Neuroimaging insights into the neurophysiological subtypes of major depressive disorder

Major depressive disorder (MDD) is a highly heterogeneous condition that limits the reliability of symptom-based diagnosis and treatment selection. In response, neuroimaging-based subtyping has emerged as a promising strategy to address this heterogeneity and advance precision psychiatry. Here, we provide a critical synthesis of neuroimaging-based subtyping research in MDD with four central contributions. First, we integrate recent methodological advances-including unsupervised and semi-supervised clustering, deep learning, and normative modeling-that move the field beyond group-level averages toward individualized deviation profiles. Second, we compare convergent and divergent subtype patterns across functional, structural, diffusion, and multimodal imaging, highlighting both shared organizational principles and modality-specific dimensions of heterogeneity. Third, we evaluate emerging evidence linking neurophysiological subtypes to symptom dimensions, illness trajectories, and treatment responses and outline a translational framework for clinical implementation. Finally, we identify key challenges and actionable future directions, including the creation of large-scale harmonized datasets, rigorous validation, and integration with physiological, genetic, and environmental data. Together, this review clarifies the current state of the neuroimaging-based subtyping of MDD and delineates a roadmap for translating brain-based heterogeneity into clinically meaningful advances.