This repository delivers an end-to-end, research-grade platform for personalized mental health optimization, developed in the context of a Ghent University master’s thesis; integrating a custom ontology, hierarchical time-series analytics and agentic reasoning for actionable intervention design.

🧠 Machine learning analysis for the paper named "Predicting 3-year persistent or recurrent major depressive episode using machine learning techniques".

Research code for DeCoDE brain-behavior biotyping analyses.

Research code for multimodal graph modeling in major depressive disorder.

Research code for contrastive learning analyses in major depressive disorder.

Research code for target-oriented multimodal fusion models for antidepressant response prediction.