Mohamad Habes

A review on neuroimaging-based classification studies and associated feature extraction methods for Alzheimer's disease and its prodromal stages

ABSTRACT Neuroimaging has made it possible to measure pathological brain changes associated with Alzheimers disease (AD) in vivo. Over the past decade, these measures have been increasingly integrated into imaging signatures of AD by means of classification frameworks, offering promising tools for individualized diagnosis and prognosis. We reviewed neuroimaging‐based studies for AD and mild cognitive impairment classification, selected after online database searches in Google Scholar and PubMed (January, 1985–June, 2016). We categorized these studies based on the following neuroimaging modalities (and sub‐categorized based on features extracted as a post‐processing step from these modalities): i) structural magnetic resonance imaging [MRI] (tissue density, cortical surface, and hippocampal measurements), ii) functional MRI (functional coherence of different brain regions, and the strength of the functional connectivity), iii) diffusion tensor imaging (patterns along the white matter fibers), iv) fluorodeoxyglucose positron emission tomography (FDG‐PET) (metabolic rate of cerebral glucose), and v) amyloid‐PET (amyloid burden). The studies reviewed indicate that the classification frameworks formulated on the basis of these features show promise for individualized diagnosis and prediction of clinical progression. Finally, we provided a detailed account of AD classification challenges and addressed some future research directions. HIGHLIGHTSWe reviewed Alzheimers disease neuroimaging‐based classification studies.We covered structural MRI, fMRI, DTI, amyloid‐PET, FDG‐PET, and multimodalities.The reported studies were validated through appropriate cross‐validation.We categorized the studies based on feature extraction methods.We discussed challenges, disparities in experimental conditions and future directions.

White matter hyperintensities and imaging patterns of brain ageing in the general population

White matter hyperintensities are associated with increased risk of dementia and cognitive decline. The current study investigates the relationship between white matter hyperintensities burden and patterns of brain atrophy associated with brain ageing and Alzheimers disease in a large populatison-based sample (n = 2367) encompassing a wide age range (20-90 years), from the Study of Health in Pomerania. We quantified white matter hyperintensities using automated segmentation and summarized atrophy patterns using machine learning methods resulting in two indices: the SPARE-BA index (capturing age-related brain atrophy), and the SPARE-AD index (previously developed to capture patterns of atrophy found in patients with Alzheimers disease). A characteristic pattern of age-related accumulation of white matter hyperintensities in both periventricular and deep white matter areas was found. Individuals with high white matter hyperintensities burden showed significantly (P < 0.0001) lower SPARE-BA and higher SPARE-AD values compared to those with low white matter hyperintensities burden, indicating that the former had more patterns of atrophy in brain regions typically affected by ageing and Alzheimers disease dementia. To investigate a possibly causal role of white matter hyperintensities, structural equation modelling was used to quantify the effect of Framingham cardiovascular disease risk score and white matter hyperintensities burden on SPARE-BA, revealing a statistically significant (P < 0.0001) causal relationship between them. Structural equation modelling showed that the age effect on SPARE-BA was mediated by white matter hyperintensities and cardiovascular risk score each explaining 10.4% and 21.6% of the variance, respectively. The direct age effect explained 70.2% of the SPARE-BA variance. Only white matter hyperintensities significantly mediated the age effect on SPARE-AD explaining 32.8% of the variance. The direct age effect explained 66.0% of the SPARE-AD variance. Multivariable regression showed significant relationship between white matter hyperintensities volume and hypertension (P = 0.001), diabetes mellitus (P = 0.023), smoking (P = 0.002) and education level (P = 0.003). The only significant association with cognitive tests was with the immediate recall of the California verbal and learning memory test. No significant association was present with the APOE genotype. These results support the hypothesis that white matter hyperintensities contribute to patterns of brain atrophy found in beyond-normal brain ageing in the general population. White matter hyperintensities also contribute to brain atrophy patterns in regions related to Alzheimers disease dementia, in agreement with their known additive role to the likelihood of dementia. Preventive strategies reducing the odds to develop cardiovascular disease and white matter hyperintensities could decrease the incidence or delay the onset of dementia.

Association between waist circumference and gray matter volume in 2344 individuals from two adult community-based samples.

We analyzed the putative association between abdominal obesity (measured in waist circumference) and gray matter volume (Study of Health in Pomerania: SHIP-2, N=758) adjusted for age and gender by applying volumetric analysis and voxel-based morphometry (VBM) with VBM8 to brain magnetic resonance (MR) imaging. We sought replication in a second, independent population sample (SHIP-TREND, N=1586). In a combined analysis (SHIP-2 and SHIP-TREND) we investigated the impact of hypertension, type II diabetes and blood lipids on the association between waist circumference and gray matter. Volumetric analysis revealed a significant inverse association between waist circumference and gray matter volume. VBM in SHIP-2 indicated distinct inverse associations in the following structures for both hemispheres: frontal lobe, temporal lobes, pre- and postcentral gyrus, supplementary motor area, supramarginal gyrus, insula, cingulate gyrus, caudate nucleus, olfactory sulcus, para-/hippocampus, gyrus rectus, amygdala, globus pallidus, putamen, cerebellum, fusiform and lingual gyrus, (pre-) cuneus and thalamus. These areas were replicated in SHIP-TREND. More than 76% of the voxels with significant gray matter volume reduction in SHIP-2 were also distinct in TREND. These brain areas are involved in cognition, attention to interoceptive signals as satiety or reward and control food intake. Due to our cross-sectional design we cannot clarify the causal direction of the association. However, previous studies described an association between subjects with higher waist circumference and future cognitive decline suggesting a progressive brain alteration in obese subjects. Pathomechanisms may involve chronic inflammation, increased oxidative stress or cellular autophagy associated with obesity.

Advanced brain aging: relationship with epidemiologic and genetic risk factors, and overlap with Alzheimer disease atrophy patterns.

We systematically compared structural imaging patterns of advanced brain aging (ABA) in the general-population, herein defined as significant deviation from typical BA to those found in Alzheimer disease (AD). The hypothesis that ABA would show different patterns of structural change compared with those found in AD was tested via advanced pattern analysis methods. In particular, magnetic resonance images of 2705 participants from the Study of Health in Pomerania (aged 20–90 years) were analyzed using an index that captures aging atrophy patterns (Spatial Pattern of Atrophy for Recognition of BA (SPARE-BA)), and an index previously shown to capture atrophy patterns found in clinical AD (Spatial Patterns of Abnormality for Recognition of Early Alzheimer’s Disease (SPARE-AD)). We studied the association between these indices and risk factors, including an AD polygenic risk score. Finally, we compared the ABA-associated atrophy with typical AD-like patterns. We observed that SPARE-BA had significant association with: smoking (P<0.05), anti-hypertensive (P<0.05), anti-diabetic drug use (men P<0.05, women P=0.06) and waist circumference for the male cohort (P<0.05), after adjusting for age. Subjects with ABA had spatially extensive gray matter loss in the frontal, parietal and temporal lobes (false-discovery-rate-corrected q<0.001). ABA patterns of atrophy were partially overlapping with, but notably deviating from those typically found in AD. Subjects with ABA had higher SPARE-AD values; largely due to the partial spatial overlap of associated patterns in temporal regions. The AD polygenic risk score was significantly associated with SPARE-AD but not with SPARE-BA. Our findings suggest that ABA is likely characterized by pathophysiologic mechanisms that are distinct from, or only partially overlapping with those of AD.

Effect of the interaction between childhood abuse and rs1360780 of the FKBP5 gene on gray matter volume in a general population sample

The FKBP5 gene codes for a co‐chaperone that regulates glucocorticoid receptor sensitivity and thereby impacts the reactivity of the hypothalamic–pituitary–adrenal (HPA)‐axis. Evidence suggested that subjects exposed to childhood abuse and carrying the TT genotype of the FKBP5 gene single nucleotide polymorphism (SNP) rs1360780 have an increased susceptibility to stress‐related disorders.

Genetic, psychosocial and clinical factors associated with hippocampal volume in the general population

The hippocampus—crucial for memory formation, recall and mood regulation—is involved in the pathophysiology of dementia and depressive disorders. Recent genome-wide association studies (GWAS) have identified five genetic loci associated with hippocampal volume (HV). Previous studies have described psychosocial and clinical factors (for example, smoking, type 2 diabetes and hypertension) to have an impact on HV. However, the interplay between genetic, psychosocial and clinical factors on the HV remains unclear. Still, it is likely that genetic variants and clinical or psychosocial factors jointly act in modifying HV; it might be possible they even interact. Knowledge of these factors might help to quantify ones individual risk of or rather resilience against HV loss. We investigated subjects (N=2463; 55.7% women; mean age 53 years) from the Study of Health in Pomerania (SHIP-2; SHIP-TREND-0) who underwent whole-body magnetic resonance imaging (MRI) and genotyping. HVs were estimated with FreeSurfer. For optimal nonlinear model fitting, we used regression analyses with restricted cubic splines. Genetic variants and associated psychosocial or clinical factors were jointly assessed for potential two-way interactions. We observed associations between HV and gender (P<0.0001), age (P<0.0001), body height (P<0.0001), education (P=0.0053), smoking (P=0.0058), diastolic blood pressure (P=0.0211), rs7294919 (P=0.0065), rs17178006 (P=0.0002), rs6581612 (P=0.0036), rs6741949 (P=0.0112) and rs7852872 (P=0.0451). In addition, we found three significant interactions: between rs7294919 and smoking (P=0.0473), rs7294919 and diastolic blood pressure (P=0.0447) and between rs7852872 and rs6581612 (P=0.0114). We suggest that these factors might have a role in the individual susceptibility to hippocampus-associated disorders.

A mobile and asynchronous electronic data capture system for epidemiologic studies

A Central Data Management (CDM) system based on electronic data capture (EDC) software and study specific databases is an essential component for assessment and management of large data volumes in epidemiologic studies. Conventional CDM systems using web applications for data capture depend on permanent access to a network. However, in many study settings permanent network access cannot be guaranteed, e.g. when participants/patients are visited in their homes. In such cases a different concept for data capture is needed. The utilized EDC software must be able to ensure data capture as stand-alone instance and to synchronize captured data with the server at a later point in time. This article describes the design of the mobile information capture (MInCa) system an EDC software meeting these requirements. In particular, we focus on client and server design, data synchronization, and data privacy as well as data security measures. The MInCa software has already proven its efficiency in epidemiologic studies revealing strengths and weaknesses concerning both concept and practical application which will be addressed in this article.

A priori collaboration in population imaging: The Uniform Neuro-Imaging of Virchow-Robin Spaces Enlargement consortium

Virchow‐Robin spaces (VRS), or perivascular spaces, are compartments of interstitial fluid enclosing cerebral blood vessels and are potential imaging markers of various underlying brain pathologies. Despite a growing interest in the study of enlarged VRS, the heterogeneity in rating and quantification methods combined with small sample sizes have so far hampered advancement in the field.

Microglia ablation alleviates myelin-associated catatonic signs in mice

The underlying cellular mechanisms of catatonia, an executive “psychomotor” syndrome that is observed across neuropsychiatric diseases, have remained obscure. In humans and mice, reduced expression of the structural myelin protein CNP is associated with catatonic signs in an age-dependent manner, pointing to the involvement of myelin-producing oligodendrocytes. Here, we showed that the underlying cause of catatonic signs is the low-grade inflammation of white matter tracts, which marks a final common pathway in Cnp-deficient and other mutant mice with minor myelin abnormalities. The inhibitor of CSF1 receptor kinase signaling PLX5622 depleted microglia and alleviated the catatonic symptoms of Cnp mutants. Thus, microglia and low-grade inflammation of myelinated tracts emerged as the trigger of a previously unexplained mental condition. We observed a very high (25%) prevalence of individuals with catatonic signs in a deeply phenotyped schizophrenia sample (n = 1095). Additionally, we found the loss-of-function allele of a myelin-specific gene (CNP rs2070106-AA) associated with catatonia in 2 independent schizophrenia cohorts and also associated with white matter hyperintensities in a general population sample. Since the catatonic syndrome is likely a surrogate marker for other executive function defects, we suggest that microglia-directed therapies may be considered in psychiatric disorders associated with myelin abnormalities.

Relationship between APOE Genotype and Structural MRI Measures throughout Adulthood in the Study of Health in Pomerania Population-Based Cohort

BACKGROUND AND PURPOSE: The presence of the apolipoprotein E ε4 allele is the strongest sporadic Alzheimer disease genetic risk factor. We hypothesized that apolipoprotein E ε4 carriers and noncarriers may already differ in imaging patterns in midlife. We therefore sought to identify the effect of apolipoprotein E genotype on brain atrophy across almost the entire adult age span by using advanced MR imaging–based pattern analysis. MATERIALS AND METHODS: We analyzed MR imaging scans of 1472 participants from the Study of Health in Pomerania (22–90 years of age). We studied the association among age, apolipoprotein E ε4 carrier status, and brain atrophy, which was quantified by using 2 MR imaging–based indices: Spatial Pattern of Atrophy for Recognition of Brain Aging (summarizing age-related brain atrophy) and Spatial Pattern of Abnormality for Recognition of Early Alzheimer Disease (summarizing Alzheimer disease-like brain atrophy patterns), as well as the gray matter volumes in several Alzheimer disease- and apolipoprotein E–related ROIs (lateral frontal, lateral temporal, medial frontal, and hippocampus). RESULTS: No significant association was found between apolipoprotein E ε4 carrier status and the studied ROIs or the MR imaging–based indices in linear regression models adjusted for age, sex, and education, including an interaction term between apolipoprotein E and age. CONCLUSIONS: Our study indicates that measurable apolipoprotein E–related brain atrophy does not occur in early adulthood and midlife and suggests that such atrophy may only occur more proximal to the onset of clinical symptoms of dementia.