Baseline and longitudinal patterns of brain atrophy in MCI patients, and their use in prediction of short-term conversion to AD: results from ADNI

Optimal Sparse Linear Prediction for Block-missing Multi-modality Data without Imputation

In modern scientific research, data are often collected from multiple modalities. Since different modalities could provide complementary information, statistical prediction methods using multi-modality data could deliver better prediction performance than using single modality data. However, one special challenge for using multi-modality data is related to block-missing data. In practice, due to dropouts or the high cost of measures, the observations of a certain modality can be missing completely for some subjects. In this paper, we propose a new DIrect Sparse regression procedure using COvariance from Multi-modality data (DISCOM). Our proposed DISCOM method includes two steps to find the optimal linear prediction of a continuous response variable using block-missing multi-modality predictors. In the first step, rather than deleting or imputing missing data, we make use of all available information to estimate the covariance matrix of the predictors and the cross-covariance vector between the predictors and the response variable. The proposed new estimate of the covariance matrix is a linear combination of the identity matrix, the estimates of the intra-modality covariance matrix and the cross-modality covariance matrix. Flexible estimates for both the sub-Gaussian and heavy-tailed cases are considered. In the second step, based on the estimated covariance matrix and the estimated cross-covariance vector, an extended Lasso-type estimator is used to deliver a sparse estimate of the coefficients in the optimal linear prediction. The number of samples that are effectively used by DISCOM is the minimum number of samples with available observations from two modalities, which can be much larger than the number of samples with complete observations from all modalities. The effectiveness of the proposed method is demonstrated by theoretical studies, simulated examples, and a real application from the Alzheimer's Disease Neuroimaging Initiative. The comparison between DISCOM and some existing methods also indicates the advantages of our proposed method.

Protein-based amide proton transfer-weighted MR imaging of amnestic mild cognitive impairment

Amide proton transfer-weighted (APTw) MRI is a novel molecular imaging technique that can noninvasively detect endogenous cellular proteins and peptides in tissue. Here, we demonstrate the feasibility of protein-based APTw MRI in characterizing amnestic mild cognitive impairment (aMCI). Eighteen patients with confirmed aMCI and 18 matched normal controls were scanned at 3 Tesla. The APTw, as well as conventional magnetization transfer ratio (MTR), signal differences between aMCI and normal groups were assessed by the independent samples t-test, and the receiver-operator-characteristic analysis was used to assess the diagnostic performance of APTw. When comparing the normal control group, aMCI brains typically had relatively higher APTw signals. Quantitatively, APTw intensity values were significantly higher in nine of 12 regions of interest in aMCI patients than in normal controls. The largest areas under the receiver-operator-characteristic curves were 0.88 (gray matter in occipital lobe) and 0.82 (gray matter in temporal lobe, white matter in occipital lobe) in diagnosing aMCI patients. On the contrary, MTR intensity values were significantly higher in only three of 12 regions of interest in the aMCI group. Additionally, the age dependency analyses revealed that these cross-sectional APTw/MTR signals had an increasing trend with age in most brain regions for normal controls, but a decreasing trend with age in most brain regions for aMCI patients. Our early results show the potential of the APTw signal as a new imaging biomarker for the noninvasive molecular diagnosis of aMCI.

Altered Directed Functional Connectivity of the Hippocampus in Mild Cognitive Impairment and Alzheimer's Disease: A Resting-State fMRI Study

The hippocampus is generally reported as one of the regions most impacted by Alzheimer's disease (AD) and is closely associated with memory function and orientation. Undirected functional connectivity (FC) alterations occur in patients with mild cognitive impairment (MCI) and AD, and these alterations have been the subject of many studies. However, abnormal patterns of directed FC remain poorly understood. In this study, to identify changes in directed FC between the hippocampus and other brain regions, Granger causality analysis (GCA) based on voxels was applied to resting-state functional magnetic resonance imaging (rs-fMRI) data from 29 AD, 65 MCI, and 30 normal control (NC) subjects. The results showed significant differences in the patterns of directed FC among the three groups. There were fewer brain regions showing changes in directed FC with the hippocampus in the MCI group than the NC group, and these regions were mainly located in the temporal lobe, frontal lobe, and cingulate cortex. However, regarding the abnormalities in directed FC in the AD group, the number of affected voxels was greater, the size of the clusters was larger, and the distribution was wider. Most of the abnormal connections were unidirectional and showed hemispheric asymmetry. In addition, we also investigated the correlations between the abnormal directional FCs and cognitive and clinical measurement scores in the three groups and found that some of them were significantly correlated. This study revealed abnormalities in the transmission and reception of information in the hippocampus of MCI and AD patients and offer insight into the neurophysiological mechanisms underlying MCI and AD.

Differentiating Between Healthy Control Participants and Those with Mild Cognitive Impairment Using Volumetric MRI Data

OBJECTIVE

To determine whether volumetric measures of the hippocampus, entorhinal cortex, and other cortical measures can differentiate between cognitively normal individuals and subjects with mild cognitive impairment (MCI).

METHOD

Magnetic resonance imaging (MRI) data from 46 cognitively normal subjects and 50 subjects with MCI as part of the Boston University Alzheimer's Disease Center research registry and the Alzheimer's Disease Neuroimaging Initiative were used in this cross-sectional study. Cortical, subcortical, and hippocampal subfield volumes were generated from each subject's MRI data using FreeSurfer v6.0. Nominal logistic regression models containing these variables were used to identify subjects as control or MCI.

RESULTS

A model containing regions of interest (superior temporal cortex, caudal anterior cingulate, pars opercularis, subiculum, precentral cortex, caudal middle frontal cortex, rostral middle frontal cortex, pars orbitalis, middle temporal cortex, insula, banks of the superior temporal sulcus, parasubiculum, paracentral lobule) fit the data best (R2 = .7310, whole model test chi-square = 97.16, p < .0001).

CONCLUSIONS

MRI data correctly classified most subjects using measures of selected medial temporal lobe structures in combination with those from other cortical areas, yielding an overall classification accuracy of 93.75%. These findings support the notion that, while volumes of medial temporal lobe regions differ between cognitively normal and MCI subjects, differences that can be used to distinguish between these two populations are present elsewhere in the brain.

Identification of the Early Stage of Alzheimer's Disease Using Structural MRI and Resting-State fMRI

Accurate prediction of the early stage of Alzheimer's disease (AD) is important but very challenging. The goal of this study was to utilize predictors for diagnosis conversion to AD based on integrating resting-state functional MRI (rs-fMRI) connectivity analysis and structural MRI (sMRI). We included 177 subjects in this study and aimed at identifying patients with mild cognitive impairment (MCI) who progress to AD, MCI converter (MCI-C), patients with MCI who do not progress to AD, MCI non-converter (MCI-NC), patients with AD, and healthy controls (HC). The graph theory was used to characterize different aspects of the rs-fMRI brain network by calculating measures of integration and segregation. The cortical and subcortical measurements, e.g., cortical thickness, were extracted from sMRI data. The rs-fMRI graph measures were combined with the sMRI measures to construct input features of a support vector machine (SVM) and classify different groups of subjects. Two feature selection algorithms [i.e., the discriminant correlation analysis (DCA) and sequential feature collection (SFC)] were used for feature reduction and selecting a subset of optimal features. Maximum accuracy of 67 and 56% for three-group (“AD, MCI-C, and MCI-NC” or “MCI-C, MCI-NC, and HC”) and four-group (“AD, MCI-C, MCI-NC, and HC”) classification, respectively, were obtained with the SFC feature selection algorithm. We also identified hub nodes in the rs-fMRI brain network which were associated with the early stage of AD. Our results demonstrated the potential of the proposed method based on integration of the functional and structural MRI for identification of the early stage of AD.

Risk estimation before progression to mild cognitive impairment and Alzheimer's disease: an AD resemblance atrophy index

To realize an individual-level risk evaluation of progression of early Alzheimer’s disease (AD), we applied an AD resemblance atrophy index (AD-RAI) to differentiate the subjects at risk of progression from normal subjects (NC) to mild cognitive impairment (MCI) and from MCI to AD. We included 183 subjects with a two-year follow-up: 50 NC stable (NCs), 23 NC-to-MCI converters (NCc), 50 MCI stable (MCIs), 35 MCI-to-AD converters (MCIc), 25 AD stable (ADs). ANCOVA analyses were used to identify baseline brain atrophy in converters compared with non-converters. To explore the relative merits of AD-RAI over individual regional volumetric measures in prediction of disease progression, we searched for the optimal cutoff for each measure in logistic regressions and plotted the longitudinal trajectories of these brain volumetric measures in converters and non-converters. Baseline AD-RAI performed the best in differentiating NCc from NCs (odds ratio 26.35, AUC 0.740) and MCIc from MCIs (odds ratio 8.91, AUC 0.771). The AD-RAI presented greater increase in the second year for NCc vs. NCs but not for MCIc vs. MCIs. Baseline AD-RAIs were also associated with CSF-based and PET-based AD biomarkers. These results showed the potential of AD-RAI in early risk estimation before progression to MCI/AD at an individual-level.

Visualizing Alzheimer's disease progression in low dimensional manifolds

While tomographic neuroimaging data is information rich, objective, and with high sensitivity in the study of brain diseases such as Alzheimer's disease (AD), its direct use in clinical practice and in regulated clinical trial (CT) still has many challenges. Taking CT as an example, unless the relevant policy and the perception of the primary outcome measures change, the need to construct univariate indices (out of the 3-D imaging data) to serve as CT's primary outcome measures will remain the focus of active research. More relevant to this current study, an overall global index that summarizes multiple complicated features from neuroimages should be developed in order to provide high diagnostic accuracy and sensitivity in tracking AD progression over time in clinical setting. Such index should also be practically intuitive and logically explainable to patients and their families. In this research, we propose a new visualization tool, derived from the manifold-based nonlinear dimension reduction of brain MRI features, to track AD progression over time. In specific, we investigate the locally linear embedding (LLE) method using a dataset from Alzheimer's Disease Neuroimaging Initiative (ADNI), which includes the longitudinal MRIs from 562 subjects. About 20% of them progressed to the next stage of dementia. Using only the baseline data of cognitively unimpaired (CU) and AD subjects, LLE reduces the feature dimension to two and a subject's AD progression path can be plotted in this low dimensional LLE feature space. In addition, the likelihood of being categorized to AD is indicated by color. This LLE map is a new data visualization tool that can assist in tracking AD progression over time.

Visualizing modules of coordinated structural brain atrophy during the course of conversion to Alzheimer's disease by applying methodology from gene co-expression analysis

Objective

We aimed to identify modularized structural atrophy of brain regions with a high degree of connectivity and its longitudinal changes associated with the progression of Alzheimer's disease (AD) using weighted gene co-expression network analysis (WGCNA), which is an unsupervised hierarchical clustering method originally used in genetic analysis.

Methods

We included participants with late mild cognitive impairment (MCI) at baseline from the Japanese Alzheimer's Disease Neuroimaging Initiative (J-ADNI) study. We imputed normalized and Z-transformed structural volume or cortical thickness data of 164 parcellated brain regions/structures based on the calculations of the FreeSurfer software. We applied the WGCNA to extract modules with highly interconnected structural atrophic patterns and examined the correlation between the identified modules and clinical AD progression.

Results

We included 204 participants from the baseline dataset, and performed a follow-up with 100 in the 36-month dataset of MCI cohort participants from the J-ADNI. In the univariate correlation or variable importance analysis, baseline atrophy in temporal lobe regions/structures significantly predicted clinical AD progression. In the WGCNA consensus analysis, co-atrophy modules associated with MCI conversion were first distributed in the temporal lobe and subsequently extended to adjacent parietal cortical regions in the following 36 months.

Conclusions

We identified coordinated modules of brain atrophy and demonstrated their longitudinal extension along with the clinical course of AD progression using WGCNA, which showed a good correspondence with previous pathological studies of the tau propagation theory. Our results suggest the potential applicability of this methodology, originating from genetic analyses, for the surrogate visualization of the underlying pathological progression in neurodegenerative diseases not limited to AD.

•

Applying origin-of-genetic unsupervised clustering to structural MRI from J-ADNI.

•

We identified highly-interconnected modules associated with AD progression.

•

Extension of co-atrophy modules corresponded with tau propagation theory in AD.

•

This method has potential for visualizing underlying pathological progression.

•

Application to other neurodegenerative diseases not limited to AD is expected.

EARLY PREDICTION OF ALZHEIMER'S DISEASE DEMENTIA BASED ON BASELINE HIPPOCAMPAL MRI AND 1-YEAR FOLLOW-UP COGNITIVE MEASURES USING DEEP RECURRENT NEURAL NETWORKS

Multi-modal biological, imaging, and neuropsychological markers have demonstrated promising performance for distinguishing Alzheimer's disease (AD) patients from cognitively normal elders. However, it remains difficult to early predict when and which mild cognitive impairment (MCI) individuals will convert to AD dementia. Informed by pattern classification studies which have demonstrated that pattern classifiers built on longitudinal data could achieve better classification performance than those built on cross-sectional data, we develop a deep learning model based on recurrent neural networks (RNNs) to learn informative representation and temporal dynamics of longitudinal cognitive measures of individual subjects and combine them with baseline hippocampal MRI for building a prognostic model of AD dementia progression. Experimental results on a large cohort of MCI subjects have demonstrated that the deep learning model could learn informative measures from longitudinal data for characterizing the progression of MCI subjects to AD dementia, and the prognostic model could early predict AD progression with high accuracy.

Detection of mild cognitive impairment in a community-dwelling population using quantitative, multiparametric MRI-based classification

Early and accurate mild cognitive impairment (MCI) detection within a heterogeneous, nonclinical population is needed to improve care for persons at risk of developing dementia. Magnetic resonance imaging (MRI)-based classification may aid early diagnosis of MCI, but has only been applied within clinical cohorts. We aimed to determine the generalizability of MRI-based classification probability scores to detect MCI on an individual basis within a general population. To determine classification probability scores, an AD, mild-AD, and moderate-AD detection model were created with anatomical and diffusion MRI measures calculated from a clinical Alzheimer's Disease (AD) cohort and subsequently applied to a population-based cohort with 48 MCI and 617 normal aging subjects. Each model's ability to detect MCI was quantified using area under the receiver operating characteristic curve (AUC) and compared with an MCI detection model trained and applied to the population-based cohort. The AD-model and mild-AD identified MCI from controls better than chance level (AUC = 0.600, p = 0.025; AUC = 0.619, p = 0.008). In contrast, the moderate-AD-model was not able to separate MCI from normal aging (AUC = 0.567, p = 0.147). The MCI-model was able to separate MCI from controls better than chance (p = 0.014) with mean AUC values comparable with the AD-model (AUC = 0.611, p = 1.0). Within our population-based cohort, classification models detected MCI better than chance. Nevertheless, classification performance rates were moderate and may be insufficient to facilitate robust MRI-based MCI detection on an individual basis. Our data indicate that multiparametric MRI-based classification algorithms, that are effective in clinical cohorts, may not straightforwardly translate to applications in a general population.

A deep learning model for early prediction of Alzheimer's disease dementia based on hippocampal magnetic resonance imaging data

INTRODUCTION

It is challenging at baseline to predict when and which individuals who meet criteria for mild cognitive impairment (MCI) will ultimately progress to Alzheimer's disease (AD) dementia.

METHODS

A deep learning method is developed and validated based on magnetic resonance imaging scans of 2146 subjects (803 for training and 1343 for validation) to predict MCI subjects' progression to AD dementia in a time-to-event analysis setting.

RESULTS

The deep-learning time-to-event model predicted individual subjects' progression to AD dementia with a concordance index of 0.762 on 439 Alzheimer's Disease Neuroimaging Initiative testing MCI subjects with follow-up duration from 6 to 78 months (quartiles: [24, 42, 54]) and a concordance index of 0.781 on 40 Australian Imaging Biomarkers and Lifestyle Study of Aging testing MCI subjects with follow-up duration from 18 to 54 months (quartiles: [18, 36, 54]). The predicted progression risk also clustered individual subjects into subgroups with significant differences in their progression time to AD dementia (P < .0002). Improved performance for predicting progression to AD dementia (concordance index = 0.864) was obtained when the deep learning-based progression risk was combined with baseline clinical measures.

DISCUSSION

Our method provides a cost effective and accurate means for prognosis and potentially to facilitate enrollment in clinical trials with individuals likely to progress within a specific temporal period.

Longitudinal score prediction for Alzheimer's disease based on ensemble correntropy and spatial-temporal constraint

Neuroimaging data has been widely used to predict clinical scores for automatic diagnosis of Alzheimer's disease (AD). For accurate clinical score prediction, one of the major challenges is high feature dimension of the imaging data. To address this issue, this paper presents an effective framework using a novel feature selection model via sparse learning. In contrast to previous approaches focusing on a single time point, this framework uses information at multiple time points. Specifically, a regularized correntropy with the spatial-temporal constraint is used to reduce the adverse effect of noise and outliers, and promote consistent and robust selection of features by exploring data characteristics. Furthermore, ensemble learning of support vector regression (SVR) is exploited to accurately predict AD scores based on the selected features. The proposed approach is extensively evaluated on the Alzheimer's disease neuroimaging initiative (ADNI) dataset. Our experiments demonstrate that the proposed approach not only achieves promising regression accuracy, but also successfully recognizes disease-related biomarkers.

Spatial Topography of Individual-Specific Cortical Networks Predicts Human Cognition, Personality, and Emotion

Resting-state functional magnetic resonance imaging (rs-fMRI) offers the opportunity to delineate individual-specific brain networks. A major question is whether individual-specific network topography (i.e., location and spatial arrangement) is behaviorally relevant. Here, we propose a multi-session hierarchical Bayesian model (MS-HBM) for estimating individual-specific cortical networks and investigate whether individual-specific network topography can predict human behavior. The multiple layers of the MS-HBM explicitly differentiate intra-subject (within-subject) from inter-subject (between-subject) network variability. By ignoring intra-subject variability, previous network mappings might confuse intra-subject variability for inter-subject differences. Compared with other approaches, MS-HBM parcellations generalized better to new rs-fMRI and task-fMRI data from the same subjects. More specifically, MS-HBM parcellations estimated from a single rs-fMRI session (10 min) showed comparable generalizability as parcellations estimated by 2 state-of-the-art methods using 5 sessions (50 min). We also showed that behavioral phenotypes across cognition, personality, and emotion could be predicted by individual-specific network topography with modest accuracy, comparable to previous reports predicting phenotypes based on connectivity strength. Network topography estimated by MS-HBM was more effective for behavioral prediction than network size, as well as network topography estimated by other parcellation approaches. Thus, similar to connectivity strength, individual-specific network topography might also serve as a fingerprint of human behavior.

Differential medial temporal lobe and default-mode network functional connectivity and morphometric changes in Alzheimer's disease

We report group level differential detection of medial temporal lobe resting-state functional connectivity disruption and morphometric changes in the transition from cognitively normal to early mild cognitive impairment in an age-, education- and gender-matched 105 subjects Alzheimer's Disease Neuroimaging Initiative dataset. In mild Alzheimer's Disease, but not early mild cognitive impairment, characteristic brain atrophy was detected in FreeSurfer estimates of subcortical and hippocampal subfield volumes and cortical thinning. By contrast, functional connectivity analysis detected earlier significant changes. In early mild cognitive impairment these changes involved medial temporal lobe regions of transentorhinal, perirhinal and entorhinal cortices (associated with the earliest stages of neurofibrillary changes in Alzheimer's Disease), hippocampus, parahippocampal gyrus and temporal pole, and cortical regions comprising or co-activated with the default-mode network, including rostral and medial prefrontal cortex, anterior cingulate cortex, precuneus and inferior temporal cortex. Key findings include: a) focal, bilaterally symmetric spatial organization of affected medial temporal lobe regions; b) mutual hyperconnectivity involving ventral medial temporal lobe structures (temporal pole, uncus); c) dorsal medial temporal lobe hypoconnectivity with anterior and posterior midline default-mode network nodes; and d) a complex pattern of transient and persistent changes in hypo- and hyper-connectivity across Alzheimer's Disease stages. These findings position medial temporal lobe resting state functional connectivity as a candidate biomarker of an Alzheimer's Disease pathophysiological cascade, potentially in advance of clinical biomarkers, and coincident with biomarkers of the earliest stages of Alzheimer's neuropathology.

Novel Effective Connectivity Inference Using Ultra-Group Constrained Orthogonal Forward Regression and Elastic Multilayer Perceptron Classifier for MCI Identification

Mild cognitive impairment (MCI) detection is important, such that appropriate interventions can be imposed to delay or prevent its progression to severe stages, including Alzheimer's disease (AD). Brain connectivity network inferred from the functional magnetic resonance imaging data has been prevalently used to identify the individuals with MCI/AD from the normal controls. The capability to detect the causal or effective connectivity is highly desirable for understanding directed functional interactions between brain regions and further helping the detection of MCI. In this paper, we proposed a novel sparse constrained effective connectivity inference method and an elastic multilayer perceptron classifier for MCI identification. Specifically, a ultra-group constrained structure detection algorithm is first designed to identify the parsimonious topology of the effective connectivity network, in which the weak derivatives of the observable data are considered. Second, based on the identified topology structure, an effective connectivity network is then constructed by using an ultra-orthogonal forward regression algorithm to minimize the shrinking effect of the group constraint-based method. Finally, the effective connectivity network is validated in MCI identification using an elastic multilayer perceptron classifier, which extracts lower to higher level information from initial input features and hence improves the classification performance. Relatively high classification accuracy is achieved by the proposed method when compared with the state-of-the-art classification methods. Furthermore, the network analysis results demonstrate that MCI patients suffer a rich club effect loss and have decreased connectivity among several brain regions. These findings suggest that the proposed method not only improves the classification performance but also successfully discovers critical disease-related neuroimaging biomarkers.

A highly predictive signature of cognition and brain atrophy for progression to Alzheimer's dementia

BACKGROUND

Clinical trials in Alzheimer's disease need to enroll patients whose cognition will decline over time, if left untreated, in order to demonstrate the efficacy of an intervention. Machine learning models used to screen for patients at risk of progression to dementia should therefore favor specificity (detecting only progressors) over sensitivity (detecting all progressors), especially when the prevalence of progressors is low. Here, we explore whether such high-risk patients can be identified using cognitive assessments and structural neuroimaging by training machine learning tools in a high-specificity regime.

RESULTS

A multimodal signature of Alzheimer's dementia was first extracted from the ADNI1 dataset. We then validated the predictive value of this signature on ADNI1 patients with mild cognitive impairment (N = 235). The signature was optimized to predict progression to dementia over 3 years with low sensitivity (55.1%) but high specificity (95.6%), resulting in only moderate accuracy (69.3%) but high positive predictive value (80.4%, adjusted for a "typical" 33% prevalence rate of true progressors). These results were replicated in ADNI2 (N = 235), with 87.8% adjusted positive predictive value (96.7% specificity, 47.3% sensitivity, 85.1% accuracy).

CONCLUSIONS

We found that cognitive measures alone could identify high-risk individuals, with structural measurements providing a slight improvement. The signature had comparable receiver operating characteristics to standard machine learning tools, yet a marked improvement in positive predictive value was achieved over the literature by selecting a high-specificity operating point. The multimodal signature can be readily applied for the enrichment of clinical trials.

Prediction of Alzheimer's disease dementia with MRI beyond the short-term: Implications for the design of predictive models

Magnetic resonance imaging (MRI) volumetric measures have become a standard tool for the detection of incipient Alzheimer's Disease (AD) dementia in mild cognitive impairment (MCI). Focused on providing an earlier and more accurate diagnosis, sophisticated MRI machine learning algorithms have been developed over the recent years, most of them learning their non-disease patterns from MCI that remained stable over 2-3 years. In this work, we analyzed whether these stable MCI over short-term periods are actually appropriate training examples of non-disease patterns. To this aim, we compared the diagnosis of MCI patients at 2 and 5 years of follow-up and investigated its impact on the predictive performance of baseline volumetric MRI measures primarily involved in AD, i.e., hippocampal and entorhinal cortex volumes. Predictive power was evaluated in terms of the area under the ROC curve (AUC), sensitivity, and specificity in a trial sample of 248 MCI patients followed-up over 5 years. We further compared the sensitivity in those MCI that converted before 2 years and those that converted after 2 years. Our results indicate that 23% of the stable MCI at 2 years progressed in the next three years and that MRI volumetric measures are good predictors of conversion to AD dementia even at the mid-term, showing a better specificity and AUC as follow-up time increases. The combination of hippocampus and entorhinal cortex yielded an AUC that was significantly higher for the 5-year follow-up (AUC = 73% at 2 years vs. AUC = 84% at 5 years), as well as for specificity (56% vs. 71%). Sensitivity showed a non-significant slight decrease (81% vs. 78%). Remarkably, the performance of this model was comparable to machine learning models at the same follow-up times. MRI correctly identified most of the patients that converted after 2 years (with sensitivity >60%), and these patients showed a similar degree of abnormalities to those that converted before 2 years. This implies that most of the MCI patients that remained stable over short periods and subsequently progressed to AD dementia had evident atrophies at baseline. Therefore, machine learning models that use these patients to learn non-disease patterns are including an important fraction of patients with evident pathological changes related to the disease, something that might result in reduced performance and lack of biological interpretability.

EARLY PREDICTION OF ALZHEIMER'S DISEASE DEMENTIA BASED ON BASELINE HIPPOCAMPAL MRI AND 1-YEAR FOLLOW-UP COGNITIVE MEASURES USING DEEP RECURRENT NEURAL NETWORKS

Multi-modal biological, imaging, and neuropsychological markers have demonstrated promising performance for distinguishing Alzheimer's disease (AD) patients from cognitively normal elders. However, it remains difficult to early predict when and which mild cognitive impairment (MCI) individuals will convert to AD dementia. Informed by pattern classification studies which have demonstrated that pattern classifiers built on longitudinal data could achieve better classification performance than those built on cross-sectional data, we develop a deep learning model based on recurrent neural networks (RNNs) to learn informative representation and temporal dynamics of longitudinal cognitive measures of individual subjects and combine them with baseline hippocampal MRI for building a prognostic model of AD dementia progression. Experimental results on a large cohort of MCI subjects have demonstrated that the deep learning model could learn informative measures from longitudinal data for characterizing the progression of MCI subjects to AD dementia, and the prognostic model could early predict AD progression with high accuracy.

The effect of ApoE ε4 on longitudinal brain region-specific glucose metabolism in patients with mild cognitive impairment: a FDG-PET study

While the ApoE ε4 allele is a known risk factor for mild cognitive impairment (MCI) and Alzheimer's disease, brain region specific effects remain elusive. In this study, we investigate whether the ApoE ε4 allele exhibits brain region specific effects in longitudinal glucose uptake among patients with MCI from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Preprocessed FDG PET images, MRIs, and demographic information were downloaded from the ADNI database. An iterative reblurred Van Cittertiteration method was used for partial volume correction (PVC) on all PET images. Structural MRIs were used for PET spatial normalization and region of interest (ROI) definition in standard space. Longitudinal changes in ROI FDG standardized uptake value ratio (SUVR) relative to cerebellum in 24 ApoE ε4 carriers and 24 age-matched ApoE ε4 non-carriers were measured for up to 84-months (median 72 months, SD = 11.2 months) and compared using a generalized linear mixed effects model controlling for gender, education, baseline age, and follow-up period. Additionally, voxelwise analysis was performed by implementing a paired t-test comparing matched baseline and 72 month FDG SUVR images in ApoE carriers and non-carriers separately. Results with PVC were compared with ones from non-PVC based analysis. After applying PVC, the superior fontal, parietal, lateral temporal, medial temporal, caudate, thalamus, and post-cingulate, and amygdala regions had greater longitudinal decreases in FDG uptake in ApoE ε4 carriers with MCI compared to non-carriers with MCI. Similar forebrain and limbic clusters were found through voxelwise analysis. Compared to the PVC based analysis, fewer significant ApoE-associated regions and clusters were found in the non-PVC based PET analysis. Our findings suggest that the ApoE ε4 genotype is associated with a longitudinal decline in glucose uptake in 8 forebrain and limbic brain regions in the context of MCI. In conclusion, this 84-months longitudinal FDG PET study demonstrates a novel ApoE ε4-associated brain-region specific glucose metabolism pattern in patients with MCI. Partial volume correction improved FDG PET quantification.

Group sparse reduced rank regression for neuroimaging genetic study

The neuroimaging genetic study usually needs to deal with high dimensionality of both brain imaging data and genetic data, so that often resulting in the issue of curse of dimensionality. In this paper, we propose a group sparse reduced rank regression model to take the relations of both the phenotypes and the genotypes for the neuroimaging genetic study. Specifically, we propose designing a graph sparsity constraint as well as a reduced rank constraint to simultaneously conduct subspace learning and feature selection. The group sparsity constraint conducts feature selection to identify genotypes highly related to neuroimaging data, while the reduced rank constraint considers the relations among neuroimaging data to conduct subspace learning in the feature selection model. Furthermore, an alternative optimization algorithm is proposed to solve the resulting objective function and is proved to achieve fast convergence. Experimental results on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset showed that the proposed method has superiority on predicting the phenotype data by the genotype data, than the alternative methods under comparison.

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.

Robust multi-label transfer feature learning for early diagnosis of Alzheimer's disease

Transfer learning has been successfully used in the early diagnosis of Alzheimer's disease (AD). In these methods, data from one single or multiple related source domain(s) are employed to aid the learning task in the target domain. However, most of the existing methods utilize data from all source domains, ignoring the fact that unrelated source domains may degrade the learning performance. Also, previous studies assume that class labels for all subjects are reliable, without considering the ambiguity of class labels caused by slight differences between early AD patients and normal control subjects. To address these issues, we propose to transform the original binary class label of a particular subject into a multi-bit label coding vector with the aid of multiple source domains. We further develop a robust multi-label transfer feature learning (rMLTFL) model to simultaneously capture a common set of features from different domains (including the target domain and all source domains) and to identify the unrelated source domains. We evaluate our method on 406 subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database with baseline magnetic resonance imaging (MRI) and cerebrospinal fluid (CSF) data. The experimental results show that the proposed rMLTFL method can effectively improve the performance of AD diagnosis, compared with several state-of-the-art methods.

Data-driven prognostic features of cognitive trajectories in patients with amnestic mild cognitive impairments

BACKGROUND

Although amnestic mild cognitive impairment (aMCI) is generally considered to be a prodromal stage of Alzheimer's disease, patients with aMCI show heterogeneous patterns of progression. Moreover, there are few studies investigating data-driven cognitive trajectory in aMCI. We therefore classified patients with aMCI based on their cognitive trajectory, measured by clinical dementia rating sum of boxes (CDR-SOB). Then, we compared the clinical and neuroimaging features among groups classified by cognitive trajectory.

METHODS

We retrospectively recruited 278 patients with aMCI who underwent three or more timepoints of neuropsychological testing. They also had magnetic resonance imaging (MRI) including structured three-dimensional volume images. Cortical thickness was measured using surface-based methods. We performed trajectory analyses to classify our aMCI patients according to their progression and investigate their cognitive trajectory using CDR-SOB.

RESULTS

Trajectory analyses showed that patients with aMCI were divided into three groups: stable (61.8%), slow decliner (31.7%), and fast decliner (6.5%). Changes throughout a mean follow-up duration of 3.7 years in the CDR-SOB for the subgroups of stable/slow/fast decliners were 1.3-, 6.4-, and 12-point increases, respectively. Decliners were older and carried apolipoprotein E4 (APOE4) genotypes more frequently than stable patients. Compared with the stable group, decliners showed a higher frequency of aMCI patients with both visual and verbal memory dysfunction, late stage aMCI, and multiple domain dysfunction. In addition, compared with the stable group, the slow decliners showed cortical thinning predominantly in bilateral parietotemporal areas, while the fast decliners showed cortical thinning predominantly in bilateral frontotemporal areas. Both decliner groups showed worse cognitive function in attention, language, visuospatial, memory, and frontal/executive domains than the stable group.

CONCLUSIONS

Our data-driven trajectory analysis provides new insights into heterogeneous cognitive trajectories of aMCI and further suggests that baseline clinical and neuroimaging profiles might predict aMCI patients with poor prognosis.

Brain-Wide Genome-Wide Association Study for Alzheimer's Disease via Joint Projection Learning and Sparse Regression Model

Brain-wide and genome-wide association (BW-GWA) study is presented in this paper to identify the associations between the brain imaging phenotypes (i.e., regional volumetric measures) and the genetic variants [i.e., single nucleotide polymorphism (SNP)] in Alzheimer's disease (AD). The main challenges of this study include the data heterogeneity, complex phenotype-genotype associations, high-dimensional data (e.g., thousands of SNPs), and the existence of phenotype outliers. Previous BW-GWA studies, while addressing some of these challenges, did not consider the diagnostic label information in their formulations, thus limiting their clinical applicability. To address these issues, we present a novel joint projection and sparse regression model to discover the associations between the phenotypes and genotypes. Specifically, to alleviate the negative influence of data heterogeneity, we first map the genotypes into an intermediate imaging-phenotype-like space. Then, to better reveal the complex phenotype-genotype associations, we project both the mapped genotypes and the original imaging phenotypes into a diagnostic-label-guided joint feature space, where the intraclass projected points are constrained to be close to each other. In addition, we use l2,1-norm minimization on both the regression loss function and the transformation coefficient matrices, to reduce the effect of phenotype outliers and also to encourage sparse feature selections of both the genotypes and phenotypes. We evaluate our method using AD neuroimaging initiative dataset, and the results show that our proposed method outperforms several state-of-the-art methods in term of the average root-mean-square error of genome-to-phenotype predictions. Besides, the associated SNPs and brain regions identified in this study have also been shown in the previous AD-related studies, thus verifying the effectiveness and potential of our proposed method in AD pathogenesis study.

Early Diagnosis of Alzheimer's Disease Based on Resting-State Brain Networks and Deep Learning

Computerized healthcare has undergone rapid development thanks to the advances in medical imaging and machine learning technologies. Especially, recent progress on deep learning opens a new era for multimedia based clinical decision support. In this paper, we use deep learning with brain network and clinical relevant text information to make early diagnosis of Alzheimer's Disease (AD). The clinical relevant text information includes age, gender, and ApoE gene of the subject. The brain network is constructed by computing the functional connectivity of brain regions using resting-state functional magnetic resonance imaging (R-fMRI) data. A targeted autoencoder network is built to distinguish normal aging from mild cognitive impairment, an early stage of AD. The proposed method reveals discriminative brain network features effectively and provides a reliable classifier for AD detection. Compared to traditional classifiers based on R-fMRI time series data, about 31.21 percent improvement of the prediction accuracy is achieved by the proposed deep learning method, and the standard deviation reduces by 51.23 percent in the best case that means our prediction model is more stable and reliable compared to the traditional methods. Our work excavates deep learning's advantages of classifying high-dimensional multimedia data in medical services, and could help predict and prevent AD at an early stage.

TENSOR GENERALIZED ESTIMATING EQUATIONS FOR LONGITUDINAL IMAGING ANALYSIS

Longitudinal neuroimaging studies are becoming increasingly prevalent, where brain images are collected on multiple subjects at multiple time points. Analyses of such data are scientifically important, but also challenging. Brain images are in the form of multidimensional arrays, or tensors, which are characterized by both ultrahigh dimensionality and a complex structure. Longitudinally repeated images and induced temporal correlations add a further layer of complexity. Despite some recent efforts, there exist very few solutions for longitudinal imaging analyses. In response to the increasing need to analyze longitudinal imaging data, we propose several tensor generalized estimating equations (GEEs). The proposed GEE approach accounts for intra-subject correlation, and an imposed low-rank structure on the coefficient tensor effectively reduces the dimensionality. We also propose a scalable estimation algorithm, establish the asymptotic properties of the solution to the tensor GEEs, and investigate sparsity regularization for the purpose of region selection. We demonstrate the proposed method using simulations and by analyzing a real data set from the Alzheimer's Disease Neuroimaging Initiative.

A Comparative Atlas-Based Recognition of Mild Cognitive Impairment With Voxel-Based Morphometry

An accurate and reliable brain partition atlas is vital to quantitatively investigate the structural and functional abnormalities in mild cognitive impairment (MCI), generally considered to be a prodromal phase of Alzheimer's disease. In this paper, we proposed an automated structural classification method to identify MCI from healthy controls (HC) and investigated whether the classification performance was dependent on the brain parcellation schemes, including Automated Anatomical Labeling (AAL-90) atlas, Brainnetome (BN-246) atlas, and AAL-1024 atlas. In detail, structural magnetic resonance imaging (sMRI) data of 69 MCI patients and 63 HC matched well on gender, age, and education level were collected and analyzed with voxel-based morphometry method first, then the volume features of every region of interest (ROI) belonging to the above-mentioned three atlases were calculated and compared between MCI and HC groups, respectively. At last, the abnormal volume features were selected as the classification features for a proposed support vector machine based identification method. After the leave-one-out cross-validation to estimate the classification performance, our results reported accuracies of 83, 92, and 89% with AAL-90, BN-246, and AAL-1024 atlas, respectively, suggesting that future studies should pay more attention to the selection of brain partition schemes in the atlas-based studies. Furthermore, the consistent atrophic brain regions among three atlases were predominately located at bilateral hippocampus, bilateral parahippocampal, bilateral amygdala, bilateral cingulate gyrus, left angular gyrus, right superior frontal gyrus, right middle frontal gyrus, left inferior frontal gyrus, and left precentral gyrus.

Callosal thickness profiles for prognosticating conversion from mild cognitive impairment to Alzheimer's disease: A classification approach

INTRODUCTION

Alzheimer's disease (AD) is the most common form of dementia. Finding biomarkers to prognosticate transition from mild cognitive impairment (MCI) to AD is important to clinical medicine. Promising imaging biomarkers of AD conversion identified so far include atrophy of the cerebral cortex and subcortical gray matter nuclei.

METHODS

This study introduces thickness and bending angle of the corpus callosum as a putative white matter marker of MCI to AD conversion. The corpus callosum is computationally less demanding to segment automatically compared to more complicated structures and a subject can be processed in a few minutes. We aimed to demonstrate that callosal shape and thickness measures provide a simple, effective, and accurate prognostication tool in ADNI dataset. Using longitudinal datasets, we classified MCI subjects based on conversion to AD assessed via cognitive testing. We evaluated the classification accuracy of callosal shape features in comparison with the existing "gold standard" cortical thickness and subcortical gray matter volume measures.

RESULTS

The callosal thickness measures were less accurate in classifying conversion status by cognitive scores compared to gray matter measures for AD.

CONCLUSIONS

While this paper presented a negative result, this method may be more suitable for a disease of the white matter.

Contourlet-based hippocampal magnetic resonance imaging texture features for multivariant classification and prediction of Alzheimer's disease

The study is aimed to assess whether the addition of contourlet-based hippocampal magnetic resonance imaging (MRI) texture features to multivariant models improves the classification of Alzheimer's disease (AD) and the prediction of mild cognitive impairment (MCI) conversion, and to evaluate whether Gaussian process (GP) and partial least squares (PLS) are feasible in developing multivariant models in this context. Clinical and MRI data of 58 patients with probable AD, 147 with MCI, and 94 normal controls (NCs) were collected. Baseline contourlet-based hippocampal MRI texture features, medical histories, symptoms, neuropsychological tests, volume-based morphometric (VBM) parameters based on MRI, and regional CMgl measurement based on fluorine-18 fluorodeoxyglucose-positron emission tomography were included to develop GP and PLS models to classify different groups of subjects. GPR1 model, which incorporated MRI texture features and was based on GPG, performed better in classifying different groups of subjects than GPR2 model, which used the same algorithm and had the same data as GPR1 except that MRI texture features were excluded. PLS model, which included the same variables as GPR1 but was based on the PLS algorithm, performed best among the three models. GPR1 accurately predicted 82.2% (51/62) of MCI convertors confirmed during the 2-year follow-up period, while this figure was 53 (85.5%) for PLS model. GPR1 and PLS models accurately predicted 58 (79.5%) vs. 61 (83.6%) of 73 patients with stable MCI, respectively. For seven patients with MCI who converted to NCs, PLS model accurately predicted all cases (100%), while GPR1 predicted six (85.7%) cases. The addition of contourlet-based MRI texture features to multivariant models can effectively improve the classification of AD and the prediction of MCI conversion to AD. Both GPR and LPS models performed well in the classification and predictive process, with the latter having significantly higher classification and predictive accuracies. Advances in knowledge: We combined contourlet-based hippocampal MRI texture features, medical histories, symptoms, neuropsychological tests, volume-based morphometric (VBM) parameters, and regional CMgl measurement to develop models using GP and PLS algorithms to classify AD patients.

Discrimination and conversion prediction of mild cognitive impairment using convolutional neural networks

Background

Recently, studies have demonstrated that machine learning techniques, particularly cutting-edge deep learning technology, have achieved significant progression on the classification of Alzheimer's disease (AD) and its prodromal phase, mild cognitive impairment (MCI). Moreover, accurate prediction of the progress and the conversion risk from MCI to probable AD has been of great importance in clinical application.

Methods

In this study, the baseline MR images and follow-up information during 3 years of 150 normal controls (NC), 150 patients with stable MCI (sMCI) and 157 converted MCI (cMCI) were collected from the Alzheimer's Disease Neuroimaging Initiative (ADNI). The deep convolutional neural networks (CNNs) were adopted to distinguish different stages of MCI from the NC group, and predict the conversion time from MCI to AD. Two CNN architectures including GoogleNet and CaffeNet were explored and evaluated in multiple classifications and estimations of conversion risk using transfer learning from pre-trained ImageNet (via fine-tuning) and five-fold cross-validation. A novel data augmentation approach using random views aggregation was applied to generate abundant image patches from the original MR scans.

Results

The GoogleNet acquired accuracies with 97.58%, 67.33% and 84.71% in three-way discrimination among the NC, sMCI and cMCI groups respectively, whereas the CaffeNet obtained promising accuracies of 98.71%, 72.04% and 92.35% in the NC, sMCI and cMCI classifications. Furthermore, the accuracy measures of conversion risk of patients with cMCI ranged from 71.25% to 83.25% in different time points using GoogleNet, whereas the CaffeNet achieved remarkable accuracy measures from 95.42% to 97.01% in conversion risk prediction.

Conclusions

The experimental results demonstrated that the proposed methods had prominent capability in classification among the 3 groups such as sMCI, cMCI and NC, and exhibited significant ability in conversion risk prediction of patients with MCI.

Voxel-wise deviations from healthy aging for the detection of region-specific atrophy

The identification of pathological atrophy in MRI scans requires specialized training, which is scarce outside dedicated centers. We sought to investigate the clinical usefulness of computer-generated representations of local grey matter (GM) loss or increased volume of cerebral fluids (CSF) as normalized deviations (z-scores) from healthy aging to either aid human visual readings or directly detect pathological atrophy.

Two experienced neuroradiologists rated atrophy in 30 patients with Alzheimer's disease (AD), 30 patients with frontotemporal dementia (FTD), 30 with dementia due to Lewy-body disease (LBD) and 30 healthy controls (HC) on a three-point scale in 10 anatomical regions as reference gold standard. Seven raters, varying in their experience with MRI diagnostics rated all cases on the same scale once with and once without computer-generated volume deviation maps that were overlaid on anatomical slices. In addition, we investigated the predictive value of the computer generated deviation maps on their own for the detection of atrophy as identified by the gold standard raters.

Inter and intra-rater agreements of the two gold standard raters were substantial (Cohen's kappa κ > 0.62). The intra-rater agreement of the other raters ranged from fair (κ = 0.37) to substantial (κ = 0.72) and improved on average by 0.13 (0.57 < κ < 0.87) when volume deviation maps were displayed. The seven other raters showed good agreement with the gold standard in regions including the hippocampus but agreement was substantially lower in e.g. the parietal cortex and did not improve with the display of atrophy scores. Rating speed increased over the course of the study and irrespective of the presentation of voxel-wise deviations.

Automatically detected large deviations of local volume were consistently associated with gold standard atrophy reading as shown by an area under the receiver operator characteristic of up to 0.95 for the hippocampus region. When applying these test characteristics to prevalences typically found in a memory clinic, we observed a positive or negative predictive value close to or above 0.9 in the hippocampus for almost all of the expected cases. The volume deviation maps derived from CSF volume increase were generally better in detecting atrophy.

Our study demonstrates an agreement of visual ratings among non-experts not further increased by displaying, region-specific deviations of volume. The high predictive value of computer generated local deviations independent from human interaction and the consistent advantages of CSF-over GM-based estimations should be considered in the development of diagnostic tools and indicate clinical utility well beyond aiding visual assessments.

•

The visual identification of atrophy is most accurate in the temporal lobe.

•

Voxel-wise deviations of tissue volume from normal aging is easy to implement.

•

Displaying voxel-wise deviations subjectively but not objectively aids readers.

•

Voxel-wise deviations themselves show high agreement with expert readings.

•

Information on tissue deviations should be provided with cerebral MRI scans.

Identification of Alzheimer's disease and mild cognitive impairment using multimodal sparse hierarchical extreme learning machine

Different modalities such as structural MRI, FDG-PET, and CSF have complementary information, which is likely to be very useful for diagnosis of AD and MCI. Therefore, it is possible to develop a more effective and accurate AD/MCI automatic diagnosis method by integrating complementary information of different modalities. In this paper, we propose multi-modal sparse hierarchical extreme leaning machine (MSH-ELM). We used volume and mean intensity extracted from 93 regions of interest (ROIs) as features of MRI and FDG-PET, respectively, and used p-tau, t-tau, and A β 42 as CSF features. In detail, high-level representation was individually extracted from each of MRI, FDG-PET, and CSF using a stacked sparse extreme learning machine auto-encoder (sELM-AE). Then, another stacked sELM-AE was devised to acquire a joint hierarchical feature representation by fusing the high-level representations obtained from each modality. Finally, we classified joint hierarchical feature representation using a kernel-based extreme learning machine (KELM). The results of MSH-ELM were compared with those of conventional ELM, single kernel support vector machine (SK-SVM), multiple kernel support vector machine (MK-SVM) and stacked auto-encoder (SAE). Performance was evaluated through 10-fold cross-validation. In the classification of AD vs. HC and MCI vs. HC problem, the proposed MSH-ELM method showed mean balanced accuracies of 96.10% and 86.46%, respectively, which is much better than those of competing methods. In summary, the proposed algorithm exhibits consistently better performance than SK-SVM, ELM, MK-SVM and SAE in the two binary classification problems (AD vs. HC and MCI vs. HC).

Modeling and prediction of clinical symptom trajectories in Alzheimer's disease using longitudinal data

Computational models predicting symptomatic progression at the individual level can be highly beneficial for early intervention and treatment planning for Alzheimer's disease (AD). Individual prognosis is complicated by many factors including the definition of the prediction objective itself. In this work, we present a computational framework comprising machine-learning techniques for 1) modeling symptom trajectories and 2) prediction of symptom trajectories using multimodal and longitudinal data. We perform primary analyses on three cohorts from Alzheimer's Disease Neuroimaging Initiative (ADNI), and a replication analysis using subjects from Australian Imaging, Biomarker & Lifestyle Flagship Study of Ageing (AIBL). We model the prototypical symptom trajectory classes using clinical assessment scores from mini-mental state exam (MMSE) and Alzheimer's Disease Assessment Scale (ADAS-13) at nine timepoints spanned over six years based on a hierarchical clustering approach. Subsequently we predict these trajectory classes for a given subject using magnetic resonance (MR) imaging, genetic, and clinical variables from two timepoints (baseline + follow-up). For prediction, we present a longitudinal Siamese neural-network (LSN) with novel architectural modules for combining multimodal data from two timepoints. The trajectory modeling yields two (stable and decline) and three (stable, slow-decline, fast-decline) trajectory classes for MMSE and ADAS-13 assessments, respectively. For the predictive tasks, LSN offers highly accurate performance with 0.900 accuracy and 0.968 AUC for binary MMSE task and 0.760 accuracy for 3-way ADAS-13 task on ADNI datasets, as well as, 0.724 accuracy and 0.883 AUC for binary MMSE task on replication AIBL dataset.

Identification of Alzheimer's Disease and Mild Cognitive Impairment Using Networks Constructed Based on Multiple Morphological Brain Features

Structural brain markers are important for characterizing the pathology of Alzheimer's disease (AD) and mild cognitive impairment (MCI). Here, we constructed a multifeature-based network (MFN) for each individual using a sparse linear regression performed on six types of morphological features to promote the structure-based autodiagnosis. The categorization performance of the MFN was evaluated in 165 normal control subjects, 221 patients with MCI, and 142 patients with AD. We achieved 96.42% and 96.37% accuracy, respectively, in distinguishing the patients with AD and MCI from the normal control subjects, and reasonable discrimination of the two disease cohorts (70.52%) and prediction of the MCI to AD progression (65.61%). The performance was further improved by combining the properties of the MFN with the morphological features. Our results demonstrate the effectiveness of the MFN in combination with morphological features obtained from single imaging modality, serving as robust biomarkers in the diagnosis of AD and MCI.

Convergent Results from Neuropsychology and from Neuroimaging in Patients with Mild Cognitive Impairment

BACKGROUND/AIMS

To investigate the correspondence between neuropsychological single measures and variation in fludeoxyglucose positron emission tomography (FDG PET) glucose metabolism and magnetic resonance imaging (MRI) cortical thickness in mild cognitive impairment (MCI) patients.

METHODS

Forty-two elderly controls and 73 MCI subjects underwent FDG PET and MRI scanning. Backward regression analyses with PET and MRI regions were used as dependent variables, while Rey Auditory Verbal Memory Test (RAVLT) recall, Trail Making Test B (TMT B), and a composite test score (RAVLT learning and immediate recall, TMT A, COWAT, and letter-number sequencing) were used as predictor variables.

RESULTS

The composite score predicted variation in cortical metabolism; supplementary analyses showed that TMT B was significantly correlated with PET metabolism as well. RAVLT and TMT B were significant predictors of variation in MRI cortical thickness.

CONCLUSION

Our results indicate that RAVLT and TMT B are sensitive to variation in Alzheimer disease neuroimaging markers.

Brain MRI analysis for Alzheimer's disease diagnosis using an ensemble system of deep convolutional neural networks

Alzheimer's disease is an incurable, progressive neurological brain disorder. Earlier detection of Alzheimer's disease can help with proper treatment and prevent brain tissue damage. Several statistical and machine learning models have been exploited by researchers for Alzheimer's disease diagnosis. Analyzing magnetic resonance imaging (MRI) is a common practice for Alzheimer's disease diagnosis in clinical research. Detection of Alzheimer's disease is exacting due to the similarity in Alzheimer's disease MRI data and standard healthy MRI data of older people. Recently, advanced deep learning techniques have successfully demonstrated human-level performance in numerous fields including medical image analysis. We propose a deep convolutional neural network for Alzheimer's disease diagnosis using brain MRI data analysis. While most of the existing approaches perform binary classification, our model can identify different stages of Alzheimer's disease and obtains superior performance for early-stage diagnosis. We conducted ample experiments to demonstrate that our proposed model outperformed comparative baselines on the Open Access Series of Imaging Studies dataset.

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.

A Systematic Review of Longitudinal Studies Which Measure Alzheimer's Disease Biomarkers

Alzheimer’s disease (AD) is a progressive and fatal neurodegenerative disease, with no effective treatment or cure. A gold standard therapy would be treatment to slow or halt disease progression; however, knowledge of causation in the early stages of AD is very limited. In order to determine effective endpoints for possible therapies, a number of quantitative surrogate markers of disease progression have been suggested, including biochemical and imaging biomarkers. The dynamics of these various surrogate markers over time, particularly in relation to disease development, are, however, not well characterized. We reviewed the literature for studies that measured cerebrospinal fluid or plasma amyloid-β and tau, or took magnetic resonance image or fluorodeoxyglucose/Pittsburgh compound B-positron electron tomography scans, in longitudinal cohort studies. We summarized the properties of the major cohort studies in various countries, commonly used diagnosis methods and study designs. We have concluded that additional studies with repeat measures over time in a representative population cohort are needed to address the gap in knowledge of AD progression. Based on our analysis, we suggest directions in which research could move in order to advance our understanding of this complex disease, including repeat biomarker measurements, standardization and increased sample sizes.

Multimodal and Multiscale Deep Neural Networks for the Early Diagnosis of Alzheimer's Disease using structural MR and FDG-PET images

Alzheimer's Disease (AD) is a progressive neurodegenerative disease where biomarkers for disease based on pathophysiology may be able to provide objective measures for disease diagnosis and staging. Neuroimaging scans acquired from MRI and metabolism images obtained by FDG-PET provide in-vivo measurements of structure and function (glucose metabolism) in a living brain. It is hypothesized that combining multiple different image modalities providing complementary information could help improve early diagnosis of AD. In this paper, we propose a novel deep-learning-based framework to discriminate individuals with AD utilizing a multimodal and multiscale deep neural network. Our method delivers 82.4% accuracy in identifying the individuals with mild cognitive impairment (MCI) who will convert to AD at 3 years prior to conversion (86.4% combined accuracy for conversion within 1-3 years), a 94.23% sensitivity in classifying individuals with clinical diagnosis of probable AD, and a 86.3% specificity in classifying non-demented controls improving upon results in published literature.

Separating Symptomatic Alzheimer's Disease from Depression based on Structural MRI

Older patients with depression or Alzheimer’s disease (AD) at the stage of early dementia or mild cognitive impairment may present with objective cognitive impairment, although the pathology and thus therapy and prognosis differ substantially. In this study, we assessed the potential of an automated algorithm to categorize a test set of 65 T1-weighted structural magnetic resonance images (MRI). A convenience sample of elderly individuals fulfilling clinical criteria of either AD (n = 28) or moderate and severe depression (n = 37) was recruited from different settings to assess the potential of the pattern recognition method to assist in the differential diagnosis of AD versus depression. We found that our algorithm learned discriminative patterns in the subject’s grey matter distribution reflected by an area under the receiver operator characteristics curve of up to 0.83 (confidence interval ranged from 0.67 to 0.92) and a balanced accuracy of 0.79 for the separation of depression from AD, evaluated by leave-one-out cross validation. The algorithm also identified consistent structural differences in a clinically more relevant scenario where the data used during training were independent from the data used for evaluation and, critically, which included five possible diagnoses (specifically AD, frontotemporal dementia, Lewy body dementia, depression, and healthy aging). While the output was insufficiently accurate to use it directly as a means for classification when multiple classes are possible, the continuous output computed by the machine learning algorithm differed between the two groups that were investigated. The automated analysis thus could complement, but not replace clinical assessments.

Conversion and time-to-conversion predictions of mild cognitive impairment using low-rank affinity pursuit denoising and matrix completion

In this paper, we aim to predict conversion and time-to-conversion of mild cognitive impairment (MCI) patients using multi-modal neuroimaging data and clinical data, via cross-sectional and longitudinal studies. However, such data are often heterogeneous, high-dimensional, noisy, and incomplete. We thus propose a framework that includes sparse feature selection, low-rank affinity pursuit denoising (LRAD), and low-rank matrix completion (LRMC) in this study. Specifically, we first use sparse linear regressions to remove unrelated features. Then, considering the heterogeneity of the MCI data, which can be assumed as a union of multiple subspaces, we propose to use a low rank subspace method (i.e., LRAD) to denoise the data. Finally, we employ LRMC algorithm with three data fitting terms and one inequality constraint for joint conversion and time-to-conversion predictions. Our framework aims to answer a very important but yet rarely explored question in AD study, i.e., when will the MCI convert to AD? This is different from survival analysis, which provides the probabilities of conversion at different time points that are mainly used for global analysis, while our time-to-conversion prediction is for each individual subject. Evaluations using the ADNI dataset indicate that our method outperforms conventional LRMC and other state-of-the-art methods. Our method achieves a maximal pMCI classification accuracy of 84% and time prediction correlation of 0.665.

Morphological Biomarker Differentiating MCI Converters from Nonconverters: Longitudinal Evidence Based on Hemispheric Asymmetry

Identifying subjects with mild cognitive impairment (MCI) who may probably progress to Alzheimer's disease (AD) is important for better understanding the disease mechanisms and facilitating early treatments. In addition to the direct volumetric and thickness measurement based on high-resolution magnetic resonance imaging (MRI), hemispheric asymmetry could be a potential index to detect morphological variations in MCI patients with a high risk of conversion to AD. The present study collected a set of longitudinal MRI data from 53 MCI converters and nonconverters and investigated the asymmetry differences between groups. Asymmetry variation was observed in the medial temporal lobe, especially in the entorhinal cortex, between converters and nonconverters 3 years before the former developed AD. The proposed asymmetry analysis was observed to be sensitive to detect morphological changes between groups as compared to the methods of voxel-based morphometry (VBM) and thickness measurement. Hemispheric asymmetry in specific brain regions as a neuroimaging biomarker can provide helpful information for prediction of MCI conversion.

Cognitive Composites Domain Scores Related to Neuroimaging Biomarkers within Probable-Amnestic Mild Cognitive Impairment-Storage Subtype

The probable-amnestic (Pr-a) mild cognitive impairment (MCI)-storage subtype is a phenotype with 8.5 times more risk of conversion to dementia, mainly Alzheimer's disease (AD), than the possible non-amnestic (Pss-na) MCI. The aim of this study was to find the optimized cognitive composites (CCs) domain scores most related to neuroimaging biomarkers within Pr-aMCI-storage subtype patients. The Fundació ACE (ACE) study with 20 Pr-aMCI-storage subtype subjects (MCI) were analyzed. All subjects underwent a neuropsychological assessment, a structural MRI, FDG-PET, and PIB-PET. The adjusted hippocampal volume (aHV) on MRI, the standard uptake value ratio (SUVR) on FDG-PET and PIB-PET SUVR measures were analyzed. The construction of the CCs domain scores, and the aHV on MRI and FDG-PET SUVR measures, were replicated in the parental AB255 study database (n = 133 MCI). Partial correlations adjusted by age, gender, and education were calculated with the associated p-value among every CC domain score and the neuroimaging biomarkers. The results were replicated in the "MCI due to AD" with memory storage impairments from ADNI. Delayed Recall CC domain score was significantly correlated with PIB-PET SUVR (β= -0.61, p = 0.003) in the ACE study and also with aHV on MRI (β= 0.27, p = 0.01) and FDG-PET SUVR (β= 0.27, p = 0.01) in the AB255 study. After a median survival time of 20.6 months, 85% from the ACE MCI converted to AD. The replication of our results in the ADNI dataset also confirmed our findings. Delayed Recall is the CC domain score best correlated with neuroimaging biomarkers associated with prodromal AD diagnosis.

Multimodal Classification of Mild Cognitive Impairment Based on Partial Least Squares

In recent years, increasing attention has been given to the identification of the conversion of mild cognitive impairment (MCI) to Alzheimer's disease (AD). Brain neuroimaging techniques have been widely used to support the classification or prediction of MCI. The present study combined magnetic resonance imaging (MRI), 18F-fluorodeoxyglucose PET (FDG-PET), and 18F-florbetapir PET (florbetapir-PET) to discriminate MCI converters (MCI-c, individuals with MCI who convert to AD) from MCI non-converters (MCI-nc, individuals with MCI who have not converted to AD in the follow-up period) based on the partial least squares (PLS) method. Two types of PLS models (informed PLS and agnostic PLS) were built based on 64 MCI-c and 65 MCI-nc from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. The results showed that the three-modality informed PLS model achieved better classification accuracy of 81.40%, sensitivity of 79.69%, and specificity of 83.08% compared with the single-modality model, and the three-modality agnostic PLS model also achieved better classification compared with the two-modality model. Moreover, combining the three modalities with clinical test score (ADAS-cog), the agnostic PLS model (independent data: florbetapir-PET; dependent data: FDG-PET and MRI) achieved optimal accuracy of 86.05%, sensitivity of 81.25%, and specificity of 90.77%. In addition, the comparison of PLS, support vector machine (SVM), and random forest (RF) showed greater diagnostic power of PLS. These results suggested that our multimodal PLS model has the potential to discriminate MCI-c from the MCI-nc and may therefore be helpful in the early diagnosis of AD.

Quantitative MRI Brain Studies in Mild Cognitive Impairment and Alzheimer's Disease: A Methodological Review

Classifying and predicting Alzheimer's disease (AD) in individuals with memory disorders through clinical and psychometric assessment is challenging, especially in mild cognitive impairment (MCI) subjects. Quantitative structural magnetic resonance imaging acquisition methods in combination with computer-aided diagnosis are currently being used for the assessment of AD. These acquisitions methods include voxel-based morphometry, volumetric measurements in specific regions of interest (ROIs), cortical thickness measurements, shape analysis, and texture analysis. This review evaluates the aforementioned methods in the classification of cases into one of the following three groups: normal controls, MCI, and AD subjects. Furthermore, the performance of the methods is assessed on the prediction of conversion from MCI to AD. In parallel, it is also assessed which ROIs are preferred in both classification and prognosis through the different states of the disease. Structural changes in the early stages of the disease are more pronounced in the medial temporal lobe, especially in the entorhinal cortex, whereas with disease progression, both entorhinal cortex and hippocampus offer similar discriminative power. However, for the conversion from MCI subjects to AD, entorhinal cortex provides better predictive accuracies rather than other structures, such as the hippocampus.