Trustworthy Visual Analytics in Clinical Gait Analysis: A Case Study for Patients with Cerebral Palsy

Alexander Rind, Djordje Slijepcevic, Matthias Zeppelzauer, Fabian Unglaube, Andreas Kranzl, Brian Horsak

View presentation: 2022-10-16T14:10:00Z GMT-0600 Change your timezone on the schedule page
2022-10-16T14:10:00Z
Exemplar figure, described by caption below
Clinical gait analysis utilizes 3D motion tracking to record complex time series datasets of patients while they are walking in a gait laboratory. gaitXplorer is a visual analytics approach that classifies cerebral palsy-related gait patterns using a CNN, generates relevance scores using Grad-CAM, and presents the data, the predicted class, and the relevance scores in a visual interface. This case study investigates how clinical gait analysts trust the automated classifications and explanations presented by gaitXplorer.

The live footage of the talk, including the Q&A, can be viewed on the session page, TREX: Session 1.

Abstract