The construction of Z-scheme heterojunction ZnIn2S4@CuO with enhanced charge transfer capability and its mechanism study for the visible light degradation of tetracycline.

In this study, copper oxide (CuO) was prepared by the microwave-assisted hydrothermal technique subsequently, CuO was grown in situ onto different rare metal compounds to prepare Z-scheme heterojunctions to improve the degradation efficiency of tetracycline (TC) in water environments. Various characterization proved the successful synthesis of all composite materials, and the formation of tight heterojunction interfaces, among which, the core-shell structure ZnIn2S4@CuO exhibited excellent photocatalytic degradation capability. Research results indicated that the degradation efficiency of ZnIn2S4@CuO for TC (50 mg/L) in the water environment reached 95.8 %, and the degradation rate is 2.41 times and 12.93 times that of CuO and ZnIn2S4 alone, respectively, the reason is because of the introduction of ZnIn2S4, Z-scheme heterojunction structures and internal electric field (IEF) is constructed and formed to extend the visible light response range of photocatalysts to improve electron-hole separation efficiency, and enhance charge transfer. In addition, ZnIn2S4@CuO-2 exhibited good stability and reproducibility, with no significant loss of activity after five cycles. Finally, the precise locations of free radical attack on TC were investigated by the combined use of high-resolution mass spectrometry (HR-MC) and frontier electron densities (FEDs), and a reasonable degradation pathway was provided. The results of this research provide a new and viable approach to overcome the limitations of conventional photocatalytic materials in terms of limited visible light absorption range and fast carrier recombination rates, which offers promising prospects for a wide range of applications in the field of wastewater purification.

Verbal memory network mapping in individual patients predicts postoperative functional impairments.

Verbal memory decline is a significant concern following temporal lobe surgeries in patients with epilepsy, emphasizing the need for precision presurgical verbal memory mapping to optimize functional outcomes. However, the inter-individual variability in functional networks and brain function-structural dissociations pose challenges when relying solely on group-level atlases or anatomical landmarks for surgical guidance. Here, we aimed to develop and validate a personalized functional mapping technique for verbal memory using precision resting-state functional MRI (rs-fMRI) and neurosurgery. A total of 38 patients with refractory epilepsy scheduled for surgical interventions were enrolled and 28 patients were analyzed in the study. Baseline 30-min rs-fMRI scanning, verbal memory and language assessments were collected for each patient before surgery. Personalized verbal memory networks (PVMN) were delineated based on preoperative rs-fMRI data for each patient. The accuracy of PVMN was assessed by comparing post-operative functional impairments and the overlapping extent between PVMN and surgical lesions. A total of 14 out of 28 patients experienced clinically meaningful declines in verbal memory after surgery. The personalized network and the group-level atlas exhibited 100% and 75.0% accuracy in predicting postoperative verbal memory declines, respectively. Moreover, six patients with extra-temporal lesions that overlapped with PVMN showed selective impairments in verbal memory. Furthermore, the lesioned ratio of the personalized network rather than the group-level atlas was significantly correlated with postoperative declines in verbal memory (personalized networks: r = -0.39, p = .038; group-level atlas: r = -0.19, p = .332). In conclusion, our personalized functional mapping technique, using precision rs-fMRI, offers valuable insights into individual variability in the verbal memory network and holds promise in precision verbal memory network mapping in individuals.

SG-Transunet: A segmentation-guided Transformer U-Net model for KRAS gene mutation status identification in colorectal cancer.

Accurately identifying the Kirsten rat sarcoma virus (KRAS) gene mutation status in colorectal cancer (CRC) patients can assist doctors in deciding whether to use specific targeted drugs for treatment. Although deep learning methods are popular, they are often affected by redundant features from non-lesion areas. Moreover, existing methods commonly extract spatial features from imaging data, which neglect important frequency domain features and may degrade the performance of KRAS gene mutation status identification. To address this deficiency, we propose a segmentation-guided Transformer U-Net (SG-Transunet) model for KRAS gene mutation status identification in CRC. Integrating the strength of convolutional neural networks (CNNs) and Transformers, SG-Transunet offers a unique approach for both lesion segmentation and KRAS mutation status identification. Specifically, for precise lesion localization, we employ an encoder-decoder to obtain segmentation results and guide the KRAS gene mutation status identification task. Subsequently, a frequency domain supplement block is designed to capture frequency domain features, integrating it with high-level spatial features extracted in the encoding path to derive advanced spatial-frequency domain features. Furthermore, we introduce a pre-trained Xception block to mitigate the risk of overfitting associated with small-scale datasets. Following this, an aggregate attention module is devised to consolidate spatial-frequency domain features with global information extracted by the Transformer at shallow and deep levels, thereby enhancing feature discriminability. Finally, we propose a mutual-constrained loss function that simultaneously constrains the segmentation mask acquisition and gene status identification process. Experimental results demonstrate the superior performance of SG-Transunet over state-of-the-art methods in discriminating KRAS gene mutation status.

SUGAR: Spherical ultrafast graph attention framework for cortical surface registration.

Cortical surface registration plays a crucial role in aligning cortical functional and anatomical features across individuals. However, conventional registration algorithms are computationally inefficient. Recently, learning-based registration algorithms have emerged as a promising solution, significantly improving processing efficiency. Nonetheless, there remains a gap in the development of a learning-based method that exceeds the state-of-the-art conventional methods simultaneously in computational efficiency, registration accuracy, and distortion control, despite the theoretically greater representational capabilities of deep learning approaches. To address the challenge, we present SUGAR, a unified unsupervised deep-learning framework for both rigid and non-rigid registration. SUGAR incorporates a U-Net-based spherical graph attention network and leverages the Euler angle representation for deformation. In addition to the similarity loss, we introduce fold and multiple distortion losses to preserve topology and minimize various types of distortions. Furthermore, we propose a data augmentation strategy specifically tailored for spherical surface registration to enhance the registration performance. Through extensive evaluation involving over 10,000 scans from 7 diverse datasets, we showed that our framework exhibits comparable or superior registration performance in accuracy, distortion, and test-retest reliability compared to conventional and learning-based methods. Additionally, SUGAR achieves remarkable sub-second processing times, offering a notable speed-up of approximately 12,000 times in registering 9,000 subjects from the UK Biobank dataset in just 32 min. This combination of high registration performance and accelerated processing time may greatly benefit large-scale neuroimaging studies.

Defect-engineering improves the activity of Metal-Organic frameworks for catalyzing hydroboration of Alkynes: A combination of experimental investigation and Density functional theory calculations.

Defect-engineered metal-organic frameworks (DEMOFs) are emerging advanced materials. The construction of DEMOFs is of great significance; however, DEMOF-based catalysis remains unexplored. (E)-vinylboronates, an important building block for asymmetric synthesis, can be synthesized via the hydroboration of alkynes. However, the lack of high-performance catalysts considerably hinders their synthesis. Herein, a series of DEHKUST-1 (HKUST = Hong Kong University of Science and Technology) (Da-f) catalysts with missing occupation of linkers at Cu nodes were designed by partially replacing benzene-1,3,5-tricarboxylate (H3BTC) with defective connectors of pyridine-3,5-dicarboxylate (PYDC) to efficiently promote the hydroboration of alkynes. Results showed that the Dd containing 0.8 doping ratio of PYDC exhibited remarkable catalytic activity than the defect-free HKUST-1. This originated from the improved accessibility for reactants towards the Lewis acid active Cu sites of DEHKUST-1 due to the presence of plenty of rooms next to the Cu sites and enhanced coordination ability in such 'defective' HKUST-1. Dd had high selectivity (>99 %) and yield (>96 %) for (E)-vinylboronates and extensive functional group compatibility for terminal alkynes. Density functional theory (DFT) calculations were performed to elucidate the mechanism of hydroboration. Compared with that of defect-free HKUST-1, the low energy barrier of DEHKUST-1 can be attributed to the lower coordination number of Cu sites and enhanced accessibility of Cu active sites towards reagents.

Role of succinylation modification in central nervous system diseases.

Diseases of the central nervous system (CNS), including stroke, brain tumors, and neurodegenerative diseases, have a serious impact on human health worldwide, especially in elderly patients. The brain, which is one of the body's most metabolically dynamic organs, lacks fuel stores and therefore requires a continuous supply of energy substrates. Metabolic abnormalities are closely associated with the pathogenesis of CNS disorders. Post-translational modifications (PTMs) are essential regulatory mechanisms that affect the functions of almost all proteins. Succinylation, a broad-spectrum dynamic PTM, primarily occurs in mitochondria and plays a crucial regulatory role in various diseases. In addition to directly affecting various metabolic cycle pathways, succinylation serves as an efficient and rapid biological regulatory mechanism that establishes a connection between metabolism and proteins, thereby influencing cellular functions in CNS diseases. This review offers a comprehensive analysis of succinylation and its implications in the pathological mechanisms of CNS diseases. The objective is to outline novel strategies and targets for the prevention and treatment of CNS conditions.

A glimpse into novel acylations and their emerging role in regulating cancer metastasis.

Metastatic cancer is a major cause of cancer-related mortality; however, the complex regulation process remains to be further elucidated. A large amount of preliminary investigations focus on the role of epigenetic mechanisms in cancer metastasis. Notably, the posttranslational modifications were found to be critically involved in malignancy, thus attracting considerable attention. Beyond acetylation, novel forms of acylation have been recently identified following advances in mass spectrometry, proteomics technologies, and bioinformatics, such as propionylation, butyrylation, malonylation, succinylation, crotonylation, 2-hydroxyisobutyrylation, lactylation, among others. These novel acylations play pivotal roles in regulating different aspects of energy mechanism and mediating signal transduction by covalently modifying histone or nonhistone proteins. Furthermore, these acylations and their modifying enzymes show promise regarding the diagnosis and treatment of tumors, especially tumor metastasis. Here, we comprehensively review the identification and characterization of 11 novel acylations, and the corresponding modifying enzymes, highlighting their significance for tumor metastasis. We also focus on their potential application as clinical therapeutic targets and diagnostic predictors, discussing the current obstacles and future research prospects.

Deep Multiview Module Adaption Transfer Network for Subject-Specific EEG Recognition.

Transfer learning is one of the popular methods to solve the problem of insufficient data in subject-specific electroencephalogram (EEG) recognition tasks. However, most existing approaches ignore the difference between subjects and transfer the same feature representations from source domain to different target domains, resulting in poor transfer performance. To address this issue, we propose a novel subject-specific EEG recognition method named deep multiview module adaption transfer (DMV-MAT) network. First, we design a universal deep multiview (DMV) network to generate different types of discriminative features from multiple perspectives, which improves the generalization performance by extensive feature sets. Second, module adaption transfer (MAT) is designed to evaluate each module by the feature distributions of source and target samples, which can generate an optimal weight sharing strategy for each target subject and promote the model to learn domain-invariant and domain-specific features simultaneously. We conduct extensive experiments in two EEG recognition tasks, i.e., motor imagery (MI) and seizure prediction, on four datasets. Experimental results demonstrate that the proposed method achieves promising performance compared with the state-of-the-art methods, indicating a feasible solution for subject-specific EEG recognition tasks. Implementation codes are available at https://github.com/YangLibuaa/DMV-MAT.

A Cross-Scale Transformer and Triple-View Attention Based Domain-Rectified Transfer Learning for EEG Classification in RSVP Tasks.

Rapid serial visual presentation (RSVP)-based brain-computer interface (BCI) is a promising target detection technique by using electroencephalogram (EEG) signals. However, existing deep learning approaches seldom considered dependencies of multi-scale temporal features and discriminative multi-view spectral features simultaneously, which limits the representation learning ability of the model and undermine the EEG classification performance. In addition, recent transfer learning-based methods generally failed to obtain transferable cross-subject invariant representations and commonly ignore the individual-specific information, leading to the poor cross-subject transfer performance. In response to these limitations, we propose a cross-scale Transformer and triple-view attention based domain-rectified transfer learning (CST-TVA-DRTL) for the RSVP classification. Specially, we first develop a cross-scale Transformer (CST) to extract multi-scale temporal features and exploit the dependencies of different scales features. Then, a triple-view attention (TVA) is designed to capture spectral features from triple views of multi-channel time-frequency images. Finally, a domain-rectified transfer learning (DRTL) framework is proposed to simultaneously obtain transferable domain-invariant representations and untransferable domain-specific representations, then utilize domain-specific information to rectify domain-invariant representations to adapt to target data. Experimental results on two public RSVP datasets suggests that our CST-TVA-DRTL outperforms the state-of-the-art methods in the RSVP classification task. The source code of our model is publicly available in https://github.com/ljbuaa/CST_TVA_DRTL.

Individual connectivity-based parcellations reflect functional properties of human auditory cortex.

Neuroimaging studies of the functional organization of human auditory cortex have focused on group-level analyses to identify tendencies that represent the typical brain. Here, we mapped auditory areas of the human superior temporal cortex (STC) in 30 participants by combining functional network analysis and 1-mm isotropic resolution 7T functional magnetic resonance imaging (fMRI). Two resting-state fMRI sessions, and one or two auditory and audiovisual speech localizer sessions, were collected on 3-4 separate days. We generated a set of functional network-based parcellations from these data. Solutions with 4, 6, and 11 networks were selected for closer examination based on local maxima of Dice and Silhouette values. The resulting parcellation of auditory cortices showed high intraindividual reproducibility both between resting state sessions (Dice coefficient: 69-78%) and between resting state and task sessions (Dice coefficient: 62-73%). This demonstrates that auditory areas in STC can be reliably segmented into functional subareas. The interindividual variability was significantly larger than intraindividual variability (Dice coefficient: 57%-68%, p<0.001), indicating that the parcellations also captured meaningful interindividual variability. The individual-specific parcellations yielded the highest alignment with task response topographies, suggesting that individual variability in parcellations reflects individual variability in auditory function. Connectional homogeneity within networks was also highest for the individual-specific parcellations. Furthermore, the similarity in the functional parcellations was not explainable by the similarity of macroanatomical properties of auditory cortex. Our findings suggest that individual-level parcellations capture meaningful idiosyncrasies in auditory cortex organization.

Deep Fusion of Multi-Template Using Spatio-Temporal Weighted Multi-Hypergraph Convolutional Networks for Brain Disease Analysis.

Conventional functional connectivity network (FCN) based on resting-state fMRI (rs-fMRI) can only reflect the relationship between pairwise brain regions. Thus, the hyper-connectivity network (HCN) has been widely used to reveal high-order interactions among multiple brain regions. However, existing HCN models are essentially spatial HCN, which reflect the spatial relevance of multiple brain regions, but ignore the temporal correlation among multiple time points. Furthermore, the majority of HCN construction and learning frameworks are limited to using a single template, while the multi-template carries richer information. To address these issues, we first employ multiple templates to parcellate the rs-fMRI into different brain regions. Then, based on the multi-template data, we propose a spatio-temporal weighted HCN (STW-HCN) to capture more comprehensive high-order temporal and spatial properties of brain activity. Next, a novel deep fusion model of multi-template called spatio-temporal weighted multi-hypergraph convolutional network (STW-MHGCN) is proposed to fuse the STW-HCN of multiple templates, which extracts the deep interrelation information between different templates. Finally, we evaluate our method on the ADNI-2 and ABIDE-I datasets for mild cognitive impairment (MCI) and autism spectrum disorder (ASD) analysis. Experimental results demonstrate that the proposed method is superior to the state-of-the-art approaches in MCI and ASD classification, and the abnormal spatio-temporal hyper-edges discovered by our method have significant significance for the brain abnormalities analysis of MCI and ASD.

Personalized brain MRI revealed distinct functional and anatomical disruptions in Creutzfeldt-Jakob disease and Alzheimer's disease.

AIMS: Creutzfeldt-Jakob disease (CJD) is a lethal neurodegenerative disorder, which leads to a rapidly progressive dementia. This study aimed to examine the cortical alterations in CJD, changes in these brain characteristics over time, and the differences between CJD and Alzheimer's disease (AD) that show similar clinical manifestations. METHODS: To obtain reliable, subject-specific functional measures, we acquired 24 min of resting-state fMRI data from each subject. We applied an individual-based approach to characterize the functional brain organization of 10 patients with CJD, 8 matched patients with AD, and 8 normal controls. We measured cortical atrophy as well as disruption in resting-state functional connectivity (rsFC) and then investigated longitudinal brain changes in a subset of CJD patients. RESULTS: CJD was associated with widespread cortical thinning and weakened rsFC. Compared with AD, CJD showed distinct atrophy patterns and greater disruptions in rsFC. Moreover, the longitudinal data demonstrated that the progressive cortical thinning and disruption in rsFC mainly affected the association rather than the primary cortex in CJD. CONCLUSIONS: CJD shows unique anatomical and functional disruptions in the cerebral cortex, distinct from AD. Rapid progression of CJD affects both the cortical thickness and rsFC in the association cortex.

Impaired visual-motor functional connectivity in first-episode medication-naïve patients with major depressive disorder.

The perceptual dysfunctions have been fundamental causes of cognitive and emotional problems in patients with major depressive disorder. However, visual system impairment in depression has been underexplored. Here, we explored functional connectivity in a large cohort of first-episode medication-naïve patients with major depressive disorder (n = 190) and compared it with age- and sex-matched healthy controls (n = 190). A recently developed individual-oriented approach was applied to parcellate the cerebral cortex into 92 regions of interest using resting-state functional magnetic resonance imaging data. Significant reductions in functional connectivities were observed between the right lateral occipitotemporal junction within the visual network and 2 regions of interest within the sensorimotor network in patients. The volume of right lateral occipitotemporal junction was also significantly reduced in major depressive disorder patients, indicating that this visual region is anatomically and functionally impaired. Behavioral correlation analysis showed that the reduced functional connectivities were significantly associated with inhibition control in visual-motor processing in patients. Taken together, our data suggest that functional connectivity between visual network and sensorimotor network already shows a significant reduction in the first episode of major depressive disorder, which may interfere with the inhibition control in visual-motor processing. The lateral occipitotemporal junction may be a hub of disconnection and may play a role in the pathophysiology of major depressive disorder.

A Multi-Graph Cross-Attention-Based Region-Aware Feature Fusion Network Using Multi-Template for Brain Disorder Diagnosis.

Functional connectivity (FC) networks based on resting-state functional magnetic imaging (rs-fMRI) are reliable and sensitive for brain disorder diagnosis. However, most existing methods are limited by using a single template, which may be insufficient to reveal complex brain connectivities. Furthermore, these methods usually neglect the complementary information between static and dynamic brain networks, and the functional divergence among different brain regions, leading to suboptimal diagnosis performance. To address these limitations, we propose a novel multi-graph cross-attention based region-aware feature fusion network (MGCA-RAFFNet) by using multi-template for brain disorder diagnosis. Specifically, we first employ multi-template to parcellate the brain space into different regions of interest (ROIs). Then, a multi-graph cross-attention network (MGCAN), including static and dynamic graph convolutions, is developed to explore the deep features contained in multi-template data, which can effectively analyze complex interaction patterns of brain networks for each template, and further adopt a dual-view cross-attention (DVCA) to acquire complementary information. Finally, to efficiently fuse multiple static-dynamic features, we design a region-aware feature fusion network (RAFFNet), which is beneficial to improve the feature discrimination by considering the underlying relations among static-dynamic features in different brain regions. Our proposed method is evaluated on both public ADNI-2 and ABIDE-I datasets for diagnosing mild cognitive impairment (MCI) and autism spectrum disorder (ASD). Extensive experiments demonstrate that the proposed method outperforms the state-of-the-art methods. Our source code is available at https://github.com/mylbuaa/MGCA-RAFFNet.

Dynamic Multi-Graph Convolution-Based Channel-Weighted Transformer Feature Fusion Network for Epileptic Seizure Prediction.

Electroencephalogram (EEG) based seizure prediction plays an important role in the closed-loop neuromodulation system. However, most existing seizure prediction methods based on graph convolution network only focused on constructing the static graph, ignoring multi-domain dynamic changes in deep graph structure. Moreover, the existing feature fusion strategies generally concatenated coarse-grained epileptic EEG features directly, leading to the suboptimal seizure prediction performance. To address these issues, we propose a novel multi-branch dynamic multi-graph convolution based channel-weighted transformer feature fusion network (MB-dMGC-CWTFFNet) for the patient-specific seizure prediction with the superior performance. Specifically, a multi-branch (MB) feature extractor is first applied to capture the temporal, spatial and spectral representations fromthe epileptic EEG jointly. Then, we design a point-wise dynamic multi-graph convolution network (dMGCN) to dynamically learn deep graph structures, which can effectively extract high-level features from the multi-domain graph. Finally, by integrating the local and global channel-weighted strategies with the multi-head self-attention mechanism, a channel-weighted transformer feature fusion network (CWTFFNet) is adopted to efficiently fuse the multi-domain graph features. The proposed MB-dMGC-CWTFFNet is evaluated on the public CHB-MIT EEG and a private intracranial sEEG datasets, and the experimental results demonstrate that our proposed method achieves outstanding prediction performance compared with the state-of-the-art methods, indicating an effective tool for patient-specific seizure warning. Our code will be available at: https://github.com/Rockingsnow/MB-dMGC-CWTFFNet.

Neurosurgical robots in China: State of the art and future prospect.

Neurosurgical robots have developed for decades and can effectively assist surgeons to carry out a variety of surgical operations, such as biopsy, stereo-electroencephalography (SEEG), deep brain stimulation (DBS), and so forth. In recent years, neurosurgical robots in China have developed rapidly. This article will focus on several key skills in neurosurgical robots, such as medical imaging systems, automatic manipulator, lesion localization techniques, multimodal image fusion technology, registration method, and vascular imaging technology; introduce the clinical application of neurosurgical robots in China, and look forward to the potential improvement points in the future based on our experience and research in the field.

Focused ultrasound thalamotomy for tremor treatment impacts the cerebello-thalamo-cortical network.

High-intensity Magnetic Resonance-guided Focused Ultrasound (MRgFUS) is a recent, non-invasive line of treatment for medication-resistant tremor. We used MRgFUS to produce small lesions in the thalamic ventral intermediate nucleus (VIM), an important node in the cerebello-thalamo-cortical tremor network, in 13 patients with tremor-dominant Parkinson's disease or essential tremor. Significant tremor alleviation in the target hand ensued (t(12) = 7.21, p < 0.001, two-tailed), which was strongly associated with the functional reorganization of the brain's hand region with the cerebellum (r = 0.91, p < 0.001, one-tailed). This reorganization potentially reflected a process of normalization, as there was a trend of increase in similarity between the hand cerebellar connectivity of the patients and that of a matched, healthy control group (n = 48) after treatment. Control regions in the ventral attention, dorsal attention, default, and frontoparietal networks, in comparison, exhibited no association with tremor alleviation and no normalization. More broadly, changes in functional connectivity were observed in regions belonging to the motor, limbic, visual, and dorsal attention networks, largely overlapping with regions connected to the lesion targets. Our results indicate that MRgFUS is a highly efficient treatment for tremor, and that lesioning the VIM may result in the reorganization of the cerebello-thalamo-cortical tremor network.

MDSCs: The Key Players in the Formation of Pre-Metastatic Niche.

The distal metastasis of tumor cells is viewed as a series of concurrent processes rather than a linear cascade of events. Accompanied with the progression of the primary tumor, a favorable microenvironment, refered as pre-metastatic niche, has been created in pre-metastatic organs and sites by primary tumors for subsequent metastases. The proposal of "pre-metastatic niche" theory brings fresh insight into our understanding of cancer metastasis. Myeloid-derived suppressor cells (MDSCs) are indispensable for the formation of pre-metastatic niche, which empower the niche to favor tumor cell colonization and promote metastasis. In this review, we aim to provide a comprehensive understanding of the regulation of pre-metastatic niche formation by MDSCs and to conceptualize the framework for understanding the related factors involved in cancer metastasis.

Personalized Functional Connectivity Based Spatio-Temporal Aggregated Attention Network for MCI Identification.

Functional connectivity (FC) networks deri- ved from resting-state magnetic resonance image (rs-fMRI) are effective biomarkers for identifying mild cognitive impairment (MCI) patients. However, most FC identification methods simply extract features from group-averaged brain templates, and neglect inter-subject functional variations. Furthermore, the existing methods generally concentrate on spatial correlation among brain regions, resulting in the inefficient capture of the fMRI temporal features. To address these limitations, we propose a novel personalized functional connectivity based dual-branch graph neural network with spatio-temporal aggregated attention (PFC-DBGNN-STAA) for MCI identification. Specifically, a personalized functional connectivity (PFC) template is firstly constructed to align 213 functional regions across samples and generate discriminative individualized FC features. Secondly, a dual-branch graph neural network (DBGNN) is conducted by aggregating features from the individual- and group-level templates with the cross-template FC, which is beneficial to improve the feature discrimination by considering dependency between templates. Finally, a spatio-temporal aggregated attention (STAA) module is investigated to capture the spatial and dynamic relationships between functional regions, which solves the limitation of insufficient temporal information utilization. We evaluate our proposed method on 442 samples from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database, and achieve the accuracies of 90.1%, 90.3%, 83.3% for normal control (NC) vs. early MCI (EMCI), EMCI vs. late MCI (LMCI), and NC vs. EMCI vs. LMCI classification tasks, respectively, indicating that our method boosts MCI identification performance and outperforms state-of-the-art methods.

Dynamic multi-site graph convolutional network for autism spectrum disorder identification.

Multi-site learning has attracted increasing interests in autism spectrum disorder (ASD) identification tasks by its efficacy on capturing data heterogeneity of neuroimaging taken from different medical sites. However, existing multi-site graph convolutional network (MSGCN) often ignores the correlations between different sites, and may obtain suboptimal identification results. Moreover, current feature extraction methods characterizing temporal variations of functional magnetic resonance imaging (fMRI) signals require the time series to be of the same length and cannot be directly applied to multi-site fMRI datasets. To address these problems, we propose a dual graph based dynamic multi-site graph convolutional network (DG-DMSGCN) for multi-site ASD identification. First, a sliding-window dual-graph convolutional network (SW-DGCN) is introduced for feature extraction, simultaneously capturing temporal and spatial features of fMRI data with different series lengths. Then we aggregate the features extracted from multiple medical sites through a novel dynamic multi-site graph convolutional network (DMSGCN), which effectively considers the correlations between different sites and is beneficial to improve identification performance. We evaluate the proposed DG-DMSGCN on public ABIDE I dataset containing data from 17 medical sites. The promising results obtained by our framework outperforms the state-of-the-art methods with increase in identification accuracy, indicating that it has a potential clinical prospect for practical ASD diagnosis. Our codes are available on https://github.com/Junling-Du/DG-DMSGCN.