Supervised Discriminative Group Sparse Representation for Mild Cognitive Impairment Diagnosis

Abnormal functional connectivity density involvement in freezing of gait and its application for subtyping Parkinson's disease

The pathophysiological mechanisms at work in Parkinson's disease (PD) patients with freezing of gait (FOG) remain poorly understood. Functional connectivity density (FCD) could provide an unbiased way to analyse connectivity across the brain. In this study, a total of 23 PD patients with FOG (PD FOG + patients), 26 PD patients without FOG (PD FOG- patients), and 22 healthy controls (HCs) were recruited, and their resting-state functional magnetic resonance imaging (rs-fMRI) images were collected. FCD mapping was first performed to identify differences between groups. Pearson correlation analysis was used to explore relationships between FCD values and the severity of FOG. Then, a machine learning model was employed to classify each pair of groups. PD FOG + patients showed significantly increased short-range FCD in the precuneus, cingulate gyrus, and fusiform gyrus and decreased long-range FCD in the frontal gyrus, temporal gyrus, and cingulate gyrus. Short-range FCD values in the middle temporal gyrus and inferior temporal gyrus were positively correlated with FOG questionnaire (FOGQ) scores, and long-range FCD values in the middle frontal gyrus were negatively correlated with FOGQ scores. Using FCD in abnormal regions as input, a support vector machine (SVM) classifier can achieve classification with good performance. The mean accuracy values were 0.895 (PD FOG + vs. HC), 0.966 (PD FOG- vs. HC), and 0.897 (PD FOG + vs. PD FOG-). This study demonstrates that PD FOG + patients showed altered short- and long-range FCD in several brain regions involved in action planning and control, motion processing, emotion, cognition, and object recognition.

Learn-Explain-Reinforce: Counterfactual Reasoning and its Guidance to Reinforce an Alzheimer's Disease Diagnosis Model

Existing studies on disease diagnostic models focus either on diagnostic model learning for performance improvement or on the visual explanation of a trained diagnostic model. We propose a novel learn-explain-reinforce (LEAR) framework that unifies diagnostic model learning, visual explanation generation (explanation unit), and trained diagnostic model reinforcement (reinforcement unit) guided by the visual explanation. For the visual explanation, we generate a counterfactual map that transforms an input sample to be identified as an intended target label. For example, a counterfactual map can localize hypothetical abnormalities within a normal brain image that may cause it to be diagnosed with Alzheimer's disease (AD). We believe that the generated counterfactual maps represent data-driven knowledge about a target task, i.e., AD diagnosis using structural MRI, which can be a vital source of information to reinforce the generalization of the trained diagnostic model. To this end, we devise an attention-based feature refinement module with the guidance of the counterfactual maps. The explanation and reinforcement units are reciprocal and can be operated iteratively. Our proposed approach was validated via qualitative and quantitative analysis on the ADNI dataset. Its comprehensibility and fidelity were demonstrated through ablation studies and comparisons with existing methods.

Accurate module induced brain network construction for mild cognitive impairment identification with functional MRI

Introduction

Functional brain networks (FBNs) estimated from functional magnetic resonance imaging (fMRI) data has become a potentially useful way for computer-aided diagnosis of neurological disorders, such as mild cognitive impairment (MCI), a prodromal stage of Alzheimer's Disease (AD). Currently, Pearson's correlation (PC) is the most widely-used method for constructing FBNs. Despite its popularity and simplicity, the conventional PC-based method usually results in dense networks where regions-of-interest (ROIs) are densely connected. This is not accordance with the biological prior that ROIs may be sparsely connected in the brain. To address this issue, previous studies proposed to employ a threshold or l_1-regularizer to construct sparse FBNs. However, these methods usually ignore rich topology structures, such as modularity that has been proven to be an important property for improving the information processing ability of the brain.

Methods

To this end, in this paper, we propose an accurate module induced PC (AM-PC) model to estimate FBNs with a clear modular structure, by including sparse and low-rank constraints on the Laplacian matrix of the network. Based on the property that zero eigenvalues of graph Laplacian matrix indicate the connected components, the proposed method can reduce the rank of the Laplacian matrix to a pre-defined number and obtain FBNs with an accurate number of modules.

Results

To validate the effectiveness of the proposed method, we use the estimated FBNs to classify subjects with MCI from healthy controls. Experimental results on 143 subjects from Alzheimer's Disease Neuroimaging Initiative (ADNI) with resting-state functional MRIs show that the proposed method achieves better classification performance than previous methods.

A Fusion-Based Technique With Hybrid Swarm Algorithm and Deep Learning for Biosignal Classification

The vital data about the electrical activities of the brain are carried by the electroencephalography (EEG) signals. The recordings of the electrical activity of brain neurons in a rhythmic and spontaneous manner from the scalp surface are measured by EEG. One of the most important aspects in the field of neuroscience and neural engineering is EEG signal analysis, as it aids significantly in dealing with the commercial applications as well. To uncover the highly useful information for neural classification activities, EEG studies incorporated with machine learning provide good results. In this study, a Fusion Hybrid Model (FHM) with Singular Value Decomposition (SVD) Based Estimation of Robust Parameters is proposed for efficient feature extraction of the biosignals and to understand the essential information it has for analyzing the brain functionality. The essential features in terms of parameter components are extracted using the developed hybrid model, and a specialized hybrid swarm technique called Hybrid Differential Particle Artificial Bee (HDPAB) algorithm is proposed for feature selection. To make the EEG more practical and to be used in a plethora of applications, the robust classification of these signals is necessary thereby relying less on the trained professionals. Therefore, the classification is done initially using the proposed Zero Inflated Poisson Mixture Regression Model (ZIPMRM) and then it is also classified with a deep learning methodology, and the results are compared with other standard machine learning techniques. This proposed flow of methodology is validated on a few standard Biosignal datasets, and finally, a good classification accuracy of 98.79% is obtained for epileptic dataset and 98.35% is obtained for schizophrenia dataset.

Sparse Granger Causality Analysis Model Based on Sensors Correlation for Emotion Recognition Classification in Electroencephalography

In recent years, affective computing based on electroencephalogram (EEG) data has attracted increased attention. As a classic EEG feature extraction model, Granger causality analysis has been widely used in emotion classification models, which construct a brain network by calculating the causal relationships between EEG sensors and select the key EEG features. Traditional EEG Granger causality analysis uses the L₂ norm to extract features from the data, and so the results are susceptible to EEG artifacts. Recently, several researchers have proposed Granger causality analysis models based on the least absolute shrinkage and selection operator (LASSO) and the L_1/2 norm to solve this problem. However, the conventional sparse Granger causality analysis model assumes that the connections between each sensor have the same prior probability. This paper shows that if the correlation between the EEG data from each sensor can be added to the Granger causality network as prior knowledge, the EEG feature selection ability and emotional classification ability of the sparse Granger causality model can be enhanced. Based on this idea, we propose a new emotional computing model, named the sparse Granger causality analysis model based on sensor correlation (SC-SGA). SC-SGA integrates the correlation between sensors as prior knowledge into the Granger causality analysis based on the L_1/2 norm framework for feature extraction, and uses L₂ norm logistic regression as the emotional classification algorithm. We report the results of experiments using two real EEG emotion datasets. These results demonstrate that the emotion classification accuracy of the SC-SGA model is better than that of existing models by 2.46–21.81%.

Identifying Internet Addiction and Evaluating the Efficacy of Treatment Based on Functional Connectivity Density: A Machine Learning Study

Although mounting neuroimaging studies have greatly improved our understanding of the neurobiological mechanism underlying internet addiction (IA), the results based on traditional group-level comparisons are insufficient in guiding individual clinical practice directly. Specific neuroimaging biomarkers are urgently needed for IA diagnosis and the evaluation of therapy efficacy. Therefore, this study aimed to develop support vector machine (SVM) models to identify IA and assess the efficacy of cognitive behavior therapy (CBT) based on unbiased functional connectivity density (FCD). Resting-state fMRI data were acquired from 27 individuals with IA before and after 8-week CBT sessions and 30 demographically matched healthy controls (HCs). The discriminative FCDs were computed as the features of the support vector classification (SVC) model to identify individuals with IA from HCs, and the changes in these discriminative FCDs after treatment were further used as features of the support vector regression (SVR) model to evaluate the efficacy of CBT. Based on the informative FCDs, our SVC model successfully differentiated individuals with IA from HCs with an accuracy of 82.5% and an area under the curve (AUC) of 0.91. Our SVR model successfully evaluated the efficacy of CBT using the FCD change ratio with a correlation efficient of 0.59. The brain regions contributing to IA classification and CBT efficacy assessment were the left inferior frontal cortex (IFC), middle frontal cortex (MFC) and angular gyrus (AG), the right premotor cortex (PMC) and middle cingulate cortex (MCC), and the bilateral cerebellum, orbitofrontal cortex (OFC) and superior frontal cortex (SFC). These findings confirmed the FCDs of hyperactive impulsive habit system, hypoactive reflecting system and sensitive interoceptive reward awareness system as potential neuroimaging biomarkers for IA, which might provide objective indexes for the diagnosis and efficacy evaluation of IA.

Diagnostic power of resting-state fMRI for detection of network connectivity in Alzheimer's disease and mild cognitive impairment: A systematic review

Resting‐state fMRI (rs‐fMRI) detects functional connectivity (FC) abnormalities that occur in the brains of patients with Alzheimer's disease (AD) and mild cognitive impairment (MCI). FC of the default mode network (DMN) is commonly impaired in AD and MCI. We conducted a systematic review aimed at determining the diagnostic power of rs‐fMRI to identify FC abnormalities in the DMN of patients with AD or MCI compared with healthy controls (HCs) using machine learning (ML) methods. Multimodal support vector machine (SVM) algorithm was the commonest form of ML method utilized. Multiple kernel approach can be utilized to aid in the classification by incorporating various discriminating features, such as FC graphs based on “nodes” and “edges” together with structural MRI‐based regional cortical thickness and gray matter volume. Other multimodal features include neuropsychiatric testing scores, DTI features, and regional cerebral blood flow. Among AD patients, the posterior cingulate cortex (PCC)/Precuneus was noted to be a highly affected hub of the DMN that demonstrated overall reduced FC. Whereas reduced DMN FC between the PCC and anterior cingulate cortex (ACC) was observed in MCI patients. Evidence indicates that the nodes of the DMN can offer moderate to high diagnostic power to distinguish AD and MCI patients. Nevertheless, various concerns over the homogeneity of data based on patient selection, scanner effects, and the variable usage of classifiers and algorithms pose a challenge for ML‐based image interpretation of rs‐fMRI datasets to become a mainstream option for diagnosing AD and predicting the conversion of HC/MCI to AD.

Estimating sparse functional connectivity networks via hyperparameter-free learning model

Functional connectivity networks (FCNs) provide a potential way for understanding the brain organizational patterns and diagnosing neurological diseases. Currently, researchers have proposed many methods for FCN construction, among which the most classic example is Pearson's correlation (PC). Despite its simplicity and popularity, PC always results in dense FCNs, and thus a thresholding strategy is usually needed in practice to sparsify the estimated FCNs prior to the network analysis, which undoubtedly causes the problem of threshold parameter selection. As an alternative to PC, sparse representation (SR) can directly generate sparse FCNs due to the l₁ regularizer in the estimation model. However, similar to the thresholding scheme used in PC, it is also challenging to determine suitable values for the regularization parameter in SR. To circumvent the difficulty of parameter selection involved in these traditional methods, we propose a hyperparameter-free method for FCN construction based on the global representation among fMRI time courses. Interestingly, the proposed method can automatically generate sparse FCNs, without any thresholding or regularization parameters. To verify the effectiveness of the proposed method, we conduct experiments to identify subjects with mild cognitive impairment (MCI) and Autism spectrum disorder (ASD) from normal controls (NCs) based on the estimated FCNs. Experimental results on two benchmark databases demonstrate that the achieved classification performance of our proposed scheme is comparable to four conventional methods.

Identifying resting-state effective connectivity abnormalities in drug-naïve major depressive disorder diagnosis via graph convolutional networks

Major depressive disorder (MDD) is a leading cause of disability; its symptoms interfere with social, occupational, interpersonal, and academic functioning. However, the diagnosis of MDD is still made by phenomenological approach. The advent of neuroimaging techniques allowed numerous studies to use resting-state functional magnetic resonance imaging (rs-fMRI) and estimate functional connectivity for brain-disease identification. Recently, attempts have been made to investigate effective connectivity (EC) that represents causal relations among regions of interest. In the meantime, to identify meaningful phenotypes for clinical diagnosis, graph-based approaches such as graph convolutional networks (GCNs) have been leveraged recently to explore complex pairwise similarities in imaging/nonimaging features among subjects. In this study, we validate the use of EC for MDD identification by estimating its measures via a group sparse representation along with a structured equation modeling approach in a whole-brain data-driven manner from rs-fMRI. To distinguish drug-naïve MDD patients from healthy controls, we utilize spectral GCNs based on a population graph to successfully integrate EC and nonimaging phenotypic information. Furthermore, we devise a novel sensitivity analysis method to investigate the discriminant connections for MDD identification in our trained GCNs. Our experimental results validated the effectiveness of our method in various scenarios, and we identified altered connectivities associated with the diagnosis of MDD.

A toolbox for brain network construction and classification (BrainNetClass)

Brain functional network has been increasingly used in understanding brain functions and diseases. While many network construction methods have been proposed, the progress in the field still largely relies on static pairwise Pearson's correlation-based functional network and group-level comparisons. We introduce a "Brain Network Construction and Classification (BrainNetClass)" toolbox to promote more advanced brain network construction methods to the filed, including some state-of-the-art methods that were recently developed to capture complex and high-order interactions among brain regions. The toolbox also integrates a well-accepted and rigorous classification framework based on brain connectome features toward individualized disease diagnosis in a hope that the advanced network modeling could boost the subsequent classification. BrainNetClass is a MATLAB-based, open-source, cross-platform toolbox with both graphical user-friendly interfaces and a command line mode targeting cognitive neuroscientists and clinicians for promoting reliability, reproducibility, and interpretability of connectome-based, computer-aided diagnosis. It generates abundant classification-related results from network presentations to contributing features that have been largely ignored by most studies to grant users the ability of evaluating the disease diagnostic model and its robustness and generalizability. We demonstrate the effectiveness of the toolbox on real resting-state functional MRI datasets. BrainNetClass (v1.0) is available at https://github.com/zzstefan/BrainNetClass.

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.

Discriminative self-representation sparse regression for neuroimaging-based alzheimer's disease diagnosis

In this paper, we propose a novel feature selection method by jointly considering (1) 'task-specific' relations between response variables (e.g., clinical labels in this work) and neuroimaging features and (2) 'self-representation' relations among neuroimaging features in a sparse regression framework. Specifically, the task-specific relation is devised to learn the relative importance of features for representation of response variables by a linear combination of the input features in a supervised manner, while the self-representation relation is used to take into account the inherent information among neuroimaging features such that any feature can be represented by a weighted sum of the other features, regardless of the label information, in an unsupervised manner. By integrating these two different relations along with a group sparsity constraint, we formulate a new sparse linear regression model for class-discriminative feature selection. The selected features are used to train a support vector machine for classification. To validate the effectiveness of the proposed method, we conducted experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset; experimental results showed superiority of the proposed method over the state-of-the-art methods considered in this work.

Improving Sparsity and Modularity of High-Order Functional Connectivity Networks for MCI and ASD Identification

High-order correlation has recently been proposed to model brain functional connectivity network (FCN) for identifying neurological disorders, such as mild cognitive impairment (MCI) and autism spectrum disorder (ASD). In practice, the high-order FCN (HoFCN) can be derived from multiple low-order FCNs that are estimated separately in a series of sliding windows, and thus it in fact provides a way of integrating dynamic information encoded in a sequence of low-order FCNs. However, the estimation of low-order FCN may be unreliable due to the fact that the use of limited volumes/samples in a sliding window can significantly reduce the statistical power, which in turn affects the reliability of the resulted HoFCN. To address this issue, we propose to enhance HoFCN based on a regularized learning framework. More specifically, we first calculate an initial HoFCN using a recently developed method based on maximum likelihood estimation. Then, we learn an optimal neighborhood network of the initially estimated HoFCN with sparsity and modularity priors as regularizers. Finally, based on the improved HoFCNs, we conduct experiments to identify MCI and ASD patients from their corresponding normal controls. Experimental results show that the proposed methods outperform the baseline methods, and the improved HoFCNs with modularity prior consistently achieve the best performance.

Multi-task fused sparse learning for mild cognitive impairment identification

BACKGROUND

Brain functional connectivity network (BFCN) has been widely applied to identify biomarkers for the brain function understanding and brain diseases analysis.

OBJECTIVE

Building a biologically meaningful brain network is a crucial work in these applications. For this task, sparse learning has been widely applied for the network construction. If multiple time-point data is added to the brain imaging application, the disease progression pattern in the longitudinal analysis can be better revealed.

METHODS

A novel longitudinal analysis for MCI classification is devised based on resting-state functional magnetic resonating imaging (rs-fMRI). Specifically, this paper proposes a novel multi-task learning method to integrate fused penalty by regularization. In addition, a novel objective function is developed for fused sparse learning via smoothness constraint.

RESULTS

The proposed method achieves the best classification performance with an accuracy of 95.74% for baseline and 93.64% for year 1 data.

CONCLUSIONS

The experimental results show that our proposed method achieves quite promising classification performance.

Canonical feature selection for joint regression and multi-class identification in Alzheimer's disease diagnosis

Fusing information from different imaging modalities is crucial for more accurate identification of the brain state because imaging data of different modalities can provide complementary perspectives on the complex nature of brain disorders. However, most existing fusion methods often extract features independently from each modality, and then simply concatenate them into a long vector for classification, without appropriate consideration of the correlation among modalities. In this paper, we propose a novel method to transform the original features from different modalities to a common space, where the transformed features become comparable and easy to find their relation, by canonical correlation analysis. We then perform the sparse multi-task learning for discriminative feature selection by using the canonical features as regressors and penalizing a loss function with a canonical regularizer. In our experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, we use Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET) images to jointly predict clinical scores of Alzheimer's Disease Assessment Scale-Cognitive subscale (ADAS-Cog) and Mini-Mental State Examination (MMSE) and also identify multi-class disease status for Alzheimer's disease diagnosis. The experimental results showed that the proposed canonical feature selection method helped enhance the performance of both clinical score prediction and disease status identification, outperforming the state-of-the-art methods.

Sparse temporally dynamic resting-state functional connectivity networks for early MCI identification

In conventional resting-state functional MRI (R-fMRI) analysis, functional connectivity is assumed to be temporally stationary, overlooking neural activities or interactions that may happen within the scan duration. Dynamic changes of neural interactions can be reflected by variations of topology and correlation strength in temporally correlated functional connectivity networks. These connectivity networks may potentially capture subtle yet short neural connectivity disruptions induced by disease pathologies. Accordingly, we are motivated to utilize disrupted temporal network properties for improving control-patient classification performance. Specifically, a sliding window approach is firstly employed to generate a sequence of overlapping R-fMRI sub-series. Based on these sub-series, sliding window correlations, which characterize the neural interactions between brain regions, are then computed to construct a series of temporal networks. Individual estimation of these temporal networks using conventional network construction approaches fails to take into consideration intrinsic temporal smoothness among successive overlapping R-fMRI sub-series. To preserve temporal smoothness of R-fMRI sub-series, we suggest to jointly estimate the temporal networks by maximizing a penalized log likelihood using a fused sparse learning algorithm. This sparse learning algorithm encourages temporally correlated networks to have similar network topology and correlation strengths. We design a disease identification framework based on the estimated temporal networks, and group level network property differences and classification results demonstrate the importance of including temporally dynamic R-fMRI scan information to improve diagnosis accuracy of mild cognitive impairment patients.

Inter-subject Similarity Guided Brain Network Modeling for MCI Diagnosis

Sparse representation-based brain network modeling, although popular, often results in relatively large inter-subject variability in network structures. This inevitably makes it difficult for inter-subject comparison, thus eventually deteriorating the generalization capability of personalized disease diagnosis. Accordingly, group sparse representation has been proposed to alleviate such limitation by jointly estimating connectivity weights for all subjects. However, the constructed brain networks based on this method often fail in providing satisfactory separability between the subjects from different groups (e.g., patients vs. normal controls), which will also affect the performance of computer-aided disease diagnosis. Based on the hypothesis that subjects from the same group should have larger similarity in their functional connectivity (FC) patterns than subjects from other groups, we propose an "inter-subject FC similarity-guided" group sparse network modeling method. In this method, we explicitly include the inter-subject FC similarity as a constraint to conduct group-wise FC network modeling, while retaining sufficient between-group differences in the resultant FC networks. This improves the separability of brain functional networks between different groups, thus facilitating better personalized brain disease diagnosis. Specifically, the inter-subject FC similarity is roughly estimated by comparing the Pearson's correlation based FC patterns of each brain region to other regions for each pair of the subjects. Then, this is implemented as an additional weighting term to ensure the adequate inter-subject FC differences between the subjects from different groups. Of note, our method retains the group sparsity constraint to ensure the overall consistency of the resultant individual brain networks. Experimental results show that our method achieves a balanced trade-off by not only generating the individually consistent FC networks, but also effectively maintaining the necessary group difference, thereby significantly improving connectomics-based diagnosis for mild cognitive impairment (MCI).

Can a Resting-State Functional Connectivity Index Identify Patients with Alzheimer's Disease and Mild Cognitive Impairment Across Multiple Sites?

Resting-state functional connectivity is one promising biomarker for Alzheimer's disease (AD) and mild cognitive impairment (MCI). However, it is still not known how accurately network analysis identifies AD and MCI across multiple sites. In this study, we examined whether resting-state functional connectivity data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) could identify patients with AD and MCI at our site. We implemented an index based on the functional connectivity frequency distribution and compared performance for AD and MCI identification with multivoxel pattern analysis. The multivoxel pattern analysis using a connectivity map of the default mode network showed good performance, with an accuracy of 81.9% for AD and MCI identification within the ADNI, but the classification model obtained from the ADNI failed to classify AD, MCI, and healthy elderly adults from our site, with an accuracy of only 43.1%. In contrast, a functional connectivity index of the medial temporal lobe based on the frequency distribution showed moderate performance, with an accuracy of 76.5-80.3% for AD identification within the ADNI. The performance of this index was similar for our data, with an accuracy of 73.9-82.6%. The frequency distribution-based index of functional connectivity could be a good biomarker for AD across multiple sites.

Hierarchical High-Order Functional Connectivity Networks and Selective Feature Fusion for MCI Classification

Conventional Functional connectivity (FC) analysis focuses on characterizing the correlation between two brain regions, whereas the high-order FC can model the correlation between two brain region pairs. To reduce the number of brain region pairs, clustering is applied to group all the brain region pairs into a small number of clusters. Then, a high-order FC network can be constructed based on the clustering result. By varying the number of clusters, multiple high-order FC networks can be generated and the one with the best overall performance can be finally selected. However, the important information contained in other networks may be simply discarded. To address this issue, in this paper, we propose to make full use of the information contained in all high-order FC networks. First, an agglomerative hierarchical clustering technique is applied such that the clustering result in one layer always depends on the previous layer, thus making the high-order FC networks in the two consecutive layers highly correlated. As a result, the features extracted from high-order FC network in each layer can be decomposed into two parts (blocks), i.e., one is redundant while the other might be informative or complementary, with respect to its previous layer. Then, a selective feature fusion method, which combines sequential forward selection and sparse regression, is developed to select a feature set from those informative feature blocks for classification. Experimental results confirm that our novel method outperforms the best single high-order FC network in diagnosis of mild cognitive impairment (MCI) subjects.

Extraction of dynamic functional connectivity from brain grey matter and white matter for MCI classification

Brain functional connectivity (FC) extracted from resting-state fMRI (RS-fMRI) has become a popular approach for diagnosing various neurodegenerative diseases, including Alzheimer's disease (AD) and its prodromal stage, mild cognitive impairment (MCI). Current studies mainly construct the FC networks between grey matter (GM) regions of the brain based on temporal co-variations of the blood oxygenation level-dependent (BOLD) signals, which reflects the synchronized neural activities. However, it was rarely investigated whether the FC detected within the white matter (WM) could provide useful information for diagnosis. Motivated by the recently proposed functional correlation tensors (FCT) computed from RS-fMRI and used to characterize the structured pattern of local FC in the WM, we propose in this article a novel MCI classification method based on the information conveyed by both the FC between the GM regions and that within the WM regions. Specifically, in the WM, the tensor-based metrics (e.g., fractional anisotropy [FA], similar to the metric calculated based on diffusion tensor imaging [DTI]) are first calculated based on the FCT and then summarized along each of the major WM fiber tracts connecting each pair of the brain GM regions. This could capture the functional information in the WM, in a similar network structure as the FC network constructed for the GM, based only on the same RS-fMRI data. Moreover, a sliding window approach is further used to partition the voxel-wise BOLD signal into multiple short overlapping segments. Then, both the FC and FCT between each pair of the brain regions can be calculated based on the BOLD signal segments in the GM and WM, respectively. In such a way, our method can generate dynamic FC and dynamic FCT to better capture functional information in both GM and WM and further integrate them together by using our developed feature extraction, selection, and ensemble learning algorithms. The experimental results verify that the dynamic FCT can provide valuable functional information in the WM; by combining it with the dynamic FC in the GM, the diagnosis accuracy for MCI subjects can be significantly improved even using RS-fMRI data alone. Hum Brain Mapp 38:5019-5034, 2017. © 2017 Wiley Periodicals, Inc.

Deep ensemble learning of sparse regression models for brain disease diagnosis

Recent studies on brain imaging analysis witnessed the core roles of machine learning techniques in computer-assisted intervention for brain disease diagnosis. Of various machine-learning techniques, sparse regression models have proved their effectiveness in handling high-dimensional data but with a small number of training samples, especially in medical problems. In the meantime, deep learning methods have been making great successes by outperforming the state-of-the-art performances in various applications. In this paper, we propose a novel framework that combines the two conceptually different methods of sparse regression and deep learning for Alzheimer's disease/mild cognitive impairment diagnosis and prognosis. Specifically, we first train multiple sparse regression models, each of which is trained with different values of a regularization control parameter. Thus, our multiple sparse regression models potentially select different feature subsets from the original feature set; thereby they have different powers to predict the response values, i.e., clinical label and clinical scores in our work. By regarding the response values from our sparse regression models as target-level representations, we then build a deep convolutional neural network for clinical decision making, which thus we call 'Deep Ensemble Sparse Regression Network.' To our best knowledge, this is the first work that combines sparse regression models with deep neural network. In our experiments with the ADNI cohort, we validated the effectiveness of the proposed method by achieving the highest diagnostic accuracies in three classification tasks. We also rigorously analyzed our results and compared with the previous studies on the ADNI cohort in the literature.

Joint Discriminative and Representative Feature Selection for Alzheimer's Disease Diagnosis

Neuroimaging data have been widely used to derive possible biomarkers for Alzheimer's Disease (AD) diagnosis. As only certain brain regions are related to AD progression, many feature selection methods have been proposed to identify informative features (i.e., brain regions) to build an accurate prediction model. These methods mostly only focus on the feature-target relationship to select features which are discriminative to the targets (e.g., diagnosis labels). However, since the brain regions are anatomically and functionally connected, there could be useful intrinsic relationships among features. In this paper, by utilizing both the feature-target and feature-feature relationships, we propose a novel sparse regression model to select informative features which are discriminative to the targets and also representative to the features. We argue that the features which are representative (i.e., can be used to represent many other features) are important, as they signify strong "connection" with other ROIs, and could be related to the disease progression. We use our model to select features for both binary and multi-class classification tasks, and the experimental results on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset show that the proposed method outperforms other comparison methods considered in this work.

Subspace Regularized Sparse Multitask Learning for Multiclass Neurodegenerative Disease Identification

The high feature-dimension and low sample-size problem is one of the major challenges in the study of computer-aided Alzheimer's disease (AD) diagnosis. To circumvent this problem, feature selection and subspace learning have been playing core roles in the literature. Generally, feature selection methods are preferable in clinical applications due to their ease for interpretation, but subspace learning methods can usually achieve more promising results. In this paper, we combine two different methodological approaches to discriminative feature selection in a unified framework. Specifically, we utilize two subspace learning methods, namely, linear discriminant analysis and locality preserving projection, which have proven their effectiveness in a variety of fields, to select class-discriminative and noise-resistant features. Unlike previous methods in neuroimaging studies that mostly focused on a binary classification, the proposed feature selection method is further applicable for multiclass classification in AD diagnosis. Extensive experiments on the Alzheimer's disease neuroimaging initiative dataset showed the effectiveness of the proposed method over other state-of-the-art methods.

Diagnosis of Autism Spectrum Disorders Using Temporally Distinct Resting-State Functional Connectivity Networks

INTRODUCTION

Resting-state functional magnetic resonance imaging (R-fMRI) is dynamic in nature as neural activities constantly change over the time and are dominated by repeating brief activations and deactivations involving many brain regions. Each region participates in multiple brain functions and is part of various functionally distinct but spatially overlapping networks. Functional connectivity computed as correlations over the entire time series always overlooks interregion interactions that often occur repeatedly and dynamically in time, limiting its application to disease diagnosis.

AIMS

We develop a novel framework that uses short-time activation patterns of brain connectivity to better detect subtle disease-induced disruptions of brain connectivity. A clustering algorithm is first used to temporally decompose R-fMRI time series into distinct clusters with similar spatial distribution of neural activity based on the assumption that functionally distinct networks should be largely temporally distinct as brain states do not simultaneously coexist in general. A Pearson correlation-based functional connectivity network is then constructed for each cluster to allow for better exploration of spatiotemporal dynamics of individual neural activity. To reduce significant intersubject variability and to remove possible spurious connections, we use a group-constrained sparse regression model to construct a backbone sparse network for each cluster and use it to weight the corresponding Pearson correlation network.

RESULTS

The proposed method outperforms the conventional static, temporally dependent fully connected correlation-based networks by at least 7% on a publicly available autism dataset. We were able to reproduce similar results using data from other centers.

CONCLUSIONS

By combining the advantages of temporal independence and group-constrained sparse regression, our method improves autism diagnosis.

State-space model with deep learning for functional dynamics estimation in resting-state fMRI

Studies on resting-state functional Magnetic Resonance Imaging (rs-fMRI) have shown that different brain regions still actively interact with each other while a subject is at rest, and such functional interaction is not stationary but changes over time. In terms of a large-scale brain network, in this paper, we focus on time-varying patterns of functional networks, i.e., functional dynamics, inherent in rs-fMRI, which is one of the emerging issues along with the network modelling. Specifically, we propose a novel methodological architecture that combines deep learning and state-space modelling, and apply it to rs-fMRI based Mild Cognitive Impairment (MCI) diagnosis. We first devise a Deep Auto-Encoder (DAE) to discover hierarchical non-linear functional relations among regions, by which we transform the regional features into an embedding space, whose bases are complex functional networks. Given the embedded functional features, we then use a Hidden Markov Model (HMM) to estimate dynamic characteristics of functional networks inherent in rs-fMRI via internal states, which are unobservable but can be inferred from observations statistically. By building a generative model with an HMM, we estimate the likelihood of the input features of rs-fMRI as belonging to the corresponding status, i.e., MCI or normal healthy control, based on which we identify the clinical label of a testing subject. In order to validate the effectiveness of the proposed method, we performed experiments on two different datasets and compared with state-of-the-art methods in the literature. We also analyzed the functional networks learned by DAE, estimated the functional connectivities by decoding hidden states in HMM, and investigated the estimated functional connectivities by means of a graph-theoretic approach.

A novel relational regularization feature selection method for joint regression and classification in AD diagnosis

In this paper, we focus on joint regression and classification for Alzheimer's disease diagnosis and propose a new feature selection method by embedding the relational information inherent in the observations into a sparse multi-task learning framework. Specifically, the relational information includes three kinds of relationships (such as feature-feature relation, response-response relation, and sample-sample relation), for preserving three kinds of the similarity, such as for the features, the response variables, and the samples, respectively. To conduct feature selection, we first formulate the objective function by imposing these three relational characteristics along with an ℓ_2,1-norm regularization term, and further propose a computationally efficient algorithm to optimize the proposed objective function. With the dimension-reduced data, we train two support vector regression models to predict the clinical scores of ADAS-Cog and MMSE, respectively, and also a support vector classification model to determine the clinical label. We conducted extensive experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset to validate the effectiveness of the proposed method. Our experimental results showed the efficacy of the proposed method in enhancing the performances of both clinical scores prediction and disease status identification, compared to the state-of-the-art methods.

A Hybrid of Deep Network and Hidden Markov Model for MCI Identification with Resting-State fMRI

In this paper, we propose a novel method for modelling functional dynamics in resting-state fMRI (rs-fMRI) for Mild Cognitive Impairment (MCI) identification. Specifically, we devise a hybrid architecture by combining Deep Auto-Encoder (DAE) and Hidden Markov Model (HMM). The roles of DAE and HMM are, respectively, to discover hierarchical non-linear relations among features, by which we transform the original features into a lower dimension space, and to model dynamic characteristics inherent in rs-fMRI, i.e., internal state changes. By building a generative model with HMMs for each class individually, we estimate the data likelihood of a test subject as MCI or normal healthy control, based on which we identify the clinical label. In our experiments, we achieved the maximal accuracy of 81.08% with the proposed method, outperforming state-of-the-art methods in the literature.

A novel matrix-similarity based loss function for joint regression and classification in AD diagnosis

Recent studies on AD/MCI diagnosis have shown that the tasks of identifying brain disease and predicting clinical scores are highly related to each other. Furthermore, it has been shown that feature selection with a manifold learning or a sparse model can handle the problems of high feature dimensionality and small sample size. However, the tasks of clinical score regression and clinical label classification were often conducted separately in the previous studies. Regarding the feature selection, to our best knowledge, most of the previous work considered a loss function defined as an element-wise difference between the target values and the predicted ones. In this paper, we consider the problems of joint regression and classification for AD/MCI diagnosis and propose a novel matrix-similarity based loss function that uses high-level information inherent in the target response matrix and imposes the information to be preserved in the predicted response matrix. The newly devised loss function is combined with a group lasso method for joint feature selection across tasks, i.e., predictions of clinical scores and a class label. In order to validate the effectiveness of the proposed method, we conducted experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, and showed that the newly devised loss function helped enhance the performances of both clinical score prediction and disease status identification, outperforming the state-of-the-art methods.

Matrix-Similarity Based Loss Function and Feature Selection for Alzheimer's Disease Diagnosis

Recent studies on Alzheimer's Disease (AD) or its prodromal stage, Mild Cognitive Impairment (MCI), diagnosis presented that the tasks of identifying brain disease status and predicting clinical scores based on neuroimaging features were highly related to each other. However, these tasks were often conducted independently in the previous studies. Regarding the feature selection, to our best knowledge, most of the previous work considered a loss function defined as an element-wise difference between the target values and the predicted ones. In this paper, we consider the problems of joint regression and classification for AD/MCI diagnosis and propose a novel matrix-similarity based loss function that uses high-level information inherent in the target response matrix and imposes the information to be preserved in the predicted response matrix. The newly devised loss function is combined with a group lasso method for joint feature selection across tasks, i.e., clinical scores prediction and disease status identification. We conducted experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, and showed that the newly devised loss function was effective to enhance the performances of both clinical score prediction and disease status identification, outperforming the state-of-the-art methods.