Real-world speed estimation using single IMU: A conceptual framework

In this study an extensive technical review was performed to identify original algorithms for walking speed (WS) estimation using single IMU. Based on the review a conceptual framework was proposed highliting different stages for accurate assesment of WS from separate cadence and step length estimation. Original algorithms and their improved versions were implemented and tested on gait dataset including large range of WS. Time-domain algorithms allowed a better step demarcation while had more error that frequency-domain for cadence estimation. Machine learning provided better results than model based approach but may suffer from overtraining. The results showed high heterogeneity between algorithms and a significant decline in performance for low WS values. Considering the higher risk of disability in slow walkers, there is a need in clinic to improve WS performance. An interesting approach would be weighted average of WS estimated with various algorithms in order to take the best from each algorithms and potentially improve robustness. This work was supported by the Mobilise-D project that has received funding from the Innovative Medicines Initiative 2 Joint Undertaking under grant agreement No. 820820. This Joint Undertaking receives support from the European Union's Horizon 2020 research and innovation program and the European Federation of Pharmaceutical Industries and Associations (EFPIA). Content in this publication reflects the authors’ view and neither IMI nor the European Union, EFPIA, or any Associated Partners are responsible for any use that may be made of the information contained herein.