Analysis of Cerebral CT Based on Supervised Machine Learning as a Predictor of Outcome After Out-of-Hospital Cardiac Arrest.

BACKGROUND AND OBJECTIVES: In light of limited intensive care capacities and a lack of accurate prognostic tools to advise caregivers and family members responsibly, this study aims to determine whether automated cerebral CT (CCT) analysis allows prognostication after out-of-hospital cardiac arrest. METHODS: In this monocentric, retrospective cohort study, a supervised machine learning classifier based on an elastic net regularized logistic regression model for gray matter alterations on nonenhanced CCT obtained after cardiac arrest was trained using 10-fold cross-validation and tested on a hold-out sample (random split 75%/25%) for outcome prediction. Following the literature, a favorable outcome was defined as a cerebral performance category of 1-2 and a poor outcome of 3-5. The diagnostic accuracy was compared with established and guideline-recommended prognostic measures within the sample, that is, gray matter-white matter ratio (GWR), neuron-specific enolase (NSE), and neurofilament light chain (NfL) in serum. RESULTS: Of 279 adult patients, 132 who underwent CCT within 14 days of cardiac arrest with good imaging quality were identified. Our approach discriminated between favorable and poor outcomes with an area under the curve (AUC) of 0.73 (95% CI 0.59-0.82). Thus, the prognostic power outperformed the GWR (AUC 0.66, 95% CI 0.56-0.76). The biomarkers NfL, measured at days 1 and 2, and NSE, measured at day 2, exceeded the reliability of the imaging markers derived from CT (AUC NfL day 1: 0.87, 95% CI 0.75-0.99; AUC NfL day 2: 0.90, 95% CI 0.79-1.00; AUC NSE day: 2 0.78, 95% CI 0.62-0.94). DISCUSSION: Our data show that machine learning-assisted gray matter analysis of CCT images offers prognostic information after out-of-hospital cardiac arrest. Thus, CCT gray matter analysis could become a reliable and time-independent addition to the standard workup with serum biomarkers sampled at predefined time points. Prospective studies are warranted to replicate these findings.

Fine-grained age-matching improves atrophy-based detection of mild cognitive impairment more than amyloid-negative reference subjects.

INTRODUCTION: In clinical practice, differentiating between age-related gray matter (GM) atrophy and neurodegeneration-related atrophy at early disease stages, such as mild cognitive impairment (MCI), remains challenging. We hypothesized that fined-grained adjustment for age effects and using amyloid-negative reference subjects could increase classification accuracy. METHODS: T1-weighted magnetic resonance imaging (MRI) data of 131 cognitively normal (CN) individuals and 91 patients with MCI from the Alzheimer's disease neuroimaging initiative (ADNI) characterized concerning amyloid status, as well as 19 CN individuals and 19 MCI patients from an independent validation sample were segmented, spatially normalized and analyzed in the framework of voxel-based morphometry (VBM). For each participant, statistical maps of GM atrophy were computed as the deviation from the GM of CN reference groups at the voxel level. CN reference groups composed with different degrees of age-matching, and mixed and strictly amyloid-negative CN reference groups were examined regarding their effect on the accuracy in distinguishing between CN and MCI. Furthermore, the effects of spatial smoothing and atrophy threshold were assessed. RESULTS: Approaches with a specific reference group for each age significantly outperformed all other age-adjustment strategies with a maximum area under the curve of 1.0 in the ADNI sample and 0.985 in the validation sample. Accounting for age in a regression-based approach improved classification accuracy over that of a single CN reference group in the age range of the patient sample. Using strictly amyloid-negative reference groups improved classification accuracy only when age was not considered. CONCLUSION: Our results demonstrate that VBM can differentiate between age-related and MCI-associated atrophy with high accuracy. Crucially, age-specific reference groups significantly increased accuracy, more so than regression-based approaches and using amyloid-negative reference groups.

Serum cortisol is negatively related to hippocampal volume, brain structure, and memory performance in healthy aging and Alzheimer's disease.

Objective: Elevated cortisol levels have been frequently reported in Alzheimer's disease (AD) and linked to brain atrophy, especially of the hippocampus. Besides, high cortisol levels have been shown to impair memory performance and increase the risk of developing AD in healthy individuals. We investigated the associations between serum cortisol levels, hippocampal volume, gray matter volume and memory performance in healthy aging and AD. Methods: In our cross-sectional study, we analyzed the relationships between morning serum cortisol levels, verbal memory performance, hippocampal volume, and whole-brain voxel-wise gray matter volume in an independent sample of 29 healthy seniors (HS) and 29 patients along the spectrum of biomarker-based AD. Results: Cortisol levels were significantly elevated in patients with AD as compared to HS, and higher cortisol levels were correlated with worse memory performance in AD. Furthermore, higher cortisol levels were significantly associated with smaller left hippocampal volumes in HS and indirectly negatively correlated to memory function through hippocampal volume. Higher cortisol levels were further related to lower gray matter volume in the hippocampus and temporal and parietal areas in the left hemisphere in both groups. The strength of this association was similar in HS and AD. Conclusion: In AD, cortisol levels are elevated and associated with worse memory performance. Furthermore, in healthy seniors, higher cortisol levels show a detrimental relationship with brain regions typically affected by AD. Thus, increased cortisol levels seem to be indirectly linked to worse memory function even in otherwise healthy individuals. Cortisol may therefore not only serve as a biomarker of increased risk for AD, but maybe even more importantly, as an early target for preventive and therapeutic interventions.

Concordance of Intrinsic Brain Connectivity Measures Is Disrupted in Alzheimer's Disease.

Background: Recently, a new resting-state functional magnetic resonance imaging (rs-fMRI) measure to evaluate the concordance between different rs-fMRI metrics has been proposed and has not been investigated in Alzheimer's disease (AD). Methods: 3T rs-fMRI data were obtained from healthy young controls (YC, n = 26), healthy senior controls (SC, n = 29), and AD patients (n = 35). The fractional amplitude of low-frequency fluctuations (fALFF), regional homogeneity (ReHo), and degree centrality (DC) were analyzed, followed by the calculation of their concordance using Kendall's W for each brain voxel across time. Group differences in the concordance were compared globally, within seven intrinsic brain networks, and on a voxel-by-voxel basis with covariates of age, sex, head motion, and gray matter volume. Results: The global concordance was lowest in AD among the three groups, with similar differences for the single metrics. When comparing AD to SC, reductions of concordance were detected in each of the investigated networks apart from the limbic network. For SC in comparison to YC, lower global concordance without any network-level difference was observed. Voxel-wise analyses revealed lower concordance in the right middle temporal gyrus in AD compared to SC and lower concordance in the left middle frontal gyrus in SC compared to YC. Lower fALFF were observed in the right angular gyrus in AD in comparison to SC, but ReHo and DC showed no group differences. Conclusions: The concordance of resting-state measures differentiates AD from healthy aging and may represent a novel imaging marker in AD.

Decreased Efficiency of Between-Network Dynamics During Early Memory Consolidation With Aging.

Aging is associated with memory decline and progressive disabilities in the activities of daily living. These deficits have a significant impact on the quality of life of the aging population and lead to a tremendous burden on societies and health care systems. Understanding the mechanisms underlying aging-related memory decline is likely to inform the development of compensatory strategies promoting independence in old age. Research on aging-related memory decline has mainly focused on encoding and retrieval. However, some findings suggest that memory deficits may at least partly be due to impaired consolidation. To date, it remains elusive whether aging-related memory decline results from defective consolidation. This study examined age effects on consolidation-related neural mechanisms and their susceptibility to interference using functional magnetic resonance imaging data from 13 younger (20-30 years, 8 female) and 16 older (49-75 years, 5 female) healthy participants. fMRI was performed before and during a memory paradigm comprised of encoding, consolidation, and retrieval phases. Consolidation was variously challenged: (1) control (no manipulation), (2) interference (repeated stimulus presentation with interfering information), and (3) reminder condition (repeated presentation without interfering information). We analyzed the fractional amplitude of low-frequency fluctuations (fALFF) to compare brain activity changes from pre- to post-encoding rest. In the control condition, fALFF was decreased in the left supramarginal gyrus, right middle temporal gyrus, and left precuneus but increased in parts of the occipital and inferior temporal cortex. Connectivity analyses between fALFF-derived seeds and network ROIs revealed an aging-related decrease in the efficiency of functional connectivity (FC) within the ventral stream network and between salience, default mode, and central executive networks during consolidation. Moreover, our results indicate increased interference susceptibility in older individuals with dynamics between salience and default mode networks as a neurophysiological correlate. Conclusively, aging-related memory decline is partly caused by inefficient consolidation. Memory consolidation requires a complex interplay between large-scale brain networks, which qualitatively decreases with age.

Finding New Communities: A Principle of Neuronal Network Reorganization in Alzheimer's Disease.

Background: Graph-theoretical analyses have been previously used to investigate changes in the functional connectome in patients with Alzheimer's disease (AD). However, these analyses generally assume static organizational principles, thereby neglecting a fundamental reconfiguration of functional connections in the face of neurodegeneration. Methods: Here, we focus on differences in the community structure of the functional connectome in young and old individuals and patients with AD. Patients with AD, moreover, underwent molecular imaging positron emission tomography by using [18F]AV1451 to measure tau burden, a major hallmark of AD. Results: Although the overall organizational principles of the community structure of the human functional connectome were preserved even in advanced healthy aging, they were considerably changed in AD. We discovered that the communities in AD are re-organized, with nodes changing their allegiance to communities, thus resulting in an overall less efficient re-organized community structure. We further discovered that nodes with a tendency to leave the communities displayed a relatively higher tau pathology burden. Discussion: Together, this study suggests that local tau pathology in AD is associated to fundamental changes in basic organizational principles of the human connectome. Our results shed new light on previous findings obtained by using the graph theory in AD and imply a general principle of the brain in response to neurodegeneration.

Graph Theory Analysis Reveals Resting-State Compensatory Mechanisms in Healthy Aging and Prodromal Alzheimer's Disease.

Several theories of cognitive compensation have been suggested to explain sustained cognitive abilities in healthy brain aging and early neurodegenerative processes. The growing number of studies investigating various aspects of task-based compensation in these conditions is contrasted by the shortage of data about resting-state compensatory mechanisms. Using our proposed criterion-based framework for compensation, we investigated 45 participants in three groups: (i) patients with mild cognitive impairment (MCI) and positive biomarkers indicative of Alzheimer's disease (AD); (ii) cognitively normal young adults; (iii) cognitively normal older adults. To increase reliability, three sessions of resting-state functional magnetic resonance imaging for each participant were performed on different days (135 scans in total). To elucidate the dimensions and dynamics of resting-state compensatory mechanisms, we used graph theory analysis along with volumetric analysis. Graph theory analysis was applied based on the Brainnetome atlas, which provides a connectivity-based parcellation framework. Comprehensive neuropsychological examinations including the Rey Auditory Verbal Learning Test (RAVLT) and the Trail Making Test (TMT) were performed, to relate graph measures of compensatory nodes to cognition. To avoid false-positive findings, results were corrected for multiple comparisons. First, we observed an increase of degree centrality in cognition related brain regions of the middle frontal gyrus, precentral gyrus and superior parietal lobe despite local atrophy in MCI and healthy aging, indicating a resting-state connectivity increase with positive biomarkers. When relating the degree centrality measures to cognitive performance, we observed that greater connectivity led to better RAVLT and TMT scores in MCI and, hence, might constitute a compensatory mechanism. The detection and improved understanding of the compensatory dynamics in healthy aging and prodromal AD is mandatory for implementing and tailoring preventive interventions aiming at preserved overall cognitive functioning and delayed clinical onset of dementia.

Entorhinal Tau Predicts Hippocampal Activation and Memory Deficits in Alzheimer's Disease.

BACKGROUND: To date, it remains unclear how amyloid plaques and neurofibrillary tangles are related to neural activation and, consequently, cognition in Alzheimer's disease (AD). Recent findings indicate that tau accumulation may drive hippocampal hyperactivity in cognitively normal aging, but it remains to be elucidated how tau accumulation is related to neural activation in AD. OBJECTIVE: To determine whether the association between tau accumulation and hippocampal hyperactivation persists in mild cognitive impairment (MCI) and mild dementia or if the two measures dissociate with disease progression, we investigated the relationship between local tau deposits and memory-related neural activation in MCI and mild dementia due to AD. METHODS: Fifteen patients with MCI or mild dementia due to AD underwent a neuropsychological assessment and performed an item memory task during functional magnetic resonance imaging. Cerebral tau accumulation was assessed using positron emission tomography and [18F]-AV-1451. RESULTS: Entorhinal, but not global tau accumulation, was highly correlated with hippocampal activation due to visual item memory encoding and predicted memory loss over time. Neural activation in the posterior cingulate cortex and the fusiform gyrus was not significantly correlated with tau accumulation. CONCLUSION: These findings extend previous observations in cognitively normal aging, demonstrating that entorhinal tau continues to be closely associated with hippocampal hyperactivity and memory performance in MCI and mild dementia due to AD. Furthermore, data suggest that this association is strongest in medial temporal lobe structures. In summary, our data provide novel insights into the relationship of tau accumulation to neural activation and memory in AD.

Spatial distributions of cholinergic impairment and neuronal hypometabolism differ in MCI due to AD.

Elucidating the relationship between neuronal metabolism and the integrity of the cholinergic system is prerequisite for a profound understanding of cholinergic dysfunction in Alzheimer's disease. The cholinergic system can be investigated specifically using positron emission tomography (PET) with [11C]N-methyl-4-piperidyl-acetate (MP4A), while neuronal metabolism is often assessed with 2-deoxy-2-[18F]fluoro-d-glucose-(FDG) PET. We hypothesised a close correlation between MP4A-perfusion and FDG-uptake, permitting inferences about metabolism from MP4A-perfusion, and investigated the patterns of neuronal hypometabolism and cholinergic impairment in non-demented AD patients. MP4A-PET was performed in 18 cognitively normal adults and 19 patients with mild cognitive impairment (MCI) and positive AD biomarkers. In nine patients with additional FDG-PET, the sum images of every combination of consecutive early MP4A-frames were correlated with FDG-scans to determine the optimal time window for assessing MP4A-perfusion. Acetylcholinesterase (AChE) activity was estimated using a 3-compartmental model. Group comparisons of MP4A-perfusion and AChE-activity were performed using the entire sample. The highest correlation between MP4A-perfusion and FDG-uptake across the cerebral cortex was observed 60-450 s after injection (r = 0.867). The patterns of hypometabolism (FDG-PET) and hypoperfusion (MP4A-PET) in MCI covered areas known to be hypometabolic early in AD, while AChE activity was mainly reduced in the lateral temporal cortex and the occipital lobe, sparing posterior midline structures. Data indicate that patterns of cholinergic impairment and neuronal hypometabolism differ significantly at the stage of MCI in AD, implying distinct underlying pathologies, and suggesting potential predictors of the response to cholinergic pharmacotherapy.

On the Extraction and Analysis of Graphs From Resting-State fMRI to Support a Correct and Robust Diagnostic Tool for Alzheimer's Disease.

The diagnosis of Alzheimer's disease (AD), especially in the early stage, is still not very reliable and the development of new diagnosis tools is desirable. A diagnosis based on functional magnetic resonance imaging (fMRI) is a suitable candidate, since fMRI is non-invasive, readily available, and indirectly measures synaptic dysfunction, which can be observed even at the earliest stages of AD. However, the results of previous attempts to analyze graph properties of resting state fMRI data are contradictory, presumably caused by methodological differences in graph construction. This comprises two steps: clustering the voxels of the functional image to define the nodes of the graph, and calculating the graph's edge weights based on a functional connectivity measure of the average cluster activities. A variety of methods are available for each step, but the robustness of results to method choice, and the suitability of the methods to support a diagnostic tool, are largely unknown. To address this issue, we employ a range of commonly and rarely used clustering and edge definition methods and analyze their graph theoretic measures (graph weight, shortest path length, clustering coefficient, and weighted degree distribution and modularity) on a small data set of 26 healthy controls, 16 subjects with mild cognitive impairment (MCI) and 14 with Alzheimer's disease. We examine the results with respect to statistical significance of the mean difference in graph properties, the sensitivity of the results to model and parameter choices, and relative diagnostic power based on both a statistical model and support vector machines. We find that different combinations of graph construction techniques yield contradicting, but statistically significant, relations of graph properties between health conditions, explaining the discrepancy across previous studies, but casting doubt on such analyses as a method to gain insight into disease effects. The production of significant differences in mean graph properties turns out not to be a good predictor of future diagnostic capacity. Highest predictive power, expressed by largest negative surprise values, are achieved for both atlas-driven and data-driven clustering (Ward clustering), as long as graphs are small and clusters large, in combination with edge definitions based on correlations and mutual information transfer.

Neural Network Connectivity During Post-encoding Rest: Linking Episodic Memory Encoding and Retrieval.

Commonly, a switch between networks mediating memory encoding and those mediating retrieval is observed. This may not only be due to differential involvement of neural resources due to distinct cognitive processes but could also reflect the formation of new memory traces and their dynamic change during consolidation. We used resting state fMRI to measure functional connectivity (FC) changes during post-encoding rest, hypothesizing that during this phase, new functional connections between encoding- and retrieval-related regions are created. Interfering and reminding tasks served as experimental modulators to corroborate that the observed FC differences indeed reflect changes specific to post-encoding rest. The right inferior occipital and fusiform gyri (active during encoding) showed increased FC with the left inferior frontal gyrus and the left middle temporal gyrus (MTG) during post-encoding rest. Importantly, the left MTG subsequently also mediated successful retrieval. This finding might reflect the formation of functional connections between encoding- and retrieval-related regions during undisturbed post-encoding rest. These connections were vulnerable to experimental modulation: Cognitive interference disrupted FC changes during post-encoding rest resulting in poorer memory performance. The presentation of reminders also inhibited FC increases but without affecting memory performance. Our results contribute to a better understanding of the mechanisms by which post-encoding rest bridges the gap between encoding- and retrieval-related networks.

In vivo Patterns of Tau Pathology, Amyloid-ß Burden, and Neuronal Dysfunction in Clinical Variants of Alzheimer's Disease.

The clinical heterogeneity of Alzheimer's disease is not reflected in the rather diffuse cortical deposition of amyloid-ß. We assessed the relationship between clinical symptoms, in vivo tau pathology, amyloid distribution, and hypometabolism in variants of Alzheimer's disease using novel multimodal PET imaging techniques. Tau pathology was primarily observed in brain regions related to clinical symptoms and overlapped with areas of hypometabolism. In contrast, amyloid-ß deposition was diffusely distributed over the entire cortex. Tau PET imaging may thus serve as a valuable biomarker for the localization of neuronal injury in vivo and may help to validate atypical subtypes of Alzheimer's disease.

Aberrant functional connectivity differentiates retrosplenial cortex from posterior cingulate cortex in prodromal Alzheimer's disease.

The posterior cingulate cortex (PCC) is a key hub of the default mode network, a resting-state network involved in episodic memory, showing functional connectivity (FC) changes in Alzheimer's disease (AD). However, PCC is a cytoarchitectonically heterogeneous region. Specifically, the retrosplenial cortex (RSC), often subsumed under the PCC, is an area functionally and microanatomically distinct from PCC. To investigate FC patterns of RSC and PCC separately, we used resting-state functional magnetic resonance imaging in healthy aging participants, patients with subjective cognitive impairment, and prodromal AD. Compared to the other 2 groups, we found higher FC from RSC to frontal cortex in subjective cognitive impairment but higher FC to occipital cortex in prodromal AD. Conversely, FC from PCC to the lingual gyrus was higher in prodromal AD. Furthermore, data indicate that RSC and PCC are characterized by differential FC patterns represented by hub-specific interactions with memory and attentions scores in prodromal AD compared to cognitively normal individuals, possibly reflecting compensatory mechanisms for RSC and neurodegenerative processes for PCC. Data thus confirm and extend previous studies suggesting that the RSC is functionally distinct from PCC.

Impact of tau and amyloid burden on glucose metabolism in Alzheimer's disease.

In a multimodal PET imaging approach, we determined the differential contribution of neurofibrillary tangles (measured with [18F]AV-1451) and beta-amyloid burden (measured with [11C]PiB) on degree of neurodegeneration (i.e., glucose metabolism measured with [18F]FDG-PET) in patients with Alzheimer's disease. Across brain regions, we observed an interactive effect of beta-amyloid burden and tau deposition on glucose metabolism which was most pronounced in the parietal lobe. Elevated beta-amyloid burden was associated with a stronger influence of tau accumulation on glucose metabolism. Our data provide the first in vivo insights into the differential contribution of Aß and tau to neurodegeneration in Alzheimer's disease.