Abstract
The accurate detection of foot-strike and toe-off is often critical in the assessment of running biomechanics. The gold standard method for step event detection requires force data which are not always available. Although kinematics-based algorithms can also be used, their accuracy and generalisability are limited, often requiring corrections for speed or foot-strike pattern. The purpose of this study was to develop FootNet, a novel kinematics and deep learning-based algorithm for the detection of step events in treadmill running. Five treadmill running datasets were gathered and processed to obtain segment and joint kinematics, and to identify the contact phase within each gait cycle using force data. The proposed algorithm is based on a long short-term memory recurrent neural network and takes the distal tibia anteroposterior velocity, ankle dorsiflexion/plantar flexion angle and the anteroposterior and vertical velocities of the foot centre of mass as input features to predict the contact phase within a given gait cycle. The chosen model architecture underwent 5-fold cross-validation and the final model was tested in a subset of participants from each dataset (30%). Non-parametric Bland-Altman analyses (bias and [95% limits of agreement]) and root mean squared error (RMSE) were used to compare FootNet against the force data step event detection method. The association between detection errors and running speed, foot-strike angle and incline were also investigated. FootNet outperformed previously published algorithms (foot-strike bias = 0 [-10, 7] ms, RMSE = 5 ms; toe-off bias = 0 [-10, 10] ms, RMSE = 6 ms; and contact time bias = 0 [-15, 15] ms, RMSE = 8 ms) and proved robust to different running speeds, foot-strike angles and inclines. We have made FootNet's source code publicly available for step event detection in treadmill running when force data are not available.
Highlights
Running is one of the most popular sports around the world [1, 2] and one of the most studied actions within human movement research
Foot-strike and toe-off are used to compute basic running metrics associated with performance and injury, two of the main foci of running biomechanics research and consumer-based running technology
We investigated the sensitivity of different kinematic variables to errors in step detection
Summary
Running is one of the most popular sports around the world [1, 2] and one of the most studied actions within human movement research. Biomechanical analyses of running technique typically involve the identification of gait cycles (i.e. strides) and rely on the accurate detection of foot-strike and toe-off within each cycle. The current “gold standard” to identify foot-strike and toe-off for both overground and treadmill running is the onset and offset of the vertical ground reaction force (vGRF) that exceeds a particular force magnitude. Force plates are typically limited to small areas in overground running studies making it difficult to collect several consecutive steps and force-instrumented treadmills are costly and only available in a few laboratories. Conventional treadmills are often used in running kinematics research and gait clinics, requiring the estimation of step events without force data, which has historically proved challenging
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
