The heterogeneity has been recognized as a large obstacle to the treatment of autism spectrum disorder (ASD). Recent studies have identified several subgroups of ASD that exhibited heterogeneous alterations in brain. However, most of them primarily depicted the pairwise similarity between individuals, relying solely on a single imaging modality. This leads to an underestimation of the complexity in inter-individual relationships and the rich information provided by multimodal images. To capture the high-order relationships among individuals, we utilized multi-task method to construct multilayer hypergraph based on brain structure and function. We then developed a novel co-optimized community detection algorithm, which jointly optimizes the modular structure across hypergraph layer, with the aim of categorizing subtypes of ASD by fusing multimodal information. By applying the proposed method on the Autism Brain Imaging Data Exchange repository data (n = 287/303, ASD/typical development [TD]), we identified two ASD subtypes with distinct alteration patterns in both brain structure and function. Distinct clinical manifestations in social and communication were observed between the two subtypes. Furthermore, subtyping significantly enhanced the diagnostic accuracy of ASD by over 10%. In addition, our method exhibited superior clustering performance that outperformed traditional community detection algorithms on graphs. Taken together, our study demonstrated the effectiveness of subtyping ASD through a multimodal multilayer hypergraph, highlighting its potential in elucidating the heterogeneity of autism and improving clinical diagnosis.

Autism spectrum disorder (ASD) is accompanied by impaired social cognition and behavior. The expense of supporting patients with ASD turns into a significant problem for society. Parsing neurobiological subtypes is a crucial way for delineating the heterogeneity in autistic brains, with significant implications for improving ASD diagnosis and promoting the development of personalized intervention models. Nevertheless, a comprehensive understanding of the heterogeneity in cortical morphology of ASD is still lacking, and the question of whether neuroanatomical subtypes remain stable during cortical development remains unclear. Here, we used T1-weighted images of 515 male patients with ASD, including 216 autistic children (6–11 years), 187 adolescents (12–17 years), and 112 young adults (18–29 years), along with 595 age and gender-matched typically developing (TD) individuals. Cortical thickness (CT), surface area (SA), and volumes of cortical (CV) and subcortical (SV) regions were extracted. A single network layer was established by calculating the covariance of each feature across brain regions between participants, thereby constructing a multilayer intersubject covariance network. Applying a community detection algorithm to multilayer networks derived from different feature combinations, we observed that the network comprising CT and CV layers exhibited the most prominent modular organization, resulting in three subtypes of ASD for each of the three age groups. Subtypes within the corresponding age group significantly differed in terms of brain morphology and clinical scales. Furthermore, the subtypes of children with ASD underwent reorganization with development, transitioning from childhood to adolescence and adulthood, rather than consistently persist. Additionally, subtype categorization largely improved the diagnostic accuracy of ASD compared to diagnosing the entire ASD cohort. These findings demonstrated distinct neuroanatomical manifestations of ASD subtypes across various developmental periods, highlighting the significance of age-related subtyping in facilitating the etiology and diagnosis of ASD.

Growing evidence has unveiled the dynamic nature of human brain networks. However, the dynamic and hierarchical organization that supports information transmission in brain networks remains unexplored across the adult lifespan. In this study, we developed an analytical framework to investigate the spatiotemporal reorganization of dynamic brain networks during adult development and aging. Specifically, using resting-state fMRI data from the Cam-CAN lifespan dataset, we examined the age effects on the topological stability of egocentric structures and the average length of temporal paths. As the egocentric structure reflects the relationships between a node’s neighbors, we further explored whether the stability of egocentric structures mediates age effects on information diffusion. The results showed that the topological stability of egocentric structures has an impact on information processing in the spatiotemporal domain. In particular, the age-related changes observed in some functional systems followed different progressive patterns from that of other systems, which might be explained by some compensation mechanisms. Taken together, the present work may provide an additional perspective for understanding the underlying neuro-mechanisms of healthy aging.

Brain anatomical age is an effective feature to assess the status of the brain, such as atypical development and aging. Although some deep learning models have been developed for estimating infant brain age, the performance of these models was unsatisfactory because few of them considered the developmental characteristics of brain anatomy during the perinatal period-the most rapid and complex developmental stage across the lifespan. The present study proposed an attention-based hemispheric relation inference network (HRINet) that takes advantage of the nature of brain structural lateralization during early development. This model captures the inter-hemispheric relationship using a graph attention mechanism and transmits lateralization information as features to describe the interactive development between bilateral hemispheres. The HRINet was used to estimate the brain age of 531 preterm and full-term neonates from the Developing Human Connectome Project (dHCP) database based on two metrics (mean curvature and sulcal depth) characterizing the folding morphology of the cortex. Our results showed that the HRINet outperformed other benchmark models in fitting the perinatal brain age, with mean absolute error of 0.53 and determination coefficient of 0.89. We also verified the generalizability of the HRINet on an extra independent dataset collected from the Gansu Provincial Maternity and Child-care Hospital. Furthermore, by applying the best-performing model to an independent dataset consisting of 47 scans of preterm infants at term-equivalent age, we showed that the predicted age was significantly lower than the chronological age, suggesting a delayed development of premature brains. Our results demonstrate the effectiveness and generalizability of the HRINet in estimating infant brain age, providing promising clinical applications for assessing neonatal brain maturity.

Functional magnetic resonance imaging (fMRI) plays a vital role in brain science as it measures and maps brain activity through the analysis of blood flow changes, offering valuable insights into cognitive functions and neural processes. However, due to the intricacy and dynamism of brain activity, conventional approaches failed to accurately predict the blood oxygen level-dependent (BOLD) time series in fMRI data. To tackle this issue, we proposed an end-to-end adversarial U-network (AUN) to verify the predictability of existing BOLD signals in primary cortex (i.e., primary visual, primary motor and primary sensory) and higher cortex(i.e., dorsolateral prefrontal and posterior cingulate). The model combined the U-network architecture and adversarial strategy to ensure that the predicted results capture both the intricate nonlinear details and the overall distribution characteristics. We performed the experiment using the Human Connectome Project (HCP) database. The results demonstrated the predictability of both primary and higher cortex, with primary cortex showing higher predictability. Additionally, the AUN performed better than other popular methods. We also found the improvement in dynamic functional connectivity (dFC) metrics through accurate prediction. The above results confirm the feasibility of predicting BOLD signals and their potential application in clinical settings.1

Effective estimation of brain network connectivity enables better unraveling of the extraordinary complexity interactions of brain regions and helps in auxiliary diagnosis of psychiatric disorders. Considering different modalities can provide comprehensive characterizations of brain connectivity, we propose the message-passing-based nonlinear network fusion (MP-NNF) algorithm to estimate multimodal brain network connectivity. In the proposed method, the initial functional and structural networks were computed from fMRI and DTI separately. Then, we update every unimodal network iteratively, making it more similar to the others in every iteration, and finally converge to one unified network. The estimated brain connectivities integrate complementary information from multiple modalities while preserving their original structure, by adding the strong connectivities present in unimodal brain networks and eliminating the weak connectivities. The effectiveness of the method was evaluated by applying the learned brain connectivity for the classification of major depressive disorder (MDD). Specifically, 82.18% classification accuracy was achieved even with the simple feature selection and classification pipeline, which significantly outperforms the competing methods. Exploration of brain connectivity contributed to MDD identification suggests that the proposed method not only improves the classification performance but also was sensitive to critical disease-related neuroimaging biomarkers.

Major depressive disorder (MDD) is accompanied by abnormal changes in functional connectivities (FC) among brain regions. However, most studies estimated the pairwise connectivity without the thorough consideration of the influence of other regions and assumed that the brain functional connectivity was static, which may be insufficient for the accurate identification of the pathological mechanisms underlying MDD. The purpose of this study was to explore the pathological mechanisms of MDD based on dynamic FC taking the influence of other regions into account. We performed time-varying connectivity analysis on resting-state functional magnetic resonance imaging (rs-fMRI) of 58 MDD patients and 63 matched healthy controls. The dynamic functional connectivity matrices were constructed using a novel Multivariate Vector Regression-based Connectivity (MVRC) method, which could regress time series of all regions while estimating the pairwise association between two regions. Then we analyzed two commonly used dynamic characteristics in brain network analysis, including dynamic FC (dFC) variability and node flexibility. Both dFC variability and node flexibility of MDD patients showed significant decreases in frontal, temporal, occipital, and parietal gyrus. Notably, we found the reduced dynamics of regions in frontal, temporal, and parietal gyrus was strongly negatively associated with depression severity. Our results demonstrated that the decreased dFC variability and node flexibility based on dynamic MVRC (dMVRC) were pathological manifestations of MDD.

Conventional hyper-network coefficients ignore the weighted hyper-edge information which could be vital in researching the specificity of brain disease. Functional hyper-networks for 64 healthy controls (HC) and 56 patients with major depressive disorder (MDD) were constructed using the least absolute shrinkage and selection operator (Lasso). Not only the classical topology metrics but also a novel hyper-edge weight (HEW) attribute were extracted as features to promote the functional-based auto-diagnosis accuracy of MDD. We compared the categorization performance of each hyper-network coefficient. A multi-feature ensemble model was applied to fuse different kinds of features. We obtained 82.15 % accuracy with the classical hyper-network clustering coefficient (HCC) and 84.08 % accuracy with the HEW attribute on the MDD dataset. The performance was further improved to 89.24% by combining all the properties of the hyper-networks. The multi-feature ensemble model combining different hyper-network coefficients provides new insights into the automatic diagnosis with diverse information of MDD.

As the ϵ4 allele of apolipoprotein E (APOE4) is proved a high risk factor of Alzheimer's disease (AD), numerous studies have used modalities of neuroimaging data to investigate the alterations of brain caused by APOE4. A recent study has shown that APOE4-related pathological changes of cortical networks during rest exist in APOE4 carriers. However, the interrelationship among the resting intrinsic networks (ICNs) has not been adequately explored. In the present study, seven ICNs were detected in both of APOE4 carriers and APOE4 non-carriers with spatial independent component analysis (ICA). Then the effective connectivity patterns of the two groups were acquired using multivariate Granger causality analysis (mGCA) the results of which reflect the information flow among ICNs. Compared with APOE4 non-carriers, significant difference in activity was found within default mode network in APOE4 carriers. Moreover, the significantly reduced intensity of causal interaction from auditory network to cerebellum network was found in APOE4 carriers. And we found the significantly causal relationship from cerebellum network to default mode network only existed in the connectivity pattern of APOE4 carriers but not in that of APOE4 non-carriers. In addition, the subcortical network and cognitive control network in APOE4 carriers were found less influenced by other ICNs. The findings in our study suggest that APOE4 indeed modulates the activities and interactions of ICNs, which might be the genetic cause of the dysfunctional process in APOE4 carriers. Our findings may shed new light on the ways how APOE4 affects the functions of nervous systems and provide a perspective to understand genetic basis of pathogenesis of AD.

Mild cognitive impairment (MCI) is a transition stage between normal aging and dementia. Brain network has been proven to occupy an important role in the study of differences in Alzheimer's disease (AD) and MCI. However, there is little knowledge about individual metabolic network abnormities which might be sensitive features in the prediction of MCI progression. In this paper, we constructed the individual metabolic network based on longitudinal Fluorodeoxyglucose positron emission tomography (FDG-PET) of 33 progress MCI (pMCI) patients and 46 stable MCI (sMCI) patients from the Alzheimer's disease Neuroimaging Initiative (ADNI). Firstly, PET images of each time point are normalized with the Yakushev normalization procedure and registered to the Brainnetome Atlas (BNA) template. Then the combination of rough distance and precision distance is utilized for accurate evaluation of between-region dissimilarity and calculated the correlation coefficient of the network. Finally, correlative feature selected by Lasso shows a significant promotion in classification performance compare with the metabolic intensity, achieving an accuracy of 89.9% and area under the receiver operating characteristic curve of 0.892. What's more, the combination of multi time points also suggests a better classification result than single time point. This finding may predict disease course in individuals with mild cognitive impairment.

Recent studies suggested that cognitive impairments and memory difficulties in cancer survivors were associated with topology changes of brain network, particularly in terms of the functional and structural abnormalities. However, little is known about the modular reconfiguration of metabolic brain network among this population. In this study, we recruited 78 patients with pre-treatment cancer and 80 age- and gender-matched normal controls (NCs), and constructed the metabolic brain networks derived from resting-state 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) to assess the alters of modularity pattern in cancer. The measurements of the participation index (PI) and mutual information (MI) were calculated for the cancer and NC groups. Compared with NC group, one module composed by the hippocampus, the amygdala and frontal and temporal regions was absented in cancer group. Moreover, cancer patients showed abnormal topology pattern in their metabolic networks (i.e., increased local efficiency and reduced global efficiency). Although node-wise PI shared positive correlated with normalized metabolism uptake in both groups, the more energy consumption were observed in metabolism network of cancer group that might be indicative of reduced capability of information processing. In addition, the between-group MIs were gradually increased over a range of thresholds. Our results suggested that modular pattern of the metabolic brain network seemed to re-shape its organization in cancer, which might uncover the neurobiological mechanisms underlying cancer-related cognitive dysfunction.

The rapid development of neuroimaging technology and brain network analysis methodologies have promoted the research of Alzheimer's disease (AD). Recently, studies on brain networks reported that AD patients showed abnormal connectivity alterations and disrupted coordinated organizations compared with normal controls (NC). However, much less knowledge is about the abnormalities of metabolic network at individual level, which might be the potential marker in promoting current AD diagnosis. In the present study, we constructed the individual metabolic network based on 18F-Fluro-Deoxyglucose Positron Emission Tomography (18F-FDG-PET) data by using cubes consisted with certain numbers of voxels. Network properties, connectivity strength and metabolic cost of cubes of 111 NCs and 111 AD patients were calculated to evaluate the performance and feasibility of the proposed network via machine learning approaches. Results showed that the features we extracted were well-performed in classification, with accuracy of 95.64% and area of 0.9915 under receiver operating characteristic curve, indicating the individual metabolic network and local metabolic information are potential powerful in AD diagnosis.

This paper presents a new image enhancement method using histogram equalization called Bi-Histogram Equalization with Non-parametric Modified Technology (BHENMT). Our proposed method consists of three steps: (i) The input original histogram is divided into two parts using the Otsu method. (ii) Then the histogram modification technique is used to control over enhancement and maximize entropy. (iii) Two sub images are enhanced by the traditional histogram equalization method using the corresponding modified histogram respectively and finally are merged into one output enhanced image. The experimental results show that BHENMT is better than other contrast enhancement methods according to subjective evaluation and various image objective evaluation measures, i.e. Entropy, AMBE and PSNR.

In this paper, we propose a novel method that can detect fingertips as well as recognize hand gestures. Firstly, we collect the hand curves with a Kinect sensor. Secondly, we detect fingertips based on the discrete curve evolution. Thirdly, we recognize hand gestures using evolved curves partitioned at the detected fingertips. Experimental results show that our method performs well in both fingertips detection and hand gesture recognition.

Recently, many researchers have used graph theory to study the aberrant brain functions in mental disorders. However, the characteristics of the brain functional network in attention deficit hyperactivity disorder (ADHD) are still largely unexplored. In this study, blood oxygen level-dependence (BOLD) functional magnetic resonance images (fMRI) were employed to construct brain functional networks in 57 children with ADHD and 59 healthy controls (HC). The results showed that both groups had similar global efficiency and the ADHD group had significantly decreased local efficiency compared with the HC. The between-group differences of degree values were found in the frontal cortex, temporal cortex and the default mode network (DMN). The ADHD group retained most of the hub regions found in the HC, but showed altered hub regions in the left posterior cingulate gyrus and the right lingual gyrus. This altered topological organization of the functional network might be associated with the underlying pathophysiology in ADHD.

The problem of learning conditional preference networks (CP-nets) from a set of examples has received great attention recently. However, because of the randomicity of the users' behaviors and the observation errors, there is always some noise making the examples inconsistent, namely, there exists at least one outcome preferred over itself (by transferring) in examples. Existing CP-nets learning methods cannot handle inconsistent examples. In this work, we introduce the model of learning consistent CP-nets from inconsistent examples and present a method to solve this model. We do not learn the CP-nets directly. Instead, we first learn a preference graph from the inconsistent examples, because dominance testing and consistency testing in preference graphs are easier than those in CP-nets. The problem of learning preference graphs is translated into a 0-1 programming and is solved by the branch-and-bound search. Then, the obtained preference graph is transformed into a CP-net equivalently, which can entail a subset of examples with maximal sum of weight. Examples are given to show that our method can obtain consistent CP-nets over both binary and multivalued variables from inconsistent examples. The proposed method is verified on both simulated data and real data, and it is also compared with existing methods.

In this study, the amplitude of low frequency (0.01-0.08Hz) fluctuation (ALFF) and functional connections were used to analyze blood oxygen level-dependent (BOLD) functional magnetic resonance imaging (fMRI) data for 57 children with attention deficit hyperactivity disorder (ADHD) and 59 healthy controls (HC) in a resting state. Compared with HC, the ADHD showed significantly altered neural activity in the prefrontal and temporal cortex. In ADHD, altered functional connections were observed mainly in prefrontal cortex, temporal cortex, default mode network (DMN) and parahippocampal gyrus (PHG). The abnormalities in prefrontal and temporal cortex might provide evidence to support that functional abnormalities are extended from the prefrontal cortex to the temporal cortex in ADHD. Additionally, dysfunction of the DMN and PHG could be related to the pathophysiological mechanisms of ADHD.

In this paper, we present a new contour point distortion measurement, called accurate distortion measurement for B-spline-based shape coding (ADMBSC). Different from existing distortion measurements containing approximation, quantization or parameterization, our distortion is defined as the shortest distance from the original B-spline to the associated contour point. This is in line with the subjective-based objective quality metric. Geometric relationships are introduced to simplify computation, followed by a hybrid admissible distortion checking algorithm to reduce execution time. Theoretical analysis and experimental results demonstrate that when the operational rate-distortion optimal shape coding framework under the minimum-maximum criterion is applied, the ADMBSC can lead to the smallest bit-rate among all the distortion measurements that can guarantee the admissible distortion. Moreover, if the original contour has NCpoints, it takes only O(NC) time for segment distortion measuring paradigms, whose computational complexity is the same as the lowest one among the existing distortion measurements.

Spatiogram is a generalization of histogram to capture higher-order spatial moments information. To apply spatiogram to object tracking, suitable similarity measure is critical. Although there is a series of work on introducing improved distance measure over the original method, their performance in object tracking is very limited due to insufficient discriminative power. In this paper, we present a symmetric KL divergence based spatiogram similarity measure and show both theoretically and experimentally that, the proposed measure gives superior discriminative power than existing methods, and achieved promising performance in tracking object from single or sequence of images.

Recently, many researchers have used graph theory to study the aberrant brain structures in Alzheimer's disease (AD) and have made great progress. However, the characteristics of the cortical network in Mild Cognitive Impairment (MCI) are still largely unexplored. In this study, the gray matter volumes obtained from magnetic resonance imaging (MRI) for all brain regions except the cerebellum were parcellated into 90 areas using the automated anatomical labeling (AAL) template to construct cortical networks for 98 normal controls (NCs), 113 MCIs and 91 ADs. The measurements of the network properties were calculated for each of the three groups respectively. We found that all three cortical networks exhibited small-world properties and those strong interhemispheric correlations existed between bilaterally homologous regions. Among the three cortical networks, we found the greatest clustering coefficient and the longest absolute path length in AD, which might indicate that the organization of the cortical network was the least optimal in AD. The small-world measures of the MCI network exhibited intermediate values. This finding is logical given that MCI is considered to be the transitional stage between normal aging and AD. Out of all the between-group differences in the clustering coefficient and absolute path length, only the differences between the AD and normal control groups were statistically significant. Compared with the normal controls, the MCI and AD groups retained their hub regions in the frontal lobe but showed a loss of hub regions in the temporal lobe. In addition, altered interregional correlations were detected in the parahippocampus gyrus, medial temporal lobe, cingulum, fusiform, medial frontal lobe, and orbital frontal gyrus in groups with MCI and AD. Similar to previous studies of functional connectivity, we also revealed increased interregional correlations within the local brain lobes and disrupted long distance interregional correlations in groups with MCI and AD.