AI meets psychiatry
The MindShift project at the University of Marburg's Faculty of Medicine is receiving over € 870,000 in funding from the state of Hesse. The aim of the project is to develop and validate personalized, AI-supported neurostimulation approaches to treat depression and anxiety disorders. With this project, Marburg University will strengthen its international visibility in the field of computational psychiatry and make an important contribution to the development of personalized, data-driven approaches in psychiatric care.

The project team (from left to right): Marco Rothermel, Bita Shariatpanahi, Dr. Felix Bernhard, Prof. Hamidreza Jamalabadi, Svenja Francke, Timo Becker, Alina Buschhüter, Dr. Sarah Alizadeh. Photo: Shakiba Bakhshian, Bita Sharaiatapanah
Bernstein members involved: Sarah Alizadeh, Hamidreza Jamalabadi
Computational psychiatry is an interdisciplinary field of research at the intersection of psychiatry, neuroscience, psychology, computer science, and mathematics. Its goal is to better understand, diagnose, and treat mental illness using formal models and data-driven methods. “MindShift” builds on numerous studies conducted at the School of Medicine, which showed that the technology developed within the project can reliably and consistently predict perceptions and specifically influence information processing in the brain. This makes it possible to efficiently correct negative cognitive distortions, such as those that occur in depression or anxiety disorders.
“We are delighted about the funding—it will enable us to scale our technology in a targeted manner and bring it closer to social and market-oriented applications,” says project lead Prof. Dr. Hamidreza Jamalabadi. “MindShift creates the necessary framework to comprehensively validate our scientific approaches and thus lay the foundation for external investment. This will enable us to take the next step: transferring our technology into clinically viable applications for those who urgently need them.”
Specifically, the project team wants to use artificial intelligence to develop methods that can be used to tailor existing brain stimulation devices to individual patients. To this end, software is being developed to help doctors find the most effective settings for different individuals and put them into practice. The focus is on three common neurostimulation systems: tVNS, tES, and TMS.
Contentwise, “MindShift” is closely embedded in the current research structures at the University of Marburg. The project is directly related to the recently approved Cluster of Excellence EXC 3066 “The Adaptive Mind” and the DFG Collaborative Research Center SFB/TRR 393 “Trajectories of Affective Disorders,” in which Jamalabadi serves as principal investigator. These networks create an integrative research environment that systematically combines basic research, clinical psychiatry, and modern AI methods.
“The MindShift project is of great importance for Marburg and its neuroscientific and psychiatric research community, as it ties in closely with existing initiatives and allows for consistent translational development with a clear social impact,” says Jamalabadi.
Computational Psychiatry research group
The Computational Psychiatry research group, led by Prof. Dr. Hamidreza Jamalabadi at Marburg University’s School of Medicine, is investigating, among other things, how neural, psychological, and behavioral processes develop dynamically in humans—and how they change in the case of mental illness. Using methods from systems theory, network analysis, and machine learning, they aim to quantify, predict, and specifically influence these processes. To this end, behavioral and symptom data are combined with modern brain imaging techniques and neurostimulation methods over extended periods of time. The long-term goal is to treat mental illness more effectively through adaptive, individually tailored treatment approaches.
Translated from German by Elena Reiriz Martínez/BCOS




