Category-Specific Brain Structural Changes in Psychiatric Disorders and Conditions Using Well-Qualified Harmonization Dataset from Multiple Sites

Background Large-scale, multi-site brain magnetic resonance imaging (MRI) studies have revealed neuroanatomical alterations in various psychiatric disorders. However, site-specific differences—such as sampling and measurement biases—and the non-linear nature of brain features can limit individual-level investigations. These constraints underscore the need for a meta-analytic approach to accurately capture and compare brain structural changes across different psychiatric conditions. Methods Structural MRI data were collected from 1736 healthy controls and 1371 participants with clinical conditions, including major depressive disorder, schizophrenia, bipolar disorder, first-episode psychosis, ultra-high risk for psychosis, and autism spectrum disorder. A harmonization procedure was applied to reduce measurement bias across sites. Subsequently, deviations from normative trajectories—modeled on healthy controls—were estimated to diminish non-disease-related variations while preserving disease-specific patterns of brain changes. Results Significant reductions in cortical thickness, surface area, and subcortical volumes were observed in most clinical groups. The effect sizes exceeded those reported in previous meta-analyses, indicating enhanced sensitivity in detecting subtle neuroanatomical alterations. Furthermore, distinct profiles of brain alterations emerged across the examined psychiatric disorders and conditions, highlighting the heterogeneous nature of structural changes. Conclusion This study provides important insights into the neuroanatomical underpinnings of psychiatric disorders. By leveraging harmonized datasets and robust statistical methods, we demonstrate improved detection of structural alterations across psychiatric disorders, underscoring the value of multi-site collaborations and advanced analytic techniques for elucidating the complex neurobiology of psychiatric conditions.