Low-order Functional Connectivity Research Articles

EAG-RS: A Novel Explainability-Guided ROI-Selection Framework for ASD Diagnosis via Inter-Regional Relation Learning.

Deep learning models based on resting-state functional magnetic resonance imaging (rs-fMRI) have been widely used to diagnose brain diseases, particularly autism spectrum disorder (ASD). Existing studies have leveraged the functional connectivity (FC) of rs-fMRI, achieving notable classification performance. However, they have significant limitations, including the lack of adequate information while using linear low-order FC as inputs to the model, not considering individual characteristics (i.e., different symptoms or varying stages of severity) among patients with ASD, and the non-explainability of the decision process. To cover these limitations, we propose a novel explainability-guided region of interest (ROI) selection (EAG-RS) framework that identifies non-linear high-order functional associations among brain regions by leveraging an explainable artificial intelligence technique and selects class-discriminative regions for brain disease identification. The proposed framework includes three steps: (i) inter-regional relation learning to estimate non-linear relations through random seed-based network masking, (ii) explainable connection-wise relevance score estimation to explore high-order relations between functional connections, and (iii) non-linear high-order FC-based diagnosis-informative ROI selection and classifier learning to identify ASD. We validated the effectiveness of our proposed method by conducting experiments using the Autism Brain Imaging Database Exchange (ABIDE) dataset, demonstrating that the proposed method outperforms other comparative methods in terms of various evaluation metrics. Furthermore, we qualitatively analyzed the selected ROIs and identified ASD subtypes linked to previous neuroscientific studies.

Topological disruption of low- and high-order functional networks in presbycusis.

Prior efforts have manifested that functional connectivity (FC) network disruptions are concerned with cognitive disorder in presbycusis. The present research was designed to investigate the topological reorganization and classification performance of low-order functional connectivity (LOFC) and high-order functional connectivity (HOFC) networks in patients with presbycusis. Resting-state functional magnetic resonance imaging (Rs-fMRI) data were obtained in 60 patients with presbycusis and 50 matched healthy control subjects (HCs). LOFC and HOFC networks were then constructed, and the topological metrics obtained from the constructed networks were compared to evaluate topological differences in global, nodal network metrics, modularity and rich-club organization between patients with presbycusis and HCs. The use of HOFC profiles boosted presbycusis classification accuracy, sensitivity and specificity compared to that using LOFC profiles. The brain networks in both patients with presbycusis and HCs exhibited small-world properties within the given threshold range, and striking differences between groups in topological metrics were discovered in the constructed networks (LOFC and HOFC). NBS analysis identified a subnetwork involving 26 nodes and 23 signally altered internodal connections in patients with presbycusis in comparison to HCs in HOFC networks. This study highlighted the topological differences between LOFC and HOFC networks in patients with presbycusis, suggesting that HOFC profiles may help to further identify brain network abnormalities in presbycusis.

Fusing multi-scale fMRI features using a brain-inspired multi-channel graph neural network for major depressive disorder diagnosis

Depression stands as one of the most pernicious mental disorders in contemporary society, characterized by a highly intricate pathological mechanism. Specifically, individuals suffering from Major Depressive Disorder (MDD) exhibit heightened vulnerability to suicidal tendencies. Currently, healthcare practitioners often encounter challenges related to the misdiagnosis and underdiagnosis of depression during clinical assessments. Consequently, it is of paramount importance to develop highly accurate auxiliary diagnostic tools for depression. Unfortunately, traditional machine learning and deep learning methodologies frequently neglect the integration of multi-source data and disregard the intricate topological structure and high-order attributes of brain networks. In this study, a multi-scale feature fusion classification framework is proposed to distinguish between MDD patients and healthy controls. Within the proposed model, a novel method, the Cross-Level High-Order Interaction (CLHOI), is introduced and implemented on a low-order functional connectivity (LOFC) matrix to derive two distinct high-order functional connectivity (HOFC) matrices. Subsequently, a Multi-Channel Fusion Graph Convolutional Network (MFGCN) is trained by integrating high-order and low-order brain graph features along with phenotypic information. The results of 10-fold cross-validation experiments conducted on the publicly available REST-meta-MDD dataset indicate that the fusion of multi-scale features improves the average accuracy by approximately 3%, resulting in an accuracy rate of 77.6%. Simultaneously, the hypothesis asserting the existence of intricate information interactions at various levels within brain connectivity networks is validated. Moreover, our model exhibits strong explanatory capabilities, effectively identifying brain regions closely associated with MDD, including the Precentral gyrus, Superior frontal gyrus, Cuneus, Lingual gyrus, and Fusiform gyrus. In comparison to numerous advanced studies within the same domain, our approach has produced competitive results. Furthermore, our proposed method can be readily extended to facilitate the diagnosis of various neurological diseases.

Intelligent diagnosis of major depression disease based on multi-layer brain network.

Resting-state brain network with physiological and pathological basis has always been the ideal data for intelligent diagnosis of major depression disease (MDD). Brain networks are divided into low-order networks and high-order networks. Most of the studies only use a single-level network to classify while ignoring that the brain works cooperatively with different levels of networks. This study hopes to find out whether varying levels of networks will provide complementary information in the process of intelligent diagnosis and what impact will be made on the final classification results by combining the characteristics of different networks. Our data are from the REST-meta-MDD project. After the screening, 1,160 subjects from ten sites were included in this study (597 MDD and 563 normal controls). For each subject, we constructed three different levels of networks according to the brain atlas: the traditional low-order network based on Pearson's correlation (low-order functional connectivity, LOFC), the high-order network based on topographical profile similarity (topographical information-based high-order functional connectivity, tHOFC) and the associated network between them (aHOFC). Two sample t-test is used for feature selection, and then features from different sources are fused. Finally, the classifier is trained by a multi-layer perceptron or support vector machine. The performance of the classifier was evaluated using the leave-one-site cross-validation method. The classification ability of LOFC is the highest among the three networks. The classification accuracy of the three networks combined is similar to the LOFC network. These are seven features chosen in all networks. In the aHOFC classification, six features were selected in each round but not seen in other classifications. In the tHOFC classification, five features were selected in each round but were unique. These new features have crucial pathological significance and are essential supplements to LOFC. A high-order network can provide auxiliary information for low-order networks but cannot improve classification accuracy.

Multi-Level Functional Connectivity Fusion Classification Framework for Brain Disease Diagnosis.

Brain disease diagnosis is a new hotspot in the cross research of artificial intelligence and neuroscience. Quantitative analysis of functional magnetic resonance imaging (fMRI) data can provide valuable biomarkers that contributes to clinical diagnosis, and the analysis of functional connectivity (FC) has become the primary method. However, previous studies mainly focus on brain disease classification based on the low-order FC features, ignoring the potential role of high-order functional relationships among brain regions. To solve this problem, this study proposed a novel multi-level FC fusion classification framework (MFC) for brain disease diagnosis. We firstly designed a deep neural network (DNN) model to extract and learn abstract feature representations for the constructed low-order and high-order FC patterns. Both unsupervised and supervised learning steps were performed during the DNN model training, and the prototype learning was introduced in the supervised fine-tuning to improve the intra-class compactness and inter-class separability of the feature representation. Then, we combined the learned multi-level abstract FC features and trained an ensemble classifier with a hierarchical stacking learning strategy for the brain disease classification. Systematic experiments were conducted on two real large-scale fMRI datasets. Results showed that the proposed MFC model obtained robust classification performance for different preprocessing pipelines, different brain parcellations, and different cross-validation schemes, suggesting the effectiveness and generality of the proposed MFC model. Overall, this study provides a promising solution to combine the informative low-order and high-order FC patterns to further promote the classification of brain diseases.

Predicting Post-Stroke Somatosensory Function from Resting-State Functional Connectivity: A Feasibility Study.

Accumulating evidence shows that brain functional deficits may be impacted by damage to remote brain regions. Recent advances in neuroimaging suggest that stroke impairment can be better predicted based on disruption to brain networks rather than from lesion locations or volumes only. Our aim was to explore the feasibility of predicting post-stroke somatosensory function from brain functional connectivity through the application of machine learning techniques. Somatosensory impairment was measured using the Tactile Discrimination Test. Functional connectivity was employed to model the global brain function. Behavioral measures and MRI were collected at the same timepoint. Two machine learning models (linear regression and support vector regression) were chosen to predict somatosensory impairment from disrupted networks. Along with two feature pools (i.e., low-order and high-order functional connectivity, or low-order functional connectivity only) engineered, four predictive models were built and evaluated in the present study. Forty-three chronic stroke survivors participated this study. Results showed that the regression model employing both low-order and high-order functional connectivity can predict outcomes based on correlation coefficient of r = 0.54 (p = 0.0002). A machine learning predictive approach, involving high- and low-order modelling, is feasible for the prediction of residual somatosensory function in stroke patients using functional brain networks.

Multi-Order Brain Functional Connectivity Network-Based Machine Learning Method for Recognition of Delayed Neurocognitive Recovery in Older Adults Undergoing Non-cardiac Surgery.

Objectives: Delayed neurocognitive recovery (DNR) seriously affects the post-operative recovery of elderly surgical patients, but there is still a lack of effective methods to recognize high-risk patients with DNR. This study proposed a machine learning method based on a multi-order brain functional connectivity (FC) network to recognize DNR.Method: Seventy-four patients who completed assessments were included in this study, in which 16/74 (21.6%) had DNR following surgery. Based on resting-state functional magnetic resonance imaging (rs-fMRI), we first constructed low-order FC networks of 90 brain regions by calculating the correlation of brain region signal changing in the time dimension. Then, we established high-order FC networks by calculating correlations among each pair of brain regions. Afterward, we built sparse representation-based machine learning model to recognize DNR on the extracted multi-order FC network features. Finally, an independent testing was conducted to validate the established recognition model.Results: Three hundred ninety features of FC networks were finally extracted to identify DNR. After performing the independent-sample T test between these features and the categories, 15 features showed statistical differences (P < 0.05) and 3 features had significant statistical differences (P < 0.01). By comparing DNR and non-DNR patients’ brain region connection matrices, it is found that there are more connections among brain regions in DNR patients than in non-DNR patients. For the machine learning recognition model based on multi-feature combination, the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity of the classifier reached 95.61, 92.00, 66.67, and 100.00%, respectively.Conclusion: This study not only reveals the significance of preoperative rs-fMRI in recognizing post-operative DNR in elderly patients but also establishes a promising machine learning method to recognize DNR.

Recognition of Cognitive Impairment in Adult Moyamoya Disease: A Classifier Based on High-Order Resting-State Functional Connectivity Network.

Objective: Vascular cognitive impairment (VCI) is a common complication in adult patients with moyamoya disease (MMD), and is reversible by surgical revascularization in its early stage of mild VCI. However, accurate diagnosis of mild VCI is difficult based on neuropsychological examination alone. This study proposed a method of dynamic resting-state functional connectivity (FC) network to recognize global cognitive impairment in MMD.Methods: For MMD, 36 patients with VCI and 43 patients with intact cognition (Non-VCI) were included, as well as 26 normal controls (NCs). Using resting-state fMRI, dynamic low-order FC networks were first constructed with multiple brain regions which were generated through a sliding window approach and correlated in temporal dimension. In order to obtain more information of network interactions along the time, high-order FC networks were established by calculating correlations among each pair of brain regions. Afterwards, a sparse representation-based classifier was constructed to recognize MMD (experiment 1) and its cognitive impairment (experiment 2) with features extracted from both low- and high-order FC networks. Finally, the ten-fold cross-validation strategy was proposed to train and validate the performance of the classifier.Results: The three groups did not differ significantly in demographic features (p > 0.05), while the VCI group exhibited the lowest MMSE scores (p = 0.001). The Non-VCI and NCs groups did not differ significantly in MMSE scores (p = 0.054). As for the classification between MMD and NCs, the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity of the classifier reached 90.70, 88.57, 93.67, and 73.08%, respectively. While for the classification between VCI and Non-VCI, the AUC, accuracy, sensitivity, and specificity of the classifier reached 91.02, 84.81, 80.56, and 88.37%, respectively.Conclusion: This study not only develops a promising classifier to recognize VCI in adult MMD in its early stage, but also implies the significance of time-varying properties in dynamic FC networks.

Inter-Network High-Order Functional Connectivity (IN-HOFC) and its Alteration in Patients with Mild Cognitive Impairment.

Little is known about the high-order interactions among brain regions measured by the similarity of higher-order features (other than the raw blood-oxygen-level-dependent signals) which can characterize higher-level brain functional connectivity (FC). Previously, we proposed FC topographical profile-based high-order FC (HOFC) and found that this metric could provide supplementary information to traditional FC for early Alzheimer's disease (AD) detection. However, whether such findings apply to network-level brain functional integration is unknown. In this paper, we propose an extended HOFC method, termed inter-network high-order FC (IN-HOFC), as a useful complement to the traditional inter-network FC methods, for characterizing more complex organizations among the large-scale brain networks. In the IN-HOFC, both network definition and inter-network FC are defined in a high-order manner. To test whether IN-HOFC is more sensitive to cognition decline due to brain diseases than traditional inter-network FC, 77 mild cognitive impairments (MCIs) and 89 controls are compared among the conventional methods and our IN-HOFC. The result shows that IN-HOFCs among three temporal lobe-related high-order networks are dampened in MCIs. The impairment of IN-HOFC is especially found between the anterior and posterior medial temporal lobe and could be a potential MCI biomarker at the network level. The competing network-level low-order FC methods, however, either revealing less or failing to detect any group difference. This work demonstrates the biological meaning and potential diagnostic value of the IN-HOFC in clinical neuroscience studies.

Strength and similarity guided group-level brain functional network construction for MCI diagnosis

Sparse representation-based brain functional network modeling often results in large inter-subject variability in the network structure. This could reduce the statistical power in group comparison, or even deteriorate the generalization capability of the individualized diagnosis of brain diseases. Although group sparse representation (GSR) can alleviate such a limitation by increasing network similarity across subjects, it could, in turn, fail in providing satisfactory separability between the subjects from different groups (e.g., patients vs. controls). In this study, we propose to integrate individual functional connectivity (FC) information into the GSR-based network construction framework to achieve higher between-group separability while maintaining the merit of within-group consistency. Our method was based on an observation that the subjects from the same group have generally more similar FC patterns than those from different groups. To this end, we propose our new method, namely “strength and similarity guided GSR (SSGSR)”, which exploits both BOLD signal temporal correlation-based “low-order” FC (LOFC) and inter-subject LOFC-profile similarity-based “high-order” FC (HOFC) as two priors to jointly guide the GSR-based network modeling. Extensive experimental comparisons are carried out, with the rs-fMRI data from mild cognitive impairment (MCI) subjects and healthy controls, between the proposed algorithm and other state-of-the-art brain network modeling approaches. Individualized MCI identification results show that our method could achieve a balance between the individually consistent brain functional network construction and the adequately maintained inter-group brain functional network distinctions, thus leading to a more accurate classification result. Our method also provides a promising and generalized solution for the future connectome-based individualized diagnosis of brain disease.

Diagnosis of Autism Spectrum Disorders Using Multi-Level High-Order Functional Networks Derived From Resting-State Functional MRI.

Functional brain networks derived from resting-state functional magnetic resonance imaging (rs-fMRI) have been widely used for Autism Spectrum Disorder (ASD) diagnosis. Typically, these networks are constructed by calculating functional connectivity (FC) between any pair of brain regions of interest (ROIs), i.e., using Pearson's correlation between rs-fMRI time series. However, this can only be called as a low-order representation of the functional interaction, because the relationship is investigated just between two ROIs. Brain disorders might not only affect low-order FC, but also high-order FC, i.e., the higher-level relationship among multiple brain regions, which might be more crucial for diagnosis. To comprehensively characterize such relationship for better diagnosis of ASD, we propose a multi-level, high-order FC network representation that can nicely capture complex interactions among brain regions. Then, we design a feature selection method to identify those discriminative multi-level, high-order FC features for ASD diagnosis. Finally, we design an ensemble classifier with multiple linear SVMs, each trained on a specific level of FC networks, for boosting the final classification accuracy. Experimental results show that the integration of both low-order and first-level high-order FC networks achieves the best ASD diagnostic accuracy (81%). We further investigated those selected discriminative low-order and high-order FC features and found that the high-order FC features can provide complementary information to the low-order FC features in the ASD diagnosis.

Test-Retest Reliability of "High-Order" Functional Connectivity in Young Healthy Adults.

Functional connectivity (FC) has become a leading method for resting-state functional magnetic resonance imaging (rs-fMRI) analysis. However, the majority of the previous studies utilized pairwise, temporal synchronization-based FC. Recently, high-order FC (HOFC) methods were proposed with the idea of computing “correlation of correlations” to capture high-level, more complex associations among the brain regions. There are two types of HOFC. The first type is topographical profile similarity-based HOFC (tHOFC) and its variant, associated HOFC (aHOFC), for capturing different levels of HOFC. Instead of measuring the similarity of the original rs-fMRI signals with the traditional FC (low-order FC, or LOFC), tHOFC measures the similarity of LOFC profiles (i.e., a set of LOFC values between a region and all other regions) between each pair of brain regions. The second type is dynamics-based HOFC (dHOFC) which defines the quadruple relationship among every four brain regions by first calculating two pairwise dynamic LOFC “time series” and then measuring their temporal synchronization (i.e., temporal correlation of the LOFC fluctuations, not the BOLD fluctuations). Applications have shown the superiority of HOFC in both disease biomarker detection and individualized diagnosis than LOFC. However, no study has been carried out for the assessment of test-retest reliability of different HOFC metrics. In this paper, we systematically evaluate the reliability of the two types of HOFC methods using test-retest rs-fMRI data from 25 (12 females, age 24.48 ± 2.55 years) young healthy adults with seven repeated scans (with interval = 3–8 days). We found that all HOFC metrics have satisfactory reliability, specifically (1) fair-to-good for tHOFC and aHOFC, and (2) fair-to-moderate for dHOFC with relatively strong connectivity strength. We further give an in-depth analysis of the biological meanings of each HOFC metric and highlight their differences compared to the LOFC from the aspects of cross-level information exchanges, within-/between-network connectivity, and modulatory connectivity. In addition, how the dynamic analysis parameter (i.e., sliding window length) affects dHOFC reliability is also investigated. Our study reveals unique functional associations characterized by the HOFC metrics. Guidance and recommendations for future applications and clinical research using HOFC are provided. This study has made a further step toward unveiling more complex human brain connectome.

Hybrid High-order Functional Connectivity Networks Using Resting-state Functional MRI for Mild Cognitive Impairment Diagnosis

Conventional functional connectivity (FC), referred to as low-order FC, estimates temporal correlation of the resting-state functional magnetic resonance imaging (rs-fMRI) time series between any pair of brain regions, simply ignoring the potentially high-level relationship among these brain regions. A high-order FC based on “correlation’s correlation” has emerged as a new approach for abnormality detection of brain disease. However, separate construction of the low- and high-order FC networks overlooks information exchange between the two FC levels. Such a higher-level relationship could be more important for brain diseases study. In this paper, we propose a novel framework, namely “hybrid high-order FC networks” by exploiting the higher-level dynamic interaction among brain regions for early mild cognitive impairment (eMCI) diagnosis. For each sliding window-based rs-fMRI sub-series, we construct a whole-brain associated high-order network, by estimating the correlations between the topographical information of the high-order FC sub-network from one brain region and that of the low-order FC sub-network from another brain region. With multi-kernel learning, complementary features from multiple time-varying FC networks constructed at different levels are fused for eMCI classification. Compared with other state-of-the-art methods, the proposed framework achieves superior diagnosis accuracy, and hence could be promising for understanding pathological changes of brain connectome.

Constructing Multi-frequency High-Order Functional Connectivity Network for Diagnosis of Mild Cognitive Impairment.

Human brain functional connectivity (FC) networks, estimated based on resting-state functional magnetic resonance imaging (rs-fMRI), has become a promising tool for imaging-based brain disease diagnosis. Conventional low-order FC network (LON) usually characterizes pairwise temporal correlation of rs-fMRI signals between any pair of brain regions. Meanwhile, high-order FC network (HON) has provided an alternative brain network modeling strategy, characterizing more complex interactions among low-order FC sub-networks that involve multiple brain regions. However, both LON and HON are usually constructed within a fixed and relatively wide frequency band, which may fail in capturing (sensitive) frequency-specific FC changes caused by pathological attacks. To address this issue, we propose a novel "multi-frequency HON construction" method. Specifically, we construct not only multiple frequency-specific HONs (intra-spectrum HONs), but also a series of cross-frequency interaction-based HONs (inter-spectrum HONs) based on the low-order FC sub-networks constructed at different frequency bands. Both types of these HONs, together with the frequency-specific LONs, are used for the complex network analysis-based feature extraction, followed by sparse regression-based feature selection and the classification between mild cognitive impairment (MCI) patients and normal aging subjects using a support vector machine. Compared with the previous methods, our proposed method achieves the best diagnosis accuracy in early diagnosis of Alzheimer's disease.

High-order resting-state functional connectivity network for MCI classification.

Brain functional connectivity (FC) network, estimated with resting-state functional magnetic resonance imaging (RS-fMRI) technique, has emerged as a promising approach for accurate diagnosis of neurodegenerative diseases. However, the conventional FC network is essentially low-order in the sense that only the correlations among brain regions (in terms of RS-fMRI time series) are taken into account. The features derived from this type of brain network may fail to serve as an effective disease biomarker. To overcome this drawback, we propose extraction of novel high-order FC correlations that characterize how the low-order correlations between different pairs of brain regions interact with each other. Specifically, for each brain region, a sliding window approach is first performed over the entire RS-fMRI time series to generate multiple short overlapping segments. For each segment, a low-order FC network is constructed, measuring the short-term correlation between brain regions. These low-order networks (obtained from all segments) describe the dynamics of short-term FC along the time, thus also forming the correlation time series for every pair of brain regions. To overcome the curse of dimensionality, we further group the correlation time series into a small number of different clusters according to their intrinsic common patterns. Then, the correlation between the respective mean correlation time series of different clusters is calculated to represent the high-order correlation among different pairs of brain regions. Finally, we design a pattern classifier, by combining features of both low-order and high-order FC networks. Experimental results verify the effectiveness of the high-order FC network on disease diagnosis. Hum Brain Mapp 37:3282-3296, 2016. © 2016 Wiley Periodicals, Inc.