Extracting patterns of morphometry distinguishing HIV associated neurodegeneration from mild cognitive impairment via group cardinality constrained classification

The association of microbial infection and adaptive immune cell activation in Alzheimer's disease

Alzheimer's disease (AD) is a progressive neurodegenerative disorder and the most common form of dementia. Early symptoms include the loss of memory and mild cognitive ability; however, as the disease progresses, these symptoms can present with increased severity manifesting as mood and behaviour changes, disorientation, and a loss of motor/body control. AD is one of the leading causes of death in the UK, and with an ever-increasing ageing society, patient numbers are predicted to rise posing a significant global health emergency. AD is a complex neurophysiological disorder where pathology is characterized by the deposition and aggregation of misfolded amyloid-beta (Aβ)-protein that in-turn promotes excessive tau-protein production which together drives neuronal cell dysfunction, neuroinflammation, and neurodegeneration. It is widely accepted that AD is driven by a combination of both genetic and immunological processes with recent data suggesting that adaptive immune cell activity within the parenchyma occurs throughout disease. The mechanisms behind these observations remain unclear but suggest that manipulating the adaptive immune response during AD may be an effective therapeutic strategy. Using immunotherapy for AD treatment is not a new concept as the only two approved treatments for AD use antibody-based approaches to target Aβ. However, these have been shown to only temporarily ease symptoms or slow progression highlighting the urgent need for newer treatments. This review discusses the role of the adaptive immune system during AD, how microbial infections may be contributing to inflammatory immune activity and suggests how adaptive immune processes can pose as therapeutic targets for this devastating disease.

A Penalty Approach for Normalizing Feature Distributions to Build Confounder-Free Models

Translating the use of modern machine learning algorithms into clinical applications requires settling challenges related to explain-ability and management of nuanced confounding factors. To suitably interpret the results, removing or explaining the effect of confounding variables (or metadata) is essential. Confounding variables affect the relationship between input training data and target outputs. Accordingly, when we train a model on such data, confounding variables will bias the distribution of the learned features. A recent promising solution, Meta-Data Normalization (MDN), estimates the linear relationship between the metadata and each feature based on a non-trainable closed-form solution. However, this estimation is confined by the sample size of a mini-batch and thereby may result in an oscillating performance. In this paper, we extend the MDN method by applying a Penalty approach (referred to as PDMN). We cast the problem into a bi-level nested optimization problem. We then approximate that objective using a penalty method so that the linear parameters within the MDN layer are trainable and learned on all samples. This enables PMDN to be plugged into any architectures, even those unfit to run batch-level operations such as transformers and recurrent models. We show improvement in model accuracy and independence from the confounders using PMDN over MDN in a synthetic experiment and a multi-label, multi-site classification of magnetic resonance images.

Prior test experience confounds longitudinal tracking of adolescent cognitive and motor development

Background

Accurate measurement of trajectories in longitudinal studies, considered the gold standard method for tracking functional growth during adolescence, decline in aging, and change after head injury, is subject to confounding by testing experience.

Methods

We measured change in cognitive and motor abilities over four test sessions (baseline and three annual assessments) in 154 male and 165 female participants (baseline age 12–21 years) from the National Consortium on Alcohol and NeuroDevelopment in Adolescence (NCANDA) study. At each of the four test sessions, these participants were given a test battery using computerized administration and traditional pencil and paper tests that yielded accuracy and speed measures for multiple component cognitive (Abstraction, Attention, Emotion, Episodic memory, Working memory, and General Ability) and motor (Ataxia and Speed) functions. The analysis aim was to dissociate neurodevelopment from testing experience by using an adaptation of the twice-minus-once tested method, which calculated the difference between longitudinal change (comprising developmental plus practice effects) and practice-free initial cross-sectional performance for each consecutive pairs of test sessions. Accordingly, the first set of analyses quantified the effects of learning (i.e., prior test experience) on accuracy and after speed domain scores. Then developmental effects were  determined for each domain for accuracy and speed having removed the measured learning effects.

Results

The greatest gains in performance occurred between the first and second sessions, especially in younger participants, regardless of sex, but practice gains continued to accrue thereafter for several functions. For all 8 accuracy composite scores, the developmental effect after accounting for learning was significant across age and was adequately described by linear fits. The learning-adjusted developmental effects for speed were adequately described by linear fits for Abstraction, Emotion, Episodic Memory, General Ability, and Motor scores, although a nonlinear fit was better for Attention, Working Memory, and Average Speed scores.

Conclusion

Thus, what appeared as accelerated cognitive and motor development was, in most cases, attributable to learning. Recognition of the substantial influence of prior testing experience is critical for accurate characterization of normal development and for developing norms for clinical neuropsychological investigations of conditions affecting the brain.

Multi-label, multi-domain learning identifies compounding effects of HIV and cognitive impairment

Older individuals infected by Human Immunodeficiency Virus (HIV) are at risk for developing HIV-Associated Neurocognitive Disorder (HAND), i.e., from reduced cognitive functioning similar to HIV-negative individuals with Mild Cognitive Impairment (MCI) or to Alzheimer's Disease (AD) if more severely affected. Incompletely understood is how brain structure can serve to differentiate cognitive impairment (CI) in the HIV-positive (i.e., HAND) from the HIV-negative cohort (i.e., MCI and AD). To that end, we designed a multi-label classifier that labels the structural magnetic resonance images (MRI) of individuals by their HIV and CI status via two binary variables. Proper training of such an approach traditionally requires well-curated datasets containing large number of samples for each of the corresponding four cohorts (healthy controls, CI HIV-negative adults a.k.a. CI-only, HIV-positive patients without CI a.k.a. HIV-only, and HAND). Because of the rarity of such datasets, we proposed to improve training of the multi-label classifier via a multi-domain learning scheme that also incorporates domain-specific classifiers on auxiliary single-label datasets specific to either binary label. Specifically, we complement the training dataset of MRIs of the four cohorts (Control: 156, CI-only: 335, HIV-only: 37, HAND: 145) acquired by the Memory and Aging Center at the University of California - San Francisco with a CI-specific dataset only containing MRIs of HIV-negative subjects (Controls: 229, CI-only: 397) from the Alzheimer's Disease Neuroimaging Initiative and an HIV-specific dataset (Controls: 75, HIV-only: 75) provided by SRI International. Based on cross-validation on the UCSF dataset, the multi-domain and multi-label learning strategy leads to superior classification accuracy compared with one-domain or multi-class learning approaches, specifically for the undersampled HIV-only cohort. The 'prediction logits' of CI computed by the multi-label formulation also successfully stratify motor performance among the HIV-positive subjects (including HAND). Finally, brain patterns driving the subject-level predictions across all four cohorts characterize the independent and compounding effects of HIV and CI in the HAND cohort.

Cerebrovascular Disease Correlates With Longitudinal Brain Atrophy in Virally Suppressed Older People Living With HIV

BACKGROUND

Mild cognitive difficulties and progressive brain atrophy are observed in older people living with HIV (PLWH) despite persistent viral suppression. Whether cerebrovascular disease (CVD) risk factors and white matter hyperintensity (WMH) volume correspond to the observed progressive brain atrophy is not well understood.

METHODS

Longitudinal structural brain atrophy rates and WMH volume were examined among 57 HIV-infected participants and 40 demographically similar HIV-uninfected controls over an average (SD) of 3.4 (1.7) years. We investigated associations between CVD burden (presence of diabetes, hypertension, hyperlipidemia, obesity, smoking history, and atrial fibrillation) and WMH with atrophy over time.

RESULTS

The mean (SD) age was 64.8 (4.3) years for PLWH and 66.4 (3.2) years for controls. Participants and controls were similar in age and sex (P > 0.05). PLWH were persistently suppressed (VL <375 copies/mL with 93% <75 copies/mL). The total number of CVD risk factors did not associate with atrophy rates in any regions of interests examined; however, body mass index independently associated with progressive atrophy in the right precentral gyrus (β = -0.30; P = 0.023), parietal lobe (β = -0.28; P = 0.030), and frontal lobe atrophy (β = -0.27; P = 0.026) of the HIV-infected group. No associations were found in the HIV-uninfected group. In both groups, baseline WMH was associated with progressive atrophy rates bilaterally in the parietal gray in the HIV-infected group (β = -0.30; P = 0.034) and the HIV-uninfected participants (β = -0.37; P = 0.033).

CONCLUSIONS

Body mass index and WMH are associated with atrophy in selective brain regions. However, CVD burden seems to partially contribute to progressive brain atrophy in older individuals regardless of HIV status, with similar effect sizes. Thus, CVD alone is unlikely to explain accelerated atrophy rates observed in virally suppressed PLWH. In older individuals, addressing modifiable CVD risk factors remains important to optimize brain health.

Metadata Normalization

Batch Normalization (BN) and its variants have delivered tremendous success in combating the covariate shift induced by the training step of deep learning methods. While these techniques normalize feature distributions by standardizing with batch statistics, they do not correct the influence on features from extraneous variables or multiple distributions. Such extra variables, referred to as metadata here, may create bias or confounding effects (e.g., race when classifying gender from face images). We introduce the Metadata Normalization (MDN) layer, a new batch-level operation which can be used end-to-end within the training framework, to correct the influence of metadata on feature distributions. MDN adopts a regression analysis technique traditionally used for preprocessing to remove (regress out) the metadata effects on model features during training. We utilize a metric based on distance correlation to quantify the distribution bias from the metadata and demonstrate that our method successfully removes metadata effects on four diverse settings: one synthetic, one 2D image, one video, and one 3D medical image dataset.

Logistic Regression Confined by Cardinality-Constrained Sample and Feature Selection

Many vision-based applications rely on logistic regression for embedding classification within a probabilistic context, such as recognition in images and videos or identifying disease-specific image phenotypes from neuroimages. Logistic regression, however, often performs poorly when trained on data that is noisy, has irrelevant features, or when the samples are distributed across the classes in an imbalanced setting; a common occurrence in visual recognition tasks. To deal with those issues, researchers generally rely on ad-hoc regularization techniques or model a subset of these issues. We instead propose a mathematically sound logistic regression model that selects a subset of (relevant) features and (informative and balanced) set of samples during the training process. The model does so by applying cardinality constraints (via l₀-'norm' sparsity) on the features and samples. l₀ defines sparsity in mathematical settings but in practice has mostly been approximated (e.g., via l₁ or its variations) for computational simplicity. We prove that a local minimum to the non-convex optimization problems induced by cardinality constraints can be computed by combining block coordinate descent with penalty decomposition. On synthetic, image recognition, and neuroimaging datasets, we show that the accuracy of the method is higher than alternative methods and classifiers commonly used in the literature.

Machine-learning classification of neurocognitive performance in children with perinatal HIV initiating de novo antiretroviral therapy

OBJECTIVE

To develop a predictive model of neurocognitive trajectories in children with perinatal HIV (pHIV).

DESIGN

Machine learning analysis of baseline and longitudinal predictors derived from clinical measures utilized in pediatric HIV.

METHODS

Two hundred and eighty-five children (ages 2-14 years at baseline; Mage = 6.4 years) with pHIV in Southeast Asia underwent neurocognitive assessment at study enrollment and twice annually thereafter for an average of 5.4 years. Neurocognitive slopes were modeled to establish two subgroups [above (n = 145) and below average (n = 140) trajectories). Gradient-boosted multivariate regressions (GBM) with five-fold cross validation were conducted to examine baseline (pre-ART) and longitudinal predictive features derived from demographic, HIV disease, immune, mental health, and physical health indices (i.e. complete blood count [CBC]).

RESULTS

The baseline GBM established a classifier of neurocognitive group designation with an average AUC of 79% built from HIV disease severity and immune markers. GBM analysis of longitudinal predictors with and without interactions improved the average AUC to 87 and 90%, respectively. Mental health problems and hematocrit levels also emerged as salient features in the longitudinal models, with novel interactions between mental health problems and both CD4 cell count and hematocrit levels. Average AUCs derived from each GBM model were higher than results obtained using logistic regression.

CONCLUSION

Our findings support the feasibility of machine learning to identify children with pHIV at risk for suboptimal neurocognitive development. Results also suggest that interactions between HIV disease and mental health problems are early antecedents to neurocognitive difficulties in later childhood among youth with pHIV.

Detecting cognitive impairment in HIV-infected individuals using mutual connectivity analysis of resting state functional MRI

It is estimated that more than 50% of the individuals affected with Human Immunodeficiency Virus (HIV) present deficits in multiple cognitive domains, collectively known as HIV-associated neurocognitive disorder (HAND). Early stages of brain injury may be clinically silent but potentially measurable via neuroimaging. A total of 40 subjects (20 HIV positive and 20 age-matched controls) volunteered for the study. All subjects underwent a standard battery of neuropsychological tests used for the clinical diagnosis of HAND. Fourteen HIV+ and five healthy subjects showed signs of neurological impairment. Connectivity was computed using mutual connectivity analysis (MCA) with generalized radial basis function neural network, a framework for quantifying non-linear connectivity as well as conventional correlation from 160 regional time-series that were extracted based on the Dosenbach (DOS) atlas. We subsequently applied graph theoretic as well as network analysis approaches for characterizing the connectivity matrices obtained and localizing between-group differences. We focused on trying to detect cognitive impairment using the subset of 29 (14 subjects with HAND and 15 cognitively normal controls) subjects. For the global analysis, significant differences (p < 0.05) were seen in the variance in degree, modularity and Smallworldness. Regional analysis revealed changes occurring mainly in portions of the lateral occipital cortex and the cingulate cortex. Furthermore, using Network Based Statistics (NBS), we uncovered an affected sub-network of 19 nodes comprising predominantly of regions of the default mode network. Similar analysis using the conventional correlation method revealed no significant results at a global scale, while regional analysis shows some differences spread across resting state networks. These results suggest that there is a subtle reorganization occurring in the topology of brain networks in HAND, which can be captured using improved connectivity analysis.

No Evidence for Accelerated Aging-Related Brain Pathology in Treated Human Immunodeficiency Virus: Longitudinal Neuroimaging Results From the Comorbidity in Relation to AIDS (COBRA) Project

Background

Despite successful antiretroviral therapy, people living with human immunodeficiency virus (PLWH) experience higher rates of age-related morbidity, including abnormal brain structure, brain function, and cognitive impairment. This has raised concerns that PLWH may experience accelerated aging-related brain pathology.

Methods

We performed a multicenter longitudinal study of 134 virologically suppressed PLWH (median age, 56.0 years) and 79 demographically similar human immunodeficiency virus (HIV)-negative controls (median age, 57.2 years). To measure cognitive performance and brain pathology, we conducted detailed neuropsychological assessments and multimodality neuroimaging (T1-weighted, T2-weighted, diffusion magnetic resonance imaging [MRI], resting-state functional MRI, spectroscopy, arterial spin labeling) at baseline and at 2 years. Group differences in rates of change were assessed using linear mixed effects models.

Results

One hundred twenty-three PLWH and 78 HIV-negative controls completed longitudinal assessments (median interval, 1.97 years). There were no differences between PLWH and HIV-negative controls in age, sex, years of education, smoking or alcohol use. At baseline, PLWH had poorer global cognitive performance (P < .01), lower gray matter volume (P = .04), higher white matter hyperintensity load (P = .02), abnormal white matter microstructure (P < .005), and greater brain-predicted age difference (P = .01). Longitudinally, there were no significant differences in rates of change in any neuroimaging measure between PLWH and HIV-negative controls (P > .1). Cognitive performance was longitudinally stable in both groups.

Conclusions

We found no evidence that middle-aged PLWH, when receiving successful treatment, are at increased risk of accelerated aging-related brain changes or cognitive decline over 2 years.

Longitudinal brain atrophy patterns and neuropsychological performance in older adults with HIV-associated neurocognitive disorder compared with early Alzheimer's disease

Older HIV-infected patients are at risk for both HIV-associated neurocognitive disorder (HAND) and Alzheimer's disease. We investigated neuroimaging and neuropsychological performance of 61 virally suppressed older adults with HAND (mean (SD) age 64.3 (3.9) years), 53 demographically matched individuals with mild cognitive impairment of the Alzheimer's type (MCI-AD; 65.0 [4.8]), and 89 healthy controls (65.0 [4.3]) cross-sectionally and over 20 months. At the baseline, both disease groups exhibited lower volumes in multiple cortical and subcortical regions compared with controls. Hippocampal volume differentiated MCI-AD from HAND. Cognitively, MCI-AD performed worse on memory and language compared with HAND. Adjusted longitudinal models revealed greater diffuse brain atrophy in MCI-AD compared with controls, whereas HAND showed greater atrophy in frontal gray matter and cerebellum compared with controls. Comparing HAND with MCI-AD showed similar atrophy rates in all brain regions explored, with no significant findings. MCI-AD exhibited more pronounced language decline compared with HAND. These findings reveal the need for further work on unique cognitive phenotypes and neuroimaging signatures of HAND compared with early AD, providing preliminary clinical insight for differential diagnosis of age-related brain dysfunction in geriatric neuroHIV.

Novel Machine Learning Identifies Brain Patterns Distinguishing Diagnostic Membership of Human Immunodeficiency Virus, Alcoholism, and Their Comorbidity of Individuals

The incidence of alcohol use disorder (AUD) in human immunodeficiency virus (HIV) infection is twice that of the rest of the population. This study documents complex radiologically identified, neuroanatomical effects of AUD+HIV comorbidity by identifying structural brain systems that predicted diagnosis on an individual basis. Applying novel machine learning analysis to 549 participants (199 control subjects, 222 with AUD, 68 with HIV, 60 with AUD+HIV), 298 magnetic resonance imaging brain measurements were automatically reduced to small subsets per group. Significance of each diagnostic pattern was inferred from its accuracy in predicting diagnosis and performance on six cognitive measures. While all three diagnostic patterns predicted the learning and memory score, the AUD+HIV pattern was the largest and had the highest predication accuracy (78.1%). Providing a roadmap for analyzing large, multimodal datasets, the machine learning analysis revealed imaging phenotypes that predicted diagnostic membership of magnetic resonance imaging scans of individuals with AUD, HIV, and their comorbidity.

The current understanding of overlap between characteristics of HIV-associated neurocognitive disorders and Alzheimer's disease

The advent of effective antiretroviral medications (ARVs) has led to an aging of the HIV population with approximately 50% of people with HIV (PWH) being over the age of 50 years. Neurocognitive complications, typically known as HIV-associated neurocognitive disorders (HAND), persist in the era of ARVs and, in addition to risk of HAND, older PWH are also at risk for age-associated, neurodegenerative disorders including Alzheimer's disease (AD). It has been postulated that risk for AD may be greater among PWH due to potential compounding effects of HIV and aging on mechanisms of neural insult. We are now faced with the challenge of disentangling AD from HAND, which has important prognostic and treatment implications given the more rapidly debilitating trajectory of AD. Herein, we review the evidence to date demonstrating both parallels and differences in the profiles of HAND and AD. We specifically address similarities and difference of AD and HAND as it relates to (1) neuropsychological profiles (cross-sectional/longitudinal), (2) AD-associated neuropathological features as evidenced from neuropathological, cerebrospinal fluid and neuroimaging assessments, (3) biological mechanisms underlying cortical amyloid deposition, (4) parallels in mechanisms of neural insult, and (5) common risk factors. Our current understanding of the similarities and dissimilarities of AD and HAND should be further delineated and leveraged in the development of differential diagnostic methods that will allow for the early identification of AD and more suitable and effective treatment interventions among graying PWH.

Chained regularization for identifying brain patterns specific to HIV infection

Human Immunodeficiency Virus (HIV) infection continues to have major adverse public health and clinical consequences despite the effectiveness of combination Antiretroviral Therapy (cART) in reducing HIV viral load and improving immune function. As successfully treated individuals with HIV infection age, their cognition declines faster than reported for normal aging. This phenomenon underlines the importance of improving long-term care, which requires a better understanding of the impact of HIV on the brain. In this paper, automated identification of patients and brain regions affected by HIV infection are modeled as a classification problem, whose solution is determined in two steps within our proposed Chained-Regularization framework. The first step focuses on selecting the HIV pattern (i.e., the most informative constellation of brain region measurements for distinguishing HIV infected subjects from healthy controls) by constraining the search for the optimal parameter setting of the classifier via group sparsity (ℓ_2,1-norm). The second step improves classification accuracy by constraining the parameterization with respect to the selected measurements and the Euclidean regularization (ℓ₂-norm). When applied to the cortical and subcortical structural Magnetic Resonance Images (MRI) measurements of 65 controls and 65 HIV infected individuals, this approach is more accurate in distinguishing the two cohorts than more common models. Finally, the brain regions of the identified HIV pattern concur with the HIV literature that uses traditional group analysis models.

Multivariate Pattern Analysis of Volumetric Neuroimaging Data and Its Relationship With Cognitive Function in Treated HIV Disease

BACKGROUND

Accurate prediction of longitudinal changes in cognitive function would potentially allow for targeted intervention in those at greatest risk of cognitive decline. We sought to build a multivariate model using volumetric neuroimaging data alone to accurately predict cognitive function.

METHODS

Volumetric T1-weighted neuroimaging data from virally suppressed HIV-positive individuals from the CHARTER cohort (n = 139) were segmented into gray and white matter and spatially normalized before entering into machine learning models. Prediction of cognitive function at baseline and longitudinally was determined using leave-one-out cross-validation. In addition, a multivariate model of brain aging was used to measure the deviation of apparent brain age from chronological age and assess its relationship with cognitive function.

RESULTS

Cognitive impairment, defined using the global deficit score, was present in 37.4%. However, it was generally mild and occurred more commonly in those with confounding comorbidities (P < 0.001). Although multivariate prediction of cognitive impairment as a dichotomous variable at baseline was poor (area under the receiver operator curve 0.59), prediction of the global T-score was better than a comparable linear model (adjusted R = 0.08, P < 0.01 vs. adjusted R = 0.01, P = 0.14). Accurate prediction of longitudinal changes in cognitive function was not possible (P = 0.82). Brain-predicted age exceeded chronological age by mean (95% confidence interval) 1.17 (-0.14 to 2.53) years but was greatest in those with confounding comorbidities [5.87 (1.74 to 9.99) years] and prior AIDS [3.03 (0.00 to 6.06) years].

CONCLUSION

Accurate prediction of cognitive impairment using multivariate models using only T1-weighted data was not achievable, which may reflect the small sample size, heterogeneity of the data, or that impairment was usually mild.

Alcohol use effects on adolescent brain development revealed by simultaneously removing confounding factors, identifying morphometric patterns, and classifying individuals

Group analysis of brain magnetic resonance imaging (MRI) metrics frequently employs generalized additive models (GAM) to remove contributions of confounding factors before identifying cohort specific characteristics. For example, the National Consortium on Alcohol and NeuroDevelopment in Adolescence (NCANDA) used such an approach to identify effects of alcohol misuse on the developing brain. Here, we hypothesized that considering confounding factors before group analysis removes information relevant for distinguishing adolescents with drinking history from those without. To test this hypothesis, we introduce a machine-learning model that identifies cohort-specific, neuromorphometric patterns by simultaneously training a GAM and generic classifier on macrostructural MRI and microstructural diffusion tensor imaging (DTI) metrics and compare it to more traditional group analysis and machine-learning approaches. Using a baseline NCANDA MR dataset (N = 705), the proposed machine learning approach identified a pattern of eight brain regions unique to adolescents who misuse alcohol. Classifying high-drinking adolescents was more accurate with that pattern than using regions identified with alternative approaches. The findings of the joint model approach thus were (1) impartial to confounding factors; (2) relevant to drinking behaviors; and (3) in concurrence with the alcohol literature.

Differentiating HIV-Associated Neurocognitive Disorders From Alzheimer's Disease: an Emerging Issue in Geriatric NeuroHIV

PURPOSE OF REVIEW

The purpose of this review was to examine characteristics that may distinguish HIV-associated neurocognitive disorder (HAND) from early Alzheimer's disease (AD).

RECENT FINDINGS

Cerebrospinal fluid (CSF) AD biomarkers are perturbed in HIV, yet these alterations may be limited to settings of advanced dementia or unsuppressed plasma HIV RNA. Neuropsychological testing will require extensive batteries to maximize utility. Structural imaging is limited for early AD detection in the setting of HIV, but proper studies are absent. While positron-emission tomography (PET) amyloid imaging has altered the landscape of differential diagnosis for age-associated neurodegenerative disorders, costs are prohibitive. Risk for delayed AD diagnosis in the aging HIV-infected population is now among the most pressing issues in geriatric neuroHIV. While clinical, imaging, and biomarker characterizations of AD are extensively defined, fewer data define characteristics of HIV-associated neurocognitive disorder in the setting of suppressed plasma HIV RNA. Data needed to inform the phenotype of AD in the setting of HIV are equally few.