Quantifying brain-functional dynamics using deep dynamical systems: Technical considerations.

Both mental health and mental illness unfold in complex and unpredictable ways. Novel artificial intelligence approaches from the area of dynamical systems reconstruction can characterize such dynamics and help understand the underlying brain mechanisms, which can also be used as potential biomarkers. However, applying deep learning to model dynamical systems at the individual level must overcome numerous computational challenges to be reproducible and clinically useful. In this study, we performed an extensive analysis of these challenges using generative modeling of brain dynamics from fMRI data as an example and demonstrated their impact on classifying patients with schizophrenia and major depression. This study highlights the tendency of deep learning models to identify functionally unique solutions during parameter optimization, which severely impacts the reproducibility of downstream predictions. We hope this study guides the future development of individual-level generative models and similar machine learning approaches aimed at identifying reproducible biomarkers of mental illness.

Informations de copyright

© 2024 The Author(s).

Déclaration de conflit d'intérêts

A.M.-L. has received consultant fees from Agence Nationale de la Recherche, Brain Mind Institute, Brainsway, CISSN (Catania International Summer School of Neuroscience), Daimler und Benz Stiftung, Fondation FondaMental, Hector Stiftung II, Janssen-Cilag GmbH, Lundbeck A/S, Lundbeckfonden, Lundbeck Int. Neuroscience Foundation, MedinCell, Sage Therapeutics, Techspert.io, The LOOP Zürich, University Medical Center Utrecht, von Behring Röntgen Stiftung. A.M.-L. has received speaker fees from Arztekammer Nordrhein, BAG Psychiatrie Oberbayern, Biotest AG, Forum Werkstatt Karlsruhe, International Society of Psychiatric Genetics, Brentwood, Klinik für Psychiatrie und Psychotherapie Ingolstadt, Lundbeck SAS France, med Update GmbH, Merz-Stiftung, Siemens Healthineers, Society of Biological Psychiatry. A.M.-L. has received editorial fees from American Association for the Advancement of Science, Elsevier, Thieme Verlag. E.S. received speaker fees from bfd buchholz-fachingormationsdienst GmbH and editorial fees from Lundbeckfonden.