Population modeling with machine learning can enhance measures of mental health

Kamalaker Dadi (1) , Gaël Varoquaux (1, 2, 3) , Josselin Houenou (4, 5, 6) , Danilo Bzdok (1, 3, 7) , Bertrand Thirion (1) , Denis Engemann (1, 8)

1 PARIETAL - Modelling brain structure, function and variability based on high-field MRI data
2 Montreal Neurological Institute and Hospital
3 MILA - Montreal Institute for Learning Algorithms [Montréal]
4 UNIACT - Unité de recherche en NeuroImagerie Applicative Clinique et Translationnelle
5 U955 Inserm - UPEC - IMRB - "Neuropsychiatrie translationnelle" [Créteil]
6 CHU Henri Mondor [Créteil]
7 McGill University = Université McGill [Montréal, Canada]
8 IMPNSC - Max Planck Institute for Human Cognitive and Brain Sciences [Leipzig]

Background Biological aging is revealed by physical measures, e.g., DNA probes or brain scans. Instead, individual differences in mental function are explained by psychological constructs, e.g., intelligence or neuroticism. These constructs are typically assessed by tailored neuropsychological tests that build on expert judgement and require careful interpretation. Could machine learning on large samples from the general population be used to build proxy measures of these constructs that do not require human intervention? Results Here, we built proxy measures by applying machine learning on multimodal MR images and rich sociodemographic information from the largest biomedical cohort to date: the UK Biobank. Objective model comparisons revealed that all proxies captured the target constructs and were as useful, and sometimes more useful than the original measures for characterizing real-world health behavior (sleep, exercise, tobacco, alcohol consumption). We observed this complementarity of proxy measures and original measures when modeling from brain signals or sociodemographic data, capturing multiple health-related constructs. Conclusions Population modeling with machine learning can derive measures of mental health from brain signals and questionnaire data, which may complement or even substitute for psychometric assessments in clinical populations.

Mots clés: Mental Health; Proxy Measures; Machine Learning; Sociodemographic Factors; Brain Imaging

Domaines: Machine Learning [stat.ML]; Psychiatrie et santé mentale

Kamalaker Dadi, Gaël Varoquaux, Josselin Houenou, Danilo Bzdok, Bertrand Thirion, et al.. Population modeling with machine learning can enhance measures of mental health. GigaScience, 2021, ⟨10.1101/2020.08.25.266536⟩. ⟨hal-03470466⟩