We are developing and validating imaging-based machine-learning models that integrate structural and functional brain data with clinical information. This combination allows us to capture the complexity of mental disorders from multiple perspectives, identifying subtle patterns that are not visible through traditional methods. By leveraging large and diverse datasets, we aim to build models that are robust, generalizable, and clinically meaningful.

These tools are designed to improve the accuracy of diagnosis and to predict the course and treatment outcomes of psychiatric conditions, particularly mood- and anxiety-related disorders. By combining advanced neuroimaging with artificial intelligence, our ultimate goal is to equip clinicians with decision-support systems that can guide more personalized and effective interventions, moving the field closer to true precision psychiatry.