Can we diagnose mental disorders in children? A large-scale assessment of machine learning on structural neuroimaging of 6916 children in the adolescent brain cognitive development study.

Background: Prediction of mental disorders based on neuroimaging is an emerging area of research with promising first results in adults. However, research on the unique demographic of children is underrepresented and it is doubtful whether findings obtained on adults can be transferred to children. Methods: Using data from 6916 children aged 9-10 in the multicenter Adolescent Brain Cognitive Development study, we extracted 136 regional volume and thickness measures from structural magnetic resonance images to rigorously evaluate the capabilities of machine learning to predict 10 different psychiatric disorders: major depressive disorder, bipolar disorder (BD), psychotic symptoms, attention deficit hyperactivity disorder (ADHD), oppositional defiant disorder, conduct disorder, post-traumatic stress disorder, obsessive-compulsive disorder, generalized anxiety disorder, and social anxiety disorder. For each disorder, we performed cross-validation and assessed whether models discovered a true pattern in the data via permutation testing. Results: Two of 10 disorders can be detected with statistical significance when using advanced models that (i) allow for non-linear relationships between neuroanatomy and disorder, (ii) model interdependencies between disorders, and (iii) avoid confounding due to sociodemographic factors: ADHD (AUROC = 0.567, p = 0.002) and BD (AUROC = 0.551, p = 0.002). In contrast, traditional models perform consistently worse and predict only ADHD with statistical significance (AUROC = 0.529, p = 0.002). Conclusion: While the modest absolute classification performance does not warrant application in the clinic, our results provide empirical evidence that embracing and explicitly accounting for the complexities of mental disorders via advanced machine learning models can discover patterns that would remain hidden with traditional models.

Reduced cortical neuron number and neuron density in schizophrenia with focus on area 24: a post-mortem case-control study.

Structural and functional abnormalities of the anterior cingulate cortex (ACC) have frequently been identified in schizophrenia. Alterations of von Economo neurons (VENs), a class of specialized projection neurons, have been found in different neuropsychiatric disorders and are also suspected in schizophrenia. To date, however, no definitive conclusions can be drawn about quantitative histologic changes in the ACC in schizophrenia because of a lack of rigorous, design-based stereologic studies. In the present study, the volume, total neuron number and total number of VENs in layer V of area 24 were determined in both hemispheres of postmortem brains from 12 male patients with schizophrenia and 11 age-matched male controls. To distinguish global from local effects, volume and total neuron number were also determined in the whole area 24 and whole cortical gray matter (CGM). Measurements were adjusted for hemisphere, age, postmortem interval and fixation time using an ANCOVA model. Compared to controls, patients with schizophrenia showed alterations, with lower mean total neuron number in CGM (- 14.9%, P = 0.007) and in layer V of area 24 (- 21.1%, P = 0.002), and lower mean total number of VENs (- 28.3%, P = 0.027). These data provide evidence for ACC involvement in the pathophysiology of schizophrenia, and complement neuroimaging findings of impaired ACC connectivity in schizophrenia. Furthermore, these results support the hypothesis that the clinical presentation of schizophrenia, particularly deficits in social cognition, is associated with pathology of VENs.