Probabilistic motion capture enhances clinical gait analysis with uncertainty quantification

Researchers have developed a probabilistic markerless motion capture method to improve the trustworthiness of clinical gait analysis. This approach quantifies epistemic uncertainty, allowing it to identify unreliable outputs without additional instrumentation. The model demonstrated reliable calibration with Expected Calibration Error (ECE) values generally below 0.1 for step and stride length, and bias-corrected gait kinematics. Errors for step and stride length were approximately 16 mm and 12 mm, respectively, with kinematic errors ranging from 1.5 to 3.8 degrees. AI

IMPACT This research offers a more reliable method for clinical gait analysis by quantifying uncertainty, potentially improving diagnostic accuracy and trust in AI-driven systems.

RANK_REASON The cluster contains an academic paper detailing a new methodology for gait analysis. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• Seth Donahue

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →