This repository accompanies the work "Topological Gait Analysis: A New Framework and its Application to the Study of Human Gait" by Shreyam Mishra, Debasish Chatterjee, and Neeta Kanekar (DOI: 10.1109/JBHI.2024.3427700).
The folder PDE-videos contain video files of the temporal evolution of the probability density estimates (PDEs) of selected subjects.
The PDF file titled TGA_Additional_Information (TGA: Topological Gait Analysis) covers a comprehensive theoretical overview of cubical homology, detailed clinical information about the subjects from the dataset under study and an explanation of the heuristic method used for hyperparameter tuning. It also discusses the minimum amount of gait data (in terms of minutes and number of strides) required for the proposed TGA framework to yield accurate results and how its applicability can be extended to subjects with severe conditions who face significant challenges in walking for extended periods. Furthermore, the repository explores an additional dataset comprising subjects with neurodegenerative conditions and aging, highlighting the generalizability of the proposed TGA framework.