Diagnostic automatique de l'état dépressif

Abstract : Psychosocial disorders are a major public health problem that can have serious consequences in the short or long term, on a professional, personal or family level. The diagnosis of these disorders must be made by a professional. However, AI (Artificial Intelligence) can make a contribution by providing the practitioner with diagnostic assistance, and the patient with rapid and inexpensive ongoing follow-up. We propose an approach towards an automatic ,diagnosis of the depressive state based on real-time facial observations, using a simple webcam. From videos of the AVEC’2014 challenge, we trained a neural classifier to extract prototypes of faces according to different values of Beck’s depression score (BDI-II).
Complete list of metadatas

https://hal.univ-antilles.fr/hal-01802764
Contributor : Stephane Cholet <>
Submitted on : Tuesday, May 29, 2018 - 5:11:09 PM
Last modification on : Monday, July 23, 2018 - 5:27:03 PM

Licence


Distributed under a Creative Commons Attribution - NonCommercial - NoDerivatives 4.0 International License

Identifiers

  • HAL Id : hal-01802764, version 1

Collections

Citation

Stephane Cholet, Helene Paugam-Moisy. Diagnostic automatique de l'état dépressif. Conférence Nationale d'Intelligence Artificielle, Plateforme Intelligence Artificielle, Jul 2018, Nancy, France. pp.102-109. ⟨hal-01802764⟩

Share

Metrics

Record views

319