A neural network method to predict task- and step-specific ground reaction force magnitudes from trunk accelerations during running activities. (April 2020)

• Record Type: Journal Article
• Authors: Pogson, Mark; Verheul, Jasper; Robinson, Mark A.; Vanrenterghem, Jos; Lisboa, Paulo
• Is Part Of: Medical engineering & physics. Volume 78(2020)
• Page Start: 82
• Page End: 89
• Publication Date: 2020-04
• Subjects: Ground reaction force -- Trunk accelerometry -- Multilayer perceptron -- Biomechanical load; Biomedical engineering -- Periodicals; Biomedical Engineering -- Periodicals; Physics -- Periodicals; Génie biomédical -- Périodiques; Biomedical engineering; Electronic journals; Periodicals; 610.28
• DOI: 10.1016/j.medengphy.2020.02.002
• Languages: English
• ISSNs: 1350-4533

Abstract

Highlights: Predicting ground reaction force (GRF) from trunk acceleration is desirable. We show a machine learning approach is effective in this aim. Unlike other methods, it uses existing sensors to predict all GRF characteristics. Accuracy is high, with scope for refinement and personalisation. This can help to guide training intensity and prevent injury in athletes. Abstract: Prediction of ground reaction force (GRF) magnitudes during running-based sports has several important applications, including optimal load prescription and injury prevention in athletes. Existing methods typically require information from multiple body-worn sensors, limiting their ecological validity, or aim to estimate discrete force parameters, limiting their ability to assess overall biomechanical load. This paper presents a neural network method to predict GRF time series from a single, commonly used, trunk-mounted accelerometer. The presented method uses a principal component analysis and multilayer perceptron (MLP) to obtain predictions. Time-series r 2 coefficients with test data averaged around 0.9 for each impact, comparing favourably with alternative approaches which require additional sensors. For the impact peak, r 2 was 0.74 across activities, comparing favourably with correlation analysis approaches. Several modifications, such as subject-specific training of the MLP, may help to improve results further, but the presented method can accurately predict GRF from trunk accelerometry data without requiring additional information. Results demonstrate the scope of machine learning to exploit common wearable technologies to estimate GRF in sport-specific environments.