Highlights We develop a dual-branch combination network (DCN) for combined segmentation and classification of COVID-19 using CT images. Inspired by the attention mechanism, we propose a lesion attention (LA) module to improve the sensitivity to CT images with small lesions and facilitate early screening of COVID-19. The LA module provide accurate attention maps to improve the interpretability of the network and contribute to further assessment of the classification result.

Recently, resting-state functional magnetic resonance imaging (fMRI) studies have been extended to explore fluctuations in correlations over shorter timescales, referred to as dynamic functional connectivity (dFC). However, the impact of global signal regression (GSR) on dFC is not well established, despite the intensive investigations of the influence of GSR on static functional connectivity (sFC). This study aimed to examine the effect of GSR on the performance of the sliding-window correlation, a commonly used method for capturing functional connectivity (FC) dynamics based on resting-state fMRI and simultaneous electroencephalograph (EEG)-fMRI data. The results revealed that the impact of GSR on dFC was spatially heterogeneous, with some susceptible regions including the occipital cortex, sensorimotor area, precuneus, posterior insula and superior temporal gyrus, and that the impact was temporally modulated by the mean global signal (GS) magnitude across windows. Furthermore, GSR substantially changed the connectivity structures of the FC states responding to a high GS magnitude, as well as their temporal features, and even led to the emergence of new FC states. Conversely, those FC states marked by obvious anti-correlation structures associated with the default model network (DMN) were largely unaffected by GSR. Finally, we reported an association between the fluctuations in the windowed magnitude of GS and the time-varying EEG power within subjects, which implied changes in mental states underlying GS dynamics. Overall, this study suggested a potential neuropsychological basis, in addition to nuisance sources, for GS dynamics and highlighted the need for caution in applying GSR to sliding-window correlation analyses. At a minimum, the mental fluctuations of an individual subject, possibly related to ongoing vigilance, should be evaluated during the entire scan when the dynamics of FC is estimated.

Previous static resting-state functional connectivity (FC) MRI (rs-fcMRI) studies have suggested certain heredity characteristics of schizophrenia. Recently, dynamic rs-fcMRI analysis, which can better characterize the time-varying nature of intrinsic activity and connectivity and may therefore unveil the special connectivity patterns that are always lost in static FC analysis, has shown a potential neuroendophenotype of schizophrenia. In this study, we have extended previous static rs-fcMRI studies to a dynamic study by exploring whether healthy siblings share aberrant dynamic FC patterns with schizophrenic patients, which may imply a potential risk for siblings to develop schizophrenia. We utilized the dynamic rs-fcMRI method using a sliding window approach to evaluate FC discrepancies within transient states across schizophrenic patients, unaffected siblings, and matched healthy controls. Statistical analysis showed five trait-related connections that are related to cingulo-opercular, occipital, and default mode networks, reflecting the shared connectivity alterations between schizophrenic patients and their unaffected siblings. The findings suggested that schizophrenic patients and their unaffected siblings shared common transient functional disconnectivity, implying a potential risk for the healthy siblings of developing schizophrenia.

Increasing evidence has suggested that the dynamic properties of functional brain networks are related to individual behaviors and cognition traits. However, current fMRI-based approaches mostly focus on statistical characteristics of the windowed correlation time course, potentially overlooking subtle time-varying patterns in dynamic functional connectivity (dFC). Here, we proposed the use of an end-to-end deep learning model that combines the convolutional neural network (CNN) and long short-term memory (LSTM) network to capture temporal and spatial features of functional connectivity sequences simultaneously. The results on a large cohort (Human Connectome Project, n = 1,050) demonstrated that our model could achieve a high classification accuracy of about 93% in a gender classification task and prediction accuracies of 0.31 and 0.49 (Pearson's correlation coefficient) in fluid and crystallized intelligence prediction tasks, significantly outperforming previously reported models. Furthermore, we demonstrated that our model could effectively learn spatiotemporal dynamics underlying dFC with high statistical significance based on the null hypothesis estimated using surrogate data. Overall, this study suggests the advantages of a deep learning model in making full use of dynamic information in resting-state functional connectivity, and highlights the potential of time-varying connectivity patterns in improving the prediction of individualized characterization of demographics and cognition traits.

Objectives To develop and validate a machine learning model for the prediction of adverse outcomes in hospitalized patients with COVID-19. Methods We included 424 patients with non-severe COVID-19 on admission from January 17, 2020, to February 17, 2020, in the primary cohort of this retrospective multicenter study. The extent of lung involvement was quantified on chest CT images by a deep learning–based framework. The composite endpoint was the occurrence of severe or critical COVID-19 or death during hospitalization. The optimal machine learning classifier and feature subset were selected for model construction. The performance was further tested in an external validation cohort consisting of 98 patients. Results There was no significant difference in the prevalence of adverse outcomes (8.7% vs. 8.2%, p = 0.858) between the primary and validation cohorts. The machine learning method extreme gradient boosting (XGBoost) and optimal feature subset including lactic dehydrogenase (LDH), presence of comorbidity, CT lesion ratio (lesion%), and hypersensitive cardiac troponin I (hs-cTnI) were selected for model construction. The XGBoost classifier based on the optimal feature subset performed well for the prediction of developing adverse outcomes in the primary and validation cohorts, with AUCs of 0.959 (95% confidence interval [CI]: 0.936–0.976) and 0.953 (95% CI: 0.891–0.986), respectively. Furthermore, the XGBoost classifier also showed clinical usefulness. Conclusions We presented a machine learning model that could be effectively used as a predictor of adverse outcomes in hospitalized patients with COVID-19, opening up the possibility for patient stratification and treatment allocation. Key Points • Developing an individually prognostic model for COVID-19 has the potential to allow efficient allocation of medical resources. • We proposed a deep learning–based framework for accurate lung involvement quantification on chest CT images. • Machine learning based on clinical and CT variables can facilitate the prediction of adverse outcomes of COVID-19. Supplementary Information The online version contains supplementary material available at 10.1007/s00330-021-07957-z.

Until now, dynamic functional connectivity (dFC) based on functional magnetic resonance imaging is typically estimated on a set of predefined regions of interest (ROIs) derived from an anatomical or static functional atlas which follows an implicit assumption of functional homogeneity within ROIs underlying temporal fluctuation of functional coupling, potentially leading to biases or underestimation of brain network dynamics. Here, we presented a novel computational method based on dynamic functional connectivity degree (dFCD) to derive meaningful brain parcellations that can capture functional homogeneous regions in temporal variance of functional connectivity. Several spatially distributed but functionally meaningful areas that are well consistent with known intrinsic connectivity networks were identified through independent component analysis (ICA) of time‐varying dFCD maps. Furthermore, a systematical comparison with commonly used brain atlases, including the Anatomical Automatic Labeling template, static ICA‐driven parcellation and random parcellation, demonstrated that the ROI‐definition strategy based on the proposed dFC‐driven parcellation could better capture the interindividual variability in dFC and predict observed individual cognitive performance (e.g., fluid intelligence, cognitive flexibility, and sustained attention) based on chronnectome. Together, our findings shed new light on the functional organization of resting brains at the timescale of seconds and emphasized the significance of a dFC‐driven and voxel‐wise functional homogeneous parcellation for network dynamics analyses in neuroscience.

The current study aimed to investigate the altered cerebellar-cerebral functional connectivity (FC) induced by radiotherapy to nasopharyngeal carcinoma (NPC) patients. Twenty-four NPC patients without treatment, and 35 NPC patients receiving radiotherapy underwent functional MRI scanning. Montreal cognitive assessment (MoCA) was performed to evaluate the cognitive status of all participants. FC between 10 predefined cerebellar seeds, which were demonstrated to be involved in different brain functional networks, and all brain voxels was obtained for each participant. Using a second-level two-sample t-test, three significantly different FCs between the two patient groups were found, including the connections between the left lobule VIII and the right medial frontal gyrus, the left lobule VIII and the right crus I, and the right lobule VIIb and the right fusiform gyrus. The altered cerebellar-cerebral FCs were also significantly correlated to the MoCA score, as well as the attention score, one of the seven subscores in MoCA. We suggested that the altered cerebellar-cerebral FCs may underlie the radiation-induced cognitive deficits in NPC patients, especially in the domain of attention. Furthermore, considering the functional networks in which the altered connections involved, the anticorrelation between the default network and dorsal attention network may be impaired, and the mediating function of the frontoparietal network to dorsal attention network may be disrupted. The significantly altered cerebellar-cerebral FC may serve as the potential biomarker in revealing the radiation-induced functional abnormalities and may help in the early intervention to the cognitive impairment.

Limited sample size hinders the application of deep learning in brain image analysis, and transfer learning is a possible solution. However, most pretrained models are 2D based and cannot be applied directly to 3D brain images. In this study, we propose a novel framework to apply 2D pretrained models to 3D brain images by projecting surface-based cortical morphometry into planar images using computational geometry mapping. Firstly, 3D cortical meshes are reconstructed from magnetic resonance imaging (MRI) using FreeSurfer and projected into 2D planar meshes with topological preservation based on area-preserving geometry mapping. Then, 2D deep models pretrained on ImageNet are adopted and fine-tuned for cortical image classification on morphometric shape metrics. We apply the framework to sex classification on the Human Connectome Project dataset and autism spectrum disorder (ASD) classification on the Autism Brain Imaging Data Exchange dataset. Moreover, a 2-stage transfer learning strategy is suggested to boost the ASD classification performance by using the sex classification as an intermediate task. Our framework brings significant improvement in sex classification and ASD classification with transfer learning. In summary, the proposed framework builds a bridge between 3D cortical data and 2D models, making 2D pretrained models available for brain image analysis in cognitive and psychiatric neuroscience.