Development and validation of FootNet, a new kinematic and deep learning-based algorithm to detect foot-strike and toe-off in treadmill running

Adrian Rodriguez Rivadulla, Xi Chen, Gillian Weir, Dario Cazzola, Grant Trewartha, Joseph Hamill, Ezio Preatoni

Research output: Chapter in Book/Report/Conference proceedingConference contribution


Foot-strike and toe-off detection is often critical in the assessment of running biomechanics. The onset and offset of the vertical ground reaction force is regarded as the “gold standard” method for step event detection, but several kinematics-based algorithms have been proposed to detect foot-strike and toe-off in the absence of force plates. However, the accuracy and generalisability of kinematics-based methods are often limited. Therefore, we developed FootNet, an algorithm using kinematic input and deep learning, to improve the detection of foot-strike and toe-off events during treadmill running in a variety of speed, foot-strike angle and incline conditions.
Original languageEnglish
Title of host publicationAbstract Book of The 28th Congress of the International Society of Biomechanics
Number of pages1
Publication statusAcceptance date - 23 Jun 2021
EventCongress of the International Society of Biomechanics (ISB) - Stockholm, Sweden
Duration: 25 Jul 202129 Jul 2021
Conference number: 28th


ConferenceCongress of the International Society of Biomechanics (ISB)
Internet address

Cite this