Multi-modal imaging predicts memory performance in normal aging and cognitive decline

Predicting conversion from mild cognitive impairment to Alzheimer's disease: a multimodal approach

Successively predicting whether mild cognitive impairment patients will progress to Alzheimer's disease is of significant clinical relevance. This ability may provide information that can be leveraged by emerging intervention approaches and thus mitigate some of the negative effects of the disease. Neuroimaging biomarkers have gained some attention in recent years and may be useful in predicting the conversion of mild cognitive impairment to Alzheimer's disease. We implemented a novel multi-modal approach that allowed us to evaluate the potential of different imaging modalities, both alone and in different degrees of combinations, in predicting the conversion to Alzheimer's disease of mild cognitive impairment patients. We applied this approach to the imaging data from the Alzheimer's Disease Neuroimaging Initiative that is a multi-modal imaging dataset comprised of MRI, Fluorodeoxyglucose PET, Florbetapir PET and diffusion tensor imaging. We included a total of 480 mild cognitive impairment patients that were split into two groups: converted and stable. Imaging data were segmented into atlas-based regions of interest, from which relevant features were extracted for the different imaging modalities and used to construct machine-learning models to classify mild cognitive impairment patients into converted or stable, using each of the different imaging modalities independently. The models were then combined, using a simple weight fusion ensemble strategy, to evaluate the complementarity of different imaging modalities and their contribution to the prediction accuracy of the models. The single-modality findings revealed that the model, utilizing features extracted from Florbetapir PET, demonstrated the highest performance with a balanced accuracy of 83.51%. Concerning multi-modality models, not all combinations enhanced mild cognitive impairment conversion prediction. Notably, the combination of MRI with Fluorodeoxyglucose PET emerged as the most promising, exhibiting an overall improvement in predictive capabilities, achieving a balanced accuracy of 78.43%. This indicates synergy and complementarity between the two imaging modalities in predicting mild cognitive impairment conversion. These findings suggest that β-amyloid accumulation provides robust predictive capabilities, while the combination of multiple imaging modalities has the potential to surpass certain single-modality approaches. Exploring modality-specific biomarkers, we identified the brainstem as a sensitive biomarker for both MRI and Fluorodeoxyglucose PET modalities, implicating its involvement in early Alzheimer's pathology. Notably, the corpus callosum and adjacent cortical regions emerged as potential biomarkers, warranting further study into their role in the early stages of Alzheimer's disease.

Depth- and curvature-based quantitative susceptibility mapping analyses of cortical iron in Alzheimer's disease

In addition to amyloid beta plaques and neurofibrillary tangles, Alzheimer's disease (AD) has been associated with elevated iron in deep gray matter nuclei using quantitative susceptibility mapping (QSM). However, only a few studies have examined cortical iron, using more macroscopic approaches that cannot assess layer-specific differences. Here, we conducted column-based QSM analyses to assess whether AD-related increases in cortical iron vary in relation to layer-specific differences in the type and density of neurons. We obtained global and regional measures of positive (iron) and negative (myelin, protein aggregation) susceptibility from 22 adults with AD and 22 demographically matched healthy controls. Depth-wise analyses indicated that global susceptibility increased from the pial surface to the gray/white matter boundary, with a larger slope for positive susceptibility in the left hemisphere for adults with AD than controls. Curvature-based analyses indicated larger global susceptibility for adults with AD versus controls; the right hemisphere versus left; and gyri versus sulci. Region-of-interest analyses identified similar depth- and curvature-specific group differences, especially for temporo-parietal regions. Finding that iron accumulates in a topographically heterogenous manner across the cortical mantle may help explain the profound cognitive deterioration that differentiates AD from the slowing of general motor processes in healthy aging.

Self-supervised multimodal learning for group inferences from MRI data: Discovering disorder-relevant brain regions and multimodal links

In recent years, deep learning approaches have gained significant attention in predicting brain disorders using neuroimaging data. However, conventional methods often rely on single-modality data and supervised models, which provide only a limited perspective of the intricacies of the highly complex brain. Moreover, the scarcity of accurate diagnostic labels in clinical settings hinders the applicability of the supervised models. To address these limitations, we propose a novel self-supervised framework for extracting multiple representations from multimodal neuroimaging data to enhance group inferences and enable analysis without resorting to labeled data during pre-training. Our approach leverages Deep InfoMax (DIM), a self-supervised methodology renowned for its efficacy in learning representations by estimating mutual information without the need for explicit labels. While DIM has shown promise in predicting brain disorders from single-modality MRI data, its potential for multimodal data remains untapped. This work extends DIM to multimodal neuroimaging data, allowing us to identify disorder-relevant brain regions and explore multimodal links. We present compelling evidence of the efficacy of our multimodal DIM analysis in uncovering disorder-relevant brain regions, including the hippocampus, caudate, insula, - and multimodal links with the thalamus, precuneus, and subthalamus hypothalamus. Our self-supervised representations demonstrate promising capabilities in predicting the presence of brain disorders across a spectrum of Alzheimer's phenotypes. Comparative evaluations against state-of-the-art unsupervised methods based on autoencoders, canonical correlation analysis, and supervised models highlight the superiority of our proposed method in achieving improved classification performance, capturing joint information, and interpretability capabilities. The computational efficiency of the decoder-free strategy enhances its practical utility, as it saves compute resources without compromising performance. This work offers a significant step forward in addressing the challenge of understanding multimodal links in complex brain disorders, with potential applications in neuroimaging research and clinical diagnosis.

Multimodal transformer network for incomplete image generation and diagnosis of Alzheimer's disease

Multimodal images such as magnetic resonance imaging (MRI) and positron emission tomography (PET) could provide complementary information about the brain and have been widely investigated for the diagnosis of neurodegenerative disorders such as Alzheimer's disease (AD). However, multimodal brain images are often incomplete in clinical practice. It is still challenging to make use of multimodality for disease diagnosis with missing data. In this paper, we propose a deep learning framework with the multi-level guided generative adversarial network (MLG-GAN) and multimodal transformer (Mul-T) for incomplete image generation and disease classification, respectively. First, MLG-GAN is proposed to generate the missing data, guided by multi-level information from voxels, features, and tasks. In addition to voxel-level supervision and task-level constraint, a feature-level auto-regression branch is proposed to embed the features of target images for an accurate generation. With the complete multimodal images, we propose a Mul-T network for disease diagnosis, which can not only combine the global and local features but also model the latent interactions and correlations from one modality to another with the cross-modal attention mechanism. Comprehensive experiments on three independent datasets (i.e., ADNI-1, ADNI-2, and OASIS-3) show that the proposed method achieves superior performance in the tasks of image generation and disease diagnosis compared to state-of-the-art methods.

Latent feature representation learning for Alzheimer's disease classification

Early detection and treatment of Alzheimer's Disease (AD) are significant. Recently, multi-modality imaging data have promoted the development of the automatic diagnosis of AD. This paper proposes a method based on latent feature fusion to make full use of multi-modality image data information. Specifically, we learn a specific projection matrix for each modality by introducing a binary label matrix and local geometry constraints and then project the original features of each modality into a low-dimensional target space. In this space, we fuse latent feature representations of different modalities for AD classification. The experimental results on Alzheimer's Disease Neuroimaging Initiative database demonstrate the proposed methods effectiveness in classifying AD.

Discriminating Aging Cognitive Decline Spectrum Using PET and Magnetic Resonance Image Features

Background: The population aging increased the prevalence of brain diseases, like Alzheimer’s disease (AD), and early identification of individuals with higher odds of cognitive decline is essential to maintain quality of life. Imaging evaluation of individuals at risk of cognitive decline includes biomarkers extracted from brain positron emission tomography (PET) and structural magnetic resonance imaging (MRI). Objective: We propose investigating ensemble models to classify groups in the aging cognitive decline spectrum by combining features extracted from single imaging modalities and combinations of imaging modalities (FDG+AMY+MRI, and a PET ensemble). Methods: We group imaging data of 131 individuals into four classes related to the individuals’ cognitive assessment in baseline and follow-up: stable cognitive non-impaired; individuals converting to mild cognitive impairment (MCI) syndrome; stable MCI; and Alzheimer’s clinical syndrome. We assess the performance of four algorithms using leave-one-out cross-validation: decision tree classifier, random forest (RF), light gradient boosting machine (LGBM), and categorical boosting (CAT). The performance analysis of models is evaluated using balanced accuracy before and after using Shapley Additive exPlanations with recursive feature elimination (SHAP-RFECV) method. Results: Our results show that feature selection with CAT or RF algorithms have the best overall performance in discriminating early cognitive decline spectrum mainly using MRI imaging features. Conclusion: Use of CAT or RF algorithms with SHAP-RFECV shows good discrimination of early stages of aging cognitive decline, mainly using MRI image features. Further work is required to analyze the impact of selected brain regions and their correlation with cognitive decline spectrum.

Diagnostic Utility of Hippocampal Volumetric Data in a Memory Disorder Clinic Setting

BACKGROUND

Hippocampal volumetric data are widely used in research but are rarely examined in clinical populations in regard to aiding diagnosis or correlating with objective memory test scores.

OBJECTIVE

To replicate and expand on the few prior clinical examinations of the utility of hippocampal volumetric data. We evaluated MRI volumetric data to determine (a) the degree of hippocampal loss across diagnostic groups compared with a cognitively intact group, (b) if total or lateralized hippocampal volumes predict diagnostic group membership, and (c) how total and lateralized volumes correlate with memory tests.

METHOD

We retrospectively examined hippocampal volumetric data and memory test scores for 294 individuals referred to a memory clinic.

RESULTS

Individuals with mild cognitive impairment or Alzheimer disease had smaller hippocampal volumes compared with cognitively intact individuals. The raw and normalized total and lateralized hippocampal volumes were essentially equal for predicting diagnostic group membership, and notably low hippocampal volumes evidenced greater specificity than sensitivity. All of the volumetric data correlated with the memory test scores, with the total and left hippocampal volumes accounting for the slightly more variance in the diagnostic groups.

CONCLUSION

The diagnostic groups exhibited hippocampal volume loss, which can be a potential biomarker for neurodegenerative disease in clinical practice. However, solely using hippocampal volumetric data to predict diagnostic group membership or memory test failure was not supported. While extreme hippocampal volume loss was rare in the cognitively intact group, the sensitivity of these volumetric data suggests a need for supplementation by other tools when making a diagnosis.

Corpus Callosum Atrophy in Detection of Mild and Moderate Alzheimer's Disease Using Brain Magnetic Resonance Image Processing and Machine Learning Techniques

Background:

The total number of people with dementia is projected to reach 82 million in 2030 and 152 in 2050. Early and accurate identification of the underlying causes of dementia, such as Alzheimer’s disease (AD) is of utmost importance. A large body of research has shown that imaging techniques are most promising technologies to improve subclinical and early diagnosis of dementia. Morphological changes, especially atrophy in various structures like cingulate gyri, caudate nucleus, hippocampus, frontotemporal lobe, etc., have been established as markers for AD. Being the largest white matter structure with a high demand of blood supply from several main arterial systems, anatomical alterations of the corpus callosum (CC) may serve as potential indication neurodegenerative disease.

Objective:

To detect mild and moderate AD using brain magnetic resonance image (MRI) processing and machine learning techniques.

Methods:

We have performed automatic detection and segmentation of the CC and calculated its morphological features to feed into a multivariate pattern analysis using support vector machine (SVM) learning techniques.

Results:

Our results using large patients’ cohort show CC atrophy-based features are capable of distinguishing healthy and mild/moderate AD patients. Our classifiers obtain more than 90%sensitivity and specificity in differentiating demented patients from healthy cohorts and importantly, achieved more than 90%sensitivity and > 80%specificity in detecting mild AD patients.

Conclusion:

Results from this analysis are encouraging and advocate development of an image analysis software package to detect dementia from brain MRI using morphological alterations of the CC.

Subjective hearing handicap is associated with processing speed and visuospatial performance in older adults without severe hearing handicap

OBJECTIVE

Age-related hearing loss is a common disorder with significant consequences for quality of life. This study assessed the Hearing Handicap Inventory for the Elderly (HHIE) and cognition (Mini Mental State Exam; MMSE, Logical Memory; LM, Symbol Search; SS, Stroop Test; ST, and Mental Rotation; MR) to investigate which cognitive domains are most strongly involved with hearing self-assessment in older adults.

METHODS

The HHIE and cognitive measures were administered to 196 older adults (average age = 67.7 ± 4.3 years, male 56, female 140) without cognitive impairment and without severe hearing handicap. We conducted permutation tests of multiple regression analysis of the standardized scores on the HHIE and cognitive tests.

RESULTS

HHIE showed a significant negative correlation between processing speed performance on the SS (standardized β = -0.095, adjusted p = 0.04) and visuospatial performance on the MR (standardized β = -0.145, adjusted p = 0.04), and no correlation between the scores of the HHIE and either episodic memory performance on the LM (standardized β = 0.060, adjusted p = 0.22) or executive function performance on the ST (standardized β = 0.053, adjusted p = 0.32).

CONCLUSION

People reporting higher hearing handicaps should watch for poor cognitive function in processing speed and visuospatial abilities. These results imply that higher HHIE can have adverse effects on age-related cognitive decline.

Second Language Learning in Older Adults: Effects on Brain Structure and Predictors of Learning Success

It has previously been demonstrated that short-term foreign language learning can lead to structural brain changes in younger adults. Experience-dependent brain plasticity is known to be possible also in older age, but the specific effect of foreign language learning on brain structure in language-and memory-relevant regions in the old brain remains unknown. In the present study, 160 older Swedish adults (65–75 years) were randomized to complete either an entry-level Italian course or a relaxation course, both with a total duration of 11 weeks. Structural MRI scans were conducted before and after the intervention in a subset of participants to test for differential change in gray matter in the two groups in the inferior frontal gyrus, the superior temporal gyrus, and the hippocampus, and in white matter microstructure in the superior longitudinal fasciculus (SLF), inferior longitudinal fasciculus (ILF), fronto-occipital fasciculus, and the hippocampal (HC) section of the cingulum. The study found no evidence for differential structural change following language training, independent of achieved vocabulary proficiency. However, hippocampal volume and associative memory ability before the intervention were found to be robust predictors of vocabulary proficiency at the end of the language course. The results suggest that having greater hippocampal volume and better associative memory ability benefits vocabulary learning in old age but that the very initial stage of foreign language learning does not trigger detectable changes in brain morphometry in old age.

COllaborative Neuropathology NEtwork Characterizing ouTcomes of TBI (CONNECT-TBI)

Efforts to characterize the late effects of traumatic brain injury (TBI) have been in progress for some time. In recent years much of this activity has been directed towards reporting of chronic traumatic encephalopathy (CTE) in former contact sports athletes and others exposed to repetitive head impacts. However, the association between TBI and dementia risk has long been acknowledged outside of contact sports. Further, growing experience suggests a complex of neurodegenerative pathologies in those surviving TBI, which extends beyond CTE. Nevertheless, despite extensive research, we have scant knowledge of the mechanisms underlying TBI-related neurodegeneration (TReND) and its link to dementia. In part, this is due to the limited number of human brain samples linked to robust demographic and clinical information available for research. Here we detail a National Institutes for Neurological Disease and Stroke Center Without Walls project, the COllaborative Neuropathology NEtwork Characterizing ouTcomes of TBI (CONNECT-TBI), designed to address current limitations in tissue and research access and to advance understanding of the neuropathologies of TReND. As an international, multidisciplinary collaboration CONNECT-TBI brings together multiple experts across 13 institutions. In so doing, CONNECT-TBI unites the existing, comprehensive clinical and neuropathological datasets of multiple established research brain archives in TBI, with survivals ranging minutes to many decades and spanning diverse injury exposures. These existing tissue specimens will be supplemented by prospective brain banking and contribute to a centralized route of access to human tissue for research for investigators. Importantly, each new case will be subject to consensus neuropathology review by the CONNECT-TBI Expert Pathology Group. Herein we set out the CONNECT-TBI program structure and aims and, by way of an illustrative case, the approach to consensus evaluation of new case donations.

Evaluating the association between brain atrophy, hypometabolism, and cognitive decline in Alzheimer's disease: a PET/MRI study

Glucose metabolism reduction and brain volume losses are widely reported in Alzheimer’s disease (AD). Considering that neuroimaging changes in the hippocampus and default mode network (DMN) are promising important candidate biomarkers and have been included in the research criteria for the diagnosis of AD, it is hypothesized that atrophy and metabolic changes of the abovementioned regions could be evaluated concurrently to fully explore the neural mechanisms underlying cognitive impairment in AD. Twenty-three AD patients and Twenty-four age-, sex- and education level-matched normal controls underwent a clinical interview, a detailed neuropsychological assessment and a simultaneous 18F-fluoro-2-deoxy-D-glucose positron emission tomography (18F-FDG PET)/high-resolution T1-weighted magnetic resonance imaging (MRI) scan on a hybrid GE SIGNA PET/MR scanner. Brain volume and glucose metabolism were examined in patients and controls to reveal group differences. Multiple linear regression models were employed to explore the relationship between multiple imaging features and cognitive performance in AD. The AD group had significantly reduced volume in the hippocampus and DMN regions (P < 0.001) relative to that of normal controls determined by using ROI analysis. Compared to normal controls, significantly decreased metabolism in the DMN (P < 0.001) was also found in AD patients, which still survived after controlling for gray matter atrophy (P < 0.001). These findings from ROI analysis were further confirmed by whole-brain confirmatory analysis (P < 0.001, FWE-corrected). Finally, multiple linear regression results showed that impairment of multiple cognitive tasks was significantly correlated with the combination of DMN hypometabolism and atrophy in the hippocampus and DMN regions. This study demonstrated that combining functional and structural features can better explain the cognitive decline of AD patients than unimodal FDG or brain volume changes alone. These findings may have important implications for understanding the neural mechanisms of cognitive decline in AD.

Prognostic value of complementary biomarkers of neurodegeneration in a mixed memory clinic cohort

Background

Biomarkers of neurodegeneration, e.g. MRI brain atrophy and [¹⁸F]FDG-PET hypometabolism, are often evaluated in patients suspected of neurodegenerative disease.

Objective

Our primary objective was to investigate prognostic properties of atrophy and hypometabolism.

Methods

From March 2015-June 2016, 149 patients referred to a university hospital memory clinic were included. The primary outcome was progression/stable disease course as assessed by a clinician at 12 months follow-up. Intracohort defined z-scores of baseline MRI automatic quantified volume and [¹⁸F]FDG-PET standardized uptake value ratios were calculated for all unilaterally defined brain lobes and dichotomized as pronounced atrophy (+A)/ pronounced hypometabolism (+H) at z-score <0. A logistic regression model with progression status as the outcome was carried out with number of lobes with the patterns +A/-H, -A/+H, +A/+H respectively as predictors. The model was mutually adjusted along with adjustment for age and sex. A sensitivity analysis with a z-score dichotomization at −0.1 and −0.5 and dichotomization regarding number of lobes affected at one and three lobes was done.

Results

Median follow-up time was 420 days [IQR: 387-461 days] and 50 patients progressed. Patients with two or more lobes affected by the pattern +A/+H compared to patients with 0–1 lobes affected had a statistically significant increased risk of progression (odds ratio, 95 % confidence interval: 4.33, 1.90–9.86) in a multivariable model. The model was partially robust to the applied sensitivity analysis.

Conclusion

Combined atrophy and hypometabolism as assessed by MRI and [¹⁸F]FDG-PET in patients under suspicion of neurodegenerative disease predicts progression over 1 year.

MVPANI: A Toolkit With Friendly Graphical User Interface for Multivariate Pattern Analysis of Neuroimaging Data

With the rapid development of machine learning techniques, multivariate pattern analysis (MVPA) is becoming increasingly popular in the field of neuroimaging data analysis. Several software packages have been developed to facilitate its application in neuroimaging studies. As most of these software packages are based on command lines, researchers are required to learn how to program, which has greatly limited the use of MVPA for researchers without programming skills. Moreover, lacking a graphical user interface (GUI) also hinders the standardization of the application of MVPA in neuroimaging studies and, consequently, the replication of previous studies or comparisons of results between different studies. Therefore, we developed a GUI-based toolkit for MVPA of neuroimaging data: MVPANI (MVPA for Neuroimaging). Compared with other existing software packages, MVPANI has several advantages. First, MVPANI has a GUI and is, thus, more friendly for non-programmers. Second, MVPANI offers a variety of machine learning algorithms with the flexibility of parameter modification so that researchers can test different algorithms and tune parameters to identify the most suitable algorithms and parameters for their own data. Third, MVPANI also offers the function of data fusion at two levels (feature level or decision level) to utilize complementary information contained in different measures obtained from multimodal neuroimaging techniques. In this paper, we introduce this toolkit and provide four examples to demonstrate its usage, including (1) classification between patients and controls, (2) identification of brain areas containing discriminating information, (3) prediction of clinical scores, and (4) multimodal data fusion.

Recollection-related hippocampal fMRI effects predict longitudinal memory change in healthy older adults

Prior fMRI studies have reported relationships between memory-related activity in the hippocampus and in-scanner memory performance, but whether such activity is predictive of longitudinal memory change remains unclear. Here, we administered a neuropsychological test battery to a sample of cognitively healthy older adults on three occasions, the second and third sessions occurring one month and three years after the first session. Structural and functional MRI data were acquired between the first two sessions. The fMRI data were derived from an associative recognition procedure and allowed estimation of hippocampal effects associated with both successful associative encoding and successful associative recognition (recollection). Baseline memory performance and memory change were evaluated using memory component scores derived from a principal components analysis of the neuropsychological test scores. Across participants, right hippocampal encoding effects correlated significantly with baseline memory performance after controlling for chronological age. Additionally, both left and right hippocampal associative recognition effects correlated negatively with longitudinal memory decline after controlling for age, and the relationship with the left hippocampal effect remained after also controlling for left hippocampal volume. Thus, in cognitively healthy older adults, the magnitude of hippocampal recollection effects appears to be a robust predictor of future memory change.

Recognition Memory Dysfunction Relates to Hippocampal Subfield Volume: A Study of Cognitively Normal and Mildly Impaired Older Adults

OBJECTIVES

The current study examined recognition memory dysfunction and its neuroanatomical substrates in cognitively normal older adults and those diagnosed with mild cognitive impairment (MCI).

METHODS

Participants completed the Mnemonic Similarity Task, which provides simultaneous measures of recognition memory and mnemonic discrimination. They also underwent structural neuroimaging to assess volume of medial temporal cortex and hippocampal subfields.

RESULTS

As expected, individuals diagnosed with MCI had significantly worse recognition memory performance and reduced volume across medial temporal cortex and hippocampal subfields relative to cognitively normal older adults. After controlling for diagnostic group differences, however, recognition memory was significantly related to whole hippocampus volume, and to volume of the dentate gyrus/CA3 subfield in particular. Recognition memory was also related to mnemonic discrimination, a fundamental component of episodic memory that has previously been linked to dentate gyrus/CA3 structure and function.

DISCUSSION

Results reveal that hippocampal subfield volume is sensitive to individual differences in recognition memory in older adults independent of clinical diagnosis. This supports the notion that episodic memory declines along a continuum within this age group, not just between diagnostic groups.

Memory quality modulates the effect of aging on memory consolidation during sleep: Reduced maintenance but intact gain

Successful consolidation of associative memories relies on the coordinated interplay of slow oscillations and sleep spindles during non-rapid eye movement (NREM) sleep. This enables the transfer of labile information from the hippocampus to permanent memory stores in the neocortex. During senescence, the decline of the structural and functional integrity of the hippocampus and neocortical regions is paralleled by changes of the physiological events that stabilize and enhance associative memories during NREM sleep. However, the currently available evidence is inconclusive as to whether and under which circumstances memory consolidation is impacted during aging. To approach this question, 30 younger adults (19-28 years) and 36 older adults (63-74 years) completed a memory task based on scene-word associations. By tracing the encoding quality of participants' individual memory associations, we demonstrate that previous learning determines the extent of age-related impairments in memory consolidation. Specifically, the detrimental effects of aging on memory maintenance were greatest for mnemonic contents of intermediate encoding quality, whereas memory gain of poorly encoded memories did not differ by age. Ambulatory polysomnography (PSG) and structural magnetic resonance imaging (MRI) data were acquired to extract potential predictors of memory consolidation from each participant's NREM sleep physiology and brain structure. Partial Least Squares Correlation was used to identify profiles of interdependent alterations in sleep physiology and brain structure that are characteristic for increasing age. Across age groups, both the 'aged' sleep profile, defined by decreased slow-wave activity (0.5-4.5 ​Hz), and a reduced presence of slow oscillations (0.5-1 ​Hz), slow, and fast spindles (9-12.5 ​Hz; 12.5-16 ​Hz), as well as the 'aged' brain structure profile, characterized by gray matter reductions in the medial prefrontal cortex, thalamus, entorhinal cortex, and hippocampus, were associated with reduced memory maintenance. However, inter-individual differences in neither sleep nor structural brain integrity alone qualified as the driving force behind age differences in sleep-dependent consolidation in the present study. Our results underscore the need for novel and age-fair analytic tools to provide a mechanistic understanding of age differences in memory consolidation.

Deterioration from healthy to mild cognitive impairment and Alzheimer's disease mirrored in corresponding loss of centrality in directed brain networks

OBJECTIVE

It is important to identify brain-based biomarkers that progressively deteriorate from healthy to mild cognitive impairment (MCI) to Alzheimer's disease (AD). Cortical thickness, amyloid-ß deposition, and graph measures derived from functional connectivity (FC) networks obtained using functional MRI (fMRI) have been previously identified as potential biomarkers. Specifically, in the latter case, betweenness centrality (BC), a nodal graph measure quantifying information flow, is reduced in both AD and MCI. However, all such reports have utilized BC calculated from undirected networks that characterize synchronization rather than information flow, which is better characterized using directed networks.

METHODS

Therefore, we estimated BC from directed networks using Granger causality (GC) on resting-state fMRI data (N = 132) to compare the following populations (p < 0.05, FDR corrected for multiple comparisons): normal control (NC), early MCI (EMCI), late MCI (LMCI) and AD. We used an additional metric called middleman power (MP), which not only characterizes nodal information flow as in BC, but also measures nodal power critical for information flow in the entire network.

RESULTS

MP detected more brain regions than BC that progressively deteriorated from NC to EMCI to LMCI to AD, as well as exhibited significant associations with behavioral measures. Additionally, graph measures obtained from conventional FC networks could not identify a single node, underscoring the relevance of GC.

CONCLUSION

Our findings demonstrate the superiority of MP over BC as well as GC over FC in our case. MP obtained from GC networks could serve as a potential biomarker for progressive deterioration of MCI and AD.

The effect of white matter signal abnormalities on default mode network connectivity in mild cognitive impairment

Regions within the default mode network (DMN) are particularly vulnerable to Alzheimer's disease pathology and mechanisms of DMN disruption in mild cognitive impairment (MCI) are still unclear. White matter lesions are presumed to be mechanistically linked to vascular dysfunction whereas cortical atrophy may be related to neurodegeneration. We examined associations between DMN seed‐based connectivity, white matter lesion load, and cortical atrophy in MCI and cognitively healthy controls. MCI showed decreased functional connectivity (FC) between the precuneus‐seed and bilateral lateral temporal cortex (LTC), medial prefrontal cortex (mPFC), posterior cingulate cortex, and inferior parietal lobe compared to those with controls. When controlling for white matter lesion volume, DMN connectivity differences between groups were diminished within bilateral LTC, although were significantly increased in the mPFC explained by significant regional associations between white matter lesion volume and DMN connectivity only in the MCI group. When controlling for cortical thickness, DMN FC was similarly decreased across both groups. These findings suggest that white matter lesions and cortical atrophy are differentially associated with alterations in FC patterns in MCI. Associations between white matter lesions and DMN connectivity in MCI further support at least a partial but important vascular contribution to age‐associated neural and cognitive impairment.

CRISPR/Cas9-mediated disruption of SHANK3 in monkey leads to drug-treatable autism-like symptoms

Monogenic mutations in the SHANK3 gene, which encodes a postsynaptic scaffold protein, play a causative role in autism spectrum disorder (ASD). Although a number of mouse models with Shank3 mutations have been valuable for investigating the pathogenesis of ASD, species-dependent differences in behaviors and brain structures post considerable challenges to use small animals to model ASD and to translate experimental therapeutics to the clinic. We have used clustered regularly interspersed short palindromic repeat/CRISPR-associated nuclease 9 to generate a cynomolgus monkey model by disrupting SHANK3 at exons 6 and 12. Analysis of the live mutant monkey revealed the core behavioral abnormalities of ASD, including impaired social interaction and repetitive behaviors, and reduced brain network activities detected by positron-emission computed tomography (PET). Importantly, these abnormal behaviors and brain activities were alleviated by the antidepressant fluoxetine treatment. Our findings provide the first demonstration that the genetically modified non-human primate can be used for translational research of therapeutics for ASD.

Fractionating the Rey Auditory Verbal Learning Test: Distinct roles of large-scale cortical networks in prodromal Alzheimer's disease

Successful episodic memory calls upon a number of different cognitive processes that are supported by the coordination of several large-scale cortical networks. Previous work from our group has demonstrated dissociable anatomic substrates at different stages of memory in patients with dementia due to Alzheimer's disease (AD). The aim of the current study was to extend the understanding of brain-behavior associations underlying a commonly administered neuropsychological assessment of verbal episodic memory (Rey Auditory Verbal Learning Test; RAVLT) by determining the cortical network contributions to the performance at early vs. late stages of list learning, delayed recall, and retention, in 235 very mild biomarker positive (A+/T+/N+) individuals diagnosed with amnestic mild cognitive impairment (aMCI; MMSE = 27.7). We measured cortical atrophy in four large-scale cortical networks impacted by AD: default mode (DMN), dorsal attention (DAN), frontoparietal (FPN), and language (LN) networks. We also evaluated the role of hippocampal atrophy at each stage of memory performance. Partial correlation analyses controlling for age, sex, and education and corrected for multiple comparisons revealed that early learning was most strongly associated with cortical thickness in the DAN, while late learning was most strongly associated with hippocampal volume, but also related to cortical thickness in the DAN, FPN, DMN, and LN. Delayed recall was associated most strongly with hippocampal volume, but was also related to cortical thickness in the FPN and DMN, while retention was associated only with hippocampal volume. These findings are consistent with prior models of the neural substrates of different stages of verbal list learning and retrieval, provide new insights into the cortical networks undergoing neurodegeneration even at very mild stages of prodromal AD, and inform our thinking about the networks and regions being interrogated by this kind of neuropsychological assessment of episodic memory.

Association of short-term cognitive decline and MCI-to-AD dementia conversion with CSF, MRI, amyloid- and 18F-FDG-PET imaging

Disease-modifying treatment trials are increasingly advanced to the prodromal or preclinical phase of Alzheimer's disease (AD), and inclusion criteria are based on biomarkers rather than clinical symptoms. Therefore, it is of great interest to determine which biomarkers should be combined to accurately predict conversion from mild cognitive impairment (MCI) to AD dementia. However, up to date, only few studies performed a complete A/T/N subject characterization using each of the CSF and imaging markers, or they only investigated long-term (≥ 2 years) prognosis. This study aimed to investigate the association between cerebrospinal fluid (CSF), magnetic resonance imaging (MRI), amyloid- and ¹⁸F-FDG positron emission tomography (PET) measures at baseline, in relation to cognitive changes and conversion to AD dementia over a short-term (12-month) period. We included 13 healthy controls, 49 MCI and 16 AD dementia patients with a clinical-based diagnosis and a complete A/T/N characterization at baseline. Global cortical amyloid-β (Aβ) burden was quantified using the ¹⁸F-AV45 standardized uptake value ratio (SUVR) with two different reference regions (cerebellar grey and subcortical white matter), whereas metabolism was assessed based on ¹⁸F-FDG SUVR. CSF measures included Aβ_1–42, Aβ_1–40, T-tau, P-tau₁₈₁, and their ratios, and MRI markers included hippocampal volumes (HV), white matter hyperintensities, and cortical grey matter volumes. Cognitive functioning was measured by MMSE and RBANS index scores. All statistical analyses were corrected for age, sex, education, and APOE ε4 genotype. As a result, faster cognitive decline was most strongly associated with hypometabolism (posterior cingulate) and smaller hippocampal volume (e.g., Δstory recall: β = +0.43 [p < 0.001] and + 0.37 [p = 0.005], resp.) at baseline. In addition, faster cognitive decline was significantly associated with higher baseline Aβ burden only if SUVR was referenced to the subcortical white matter (e.g., Δstory recall: β = −0.28 [p = 0.020]). Patients with MCI converted to AD dementia at an annual rate of 31%, which could be best predicted by combining neuropsychological testing (visuospatial construction skills) with either MRI-based HV or ¹⁸F-FDG-PET. Combining all three markers resulted in 96% specificity and 92% sensitivity. Neither amyloid-PET nor CSF biomarkers could discriminate short-term converters from non-converters.

•

FDG-PET and MRI HV are the strongest predictors of cognitive decline and conversion to AD.

•

Combination of visuospatial construction testing with FDG-PET or MRI HV present high predicting power of conversion.

•

CSF and amyloid-PET seem less suitable markers of disease progression.

•

Increased AV45-PET predicts short-term cognitive decline if SUVR is referenced to WM instead of CB.

Region-specific association between basal blood insulin and cerebral glucose metabolism in older adults

Background

Although previous studies have suggested that insulin plays a role in brain function, it still remains unclear whether or not insulin has a region-specific association with neuronal and synaptic activity in the living human brain. We investigated the regional pattern of association between basal blood insulin and resting-state cerebral glucose metabolism (CMglu), a proxy for neuronal and synaptic activity, in older adults.

Method

A total of 234 nondiabetic, cognitively normal (CN) older adults underwent comprehensive clinical assessment, resting-state 18F-fluodeoxyglucose (FDG)-positron emission tomography (PET) and blood sampling to determine overnight fasting blood insulin and glucose levels, as well as apolipoprotein E (APOE) genotyping.

Results

An exploratory voxel-wise analysis of FDG-PET without a priori hypothesis demonstrated a positive association between basal blood insulin levels and resting-state CMglu in specific cerebral cortices and hippocampus, rather than in non-specific overall cerebral regions, even after controlling for the effects of APOE e4 carrier status, vascular risk factor score, body mass index, fasting blood glucose, and demographic variables. Particularly, a positive association of basal blood insulin with CMglu in the right posterior hippocampus and adjacent parahippocampal region as well as in the right inferior parietal region remained significant after multiple comparison correction. Conversely, no region showed negative association between basal blood insulin and CMglu.

Conclusions

Our finding suggests that basal fasting blood insulin may have association with neuronal and synaptic activity in specific cerebral regions, particularly in the hippocampal/parahippocampal and inferior parietal regions.

•

We investigated regional pattern of association between basal blood insulin and resting-state cerebral glucose metabolism.

•

Significant clusters with positive associations were found mainly in the hippocampal and inferior parietal regions.

•

Our finding suggests a region-specific association of basal blood insulin with resting-state cerebral glucose metabolism.

•

Further studies to elucidate underlying mechanism and implication of this region-specific association will be necessary.

Young and Older Adults May Utilize Different Cognitive Abilities When Performing a Spatial Recognition Memory Test With Varying Levels of Similarity

We used signal detection theory to evaluate spatial recognition memory utilizing a behavioral test hypothesized to tax pattern separation. Correlations with standardized neuropsychological tests also were examined. Healthy young ( n = 40) and older ( n = 30) adults completed a spatial recognition memory test involving high- and low-similarity conditions. Using d’ as the dependent variable, we found that older adults were significantly impaired relative to young adults on the high- and low-similarity conditions ( ps < .05). Both groups performed significantly better in the low-similarity condition compared to the high-similarity condition ( p < .05), with young adults exhibiting greater improvement relative to older adults. We also found that young adults may rely on spatial attention abilities when performing our test, while older adults might rely on memory and executive function abilities. These findings indicate that young and older adults may utilize different cognitive abilities when performing certain spatial memory tests.

Reduced Gray Matter Volume of the Thalamus and Hippocampal Region in Elderly Healthy Adults with no Impact of APOE ɛ4: A Longitudinal Voxel-Based Morphometry Study

BACKGROUND

Many cross-sectional voxel-based morphometry (VBM) investigations have shown significant inverse correlations between chronological age and gray matter (GM) volume in several brain regions in healthy humans. However, few VBM studies have documented GM decrements in the healthy elderly with repeated MRI measurements obtained in the same subjects. Also, the extent to which the APOE ɛ4 allele influences longitudinal findings of GM reduction in the healthy elderly is unclear.

OBJECTIVE

Verify whether regional GM changes are associated with significant decrements in cognitive performance taking in account the presence of the APOE ɛ4 allele.

METHODS

Using structural MRI datasets acquired in 55 cognitively intact elderly subjects at two time-points separated by approximately three years, we searched for voxels showing significant GM reductions taking into account differences in APOE genotype.

RESULTS

We found global GM reductions as well as regional GM decrements in the right thalamus and left parahippocampal gyrus (p < 0.05, family-wise error corrected for multiple comparisons over the whole brain). These findings were not affected by APOE ɛ4.

CONCLUSIONS

Irrespective of APOE ɛ4, longitudinal VBM analyses show that the hippocampal region and thalamus are critical sites where GM shrinkage is greater than the degree of global volume reduction in healthy elderly subjects.

The Role of MRI Biomarkers and Their Interactions with Cognitive Status and APOE ε4 in Nondemented Elderly Subjects

PURPOSE

(1) To investigate atrophy patterns of hippocampal subfield volume and Alzheimer's disease (AD)-signature cortical thickness in mild cognitive impairment (MCI) patients; (2) to explore the association between the neuropsychological (NP) and the brain structure in the MCI and older normal cognition group; (3) to determine whether these associations were modified by the apolipoprotein E (APOE) ε4 gene and cognitive status.

METHODS

The FreeSurfer software was used for automated segmentation of hippocampal subfields and AD-signature cortical thickness for 22 MCI patients and 23 cognitive normal controls (NC). The volume, cortical thickness, and the neuropsychological scale were compared with two-sample t tests. Linear regression models were used to determine the association between the NP and the brain structure.

RESULTS

Compared with the NC group, MCI patients showed a decreased volume of the left presubiculum, subiculum and right CA2_3 and CA4_DG (p < 0.05, FDR corrected). The volume of these regions was positively correlated with NP scores. Of note, these associations depended on the cognitive status but not on the APOE ε4 status. The left subiculum and presubiculum volume were positively correlated with the Montreal Cognitive Assessment (MoCA) scores only in the MCI patients.

CONCLUSION

Atrophy of the hippocampal subfields may be a powerful biomarker for MCI in the Chinese population.

Relationship between Cortical Thickness and Neuropsychological Performance in Normal Older Adults and Those with Mild Cognitive Impairment

Mild cognitive impairment (MCI) has been extensively investigated in recent decades to identify groups with a high risk of dementia and to establish effective prevention methods during this period. Neuropsychological performance and cortical thickness are two important biomarkers used to predict progression from MCI to dementia. This study compares the cortical thickness and neuropsychological performance in people with MCI and cognitively healthy older adults. We further focus on the relationship between cortical thickness and neuropsychological performance in these two groups. Forty-nine participants with MCI and 40 cognitively healthy older adults were recruited. Cortical thickness was analysed with semiautomatic software, Freesurfer. The analysis reveals that the cortical thickness in the left caudal anterior cingulate (p=0.041), lateral occipital (p=0.009) and right superior temporal (p=0.047) areas were significantly thinner in the MCI group after adjustment for age and education. Almost all neuropsychological test results (with the exception of forward digit span) were significantly correlated to cortical thickness in the MCI group after adjustment for age, gender and education. In contrast, only the score on the Category Verbal Fluency Test and the forward digit span were found to have significant inverse correlations to cortical thickness in the control group of cognitively healthy older adults. The study results suggest that cortical thinning in the temporal region reflects the global change in cognition in subjects with MCI and may be useful to predict progression of MCI to Alzheimer’s disease. The different pattern in the correlation of cortical thickness to the neuropsychological performance of patients with MCI from the healthy control subjects may be explained by the hypothesis of MCI as a disconnection syndrome.

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).

Multilevel Feature Representation of FDG-PET Brain Images for Diagnosing Alzheimer's Disease

Using a single imaging modality to diagnose Alzheimer's disease (AD) or mild cognitive impairment (MCI) is a challenging task. FluoroDeoxyGlucose Positron Emission Tomography (FDG-PET) is an important and effective modality used for that purpose. In this paper, we develop a novel method by using single modality (FDG-PET) but multilevel feature, which considers both region properties and connectivities between regions to classify AD or MCI from normal control. First, three levels of features are extracted: statistical, connectivity, and graph-based features. Then, the connectivity features are decomposed into three different sets of features according to a proposed similarity-driven ranking method, which can not only reduce the feature dimension but also increase the classifier's diversity. Last, after feeding the three levels of features to different classifiers, a new classifier selection strategy, maximum Mean squared Error (mMsE), is developed to select a pair of classifiers with high diversity. In order to do the majority voting, a decision-making scheme, a nested cross validation technique is applied to choose another classifier according to the accuracy. Experiments on Alzheimer's Disease Neuroimaging Initiative database show that the proposed method outperforms most FDG-PET-based classification algorithms, especially for classifying progressive MCI (pMCI) from stable MCI (sMCI).

Multimodal cortical and hippocampal prediction of episodic-memory plasticity in young and older adults

Episodic memory can be trained in both early and late adulthood, but there is considerable variation in cognitive improvement across individuals. Which brain characteristics make some individuals benefit more than others? We used a multimodal approach to investigate whether volumetric magnetic resonance imaging (MRI) and resting-state functional MRI characteristics of the cortex and hippocampus, brain regions involved in episodic-memory function, were predictive of cognitive improvement after memory training. We hypothesized that these brain characteristics would differentially predict memory improvement in young and older adults, given the vulnerability of cortical regions as well as the hippocampus to healthy aging. Following structural and resting-state activity magnetic resonance scans, 50 young and 76 older participants completed 10 weeks of strategic episodic-memory training. Both age groups improved their memory performance, but the young adults more so than the older. Vertex-wise analyses of cortical volume showed no significant relation to memory benefit. When analyzing the two age groups separately, hippocampal volume was predictive of memory improvement in the group of older participants only. In this age group, the lower resting-state activity of the hippocampus was also predictive of memory improvement. Both volumetric and resting-state characteristics of the hippocampus explained unique variance of the improvement in the older participants suggesting that a multimodal imaging approach is valuable for the understanding of mechanisms underlying memory plasticity in aging.

Multimodal Discrimination between Normal Aging, Mild Cognitive Impairment and Alzheimer's Disease and Prediction of Cognitive Decline

Alzheimer’s Disease (AD) and mild cognitive impairment (MCI) are associated with widespread changes in brain structure and function, as indicated by magnetic resonance imaging (MRI) morphometry and 18-fluorodeoxyglucose position emission tomography (FDG PET) metabolism. Nevertheless, the ability to differentiate between AD, MCI and normal aging groups can be difficult. Thus, the goal of this study was to identify the combination of cerebrospinal fluid (CSF) biomarkers, MRI morphometry, FDG PET metabolism and neuropsychological test scores to that best differentiate between a sample of normal aging subjects and those with MCI and AD from the Alzheimer’s Disease Neuroimaging Initiative. The secondary goal was to determine the neuroimaging variables from MRI, FDG PET and CSF biomarkers that can predict future cognitive decline within each group. To achieve these aims, a series of multivariate stepwise logistic and linear regression models were generated. Combining all neuroimaging modalities and cognitive test scores significantly improved the index of discrimination, especially at the earliest stages of the disease, whereas MRI gray matter morphometry variables best predicted future cognitive decline compared to other neuroimaging variables. Overall these findings demonstrate that a multimodal approach using MRI morphometry, FDG PET metabolism, neuropsychological test scores and CSF biomarkers may provide significantly better discrimination than any modality alone.

Multimodality Neuroimaging in Mild Cognitive Impairment: A Cross-sectional Comparison Study

Background and Purpose

Mild cognitive impairment (MCI) is a focus of considerable research. The present study aimed to test the utility of a logistic regression-derived classifier, combining specific quantitative multimodal magnetic resonance imaging (MRI) data for the early objective phenotyping of MCI in the clinic, over structural MRI data.

Methods

Thirty-three participants with cognitively stable amnestic MCI; 15 MCI converters to early Alzheimer's disease (AD; diseased controls) and 20 healthy controls underwent high-resolution T1-weighted volumetric MRI, diffusion tensor imaging (DTI), and proton magnetic resonance spectroscopy (¹H MR spectroscopy). The regional volumes were obtained from T1-weighted MRI. The fractional anisotropy and mean diffusivity maps were derived from DTI over multiple white matter regions. The ¹H MRS voxels were placed over posterior cingulate gyri, and N-acetyl aspartate (NAA)/creatine (Cr), choline (Cho)/Cr, myoinositol (mI/Cr), and NAA/mI ratios were obtained. A multimodal classifier comprising MR volumetry, DTI, and MRS was prepared. A cutoff point was arrived based on receiver operator characteristics analysis. Results were considered significant, if P < 0.05.

Results

The most sensitive individual marker to discriminate MCI from controls was DTI (90.9%), with a specificity of 50%. For classifying MCI from AD, the best individual modality was DTI (72.7%), with a high specificity of 87.9%. The multimodal classifier approach for MCI control classification achieved an area under curve (AUC) (AUC = 0.89; P < 0.001), with 93.9% sensitivity and 70% specificity. The combined classifier for MCI-AD achieved a highest AUC (AUC = 0.93; P < 0.001), with 93% sensitivity and 85.6% specificity.

Conclusions

The combined method of gray matter atrophy, white matter tract changes, and metabolite variation achieved a better performance at classifying MCI compared to the application of individual MRI biomarkers.

Discriminative multi-task feature selection for multi-modality classification of Alzheimer's disease

Recently, multi-task based feature selection methods have been used in multi-modality based classification of Alzheimer's disease (AD) and its prodromal stage, i.e., mild cognitive impairment (MCI). However, in traditional multi-task feature selection methods, some useful discriminative information among subjects is usually not well mined for further improving the subsequent classification performance. Accordingly, in this paper, we propose a discriminative multi-task feature selection method to select the most discriminative features for multi-modality based classification of AD/MCI. Specifically, for each modality, we train a linear regression model using the corresponding modality of data, and further enforce the group-sparsity regularization on weights of those regression models for joint selection of common features across multiple modalities. Furthermore, we propose a discriminative regularization term based on the intra-class and inter-class Laplacian matrices to better use the discriminative information among subjects. To evaluate our proposed method, we perform extensive experiments on 202 subjects, including 51 AD patients, 99 MCI patients, and 52 healthy controls (HC), from the baseline MRI and FDG-PET image data of the Alzheimer's Disease Neuroimaging Initiative (ADNI). The experimental results show that our proposed method not only improves the classification performance, but also has potential to discover the disease-related biomarkers useful for diagnosis of disease, along with the comparison to several state-of-the-art methods for multi-modality based AD/MCI classification.

Multi-modal discriminative dictionary learning for Alzheimer's disease and mild cognitive impairment

BACKGROUND AND OBJECTIVE

The differentiation of mild cognitive impairment (MCI), which is the prodromal stage of Alzheimer's disease (AD), from normal control (NC) is important as the recent research emphasis on early pre-clinical stage for possible disease abnormality identification, intervention and even possible prevention.

METHODS

The current study puts forward a multi-modal supervised within-class-similarity discriminative dictionary learning algorithm (SCDDL) we introduced previously for distinguishing MCI from NC. The proposed new algorithm was based on weighted combination and named as multi-modality SCDDL (mSCDDL). Structural magnetic resonance imaging (sMRI), fluorodeoxyglucose (FDG) positron emission tomography (PET) and florbetapir PET data of 113 AD patients, 110 MCI patients and 117 NC subjects from the Alzheimer's disease Neuroimaging Initiative database were adopted for classification between MCI and NC, as well as between AD and NC.

RESULTS

Adopting mSCDDL, the classification accuracy achieved 98.5% for AD vs. NC and 82.8% for MCI vs. NC, which were superior to or comparable with the results of some other state-of-the-art approaches as reported in recent multi-modality publications.

CONCLUSIONS

The mSCDDL procedure was a promising tool in assisting early diseases diagnosis using neuroimaging data.

Early identification of MCI converting to AD: a FDG PET study

PURPOSE

Mild cognitive impairment (MCI) is a transitional pathological stage between normal ageing (NA) and Alzheimer's disease (AD). Although subjects with MCI show a decline at different rates, some individuals remain stable or even show an improvement in their cognitive level after some years. We assessed the accuracy of FDG PET in discriminating MCI patients who converted to AD from those who did not.

METHODS

FDG PET was performed in 42 NA subjects, 27 MCI patients who had not converted to AD at 5 years (nc-MCI; mean follow-up time 7.5 ± 1.5 years), and 95 MCI patients who converted to AD within 5 years (MCI-AD; mean conversion time 1.8 ± 1.1 years). Relative FDG uptake values in 26 meta-volumes of interest were submitted to ANCOVA and support vector machine analyses to evaluate regional differences and discrimination accuracy.

RESULTS

The MCI-AD group showed significantly lower FDG uptake values in the temporoparietal cortex than the other two groups. FDG uptake values in the nc-MCI group were similar to those in the NA group. Support vector machine analysis discriminated nc-MCI from MCI-AD patients with an accuracy of 89% (AUC 0.91), correctly detecting 93% of the nc-MCI patients.

CONCLUSION

In MCI patients not converting to AD within a minimum follow-up time of 5 years and MCI patients converting within 5 years, baseline FDG PET and volume-based analysis identified those who converted with an accuracy of 89%. However, further analysis is needed in patients with amnestic MCI who convert to a dementia other than AD.

Role of testosterone and Y chromosome genes for the masculinization of the human brain

Women with complete androgen insensitivity syndrome (CAIS) have a male (46,XY) karyotype but no functional androgen receptors. Their condition, therefore, offers a unique model for studying testosterone effects on cerebral sex dimorphism. We present MRI data from 16 women with CAIS and 32 male (46,XY) and 32 female (46,XX) controls.

METHODS

FreeSurfer software was employed to measure cortical thickness and subcortical structural volumes. Axonal connections, indexed by fractional anisotropy, (FA) were measured with diffusion tensor imaging, and functional connectivity with resting state fMRI.

RESULTS

Compared to men, CAIS women displayed a "female" pattern by having thicker parietal and occipital cortices, lower FA values in the right corticospinal, superior and inferior longitudinal tracts, and corpus callosum. Their functional connectivity from the amygdala to the medial prefrontal cortex, was stronger and amygdala-connections to the motor cortex weaker than in control men. CAIS and control women also showed stronger posterior cingulate and precuneus connections in the default mode network. Thickness of the motor cortex, the caudate volume, and the FA in the callosal body followed, however, a "male" pattern.

CONCLUSION

Altogether, these data suggest that testosterone modulates the microstructure of somatosensory and visual cortices and their axonal connections to the frontal cortex. Testosterone also influenced functional connections from the amygdala, whereas the motor cortex could, in agreement with our previous reports, be moderated by processes linked to X-chromosome gene dosage. These data raise the question about other genetic factors masculinizing the human brain than the SRY gene and testosterone. Hum Brain Mapp 38:1801-1814, 2017. © 2017 Wiley Periodicals, Inc.

Prediction of Progressive Mild Cognitive Impairment by Multi-Modal Neuroimaging Biomarkers

For patients with mild cognitive impairment (MCI), the likelihood of progression to probable Alzheimer's disease (AD) is important not only for individual patient care, but also for the identification of participants in clinical trial, so as to provide early interventions. Biomarkers based on various neuroimaging modalities could offer complementary information regarding different aspects of disease progression. The current study adopted a weighted multi-modality sparse representation-based classification method to combine data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database, from three imaging modalities: Volumetric magnetic resonance imaging (MRI), fluorodeoxyglucose (FDG) positron emission tomography (PET), and florbetapir PET. We included 117 normal controls (NC) and 110 MCI patients, 27 of whom progressed to AD within 36 months (pMCI), while the remaining 83 remained stable (sMCI) over the same time period. Modality-specific biomarkers were identified to distinguish MCI from NC and to predict pMCI among MCI. These included the hippocampus, amygdala, middle temporal and inferior temporal regions for MRI, the posterior cingulum, precentral, and postcentral regions for FDG-PET, and the hippocampus, amygdala, and putamen for florbetapir PET. Results indicated that FDG-PET may be a more effective modality in discriminating MCI from NC and in predicting pMCI than florbetapir PET and MRI. Combining modality-specific sensitive biomarkers from the three modalities boosted the discrimination accuracy of MCI from NC (76.7%) and the prediction accuracy of pMCI (82.5%) when compared with the best single-modality results (73.6% for MCI and 75.6% for pMCI with FDG-PET).

Label-aligned multi-task feature learning for multimodal classification of Alzheimer's disease and mild cognitive impairment

Multimodal classification methods using different modalities of imaging and non-imaging data have recently shown great advantages over traditional single-modality-based ones for diagnosis and prognosis of Alzheimer's disease (AD), as well as its prodromal stage, i.e., mild cognitive impairment (MCI). However, to the best of our knowledge, most existing methods focus on mining the relationship across multiple modalities of the same subjects, while ignoring the potentially useful relationship across different subjects. Accordingly, in this paper, we propose a novel learning method for multimodal classification of AD/MCI, by fully exploring the relationships across both modalities and subjects. Specifically, our proposed method includes two subsequent components, i.e., label-aligned multi-task feature selection and multimodal classification. In the first step, the feature selection learning from multiple modalities are treated as different learning tasks and a group sparsity regularizer is imposed to jointly select a subset of relevant features. Furthermore, to utilize the discriminative information among labeled subjects, a new label-aligned regularization term is added into the objective function of standard multi-task feature selection, where label-alignment means that all multi-modality subjects with the same class labels should be closer in the new feature-reduced space. In the second step, a multi-kernel support vector machine (SVM) is adopted to fuse the selected features from multi-modality data for final classification. To validate our method, we perform experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) database using baseline MRI and FDG-PET imaging data. The experimental results demonstrate that our proposed method achieves better classification performance compared with several state-of-the-art methods for multimodal classification of AD/MCI.

Magnetic resonance imaging in Alzheimer's Disease Neuroimaging Initiative 2

INTRODUCTION

Alzheimer's Disease Neuroimaging Initiative (ADNI) is now in its 10th year. The primary objective of the magnetic resonance imaging (MRI) core of ADNI has been to improve methods for clinical trials in Alzheimer's disease (AD) and related disorders.

METHODS

We review the contributions of the MRI core from present and past cycles of ADNI (ADNI-1, -Grand Opportunity and -2). We also review plans for the future-ADNI-3.

RESULTS

Contributions of the MRI core include creating standardized acquisition protocols and quality control methods; examining the effect of technical features of image acquisition and analysis on outcome metrics; deriving sample size estimates for future trials based on those outcomes; and piloting the potential utility of MR perfusion, diffusion, and functional connectivity measures in multicenter clinical trials.

DISCUSSION

Over the past decade the MRI core of ADNI has fulfilled its mandate of improving methods for clinical trials in AD and will continue to do so in the future.

Different partial volume correction methods lead to different conclusions: An (18)F-FDG-PET study of aging

A cross-sectional group study of the effects of aging on brain metabolism as measured with (18)F-FDG-PET was performed using several different partial volume correction (PVC) methods: no correction (NoPVC), Meltzer (MZ), Müller-Gärtner (MG), and the symmetric geometric transfer matrix (SGTM) using 99 subjects aged 65-87years from the Harvard Aging Brain study. Sensitivity to parameter selection was tested for MZ and MG. The various methods and parameter settings resulted in an extremely wide range of conclusions as to the effects of age on metabolism, from almost no changes to virtually all of cortical regions showing a decrease with age. Simulations showed that NoPVC had significant bias that made the age effect on metabolism appear to be much larger and more significant than it is. MZ was found to be the same as NoPVC for liberal brain masks; for conservative brain masks, MZ showed few areas correlated with age. MG and SGTM were found to be similar; however, MG was sensitive to a thresholding parameter that can result in data loss. CSF uptake was surprisingly high at about 15% of that in gray matter. The exclusion of CSF from SGTM and MG models, which is almost universally done, caused a substantial loss in the power to detect age-related changes. This diversity of results reflects the literature on the metabolism of aging and suggests that extreme care should be taken when applying PVC or interpreting results that have been corrected for partial volume effects. Using the SGTM, significant age-related changes of about 7% per decade were found in frontal and cingulate cortices as well as primary visual and insular cortices.

Patterns of effective connectivity during memory encoding and retrieval differ between patients with mild cognitive impairment and healthy older adults

Previous research has shown that there is considerable overlap in the neural networks mediating successful memory encoding and retrieval. However, little is known about how the relevant human brain regions interact during these distinct phases of memory or how such interactions are affected by memory deficits that characterize mild cognitive impairment (MCI), a condition that often precedes dementia due to Alzheimer's disease. Here we employed multivariate Granger causality analysis using autoregressive modeling of inferred neuronal time series obtained by deconvolving the hemodynamic response function from measured blood oxygenation level-dependent (BOLD) time series data, in order to examine the effective connectivity between brain regions during successful encoding and/or retrieval of object location associations in MCI patients and comparable healthy older adults. During encoding, healthy older adults demonstrated a left hemisphere dominant pattern where the inferior frontal junction, anterior intraparietal sulcus (likely involving the parietal eye fields), and posterior cingulate cortex drove activation in most left hemisphere regions and virtually every right hemisphere region tested. These regions are part of a frontoparietal network that mediates top-down cognitive control and is implicated in successful memory formation. In contrast, in the MCI patients, the right frontal eye field drove activation in every left hemisphere region examined, suggesting reliance on more basic visual search processes. Retrieval in the healthy older adults was primarily driven by the right hippocampus with lesser contributions of the right anterior thalamic nuclei and right inferior frontal sulcus, consistent with theoretical models holding the hippocampus as critical for the successful retrieval of memories. The pattern differed in MCI patients, in whom the right inferior frontal junction and right anterior thalamus drove successful memory retrieval, reflecting the characteristic hippocampal dysfunction of these patients. These findings demonstrate that neural network interactions differ markedly between MCI patients and healthy older adults. Future efforts will investigate the impact of cognitive rehabilitation of memory on these connectivity patterns.

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.

Development and aging of cortical thickness correspond to genetic organization patterns

There is a growing realization that early life influences have lasting impact on brain function and structure. Recent research has demonstrated that genetic relationships in adults can be used to parcellate the cortex into regions of maximal shared genetic influence, and a major hypothesis is that genetically programmed neurodevelopmental events cause a lasting impact on the organization of the cerebral cortex observable decades later. Here we tested how developmental and lifespan changes in cortical thickness fit the underlying genetic organizational principles of cortical thickness in a longitudinal sample of 974 participants between 4.1 and 88.5 y of age with a total of 1,633 scans, including 773 scans from children below 12 y. Genetic clustering of cortical thickness was based on an independent dataset of 406 adult twins. Developmental and adult age-related changes in cortical thickness followed closely the genetic organization of the cerebral cortex, with change rates varying as a function of genetic similarity between regions. Cortical regions with overlapping genetic architecture showed correlated developmental and adult age change trajectories and vice versa for regions with low genetic overlap. Thus, effects of genes on regional variations in cortical thickness in middle age can be traced to regional differences in neurodevelopmental change rates and extrapolated to further adult aging-related cortical thinning. This finding suggests that genetic factors contribute to cortical changes through life and calls for a lifespan perspective in research aimed at identifying the genetic and environmental determinants of cortical development and aging.

Multi-modality sparse representation-based classification for Alzheimer's disease and mild cognitive impairment

BACKGROUND AND OBJECTIVE

The discrimination of Alzheimer's disease (AD) and its prodromal stage known as mild cognitive impairment (MCI) from normal control (NC) is important for patients' timely treatment. The simultaneous use of multi-modality data has been demonstrated to be helpful for more accurate identification. The current study focused on extending a multi-modality algorithm and evaluating the method by identifying AD/MCI.

METHODS

In this study, sparse representation-based classification (SRC), a well-developed method in pattern recognition and machine learning, was extended to a multi-modality classification framework named as weighted multi-modality SRC (wmSRC). Data including three modalities of volumetric magnetic resonance imaging (MRI), fluorodeoxyglucose (FDG) positron emission tomography (PET) and florbetapir PET from the Alzheimer's disease Neuroimaging Initiative database were adopted for AD/MCI classification (113 AD patients, 110 MCI patients and 117 NC subjects).

RESULTS

Adopting wmSRC, the classification accuracy achieved 94.8% for AD vs. NC, 74.5% for MCI vs. NC, and 77.8% for progressive MCI vs. stable MCI, superior to or comparable with the results of some other state-of-the-art models in recent multi-modality researches.

CONCLUSIONS

The wmSRC method is a promising tool for classification with multi-modality data. It could be effective for identifying diseases from NC with neuroimaging data, which could be helpful for the timely diagnosis and treatment of diseases.

Construction and comparative evaluation of different activity detection methods in brain FDG-PET

Aim

We constructed and evaluated reference brain FDG-PET databases for usage by three software programs (Computer-aided diagnosis for dementia (CAD4D), Statistical Parametric Mapping (SPM) and NEUROSTAT), which allow a user-independent detection of dementia-related hypometabolism in patients’ brain FDG-PET.

Methods

Thirty-seven healthy volunteers were scanned in order to construct brain FDG reference databases, which reflect the normal, age-dependent glucose consumption in human brain, using either software. Databases were compared to each other to assess the impact of different stereotactic normalization algorithms used by either software package. In addition, performance of the new reference databases in the detection of altered glucose consumption in the brains of patients was evaluated by calculating statistical maps of regional hypometabolism in FDG-PET of 20 patients with confirmed Alzheimer’s dementia (AD) and of 10 non-AD patients. Extent (hypometabolic volume referred to as cluster size) and magnitude (peak z-score) of detected hypometabolism was statistically analyzed.

Results

Differences between the reference databases built by CAD4D, SPM or NEUROSTAT were observed. Due to the different normalization methods, altered spatial FDG patterns were found. When analyzing patient data with the reference databases created using CAD4D, SPM or NEUROSTAT, similar characteristic clusters of hypometabolism in the same brain regions were found in the AD group with either software. However, larger z-scores were observed with CAD4D and NEUROSTAT than those reported by SPM. Better concordance with CAD4D and NEUROSTAT was achieved using the spatially normalized images of SPM and an independent z-score calculation. The three software packages identified the peak z-scores in the same brain region in 11 of 20 AD cases, and there was concordance between CAD4D and SPM in 16 AD subjects.

Conclusion

The clinical evaluation of brain FDG-PET of 20 AD patients with either CAD4D-, SPM- or NEUROSTAT-generated databases from an identical reference dataset showed similar patterns of hypometabolism in the brain regions known to be involved in AD. The extent of hypometabolism and peak z-score appeared to be influenced by the calculation method used in each software package rather than by different spatial normalization parameters.