Artificial Intelligence methods to help measure sensory disorders : approaches and perspectives
Résumé
The objective of this article is to present an approach aimed at aiding and automating the diagnosis of sensory particularities. We hypothesize that observing patient reactions using connected objects, combined with artificial intelligence, can objectively measure these particularities over time. Based on the comparison of questionnaires traditionally used to establish a sensory profile, we propose a meta-model that covers all possible sensory particularities in an individual. Using this meta-model, a device composed of connected objects enabled us to identify each action and associate them with a signature in the sensor signals. The benefits are numerous: contributing to the development of regular and frequent profiles, assisting practitioners in considering the temporal variation of sensory profiles in their decision-making, and objectifying the measurements.