ADHDetect

Helping early-detection, diagnostics and analysis of ADHD thanks to artificial intelligence

The ADHDetect initiative is an interdisciplinary research project at the intersection of mental health, language sciences, and artificial intelligence. Its aim is to develop an automated method to support ADHD screening in adolescents, based on the analysis of written and oral personal narratives collected from French-speaking participants aged 12 to 17. The project involves building an anonymized corpus, extracting linguistic features, training and validating explainable classification models, and making the data and results available as open data. The work will be carried out in collaboration with the University of Geneva. The project aims to develop an ethical, fast, and accessible tool for the early identification of ADHD, in order to facilitate referral to appropriate care pathways. It includes a comparison of written and oral markers, the study of influential factors such as gender, IQ, and comorbidities, and the design of a demonstrator.