Background Previous analyses of grey and white matter volumes have reported that schizophrenia is associated with structural changes. Deep learning is a data-driven approach that can capture highly compact hierarchical non-linear relationships among high-dimensional features, and therefore can facilitate the development of clinical tools for making a more accurate and earlier diagnosis of schizophrenia. Aims To identify consistent grey matter abnormalities in patients with schizophrenia, 662 people with schizophrenia and 613 healthy controls were recruited from eight centres across China, and the data from these independent sites were used to validate deep-learning classifiers. Method We used a prospective image-based meta-analysis of whole-brain voxel-based morphometry. We also automatically differentiated patients with schizophrenia from healthy controls using combined grey matter, white matter and cerebrospinal fluid volumetric features, incorporated a deep neural network approach on an individual basis, and tested the generalisability of the classification models using independent validation sites. Results We found that statistically reliable schizophrenia-related grey matter abnormalities primarily occurred in regions that included the superior temporal gyrus extending to the temporal pole, insular cortex, orbital and middle frontal cortices, middle cingulum and thalamus. Evaluated using leave-one-site-out cross-validation, the performance of the classification of schizophrenia achieved by our findings from eight independent research sites were: accuracy, 77.19–85.74%; sensitivity, 75.31–89.29% and area under the receiver operating characteristic curve, 0.797–0.909. Conclusions These results suggest that, by using deep-learning techniques, multidimensional neuroanatomical changes in schizophrenia are capable of robustly discriminating patients with schizophrenia from healthy controls, findings which could facilitate clinical diagnosis and treatment in schizophrenia.

Consciousness emerges from the spatiotemporal dynamics of neural activity. However, how such an extraordinary phenomenon is supported by neural flexibility and regional specialization across the cerebral cortex remains elusive. Here, using several functional magnetic resonance imaging (fMRI) paradigms (e.g., anesthesia, sleep, and drowsiness), we show a consciousness-related signature characterized by shifting spontaneous fluctuations along a unimodal-transmodal cortical organizational axis. The signature is simple and sensitive to altered states of consciousness in single individuals, exhibiting abnormal elevation under psychedelics as well as in individuals with psychosis. Under task-free conditions, this hierarchical dynamic reflects ongoing brain state changes in global integration and connectome diversity. Applying quasi-periodic pattern detection during different vigilance states, we found that hierarchical heterogeneity also manifested as spatiotemporally propagation waves related to arousal. A similar pattern was observed in neural activity measured with cortex-wide electrocorticography (ECoG) in macaques. Last, we observed that the spatial distribution of the principle cortical gradient not only preferentially recapitulated the genetic transcription levels of the histaminergic system but also that of the functional connectome mapping of the tuberomammillary nucleus (TMN), which promotes wakefulness. Combining behavioral, neuroimaging, electrophysiological, and transcriptomic evidence, we suggest that global state of consciousness is supported by efficient hierarchical processing that can be constrained along a low-dimensional macroscale gradient.

We aimed to investigate the relationship between spatiotemporal changes of amyloid deposition and Alzheimer’s disease (AD) profiles in cognitively normal (CN) and those with mild cognitive impairment (MCI). Using a data-driven method and amyloid-PET data, we identified and validated two subtypes in two independent datasets (discovery dataset: N = 548, age = 72.4 ± 6.78, 49% female; validation dataset: N = 348, age = 74.9 ± 8.16, 47% female) from the Alzheimer’s Disease Neuroimaging Initiative across a range of individuals who were CN or had MCI. The two subtypes showed distinct regional progression patterns and presented distinct genetic, clinical and biomarker characteristics. The cortex-priority subtype was more likely to show typical clinical syndromes of symptomatic AD and vice versa. Furthermore, the regional progression patterns were associated with clinical and biomarker profiles. In sum, our findings suggest that the spatiotemporal variants of amyloid depositions are in close association with disease trajectories; these findings may provide insight into the disease monitoring and enrollment of therapeutic trials in AD.