Leveraging Brain Modularity Prior for Interpretable Representation Learning of fMRI.

Resting-state functional magnetic resonance imaging (rs-fMRI) can reflect spontaneous neural activities in the brain and is widely used for brain disorder analysis. Previous studies focus on extracting fMRI representations using machine/deep learning methods, but these features typically lack biological interpretability. The human brain exhibits a remarkable modular structure in spontaneous brain functional networks, with each module comprised of functionally interconnected brain regions-of-interest (ROIs). However, existing learning-based methods cannot adequately utilize such brain modularity prior. In this paper, we propose a brain modularity-constrained dynamic representation learning framework for interpretable fMRI analysis, consisting of dynamic graph construction, dynamic graph learning via a novel modularity-constrained graph neural network (MGNN), and prediction and biomarker detection. The designed MGNN is constrained by three core neurocognitive modules (i.e., salience network, central executive network, and default mode network), encouraging ROIs within the same module to share similar representations. To further enhance discriminative ability of learned features, we encourage the MGNN to preserve network topology of input graphs via a graph topology reconstruction constraint. Experimental results on 534 subjects with rs-fMRI scans from two datasets validate the effectiveness of the proposed method. The identified discriminative brain ROIs and functional connectivities can be regarded as potential fMRI biomarkers to aid in clinical diagnosis.

Development and validation of a clinical prediction model for glioma grade using machine learning.

BACKGROUND: Histopathological evaluation is currently the gold standard for grading gliomas; however, this technique is invasive. OBJECTIVE: This study aimed to develop and validate a diagnostic prediction model for glioma by employing multiple machine learning algorithms to identify risk factors associated with high-grade glioma, facilitating the prediction of glioma grading. METHODS: Data from 1114 eligible glioma patients were obtained from The Cancer Genome Atlas (TCGA) database, which was divided into a training set (n= 781) and a test set (n= 333). Fifty machine learning algorithms were employed, and the optimal algorithm was selected to construct a prediction model. The performance of the machine learning prediction model was compared to the clinical prediction model in terms of discrimination, calibration, and clinical validity to assess the performance of the prediction model. RESULTS: The area under the curve (AUC) values of the machine learning prediction models (training set: 0.870 vs. 0.740, test set: 0.863 vs. 0.718) were significantly improved from the clinical prediction models. Furthermore, significant improvement in discrimination was observed for the Integrated Discrimination Improvement (IDI) (training set: 0.230, test set: 0.270) and Net Reclassification Index (NRI) (training set: 0.170, test set: 0.170) from the clinical prognostic model. Both models showed a high goodness of fit and an increased net benefit. CONCLUSION: A strong prediction accuracy model can be developed using machine learning algorithms to screen for high-grade glioma risk predictors, which can serve as a non-invasive prediction tool for preoperative diagnostic grading of glioma.

Preserving specificity in federated graph learning for fMRI-based neurological disorder identification.

Resting-state functional magnetic resonance imaging (rs-fMRI) offers a non-invasive approach to examining abnormal brain connectivity associated with brain disorders. Graph neural network (GNN) gains popularity in fMRI representation learning and brain disorder analysis with powerful graph representation capabilities. Training a general GNN often necessitates a large-scale dataset from multiple imaging centers/sites, but centralizing multi-site data generally faces inherent challenges related to data privacy, security, and storage burden. Federated Learning (FL) enables collaborative model training without centralized multi-site fMRI data. Unfortunately, previous FL approaches for fMRI analysis often ignore site-specificity, including demographic factors such as age, gender, and education level. To this end, we propose a specificity-aware federated graph learning (SFGL) framework for rs-fMRI analysis and automated brain disorder identification, with a server and multiple clients/sites for federated model aggregation and prediction. At each client, our model consists of a shared and a personalized branch, where parameters of the shared branch are sent to the server while those of the personalized branch remain local. This can facilitate knowledge sharing among sites and also helps preserve site specificity. In the shared branch, we employ a spatio-temporal attention graph isomorphism network to learn dynamic fMRI representations. In the personalized branch, we integrate vectorized demographic information (i.e., age, gender, and education years) and functional connectivity networks to preserve site-specific characteristics. Representations generated by the two branches are then fused for classification. Experimental results on two fMRI datasets with a total of 1218 subjects suggest that SFGL outperforms several state-of-the-art approaches.

Unsupervised contrastive graph learning for resting-state functional MRI analysis and brain disorder detection.

Resting-state functional magnetic resonance imaging (rs-fMRI) helps characterize regional interactions that occur in the human brain at a resting state. Existing research often attempts to explore fMRI biomarkers that best predict brain disease progression using machine/deep learning techniques. Previous fMRI studies have shown that learning-based methods usually require a large amount of labeled training data, limiting their utility in clinical practice where annotating data is often time-consuming and labor-intensive. To this end, we propose an unsupervised contrastive graph learning (UCGL) framework for fMRI-based brain disease analysis, in which a pretext model is designed to generate informative fMRI representations using unlabeled training data, followed by model fine-tuning to perform downstream disease identification tasks. Specifically, in the pretext model, we first design a bi-level fMRI augmentation strategy to increase the sample size by augmenting blood-oxygen-level-dependent (BOLD) signals, and then employ two parallel graph convolutional networks for fMRI feature extraction in an unsupervised contrastive learning manner. This pretext model can be optimized on large-scale fMRI datasets, without requiring labeled training data. This model is further fine-tuned on to-be-analyzed fMRI data for downstream disease detection in a task-oriented learning manner. We evaluate the proposed method on three rs-fMRI datasets for cross-site and cross-dataset learning tasks. Experimental results suggest that the UCGL outperforms several state-of-the-art approaches in automated diagnosis of three brain diseases (i.e., major depressive disorder, autism spectrum disorder, and Alzheimer's disease) with rs-fMRI data.

Multi-Scale Dynamic Graph Learning for Brain Disorder Detection With Functional MRI.

Resting-state functional magnetic resonance imaging (rs-fMRI) has been widely used in the detection of brain disorders such as autism spectrum disorder based on various machine/deep learning techniques. Learning-based methods typically rely on functional connectivity networks (FCNs) derived from blood-oxygen-level-dependent time series of rs-fMRI data to capture interactions between brain regions-of-interest (ROIs). Graph neural networks have been recently used to extract fMRI features from graph-structured FCNs, but cannot effectively characterize spatiotemporal dynamics of FCNs, e.g., the functional connectivity of brain ROIs is dynamically changing in a short period of time. Also, many studies usually focus on single-scale topology of FCN, thereby ignoring the potential complementary topological information of FCN at different spatial resolutions. To this end, in this paper, we propose a multi-scale dynamic graph learning (MDGL) framework to capture multi-scale spatiotemporal dynamic representations of rs-fMRI data for automated brain disorder diagnosis. The MDGL framework consists of three major components: 1) multi-scale dynamic FCN construction using multiple brain atlases to model multi-scale topological information, 2) multi-scale dynamic graph representation learning to capture spatiotemporal information conveyed in fMRI data, and 3) multi-scale feature fusion and classification. Experimental results on two datasets show that MDGL outperforms several state-of-the-art methods.

Learning to Fuse Multiple Brain Functional Networks for Automated Autism Identification.

Functional connectivity network (FCN) has become a popular tool to identify potential biomarkers for brain dysfunction, such as autism spectrum disorder (ASD). Due to its importance, researchers have proposed many methods to estimate FCNs from resting-state functional MRI (rs-fMRI) data. However, the existing FCN estimation methods usually only capture a single relationship between brain regions of interest (ROIs), e.g., linear correlation, nonlinear correlation, or higher-order correlation, thus failing to model the complex interaction among ROIs in the brain. Additionally, such traditional methods estimate FCNs in an unsupervised way, and the estimation process is independent of the downstream tasks, which makes it difficult to guarantee the optimal performance for ASD identification. To address these issues, in this paper, we propose a multi-FCN fusion framework for rs-fMRI-based ASD classification. Specifically, for each subject, we first estimate multiple FCNs using different methods to encode rich interactions among ROIs from different perspectives. Then, we use the label information (ASD vs. healthy control (HC)) to learn a set of fusion weights for measuring the importance/discrimination of those estimated FCNs. Finally, we apply the adaptively weighted fused FCN on the ABIDE dataset to identify subjects with ASD from HCs. The proposed FCN fusion framework is straightforward to implement and can significantly improve diagnostic accuracy compared to traditional and state-of-the-art methods.

An end-to-end seizure prediction approach using long short-term memory network.

There are increasing epilepsy patients suffering from the pain of seizure onsets, and effective prediction of seizures could improve their quality of life. To obtain high sensitivity for epileptic seizure prediction, current studies generally need complex feature extraction operations, which heavily depends on the artificial experience (or domain knowledge) and is highly subjective. To address these issues, in this paper we propose an end-to-end epileptic seizure prediction approach based on the long short-term memory network (LSTM). In the new method, only the gamma band of raw electroencephalography (EEG) signals is extracted as network input directly for seizure prediction, thus avoiding subjective and expensive feature design process. Despite its simplicity, the proposed method achieves the mean sensitivity of 91.76% and false prediction rate (FPR) of 0.29/h on Children's Hospital Boston-MIT (CHB-MIT) scalp EEG Database, respectively, when identifying the preictal stage from the EEG signals. Furthermore, different from traditional methods that only consider the classification of preictal and interictal EEG, we introduce the postictal stage as an extra class in the proposed method. As a result, the performance of seizure prediction is further improved, obtaining a higher sensitivity of 92.17% and a low FPR of 0.27/h. The mean warning time is 44.46 min, which suggests that sufficient time is reserved for patients to take intervention measures by this prediction method.

The effect of node features on GCN-based brain network classification: an empirical study.

Brain functional network (BFN) analysis has become a popular technique for identifying neurological/mental diseases. Due to the fact that BFN is a graph, a graph convolutional network (GCN) can be naturally used in the classification of BFN. Different from traditional methods that directly use the adjacency matrices of BFNs to train a classifier, GCN requires an additional input-node features. To our best knowledge, however, there is no systematic study to analyze their influence on the performance of GCN-based brain disorder classification. Therefore, in this study, we conduct an empirical study on various node feature measures, including (1) original fMRI signals, (2) one-hot encoding, (3) node statistics, (4) node correlation, and (5) their combination. Experimental results on two benchmark databases show that different node feature inputs to GCN significantly affect the brain disease classification performance, and node correlation usually contributes higher accuracy compared to original signals and manually extracted statistical features.

Fusing Multiview Functional Brain Networks by Joint Embedding for Brain Disease Identification.

Background: Functional brain networks (FBNs) derived from resting-state functional MRI (rs-fMRI) have shown great potential in identifying brain disorders, such as autistic spectrum disorder (ASD). Therefore, many FBN estimation methods have been proposed in recent years. Most existing methods only model the functional connections between brain regions of interest (ROIs) from a single view (e.g., by estimating FBNs through a specific strategy), failing to capture the complex interactions among ROIs in the brain. Methods: To address this problem, we propose fusion of multiview FBNs through joint embedding, which can make full use of the common information of multiview FBNs estimated by different strategies. More specifically, we first stack the adjacency matrices of FBNs estimated by different methods into a tensor and use tensor factorization to learn the joint embedding (i.e., a common factor of all FBNs) for each ROI. Then, we use Pearson's correlation to calculate the connections between each embedded ROI in order to reconstruct a new FBN. Results: Experimental results obtained on the public ABIDE dataset with rs-fMRI data reveal that our method is superior to several state-of-the-art methods in automated ASD diagnosis. Moreover, by exploring FBN "features" that contributed most to ASD identification, we discovered potential biomarkers for ASD diagnosis. The proposed framework achieves an accuracy of 74.46%, which is generally better than the compared individual FBN methods. In addition, our method achieves the best performance compared to other multinetwork methods, i.e., an accuracy improvement of at least 2.72%. Conclusions: We present a multiview FBN fusion strategy through joint embedding for fMRI-based ASD identification. The proposed fusion method has an elegant theoretical explanation from the perspective of eigenvector centrality.

Modularity-Constrained Dynamic Representation Learning for Interpretable Brain Disorder Analysis with Functional MRI.

Resting-state functional MRI (rs-fMRI) is increasingly used to detect altered functional connectivity patterns caused by brain disorders, thereby facilitating objective quantification of brain pathology. Existing studies typically extract fMRI features using various machine/deep learning methods, but the generated imaging biomarkers are often challenging to interpret. Besides, the brain operates as a modular system with many cognitive/topological modules, where each module contains subsets of densely inter-connected regions-of-interest (ROIs) that are sparsely connected to ROIs in other modules. However, current methods cannot effectively characterize brain modularity. This paper proposes a modularity-constrained dynamic representation learning (MDRL) framework for interpretable brain disorder analysis with rs-fMRI. The MDRL consists of 3 parts: (1) dynamic graph construction, (2) modularity-constrained spatiotemporal graph neural network (MSGNN) for dynamic feature learning, and (3) prediction and biomarker detection. In particular, the MSGNN is designed to learn spatiotemporal dynamic representations of fMRI, constrained by 3 functional modules (i.e., central executive network, salience network, and default mode network). To enhance discriminative ability of learned features, we encourage the MSGNN to reconstruct network topology of input graphs. Experimental results on two public and one private datasets with a total of 1,155 subjects validate that our MDRL outperforms several state-of-the-art methods in fMRI-based brain disorder analysis. The detected fMRI biomarkers have good explainability and can be potentially used to improve clinical diagnosis.

Epileptic seizure prediction using successive variational mode decomposition and transformers deep learning network.

As one of the most common neurological disorders, epilepsy causes great physical and psychological damage to the patients. The long-term recurrent and unprovoked seizures make the prediction necessary. In this paper, a novel approach for epileptic seizure prediction based on successive variational mode decomposition (SVMD) and transformers is proposed. SVMD is extended to multidimensional form for time-frequency analysis of multi-channel signals. It could adaptively extract common band-limited intrinsic modes among all channels on different time scales by solving a variational optimization problem. In the proposed seizure prediction method, data are first decomposed into multiple modes on different time scales by multivariate SVMD, and then, irrelevant modes are removed for preprocessing. Finally, power spectrum of denoised data is input to a pre-trained bidirectional encoder representations from transformers (BERTs) for prediction. The BERT could identify the mode information related to epileptic seizures in time-frequency domain. It shows fair prediction performance on an intracranial EEG dataset with the average sensitivity of 0.86 and FPR of 0.18/h.

Resting-State Functional MRI Adaptation with Attention Graph Convolution Network for Brain Disorder Identification.

Multi-site resting-state functional magnetic resonance imaging (rs-fMRI) data can facilitate learning-based approaches to train reliable models on more data. However, significant data heterogeneity between imaging sites, caused by different scanners or protocols, can negatively impact the generalization ability of learned models. In addition, previous studies have shown that graph convolution neural networks (GCNs) are effective in mining fMRI biomarkers. However, they generally ignore the potentially different contributions of brain regions- of-interest (ROIs) to automated disease diagnosis/prognosis. In this work, we propose a multi-site rs-fMRI adaptation framework with attention GCN (A2GCN) for brain disorder identification. Specifically, the proposed A2GCN consists of three major components: (1) a node representation learning module based on GCN to extract rs-fMRI features from functional connectivity networks, (2) a node attention mechanism module to capture the contributions of ROIs, and (3) a domain adaptation module to alleviate the differences in data distribution between sites through the constraint of mean absolute error and covariance. The A2GCN not only reduces data heterogeneity across sites, but also improves the interpretability of the learning algorithm by exploring important ROIs. Experimental results on the public ABIDE database demonstrate that our method achieves remarkable performance in fMRI-based recognition of autism spectrum disorders.

Joint selection of brain network nodes and edges for MCI identification.

BACKGROUND AND OBJECTIVE: Functional brain graph (FBG), by describing the interactions between different brain regions, provides an effective representation of fMRI data for identifying mild cognitive impairment (MCI), an early stage of Alzheimer's Disease (AD). Prior to the identification task, selecting features from the estimated FBG is a necessary step for reducing computational cost, alleviating the risk of overfitting, and finding potential biomarkers of brain diseases. In practice, either node-based features (e.g., local clustering coefficients) or edge-based features (e.g., adjacency weights) are generally considered in current studies. Despite their popularity, these schemes can only capture one granularity (node or edge) of information in the FBG, which might be insufficient for the classification task and the interpretation of the classification result. METHODS: To address this issue, in this paper, we propose to jointly select nodes and edges from the estimated FBGs. Specifically, we first assign the edges to different node groups. Then, sparse group least absolute shrinkage and selection operator (sgLASSO) is used to select groups (nodes) and edges in the groups towards a better classification performance. Such a technique enables us to simultaneously locate discriminative brain regions, as well as connections between these brain regions, making the classification results more interpretable. RESULTS: Experimental results show that the proposed method achieves better classification performance than state-of-the-art methods. Moreover, by exploring brain network "features" that contributed most to MCI identification, we discover potential biomarkers for MCI diagnosis. CONCLUSION: A novel method for jointly selecting nodes and edges from the estimated functional brain graphs (FBGs) is proposed.

Selecting Multiple Node Statistics Jointly from Functional Connectivity Networks for Brain Disorders Identification.

Functional connectivity networks (FCN) analysis is instructive for the diagnosis of brain diseases, such as mild cognitive impairment (MCI) and major depressive disorder (MDD) at their early stages. As the critical step of FCN analysis, feature representation provides the basis for finding potential biomarkers of brain diseases. In previous studies, different node statistics (e.g. local efficiency and local clustering coefficients) are usually extracted from FCNs as features for the diagnosis/classification task, which can specifically locate disease-related regions on the node level, so as to help us understand the neurodevelopmental roots of brain disorders. However, each node statistic is proposed only considering a kind of specific network property, which has one-sidedness and limitations. As a result, it is incomplete to represent a node with only one statistic. To resolve this issue, we put forward a novel scheme to select multiple node statistics jointly from the estimated FCNs for automated classification, called multiple node statistics feature selection (MNSFS). Specifically, we first extract multiple statistics from the same nodes and assign each kind of statistic into a group. Then, sparse group least absolute shrinkage and selection operator (sgLASSO) is used to select groups (nodes) and statistics in the groups towards a better classification performance. Such a technique enables us to simultaneously locate the discriminative brain regions, as well as the specific statistics associated with these brain regions, making the classification results more interpretable. We conducted our scheme on two public databases for identifying subjects with MCI and MDD from normal controls. Experimental results show that the proposed scheme achieves superior classification accuracy and features interpreted on the benchmark datasets.

Estimating high-order brain functional networks by correlation-preserving embedding.

Brain functional network (FN) has emerged as a potential tool for identifying mental and neurological diseases. Traditional FN estimation methods such as Pearson's correlation (PC) and sparse representation (SR), despite their popularity, can only model low-order relationships between brain regions (i.e., nodes of FN), thus failing to capture more complex interaction in the brain. Recently, researchers proposed to estimate high-order FN (HoFN) and successfully used them in the early diagnosis of neurological diseases. In practice, however, such HoFN is constructed by directly considering the columns (or rows) of the adjacency matrix of low-order FN (LoFN) as node feature vectors that may contain some redundant or noisy information. In addition, it is not really reflected whether the original low-order relationship is maintained during the construction of the HoFN. To address these problems, we propose correlation-preserving embedding (COPE) to re-code the LoFN prior to constructing HoFN. Specifically, we first use SR to construct traditional LoFN. Then, we embed the LoFN via COPE to generate the new node representation for removing the potentially redundant/noisy information in original node feature vectors and simultaneously maintaining the low-order relationship between brain regions. Finally, the expected HoFN is estimated by SR based on the new node representation. To verify the effectiveness of the proposed scheme, we conduct experiments on 137 subjects from the public Alzheimer's Disease Neuroimaging Initiative (ADNI) database to identify subjects with mild cognitive impairment (MCI) from normal controls. Experimental results show that the proposed scheme can achieve better performance than the baseline method. Brain functional network (FN) has emerged as a potential tool for identifying mental and neurological diseases. Traditional FN estimation methods such as Pearson's correlation (PC) and sparse representation (SR), despite their popularity, can only model low-order relationships between brain regions (i.e., nodes of FN), thus failing to capture more complex interaction in the brain. Recently, researchers proposed to estimate high-order FN (HoFN) and successfully used them in the early diagnosis of neurological diseases. In practice, however, such HoFN is constructed by directly considering the columns (or rows) of the adjacency matrix of low-order FN (LoFN) as node feature vectors that may contain some redundant or noisy information. In addition, it is not really reflected whether the original low-order relationship is maintained during the construction of the HoFN. To address these problems, we propose correlation-preserving embedding (COPE) to re-code the LoFN prior to constructing HoFN. Specifically, we first use SR to construct traditional LoFN. Then, we embed the LoFN via COPE to generate the new node representation for removing the potentially redundant/noisy information in original node feature vectors and simultaneously maintaining the low-order relationship between brain regions. Finally, the expected HoFN is estimated by SR based on the new node representation. To verify the effectiveness of the proposed scheme, we conduct experiments on 137 subjects from the public Alzheimer's Disease Neuroimaging Initiative (ADNI) database to identify subjects with mild cognitive impairment (MCI) from normal controls. Experimental results show that the proposed scheme can achieve better performance than the baseline method.

Adaptive Multimodal Neuroimage Integration for Major Depression Disorder Detection.

Major depressive disorder (MDD) is one of the most common mental health disorders that can affect sleep, mood, appetite, and behavior of people. Multimodal neuroimaging data, such as functional and structural magnetic resonance imaging (MRI) scans, have been widely used in computer-aided detection of MDD. However, previous studies usually treat these two modalities separately, without considering their potentially complementary information. Even though a few studies propose integrating these two modalities, they usually suffer from significant inter-modality data heterogeneity. In this paper, we propose an adaptive multimodal neuroimage integration (AMNI) framework for automated MDD detection based on functional and structural MRIs. The AMNI framework consists of four major components: (1) a graph convolutional network to learn feature representations of functional connectivity networks derived from functional MRIs, (2) a convolutional neural network to learn features of T1-weighted structural MRIs, (3) a feature adaptation module to alleviate inter-modality difference, and (4) a feature fusion module to integrate feature representations extracted from two modalities for classification. To the best of our knowledge, this is among the first attempts to adaptively integrate functional and structural MRIs for neuroimaging-based MDD analysis by explicitly alleviating inter-modality heterogeneity. Extensive evaluations are performed on 533 subjects with resting-state functional MRI and T1-weighted MRI, with results suggesting the efficacy of the proposed method.

Estimating High-Order Brain Functional Networks in Bayesian View for Autism Spectrum Disorder Identification.

Brain functional network (BFN) has become an increasingly important tool to understand the inherent organization of the brain and explore informative biomarkers of neurological disorders. Pearson's correlation (PC) is the most widely accepted method for constructing BFNs and provides a basis for designing new BFN estimation schemes. Particularly, a recent study proposes to use two sequential PC operations, namely, correlation's correlation (CC), for constructing the high-order BFN. Despite its empirical effectiveness in identifying neurological disorders and detecting subtle changes of connections in different subject groups, CC is defined intuitively without a solid and sustainable theoretical foundation. For understanding CC more rigorously and providing a systematic BFN learning framework, in this paper, we reformulate it in the Bayesian view with a prior of matrix-variate normal distribution. As a result, we obtain a probabilistic explanation of CC. In addition, we develop a Bayesian high-order method (BHM) to automatically and simultaneously estimate the high- and low-order BFN based on the probabilistic framework. An efficient optimization algorithm is also proposed. Finally, we evaluate BHM in identifying subjects with autism spectrum disorder (ASD) from typical controls based on the estimated BFNs. Experimental results suggest that the automatically learned high- and low-order BFNs yield a superior performance over the artificially defined BFNs via conventional CC and PC.

Identifying Individuals by fNIRS-Based Brain Functional Network Fingerprints.

Individual identification based on brain functional network (BFN) has attracted a lot of research interest in recent years, since it provides a novel biometric for identity authentication, as well as a feasible way of exploring the brain at an individual level. Previous studies have shown that an individual can be identified by its BFN fingerprint estimated from functional magnetic resonance imaging, electroencephalogram, or magnetoencephalography data. Functional near-infrared spectroscopy (fNIRS) is an emerging imaging technique that, by measuring the changes in blood oxygen concentration, can respond to cerebral activities; in this paper, we investigate whether fNIRS-based BFN could be used as a "fingerprint" to identify individuals. In particular, Pearson's correlation is first used to calculate BFN based on the preprocessed fNIRS signals, and then the nearest neighbor scheme is used to match the estimated BFNs between different individuals. Through the experiments on an open-access fNIRS dataset, we have two main findings: (1) under the cases of cross-task (i.e., resting, right-handed, left-handed finger tapping, and foot tapping), the BFN fingerprints generally work well for the individual identification, and, more interestingly, (2) the accuracy under cross-task is well above the accuracy under cross-view (i.e., oxyhemoglobin and de-oxyhemoglobin). These findings indicate that fNIRS-based BFN fingerprint is a potential biometric for identifying individual.

Extracting BOLD signals based on time-constrained multiset canonical correlation analysis for brain functional network estimation and classification.

Brain functional network (BFN), usually estimated from blood oxygen level dependent (BOLD) functional magnetic resonance imaging (fMRI), has been proven to be a powerful tool to study the organization of the brain and discover biomarkers for diagnosis of brain disorders. Prior to BFN estimation and classification, extracting representative BOLD signals from brain regions of interest (ROIs) is a critical step. Traditional extraction methods include averaging, peaking operation and dimensionality reduction, often leading to signal cancellation and information loss. In this paper, we propose a novel method, namely time-constrained multiset canonical correlation analysis (TMCCA), to extract representative BOLD signals for subsequent BFN estimation and classification. Different from traditional methods that equally treat all BOLD signals in a ROI, the proposed method assigns weights to different BOLD signals, and learns the optimal weights to make the extracted representative signals jointly maximize the multiple correlations between ROIs. Importantly, time-constraint is incorporated into our proposed method, which can effectively encode nonlinear relationship among BOLD signals. To evaluate the effectiveness of the proposed method, the extracted BOLD signals is used to estimate BFN and, in turn, identify brain disorders, including mild cognitive impairment (MCI) and autistic spectrum disorder (ASD). Experimental results demonstrate that our proposed TMCCA can lead to better performance than traditional methods.

Multi-modality relation attention network for breast tumor classification.

Automatic breast image classification plays an important role in breast cancer diagnosis, and multi-modality image fusion may improve classification performance. However, existing fusion methods ignore relevant multi-modality information in favor of improving the discriminative ability of single-modality features. To improve classification performance, this paper proposes a multi-modality relation attention network with consistent regularization for breast tumor classification using diffusion-weighted imaging (DWI) and apparent dispersion coefficient (ADC) images. Within the proposed network, a novel multi-modality relation attention module improves the discriminative ability of single-modality features by exploring the correlation information between two modalities. In addition, a module ensures the classification consistency of ADC and DWI modality, thus improving robustness to noise. Experimental results on our database demonstrate that the proposed method is effective for breast tumor classification, and outperforms existing multi-modality fusion methods. The AUC, accuracy, specificity, and sensitivity are 85.1%, 86.7%, 83.3%, and 88.9% respectively.