Uncovering Biologically Coherent Peripheral Signatures of Health and Risk for Alzheimer's Disease in the Aging Brain

Mental health phenotypes of well-controlled HIV in Uganda

Introduction

The phenotypic expression of mental health (MH) conditions among people with HIV (PWH) in Uganda and worldwide are heterogeneous. Accordingly, there has been a shift toward identifying MH phenotypes using data-driven methods capable of identifying novel insights into mechanisms of divergent MH phenotypes among PWH. We leverage the analytic strengths of machine learning combined with inferential methods to identify novel MH phenotypes among PWH and the underlying explanatory features.

Methods

A total of 277 PWH (46% female, median age = 44; 93% virally suppressed [<50copies/mL]) were included in the analyses. Participants completed the Patient Health Questionnaire (PHQ-9), Beck Anxiety Inventory (BAI), and the PTSD Checklist-Civilian (PCL-C). A clustering pipeline consisting of dimension reduction with UMAP followed by HBDScan was used to identify MH subtypes using total symptom scores. Inferential statistics compared select demographic (age, sex, education), viral load, and early life adversity between clusters.

Results

We identified four MH phenotypes. Cluster 1 (n = 76; PTSD phenotype) endorsed clinically significant PTSD symptoms (average PCL-C total score > 33). Clusters 2 (n = 32; anxiety phenotype) and 3 (n = 130; mixed anxiety/depression phenotype) reported minimal PTSD symptoms, with modest BAI (Cluster 2) and PHQ-9 (Cluster 3) elevations. Cluster 4 (n = 39; minimal symptom phenotype) reported no clinical MH symptom elevations. Comparisons revealed higher rates of sexual abuse during childhood among the PTSD phenotype vs. the minimal symptom phenotype (p = 0.03).

Discussion

We identified unique MH phenotypes among PWH and confirmed the importance of early life adversity as an early risk determinant for unfavorable MH among PWH in adulthood.

Broadband Long Wavelength Light Promotes Myopic Eye Growth and Alters Retinal Responses to Light Offset in Chick

Purpose

Prolonged exposure to broadband light with a short-wavelength (blue) or long-wavelength (orange/red) bias is known to impact eye growth and refraction, but the mechanisms underlying this response are unknown. Thus, the present study investigated the effects of broadband blue and orange lights with well-differentiated spectrums on refractive development and global flash electroretinography (gfERG) measures of retinal function in the chick myopia model.

Methods

Chicks were raised for 4 days with monocular negative lenses, or no lens, under blue, orange, or white light. Chick weight, eye dimensions, and refraction were measured at the conclusion of rearing. In a separate cohort of chicks, the effect of 4 days of colored light rearing on retinal responses to orange, blue, or white light flashes was assessed using gfERG.

Results

Chicks reared under orange light for 4 days exhibited a significantly larger myopic shift in response to negative lenses compared to those reared under blue light. Orange light rearing for 4 days increased the gfERG d-wave amplitude and implicit time in response to orange light flashes but did not alter responses to white or blue flashes. Blue and white light rearing did not affect the retina's response to light flashes of any color.

Conclusions

Orange light rearing exacerbated defocus-induced myopia relative to blue light rearing. The gfERG recordings revealed that prolonged orange light exposure increased retinal responsivity to the offset of long wavelength light flashes, suggesting a potential role for ON/OFF pathway balance in generating the refractive response that requires further electrophysiological and molecular investigation.

Integrative Transcriptome and Proteome Analyses Elucidate the Mechanism of Lens-Induced Myopia in Mice

Purpose

The purpose of this study was to investigate the underlying molecular mechanism of lens-induced myopia (LIM) through transcriptome and proteome analyses with a modified mouse myopia model.

Methods

Four-week-old C57BL/6J mice were treated with a homemade newly designed -25 diopter (D) lens mounting by a 3D printing pen before right eyes for 4 weeks. Refraction (RE) and axial dimensions were measured every 2 weeks. Retinas were analyzed by RNA-sequencing and data-independent acquisition liquid chromatography tandem mass spectrometry. Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) annotation, and STRING databases were used to identify significantly affected pathways in transcriptomic and proteomic data sets. Western blot was used to detect the expression of specific proteins.

Results

The modified model was accessible and efficient. Mice displayed a significant myopic shift (approximately 8 D) following 4 weeks' of lens treatment. Through transcriptomics and proteomics analysis, we elucidated 175 differently expressed genes (DEGs) and 646 differentially expressed proteins (DEPs) between binoculus. The transcriptomic and proteomic data showed a low correlation. Going over the mRNA protein matches, insulin like growth factor 2 mRNA binding protein 1 (Igf2bp1) was found to be a convincing biomarker of LIM, which was confirmed by Western blot. RNA-seq and proteome profiling confirmed that these two "omics" data sets complemented one another in KEGG pathways annovation. Among these, metabolic and human diseases pathways were considered to be correlated with the LIM forming process.

Conclusions

The newly constructed LIM model provides a useful tool for future myopia research. Combining transcriptomic and proteomic analysis may potentially brighten the prospects of novel therapeutic targets for patients with myopia.

Gene expression profile analyses to identify potential biomarkers for myopia

PURPOSE

Increasing evidence suggests myopia is not a simple refractive error, many other factors might also be involved. Here, we assessed myopic and normal corneas' gene expression profiles to identify possible diagnostic and therapeutic biomarkers for myopia.

MATERIALS AND METHODS

We obtained the expression profile of ten patients and seven normal control samples from the GSE112155 and GSE151631 datasets based on the Gene Expression Omnibus (GEO) database. We used the "limma" R package to determine the differentially expressed genes (DEGs) between myopic and normal corneas. Weighted gene co-expression network analysis (WGCNA) was used to identify critical co-expressed modules related to myopia, and enrichment analyses were used to annotate the function of genes encompassed in the compulsory module. We also validated these findings in two external datasets (GSE24641 and GSE136701).

RESULTS

We identified that the DEGs were significantly enriched in ultraviolet (UV) response, TNF-α signaling via NFκB, Angiogenesis, Myogenesis pathways, etc. We used 2095 genes to construct the co-expression gene modules and found five interesting modules because the eigengene expression of these modules was significantly differentially expressed between myopic and normal corneas. Notably, the enrichment analysis found that the genes encompassed in lightgreen module were significantly enriched in immune-related pathways. These findings were proved by subsequent analysis based on Xcell software. We found the component of B cells, CD4+ memory T cells, CD8+ central memory T cells, plasmacytoid dendritic cells, T helper 2 (Th2) cells, regulatory T cells (Tregs), etc. were significantly increased in myopic corneas, while CD8+ T cells, CD4+ T central memory cells, natural killer T (NKT) cells, and T helper 1 (Th1) cells were significantly decreased.

CONCLUSION

Our findings identified some markers that might detect diagnosis and treatment for myopia from cornea aspect. Future studies are warranted to verify the functional role of immune-related pathways in cornea during the pathogenesis or progression of myopia.

The foundation and architecture of precision medicine in neurology and psychiatry

Neurological and psychiatric diseases have high degrees of genetic and pathophysiological heterogeneity, irrespective of clinical manifestations. Traditional medical paradigms have focused on late-stage syndromic aspects of these diseases, with little consideration of the underlying biology. Advances in disease modeling and methodological design have paved the way for the development of precision medicine (PM), an established concept in oncology with growing attention from other medical specialties. We propose a PM architecture for central nervous system diseases built on four converging pillars: multimodal biomarkers, systems medicine, digital health technologies, and data science. We discuss Alzheimer's disease (AD), an area of significant unmet medical need, as a case-in-point for the proposed framework. AD can be seen as one of the most advanced PM-oriented disease models and as a compelling catalyzer towards PM-oriented neuroscience drug development and advanced healthcare practice.

Metabolomics facilitates the discovery of metabolic profiles and pathways for myopia: A systematic review

BACKGROUND

Myopia is one of the major eye disorders and the global burden is increasing rapidly. Our purpose is to systematically summarize potential metabolic biomarkers and pathways in myopia to facilitate the understanding of disease mechanisms as well as the discovery of novel therapeutic measures.

METHODS

Myopia-related metabolomics studies were searched in electronic databases of PubMed and Web of Science until June 2021. Information regarding clinical and demographic characteristics of included studies and metabolomics findings were extracted. Myopia-related metabolic pathways were analysed for differential metabolic profiles, and the quality of included studies was assessed based on the QUADOMICS tool. Pathway analyses of differential metabolites were performed using bioinformatics tools and online software such as the Metaboanalyst 5.0.

RESULTS

The myopia-related metabolomics studies included in this study consisted of seven human and two animal studies. The results of the study quality assessment showed that studies were all phase I studies and all met the evaluation criteria of 70% or more. The myopia-control serum study identified 23 differential metabolites with the Sphingolipid metabolism pathway beings enriched. The high myopia-cataract aqueous humour study identified 40 differential metabolites with the Arginine biosynthesis pathway being enriched. The high myopia-control serum study identified 43 differential metabolites and 4 pathways were significantly associated with metabolites including Citrate cycle; Alanine, aspartate and glutamate metabolism; Glyoxylate and dicarboxylate metabolism; Biosynthesis of unsaturated fatty acids (all P value < 0.05).

CONCLUSIONS

This study summarizes potential metabolic biomarkers and pathways in myopia, providing new clues to elucidate disease mechanisms.

Individual Differences in CD4/CD8 T-Cell Ratio Trajectories and Associated Risk Profiles Modeled From Acute HIV Infection

OBJECTIVE

We examined individual differences in CD4/CD8 T-cell ratio trajectories and associated risk profiles from acute HIV infection (AHI) through 144 weeks of antiretroviral therapy (ART) using a data-driven approach.

METHODS

A total of 483 AHI participants began ART during Fiebig I-V and completed follow-up evaluations for 144 weeks. CD4+, CD8+, and CD4/CD8 T-cell ratio trajectories were defined followed by analyses to identify associated risk variables.

RESULTS

Participants had a median viral load (VL) of 5.88 copies/ml and CD4/CD8 T-cell ratio of 0.71 at enrollment. After 144 weeks of ART, the median CD4/CD8 T-cell ratio was 1.3. Longitudinal models revealed five CD4/CD8 T-cell ratio subgroups: group 1 (3%) exhibited a ratio >1.0 at all visits; groups 2 (18%) and 3 (29%) exhibited inversion at enrollment, with normalization 4 and 12 weeks after ART, respectively; and groups 4 (31%) and 5 (18%) experienced CD4/CD8 T-cell ratio inversion due to slow CD4+ T-cell recovery (group 4) or high CD8+ T-cell count (group 5). Persistent inversion corresponded to ART onset after Fiebig II, higher VL, soluble CD27 and TIM-3, and lower eosinophil count. Individuals with slow CD4+ T-cell recovery exhibited higher VL, lower white blood cell count, lower basophil percent, and treatment with standard ART, as well as worse mental health and cognition, compared with individuals with high CD8+ T-cell count.

CONCLUSIONS

Early HIV disease dynamics predict unfavorable CD4/CD8 T-cell ratio outcomes after ART. CD4+ and CD8+ T-cell trajectories contribute to inversion risk and correspond to specific viral, immune, and psychological profiles during AHI. Adjunctive strategies to achieve immune normalization merit consideration.

Using the Alzheimer's Disease Neuroimaging Initiative to improve early detection, diagnosis, and treatment of Alzheimer's disease

INTRODUCTION

The Alzheimer's Disease Neuroimaging Initiative (ADNI) has accumulated 15 years of clinical, neuroimaging, cognitive, biofluid biomarker and genetic data, and biofluid samples available to researchers, resulting in more than 3500 publications. This review covers studies from 2018 to 2020.

METHODS

We identified 1442 publications using ADNI data by conventional search methods and selected impactful studies for inclusion.

RESULTS

Disease progression studies supported pivotal roles for regional amyloid beta (Aβ) and tau deposition, and identified underlying genetic contributions to Alzheimer's disease (AD). Vascular disease, immune response, inflammation, resilience, and sex modulated disease course. Biologically coherent subgroups were identified at all clinical stages. Practical algorithms and methodological changes improved determination of Aβ status. Plasma Aβ, phosphorylated tau181, and neurofilament light were promising noninvasive biomarkers. Prognostic and diagnostic models were externally validated in ADNI but studies are limited by lack of ethnocultural cohort diversity.

DISCUSSION

ADNI has had a profound impact in improving clinical trials for AD.

A tale of two systems: Lessons learned from female mid-life aging with implications for Alzheimer's prevention & treatment

Neurological aging is frequently viewed as a linear process of decline, whereas in reality, it is a dynamic non-linear process. The dynamic nature of neurological aging is exemplified during midlife in the female brain. To investigate fundamental mechanisms of midlife aging that underlie risk for development of Alzheimer's disease (AD) in late life, we investigated the brain at greatest risk for the disease, the aging female brain. Outcomes of our research indicate that mid-life aging in the female is characterized by the emergence of three phases: early chronological (pre-menopause), endocrinological (peri-menopause) and late chronological (post-menopause) aging. The endocrinological aging program is sandwiched between early and late chronological aging. Throughout the three stages of midlife aging, two systems of biology, metabolic and immune, are tightly integrated through a network of signaling cascades. The network of signaling between these two systems of biology underlie an orchestrated sequence of adaptative starvation responses that shift the brain from near exclusive dependence on a single fuel, glucose, to utilization of an auxiliary fuel derived from lipids, ketone bodies. The dismantling of the estrogen control of glucose metabolism during mid-life aging is a critical contributor to the shift in fuel systems and emergence of dynamic neuroimmune phenotype. The shift in fuel reliance, puts the largest reservoir of local fatty acids, white matter, at risk for catabolism as a source of lipids to generate ketone bodies through astrocytic beta oxidation. APOE4 genotype accelerates the tipping point for emergence of the bioenergetic crisis. While outcomes derived from research conducted in the female brain are not directly translatable to the male brain, the questions addressed in a female centric program of research are directly applicable to investigation of the male brain. Like females, males with AD exhibit deficits in the bioenergetic system of the brain, activation of the immune system and hallmark Alzheimer's pathologies. The drivers and trajectory of mechanisms underlying neurodegeneration in the male brain will undoubtedly share common aspects with the female in addition to factors unique to the male. Preclinical and clinical evidence indicate that midlife endocrine aging can also be a transitional bridge to autoimmune disorders. Collectively, the data indicate that endocrinological aging is a critical period "tipping point" in midlife which can initiate emergence of the prodromal stage of late-onset-Alzheimer's disease. Interventions that target both immune and metabolic shifts that occur during midlife aging have the potential to alter the trajectory of Alzheimer's risk in late life. Further, to achieve precision medicine for AD, chromosomal sex is a critical variable to consider along with APOE genotype, other genetic risk factors and stage of disease.

Transcriptional profiling of the chick retina identifies down-regulation of VIP and UTS2B genes during early lens-induced myopia

Gene expression of the chick retina was examined during the early development of lens-induced myopia (LIM) using whole transcriptome sequencing. Monocular treatment of the right eyes with −10 diopter (D)...

The complement cascade in Alzheimer's disease: a systematic review and meta-analysis

Genetic evidence implicates a causal role for the complement pathway in Alzheimer's disease (AD). Since studies have shown inconsistent differences in cerebrospinal fluid (CSF) and peripheral blood complement protein concentrations between AD patients and healthy elderly, this study sought to summarize the clinical data. Original peer-reviewed articles measuring CSF and/or blood concentrations of complement or complement regulator protein concentrations in AD and healthy elderly control (HC) groups were included. Of 2966 records identified, means and standard deviations from 86 studies were summarized as standardized mean differences (SMD) by random effects meta-analyses. In CSF, concentrations of clusterin (N_AD/N_HC = 625/577, SMD = 0.53, Z₈ = 8.81, p < 0.005; I² < 0.005%) and complement component 3 (C3; N_AD/N_HC = 299/522, SMD = 0.45, Z₃ = 3.21, p < 0.005; I² = 68.40%) were significantly higher in AD, but differences in C1q, C-reactive protein (CRP), serum amyloid protein (SAP), and factor H concentrations were not significant. In peripheral blood, concentrations of CRP were elevated in AD (N_AD/N_HC = 3404/3332, SMD = 0.44, Z₄₃ = 3.43, p < 0.005; I² = 93.81%), but differences between groups in C3, C4, C1-inhibitor, SAP, factor H and clusterin concentrations were not significant, and inconsistent between studies. Of 64 complement pathway proteins or regulators in the quantitative synthesis, trends in C1q, factor B, C4a, and late-stage complement pathway components (e.g. C9) in blood, C4 in CSF, and the membrane attack complex in blood and CSF, might be investigated further. The results collectively support elevated complement pathway activity in AD, which was best characterized by increased CSF clusterin concentrations and less consistently by CSF C3 concentrations. Complement activity related to an AD diagnosis was not reflected consistently by the peripheral blood proteins investigated.

Long-Wavelength-Filtered Light Transiently Inhibits Negative Lens-Induced Axial Eye Growth in the Chick Myopia Model

Purpose

Eye growth and myopia development in chicks, and some other animal models, can be suppressed by rearing under near-monochromatic, short-wavelength blue light. We aimed to determine whether similar effects could be achieved using glass filters that transmit a broader range of short and middle wavelengths.

Methods

On day 6 or 7 post-hatch, 169 chicks were assigned to one of three monocular lens conditions (−10 D, +10 D, plano) and reared for 7 or 10 days under one of four 201-lux lighting conditions: (1) B410 long-wavelength–filtered light, (2) B460 long-wavelength–filtered light, (3) Y48 short-wavelength–filtered light, or (4) HA50 broadband light.

Results

At 7 days, B410 (but not B460) long-wavelength–filtered light had significantly inhibited negative lens induced axial growth relative to Y48 short-wavelength–filtered light (mean difference in experimental eye = −0.249 mm; P = 0.006) and HA50 broadband light (mean difference = −0.139 mm; P = 0.038). B410 filters also inhibited the negative lens-induced increase in vitreous chamber depth relative to all other filter conditions. Corresponding changes in refraction did not occur, and biometric measurements in a separate cohort of chicks suggested that the axial dimension changes were transient and not maintained at 10 days.

Conclusions

Chromatic effects on eye growth can be achieved using filters that transmit a broad range of wavelengths even in the presence of strong cues for myopia development.

Translational Relevance

Broad-wavelength filters that provide a more “naturalistic” visual experience relative to monochromatic light have potential to alter myopia development, although the effects shown here were modest and transient and require exploration in further species.

Genome-wide analysis of retinal transcriptome reveals common genetic network underlying perception of contrast and optical defocus detection

BACKGROUND

Refractive eye development is regulated by optical defocus in a process of emmetropization. Excessive exposure to negative optical defocus often leads to the development of myopia. However, it is still largely unknown how optical defocus is detected by the retina.

METHODS

Here, we used genome-wide RNA-sequencing to conduct analysis of the retinal gene expression network underlying contrast perception and refractive eye development.

RESULTS

We report that the genetic network subserving contrast perception plays an important role in optical defocus detection and emmetropization. Our results demonstrate an interaction between contrast perception, the retinal circadian clock pathway and the signaling pathway underlying optical defocus detection. We also observe that the relative majority of genes causing human myopia are involved in the processing of optical defocus.

CONCLUSIONS

Together, our results support the hypothesis that optical defocus is perceived by the retina using contrast as a proxy and provide new insights into molecular signaling underlying refractive eye development.

Ensemble machine learning classification of daily living abilities among older people with HIV

BACKGROUND

clinically relevant methods to identify individuals at risk for impaired daily living abilities secondary to neurocognitive impairment (ADLs) remain elusive. This is especially true for complex clinical conditions such as HIV-Associated Neurocognitive Disorders (HAND). The aim of this study was to identify novel and modifiable factors that have potential to improve diagnostic accuracy of ADL risk, with the long-term goal of guiding future interventions to minimize ADL disruption.

METHODS

study participants included 79 people with HIV (PWH; mean age = 63; range = 55-80) enrolled in neuroHIV studies at University California San Francisco (UCSF) between 2016 and 2019. All participants were virally suppressed and exhibited objective evidence of neurocognitive impairment. ADL status was defined as either normative (n = 39) or at risk (n = 40) based on a task-based protocol. Gradient boosted multivariate regression (GBM) was employed to identify the combination of variables that differentiated ADL subgroup classification. Predictor variables included demographic factors, HIV disease severity indices, brain white matter integrity quantified using diffusion tensor imaging, cognitive test performance, and health co-morbidities. Model performance was examined using average Area Under the Curve (AUC) with repeated five-fold cross validation.

FINDINGS

the univariate GBM yielded an average AUC of 83% using Wide Range Achievement test 4 (WRAT-4) reading score, self-reported thought confusion and difficulty reading, radial diffusivity (RD) in the left external capsule, fractional anisotropy (FA) in the left cingulate gyrus, and Stroop performance. The model allowing for two-way interactions modestly improved classification performance (AUC of 88%) and revealed synergies between race, reading ability, cognitive performance, and neuroimaging metrics in the genu and uncinate fasciculus. Conversion of Neuropsychological Assessment Battery Daily Living Module (NAB-DLM) performance from raw scores into T scores amplified differences between White and non-White study participants.

INTERPRETATION

demographic and sociocultural factors are critical determinants of ADL risk status among older PWH who meet diagnostic criteria for neurocognitive impairment. Task-based ADL assessment that relies heavily on reading proficiency may artificially inflate the frequency/severity of ADL impairment among diverse clinical populations. Culturally relevant measures of ADL status are needed for individuals with acquired neurocognitive disorders, including HAND.

Anomaly Analysis of Alzheimer’s Disease in PET Images Using an Unsupervised Adversarial Deep Learning Model

In this study, the anomaly analysis of Alzheimer’s disease using positron emission tomography (PET) images using an unsupervised proposed adversarial model is investigated. The model consists of three parts: a parallel-network encoder, which is comprised of a convolutional pipeline and a dilated convolutional pipeline that extracts global and local features and concatenates them, a decoder that reconstructs the input image from the obtained feature vector, and a discriminator that distinguishes if the input image image is real or fake. The hypothesis is that if the proposed model is trained with only normal brain images, the corresponding construction loss for normal images should be minimal. However, if the input image belongs to a class that is designated as an anomaly that which the model is not trained with, then the construction loss will be high. This will reflect during the anomaly score comparison between the normal and the anomalous image. A multi-case analysis is performed for three major classes using the Alzheimer’s Disease Neuroimaging Initiative dataset, Alzheimer’s disease, mild cognitive impairment, and normal control. The base parallel-encoder network shows better classification accuracy than the benchmark models, and the proposed model that is built on the parallel model outperforms the benchmark anomaly detection models. The proposed model gave out 96.03% and 75.21% in classification and area under the curve score, respectively. Additionally, a qualitative evaluation done by using Fréchet inception distance gave a better score than the state-of-the-art by three points.

Good problems to have? Policy and societal implications of a disease-modifying therapy for presymptomatic late-onset Alzheimer's disease

In the United States alone, the prevalence of AD is expected to more than double from six million people in 2019 to nearly 14 million people in 2050. Meanwhile, the track record for developing treatments for AD has been marked by decades of failure. But recent progress in genetics, neuroscience and gene editing suggest that effective treatments could be on the horizon. The arrival of such treatments would have profound implications for the way we diagnose, triage, study, and allocate resources to Alzheimer's patients. Because the disease is not rare and because it strikes late in life, the development of therapies that are expensive and efficacious but less than cures, will pose particular challenges to healthcare infrastructure. We have a window of time during which we can begin to anticipate just, equitable and salutary ways to accommodate a disease-modifying therapy Alzheimer's disease. Here we consider the implications for caregivers, clinicians, researchers, and the US healthcare system of the availability of an expensive, presymptomatic treatment for a common late-onset neurodegenerative disease for which diagnosis can be difficult.

Medical Informatics Platform (MIP): A Pilot Study Across Clinical Italian Cohorts

Introduction: With the shift of research focus to personalized medicine in Alzheimer's Dementia (AD), there is an urgent need for tools that are capable of quantifying a patient's risk using diagnostic biomarkers. The Medical Informatics Platform (MIP) is a distributed e-infrastructure federating large amounts of data coupled with machine-learning (ML) algorithms and statistical models to define the biological signature of the disease. The present study assessed (i) the accuracy of two ML algorithms, i.e., supervised Gradient Boosting (GB) and semi-unsupervised 3C strategy (Categorize, Cluster, Classify-CCC) implemented in the MIP and (ii) their contribution over the standard diagnostic workup. Methods: We examined individuals coming from the MIP installed across 3 Italian memory clinics, including subjects with Normal Cognition (CN, n = 432), Mild Cognitive Impairment (MCI, n = 456), and AD (n = 451). The GB classifier was applied to best discriminate the three diagnostic classes in 1,339 subjects, and the CCC strategy was used to refine the classical disease categories. Four dementia experts provided their diagnostic confidence (DC) of MCI conversion on an independent cohort of 38 patients. DC was based on clinical, neuropsychological, CSF, and structural MRI information and again with addition of the outcome from the MIP tools. Results: The GB algorithm provided a classification accuracy of 85% in a nested 10-fold cross-validation for CN vs. MCI vs. AD discrimination. Accuracy increased to 95% in the holdout validation, with the omission of each Italian clinical cohort out in turn. CCC identified five homogeneous clusters of subjects and 36 biomarkers that represented the disease fingerprint. In the DC assessment, CCC defined six clusters in the MCI population used to train the algorithm and 29 biomarkers to improve patients staging. GB and CCC showed a significant impact, evaluated as +5.99% of increment on physicians' DC. The influence of MIP on DC was rated from "slight" to "significant" in 80% of the cases. Discussion: GB provided fair results in classification of CN, MCI, and AD. CCC identified homogeneous and promising classes of subjects via its semi-unsupervised approach. We measured the effect of the MIP on the physician's DC. Our results pave the way for the establishment of a new paradigm for ML discrimination of patients who will or will not convert to AD, a clinical priority for neurology.

Classification of Alzheimer's Disease and Mild Cognitive Impairment Based on Cortical and Subcortical Features from MRI T1 Brain Images Utilizing Four Different Types of Datasets

Alzheimer's disease (AD) is one of the most common neurodegenerative illnesses (dementia) among the elderly. Recently, researchers have developed a new method for the instinctive analysis of AD based on machine learning and its subfield, deep learning. Recent state-of-the-art techniques consider multimodal diagnosis, which has been shown to achieve high accuracy compared to a unimodal prognosis. Furthermore, many studies have used structural magnetic resonance imaging (MRI) to measure brain volumes and the volume of subregions, as well as to search for diffuse changes in white/gray matter in the brain. In this study, T1-weighted structural MRI was used for the early classification of AD. MRI results in high-intensity visible features, making preprocessing and segmentation easy. To use this image modality, we acquired four types of datasets from each dataset's server. In this work, we downloaded 326 subjects from the National Research Center for Dementia homepage, 123 subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI) homepage, 121 subjects from the Alzheimer's Disease Repository Without Borders homepage, and 131 subjects from the National Alzheimer's Coordinating Center homepage. In our experiment, we used the multiatlas label propagation with expectation–maximization-based refinement segmentation method. We segmented the images into 138 anatomical morphometry images (in which 40 features belonged to subcortical volumes and the remaining 98 features belonged to cortical thickness). The entire dataset was split into a 70 : 30 (training and testing) ratio before classifying the data. A principal component analysis was used for dimensionality reduction. Then, the support vector machine radial basis function classifier was used for classification between two groups—AD versus health control (HC) and early mild cognitive impairment (MCI) (EMCI) versus late MCI (LMCI). The proposed method performed very well for all four types of dataset. For instance, for the AD versus HC group, the classifier achieved an area under curve (AUC) of more than 89% for each dataset. For the EMCI versus LMCI group, the classifier achieved an AUC of more than 80% for every dataset. Moreover, we also calculated Cohen kappa and Jaccard index statistical values for all datasets to evaluate the classification reliability. Finally, we compared our results with those of recently published state-of-the-art methods.

Machine Learning Analysis Reveals Novel Neuroimaging and Clinical Signatures of Frailty in HIV

BACKGROUND

Frailty is an important clinical concern for the aging population of people living with HIV (PLWH). The objective of this study was to identify the combination of risk features that distinguish frail from nonfrail individuals.

SETTING

Machine learning analysis of highly dimensional risk features was performed on a clinical cohort of PLWH.

METHODS

Participants included 105 older (average age = 55.6) PLWH, with at least a 3-month history of combination antiretroviral therapy (median CD4 = 546). Predictors included demographics, HIV clinical markers, comorbid health conditions, cognition, and neuroimaging (ie, volumetrics, resting-state functional connectivity, and cerebral blood flow). Gradient-boosted multivariate regressions were implemented to establish linear and interactive classification models. Model performance was determined by sensitivity/specificity (F1 score) with 5-fold cross validation.

RESULTS

The linear gradient-boosted multivariate regression classifier included lower current CD4 count, lower psychomotor performance, and multiple neuroimaging indices (volumes, network connectivity, and blood flow) in visual and motor brain systems (F1 score = 71%; precision = 84%; and sensitivity = 66%). The interactive model identified novel synergies between neuroimaging features, female sex, symptoms of depression, and current CD4 count.

CONCLUSIONS

Data-driven algorithms built from highly dimensional clinical and brain imaging features implicate disruption to the visuomotor system in older PLWH designated as frail individuals. Interactions between lower CD4 count, female sex, depressive symptoms, and neuroimaging features suggest potentiation of risk mechanisms. Longitudinal data-driven studies are needed to guide clinical strategies capable of preventing the development of frailty as PLWH reach advanced age.

Visual Image Quality Impacts Circadian Rhythm-Related Gene Expression in Retina and in Choroid: A Potential Mechanism for Ametropias

Purpose

Stimulated by evidence implicating diurnal/circadian rhythms and light in refractive development, we studied the expression over 24 hours of selected clock and circadian rhythm-related genes in retina/retinal pigment epithelium (RPE) and choroid of experimental ametropias in chicks.

Methods

Newly hatched chicks, entrained to a 12-hour light/dark cycle for 12 to 14 days, either experienced nonrestricted vision OU (i.e., in both eyes) or received an image-blurring diffuser or a minus 10-diopter (D) or a plus 10-D defocusing lens over one eye. Starting 1 day later and at 4-hour intervals for 24 hours, the retina/RPE and choroid were separately dissected. Without pooling, total RNA was extracted, converted to cDNA, and assayed by quantitative PCR for the expression of the following genes: Opn4m, Clock, Npas2, Per3, Cry1, Arntl, and Mtnr1a.

Results

The expression of each gene in retina/RPE and in choroid of eyes with nonrestricted vision OU varied over 24 hours, with equal levels OU for most genes and times. Altered visual input influenced gene expression in complex patterns that varied by gene, visual input, time, and eye, affecting experimental eyes with altered vision and also contralateral eyes with nonrestricted vision.

Discussion

Altering visual input in ways known to induce ametropias alters the retinal/RPE and choroidal expression of circadian rhythm-related genes, further linking circadian biology with eye growth regulation. While further investigations are needed, studying circadian processes may help understand refractive mechanisms and the increasing myopia prevalence in contemporary societies where lighting patterns can desynchronize endogenous rhythms from the natural environmental light/dark cycle.

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.

Deep Learning in Alzheimer's Disease: Diagnostic Classification and Prognostic Prediction Using Neuroimaging Data

Deep learning, a state-of-the-art machine learning approach, has shown outstanding performance over traditional machine learning in identifying intricate structures in complex high-dimensional data, especially in the domain of computer vision. The application of deep learning to early detection and automated classification of Alzheimer's disease (AD) has recently gained considerable attention, as rapid progress in neuroimaging techniques has generated large-scale multimodal neuroimaging data. A systematic review of publications using deep learning approaches and neuroimaging data for diagnostic classification of AD was performed. A PubMed and Google Scholar search was used to identify deep learning papers on AD published between January 2013 and July 2018. These papers were reviewed, evaluated, and classified by algorithm and neuroimaging type, and the findings were summarized. Of 16 studies meeting full inclusion criteria, 4 used a combination of deep learning and traditional machine learning approaches, and 12 used only deep learning approaches. The combination of traditional machine learning for classification and stacked auto-encoder (SAE) for feature selection produced accuracies of up to 98.8% for AD classification and 83.7% for prediction of conversion from mild cognitive impairment (MCI), a prodromal stage of AD, to AD. Deep learning approaches, such as convolutional neural network (CNN) or recurrent neural network (RNN), that use neuroimaging data without pre-processing for feature selection have yielded accuracies of up to 96.0% for AD classification and 84.2% for MCI conversion prediction. The best classification performance was obtained when multimodal neuroimaging and fluid biomarkers were combined. Deep learning approaches continue to improve in performance and appear to hold promise for diagnostic classification of AD using multimodal neuroimaging data. AD research that uses deep learning is still evolving, improving performance by incorporating additional hybrid data types, such as-omics data, increasing transparency with explainable approaches that add knowledge of specific disease-related features and mechanisms.

Analysis of genetic networks regulating refractive eye development in collaborative cross progenitor strain mice reveals new genes and pathways underlying human myopia

BACKGROUND

Population studies suggest that genetic factors play an important role in refractive error development; however, the precise role of genetic background and the composition of the signaling pathways underlying refractive eye development remain poorly understood.

METHODS

Here, we analyzed normal refractive development and susceptibility to form-deprivation myopia in the eight progenitor mouse strains of the Collaborative Cross (CC). We used RNA-seq to analyze gene expression in the retinae of these mice and reconstruct genetic networks and signaling pathways underlying refractive eye development. We also utilized genome-wide gene-based association analysis to identify mouse genes and pathways associated with myopia in humans.

RESULTS

Genetic background strongly influenced both baseline refractive development and susceptibility to environmentally-induced myopia. Baseline refractive errors ranged from - 21.2 diopters (D) in 129S1/svlmj mice to + 22.0 D in CAST/EiJ mice and represented a continuous distribution typical of a quantitative genetic trait. The extent of induced form-deprivation myopia ranged from - 5.6 D in NZO/HILtJ mice to - 20.0 D in CAST/EiJ mice and also followed a continuous distribution. Whole-genome (RNA-seq) gene expression profiling in retinae from CC progenitor strains identified genes whose expression level correlated with either baseline refractive error or susceptibility to myopia. Expression levels of 2,302 genes correlated with the baseline refractive state of the eye, whereas 1,917 genes correlated with susceptibility to induced myopia. Genome-wide gene-based association analysis in the CREAM and UK Biobank human cohorts revealed that 985 of the above genes were associated with myopia in humans, including 847 genes which were implicated in the development of human myopia for the first time. Although the gene sets controlling baseline refractive development and those regulating susceptibility to myopia overlapped, these two processes appeared to be controlled by largely distinct sets of genes.

CONCLUSIONS

Comparison with data for other animal models of myopia revealed that the genes identified in this study comprise a well-defined set of retinal signaling pathways, which are highly conserved across different vertebrate species. These results identify major signaling pathways involved in refractive eye development and provide attractive targets for the development of anti-myopia drugs.

Retinal Defocus and Form-Deprivation Exposure Duration Affects RPE BMP Gene Expression

In the context of ocular development and eye growth regulation, retinal defocus and/or image contrast appear key variables although the nature of the signal(s) relayed from the retina to the sclera remains poorly understood. Nonetheless, under optimal visual conditions, eye length is brought into alignment with its optical power to achieve approximate emmetropia, through appropriate adjustment to eye growth. The retinal pigment epithelium (RPE), which lies between the retina and choroid/sclera, appears to play a crucial role in this process. In the investigations reported here, we used a chick model system to assess the threshold duration of exposure to lens-imposed defocus and form-deprivation necessary for conversion of evoked retinal signals into changes in BMP gene expression in the RPE. Our study provides evidence for the following: 1) close-loop, optical defocus-guided (negative and positive lenses) bidirectional BMP gene expression regulation, 2) open-loop, form-deprivation (diffusers)-induced down-regulation of BMP gene expression, and 3) early, transient up-regulation of BMP gene expression in response to both types of lens and diffuser applications. The critical exposure for accurately encoding retinal images as biological signals at the level of the RPE is in the order of minutes to hours, depending on the nature of the visual manipulations.