When artificial intelligence helps analyse walking disorders

Japanese researchers have developed an algorithm based on artificial intelligence (AI) to analyse changes in gait in several pathologies:

  • 114 patients, including 45 with neuromuscular pathologies, took part in the study,
  • the application (TDPT-GT) was developed on the iPhone using data supplied by the camera, based on a codified and standardised movement protocol (circular walk, 3 metres away from the camera lens, lasting one minute),
  • this approach made it possible to obtain the coordinates of 27 body points without using physical sensors, as is the case with conventional methods of gait analysis,
  • AI analysis of the data then enabled several types of pathological gait to be identified.

This simple, easy-to-use tool does, however, have its limitations, particularly in the case of very severe gait disturbances.

 

Artificial Intelligence Distinguishes Pathological Gait: The Analysis of Markerless Motion Capture Gait Data Acquired by an iOS Application (TDPT-GT). Iseki C, Hayasaka T, Yanagawa H et al. Sensors (Basel). 2023 Juil.