Voxelwise meta-analysis of gray matter anomalies in Alzheimer's disease and mild cognitive impairment using anatomic likelihood estimation

Different patterns of brain activation in normal aging and Alzheimer's disease from cognitional sight: meta analysis using activation likelihood estimation

Alzheimer disease (AD) is a chronic neurological disease, frequently affecting cognitional functions. Recently, a large body of neuro-imaging studies have aimed at finding reliable biomarkers of AD for early diagnosis of disease in contrast with healthy elderlies. We intended to have a meta-analytical study on recent functional neuroimaging studies to find the relationship between cognition in AD patients and normal elderlies. A systematic search was conducted to collect functional neuroimaging studies such as positron emission therapy (PET) and functional magnetic resonance imaging (fMRI) in AD patients and healthy elderlies. The coordinates of regions related to cognition were meta-analyzed using the activation likelihood estimation (ALE) method and Sleuth software. P-value map at the false discovery rate (FDR) of P<0.05 thresholds and the clusters with a minimum size of 200 mm(3) were considered. Data were visualized with MANGO software. Forty-one articles that explored the areas activated during cognition in normal elderly subjects and AD patients were found. According to the findings, left middle frontal gyrus and left precuneus are the most activated areas in cognitional tasks in healthy elderlies and AD patients respectively. In normal elderly subjects and AD patients, comparison of ALE maps and reverse contrast showed that insula and left precuneus were the most activated areas in cognitional aspects respectively. With respect to unification of left precuneus activation in cognitional tasks, it seems that this point can be a hallmark in primary differentiation of AD and healthy individuals.

Neurodegenerative disease diagnosis using incomplete multi-modality data via matrix shrinkage and completion

In this work, we are interested in predicting the diagnostic statuses of potentially neurodegenerated patients using feature values derived from multi-modality neuroimaging data and biological data, which might be incomplete. Collecting the feature values into a matrix, with each row containing a feature vector of a sample, we propose a framework to predict the corresponding associated multiple target outputs (e.g., diagnosis label and clinical scores) from this feature matrix by performing matrix shrinkage following matrix completion. Specifically, we first combine the feature and target output matrices into a large matrix and then partition this large incomplete matrix into smaller submatrices, each consisting of samples with complete feature values (corresponding to a certain combination of modalities) and target outputs. Treating each target output as the outcome of a prediction task, we apply a 2-step multi-task learning algorithm to select the most discriminative features and samples in each submatrix. Features and samples that are not selected in any of the submatrices are discarded, resulting in a shrunk version of the original large matrix. The missing feature values and unknown target outputs of the shrunk matrix is then completed simultaneously. Experimental results using the ADNI dataset indicate that our proposed framework achieves higher classification accuracy at a greater speed when compared with conventional imputation-based classification methods and also yields competitive performance when compared with the state-of-the-art methods.

Grey-matter volume as a potential feature for the classification of Alzheimer's disease and mild cognitive impairment: an exploratory study

Specific patterns of brain atrophy may be helpful in the diagnosis of Alzheimer's disease (AD). In the present study, we set out to evaluate the utility of grey-matter volume in the classification of AD and amnestic mild cognitive impairment (aMCI) compared to normal control (NC) individuals. Voxel-based morphometric analyses were performed on structural MRIs from 35 AD patients, 27 aMCI patients, and 27 NC participants. A two-sample two-tailed t-test was computed between the NC and AD groups to create a map of abnormal grey matter in AD. The brain areas with significant differences were extracted as regions of interest (ROIs), and the grey-matter volumes in the ROIs of the aMCI patients were included to evaluate the patterns of change across different disease severities. Next, correlation analyses between the grey-matter volumes in the ROIs and all clinical variables were performed in aMCI and AD patients to determine whether they varied with disease progression. The results revealed significantly decreased grey matter in the bilateral hippocampus/parahippocampus, the bilateral superior/middle temporal gyri, and the right precuneus in AD patients. The grey-matter volumes were positively correlated with clinical variables. Finally, we performed exploratory linear discriminative analyses to assess the classifying capacity of grey-matter volumes in the bilateral hippocampus and parahippocampus among AD, aMCI, and NC. Leave-one-out cross-validation analyses demonstrated that grey-matter volumes in hippocampus and parahippocampus accurately distinguished AD from NC. These findings indicate that grey-matter volumes are useful in the classification of AD.

Altered causal connectivity of resting state brain networks in amnesic MCI

Most neuroimaging studies of resting state networks in amnesic mild cognitive impairment (aMCI) have concentrated on functional connectivity (FC) based on instantaneous correlation in a single network. The purpose of the current study was to investigate effective connectivity in aMCI patients based on Granger causality of four important networks at resting state derived from functional magnetic resonance imaging data--default mode network (DMN), hippocampal cortical memory network (HCMN), dorsal attention network (DAN) and fronto-parietal control network (FPCN). Structural and functional MRI data were collected from 16 aMCI patients and 16 age, gender-matched healthy controls. Correlation-purged Granger causality analysis was used, taking gray matter atrophy as covariates, to compare the group difference between aMCI patients and healthy controls. We found that the causal connectivity between networks in aMCI patients was significantly altered with both increases and decreases in the aMCI group as compared to healthy controls. Some alterations were significantly correlated with the disease severity as measured by mini-mental state examination (MMSE), and California verbal learning test (CVLT) scores. When the whole-brain signal averaged over the entire brain was used as a nuisance co-variate, the within-group maps were significantly altered while the between-group difference maps did not. These results suggest that the alterations in causal influences may be one of the possible underlying substrates of cognitive impairments in aMCI. The present study extends and complements previous FC studies and demonstrates the coexistence of causal disconnection and compensation in aMCI patients, and thus might provide insights into biological mechanism of the disease.

Reliability measures of functional magnetic resonance imaging in a longitudinal evaluation of mild cognitive impairment

As the aging population grows, it has become increasingly important to carefully characterize amnestic mild cognitive impairment (aMCI), a preclinical stage of Alzheimer's disease (AD). Functional magnetic resonance imaging (fMRI) is a valuable tool for monitoring disease progression in selectively vulnerable brain regions associated with AD neuropathology. However, the reliability of fMRI data in longitudinal studies of older adults with aMCI is largely unexplored. To address this, aMCI participants completed two visual working tasks, a Delayed-Recognition task and a One-Back task, on three separate scanning sessions over a three-month period. Test-retest reliability of the fMRI blood oxygen level dependent (BOLD) activity was assessed using an intraclass correlation (ICC) analysis approach. Results indicated that brain regions engaged during the task displayed greater reliability across sessions compared to regions that were not utilized by the task. During task-engagement, differential reliability scores were observed across the brain such that the frontal lobe, medial temporal lobe, and subcortical structures exhibited fair to moderate reliability (ICC=0.3-0.6), while temporal, parietal, and occipital regions exhibited moderate to good reliability (ICC=0.4-0.7). Additionally, reliability across brain regions was more stable when three fMRI sessions were used in the ICC calculation relative to two fMRI sessions. In conclusion, the fMRI BOLD signal is reliable across scanning sessions in this population and thus a useful tool for tracking longitudinal change in observational and interventional studies in aMCI.

Inter-modality relationship constrained multi-modality multi-task feature selection for Alzheimer's Disease and mild cognitive impairment identification

Previous studies have demonstrated that the use of integrated information from multi-modalities could significantly improve diagnosis of Alzheimer's Disease (AD). However, feature selection, which is one of the most important steps in classification, is typically performed separately for each modality, which ignores the potentially strong inter-modality relationship within each subject. Recent emergence of multi-task learning approach makes the joint feature selection from different modalities possible. However, joint feature selection may unfortunately overlook different yet complementary information conveyed by different modalities. We propose a novel multi-task feature selection method to preserve the complementary inter-modality information. Specifically, we treat feature selection from each modality as a separate task and further impose a constraint for preserving the inter-modality relationship, besides separately enforcing the sparseness of the selected features from each modality. After feature selection, a multi-kernel support vector machine (SVM) is further used to integrate the selected features from each modality for classification. Our method is evaluated using the baseline PET and MRI images of subjects obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. Our method achieves a good performance, with an accuracy of 94.37% and an area under the ROC curve (AUC) of 0.9724 for AD identification, and also an accuracy of 78.80% and an AUC of 0.8284 for mild cognitive impairment (MCI) identification. Moreover, the proposed method achieves an accuracy of 67.83% and an AUC of 0.6957 for separating between MCI converters and MCI non-converters (to AD). These performances demonstrate the superiority of the proposed method over the state-of-the-art classification methods.

Simultaneous EEG-fMRI acquisition at low, high and ultra-high magnetic fields up to 9.4 T: perspectives and challenges

In this perspectives article we highlight the advantages of simultaneous acquisition of electroencephalography (EEG) and functional magnetic resonance imaging (fMRI). As MRI moves towards using ultra-high magnetic fields in the quest for increased signal-to-noise, the question arises whether combined EEG-fMRI measurements are feasible at magnetic fields of 7 T and higher. We describe the challenges of MRI-EEG at 1.5, 3, 7 and 9.4 T and review the proposed solutions. In an outlook, we discuss further developments such as simultaneous trimodal imaging using MR, positron emission tomography (PET) and EEG under the same physiological conditions in the same subject.