Machine Learning and Feature Ranking for Impact Fall Detection Event Using Multisensor Data

TL;DR

This work tackles accurate identification of the ground-impact moment in elderly fall events using multisensor data from the UP-FALL dataset. It presents a pipeline that combines sensor synchronization, normalization, SMV computation, and a $2g$ threshold for robust labeling, followed by Random Forest-based feature ranking to distill the most informative signals. Eight classifiers are evaluated, with SVM delivering the best overall performance (≈99.5% accuracy) and fast inference, highlighting the value of strong preprocessing and feature selection for real-time impact detection. The results advance practical fall-detection systems and point to future work in multimodal data integration and validation on additional datasets.

Abstract

Falls among individuals, especially the elderly population, can lead to serious injuries and complications. Detecting impact moments within a fall event is crucial for providing timely assistance and minimizing the negative consequences. In this work, we aim to address this challenge by applying thorough preprocessing techniques to the multisensor dataset, the goal is to eliminate noise and improve data quality. Furthermore, we employ a feature selection process to identify the most relevant features derived from the multisensor UP-FALL dataset, which in turn will enhance the performance and efficiency of machine learning models. We then evaluate the efficiency of various machine learning models in detecting the impact moment using the resulting data information from multiple sensors. Through extensive experimentation, we assess the accuracy of our approach using various evaluation metrics. Our results achieve high accuracy rates in impact detection, showcasing the power of leveraging multisensor data for fall detection tasks. This highlights the potential of our approach to enhance fall detection systems and improve the overall safety and well-being of individuals at risk of falls.

Figures (6)

• Figure 1: Comparison of impact detection using the state of the art method with our proposed method
• Figure 2: The Proposed Approach Pipeline
• Figure 3: Fall onset detection : a) with only SMV, b) with SMV and Threshold
• Figure 4: Confusion Matrix comparison: a) current pre-processing al2021towards trained with Random Forest; b) our proposed pre-processing trained with Random Forest Algorithm
• Figure 5: Feature Ranking