Fall risk assessment with machine learning

Falls are a leading cause of fracture and mortality in older adults, and represent a considerable socioeconomic burden in ageing societies. To help medical practioners and eldery care managers with falls prevention and management, a group of FHT researchers and their co-authors have come up with a study that aims to detect individuals at a high risk of falls.

The most common indicator for fall prediction is history of falling, but this is a subjective predictor and fails to detect first-time fallers simply because it is absent in such cases. In the paper, titled 'external pageIdentifying Fallers Based on Functional Parameters: A Machine Learning Approachcall_made', the researchers used functional variables extracted from multiple functional domains, and implemented several machine learning (ML) methods to classify fallers vs non-fallers retrospectively. They also performed feature importance analysis to provide an insight into the underlying features.

Performed within a cross-validation setting, the researchers identified the ML algorithm that best maps individuals’ functional measures to their fall status. They also applied this algorithm for prospective identification of fall risk. In retrospective classification, k-nearest neighbours (KNN) model achieved a sensitivity of 74% and a specificity of 75%. In prospective evaluation, it achieved sensitivity and specificity of 80%. With these results, the study reflects the superior capability of machine learning in fallers identification even with a very small dataset.

By coming up with means to detect individuals at a high risk of falls, the research hope to eventually enable implementation of targeted therapies and timely intervention, reducing a major cause of injuries among older adults.