A machine learning comparison of accelerometric and non-accelerometric factors

TL;DR

This study compares machine learning models using accelerometric and non-accelerometric data to predict fall risk in older adults, finding combined data and Bayesian methods yield the best accuracy for fall risk assessment.

Contribution

It demonstrates that integrating accelerometric and non-accelerometric data with Bayesian regression improves fall risk prediction accuracy in older adults.

Findings

Combined data models outperform single data type models.

Bayesian Ridge Regression achieved the highest accuracy.

Non-accelerometric factors like age are crucial for prediction.

Abstract

This study investigates fall risk prediction in older adults using various machine learning models trained on accelerometric, non-accelerometric, and combined data from 146 participants. Models combining both data types achieved superior performance, with Bayesian Ridge Regression showing the highest accuracy (MSE = 0.6746, R2 = 0.9941). Non-accelerometric variables, such as age and comorbidities, proved critical for prediction. Results support the use of integrated data and Bayesian approaches to enhance fall risk assessment and inform prevention strategies.