The Effect of Age Correction on Multivariate Classification in Alzheimer's Disease, with a Focus on the Characteristics of Incorrectly and Correctly Classified Subjects

Gut metagenomes of Asian octogenarians reveal metabolic potential expansion and distinct microbial species associated with aging phenotypes

While rapid demographic changes in Asia are driving the incidence of chronic aging-related diseases, the limited availability of high-quality in vivo data hampers our ability to understand complex multi-factorial contributions, including gut microbial, to healthy aging. Leveraging a well-phenotyped cohort of community-living octogenarians in Singapore, we used deep shotgun-metagenomic sequencing for high-resolution taxonomic and functional characterization of their gut microbiomes (n = 234). Joint species-level analysis with other Asian cohorts identified distinct age-associated shifts characterized by reduction in microbial richness, and specific Alistipes and Bacteroides species enrichment (e.g., Alistipes shahii and Bacteroides xylanisolvens). Functional analysis confirmed these changes correspond to metabolic potential expansion in aging towards alternate pathways synthesizing and utilizing amino-acid precursors, vis-à-vis dominant microbial guilds producing butyrate in gut from pyruvate (e.g., Faecalibacterium prausnitzii, Roseburia inulinivorans). Extending these observations to key clinical markers helped identify >10 robust microbial associations to inflammation, cardiometabolic and liver health, including potential probiotic species (e.g., Parabacteroides goldsteinii) and pathobionts (e.g., Klebsiella pneumoniae), highlighting the microbiome's role as biomarkers and potential targets for promoting healthy aging.

Predicting progression from subjective cognitive decline to mild cognitive impairment or dementia based on brain atrophy patterns

BACKGROUND

Alzheimer's disease (AD) is a progressive neurodegenerative disorder where pathophysiological changes begin decades before the onset of clinical symptoms. Analysis of brain atrophy patterns using structural MRI and multivariate data analysis are an effective tool in identifying patients with subjective cognitive decline (SCD) at higher risk of progression to AD dementia. Atrophy patterns obtained from models trained to classify advanced AD versus normal subjects, may not be optimal for subjects at an early stage, like SCD. In this study, we compared the accuracy of the SCD progression prediction using the 'severity index' generated using a standard classification model trained on patients with AD dementia versus a new model trained on β-amyloid (Aβ) positive patients with amnestic mild cognitive impairment (aMCI).

METHODS

We used structural MRI data of 504 patients from the Swedish BioFINDER-1 study cohort (cognitively normal (CN), Aβ-negative = 220; SCD, Aβ positive and negative = 139; aMCI, Aβ-positive = 106; AD dementia = 39). We applied multivariate data analysis to create two predictive models trained to discriminate CN individuals from either individuals with Aβ positive aMCI or AD dementia. Models were applied to individuals with SCD to classify their atrophy patterns as either high-risk "disease-like" or low-risk "CN-like". Clinical trajectory and model accuracy were evaluated using 8 years of longitudinal data.

RESULTS

In predicting progression from SCD to MCI or dementia, the standard, dementia-based model, reached 100% specificity but only 10.6% sensitivity, while the new, aMCI-based model, reached 72.3% sensitivity and 60.9% specificity. The aMCI-based model was superior in predicting progression from SCD to MCI or dementia, reaching a higher receiver operating characteristic area under curve (AUC = 0.72; P = 0.037) in comparison with the dementia-based model (AUC = 0.57).

CONCLUSION

When predicting conversion from SCD to MCI or dementia using structural MRI data, prediction models based on individuals with milder levels of atrophy (i.e. aMCI) may offer superior clinical value compared to standard dementia-based models.

Order selection for heterogeneous semiparametric hidden Markov models

Hidden Markov models (HMMs), which can characterize dynamic heterogeneity, are valuable tools for analyzing longitudinal data. The order of HMMs (ie, the number of hidden states) is typically assumed to be known or predetermined by some model selection criterion in conventional analysis. As prior information about the order frequently lacks, pairwise comparisons under criterion-based methods become computationally expensive with the model space growing. A few studies have conducted order selection and parameter estimation simultaneously, but they only considered homogeneous parametric instances. This study proposes a Bayesian double penalization (BDP) procedure for simultaneous order selection and parameter estimation of heterogeneous semiparametric HMMs. To overcome the difficulties in updating the order, we create a brand-new Markov chain Monte Carlo algorithm coupled with an effective adjust-bound reversible jump strategy. Simulation results reveal that the proposed BDP procedure performs well in estimation and works noticeably better than the conventional criterion-based approaches. Application of the suggested method to the Alzheimer's Disease Neuroimaging Initiative research further supports its usefulness.

In vivo microstructural heterogeneity of white matter and cognitive correlates in aging using tissue compositional analysis of diffusion magnetic resonance imaging

BACKGROUND

Age-related cognitive decline is linked to changes in the brain, particularly the deterioration of white matter (WM) microstructure that accelerates after the age of 60. WM deterioration is associated with mild cognitive impairment and dementia, but the origin and role of white matter signal abnormalities (WMSA) seen in standard MRI remain debated due to their heterogeneity. This study explores the potential of single-shell 3-tissue constrained spherical deconvolution (SS3T-CSD), a novel technique that models diffusion data in terms of gray matter (TG ), white matter (Tw ), and cerebrospinal fluid (TC ), to differentiate WMSA from normal-appearing white matter and better understand the interplay between changes in WM microstructure and decline in cognition.

METHODS

A total of 189 individuals from the GENIC cohort were included. MRI data, including T1-weighted and diffusion images, were obtained. Preprocessing steps were performed on the diffusion MRI data, followed by the SS3T-CSD. WMSA were segmented using FreeSurfer. Statistical analyses were conducted to assess the association between age, WMSA volume, 3-tissue signal fractions (Tw , TG , and TC ), and neuropsychological variables.

RESULTS

Participants above 60 years old showed worse cognitive performance and processing speed compared to those below 60 (p < .001). Age was negatively associated with Tw in normal-appearing white matter (p < .001) and positively associated with TG in both WMSA (p < .01) and normal-appearing white matter (p < .001). Age was also significantly associated with WMSA volume (p < .001). Higher processing speed was associated with lower Tw and higher TG , in normal-appearing white matter (p < .01 and p < .001, respectively), as well as increased WMSA volume (p < .001). Similarly, lower MMSE scores correlated with lower Tw and higher TG in normal-appearing white matter (p < .05). High cholesterol and hypertension were associated with higher WMSA volume (p < .05).

CONCLUSION

The microstructural heterogeneity within normal-appearing white matter and WMSA is associated with increasing age and cognitive variation, in cognitively unimpaired individuals. Furthermore, the 3-tissue signal fractions are more specific to potential white matter alterations than conventional MRI measures such as WMSA volume. These findings also support the view that the WMSA volumes may be more influenced by vascular risk factors than the 3-tissue metrics. Finally, the 3-tissue metrics were able to capture associations with cognitive tests and therefore capable of capturing subtle pathological changes in the brain in individuals who are still within the normal range of cognitive performance.

Inter-Hospital Advanced and Mild Alzheimer's Disease Classification Based on Electroencephalogram Measurements via Classical Machine Learning Algorithms

BACKGROUND

In pursuit of diagnostic tools capable of targeting distinct stages of Alzheimer's disease (AD), this study explores the potential of electroencephalography (EEG) combined with machine learning (ML) algorithms to identify patients with mild or moderate AD (ADM) and advanced AD (ADA).

OBJECTIVE

This study aims to assess the classification accuracy of six classical ML algorithms using a dataset of 668 patients from multiple hospitals.

METHODS

The dataset comprised measurements obtained from 668 patients, distributed among control, ADM, and ADA groups, collected from five distinct hospitals between 2011 and 2022. For classification purposes, six classical ML algorithms were employed: support vector machine, Bayesian linear discriminant analysis, decision tree, Gaussian Naïve Bayes, K-nearest neighbor and random forest.

RESULTS

The RF algorithm exhibited outstanding performance, achieving a remarkable balanced accuracy of 93.55% for ADA classification and 93.25% for ADM classification. The consistent reliability in distinguishing ADA and ADM patients underscores the potential of the EEG-based approach for AD diagnosis.

CONCLUSIONS

By leveraging a dataset sourced from multiple hospitals and encompassing a substantial patient cohort, coupled with the straightforwardness of the implemented models, it is feasible to attain notably robust results in AD classification.

A predictive model using the mesoscopic architecture of the living brain to detect Alzheimer's disease

Background

Alzheimer's disease, the most common cause of dementia, causes a progressive and irreversible deterioration of cognition that can sometimes be difficult to diagnose, leading to suboptimal patient care.

Methods

We developed a predictive model that computes multi-regional statistical morpho-functional mesoscopic traits from T1-weighted MRI scans, with or without cognitive scores. For each patient, a biomarker called "Alzheimer's Predictive Vector" (ApV) was derived using a two-stage least absolute shrinkage and selection operator (LASSO).

Results

The ApV reliably discriminates between people with (ADrp) and without (nADrp) Alzheimer's related pathologies (98% and 81% accuracy between ADrp - including the early form, mild cognitive impairment - and nADrp in internal and external hold-out test sets, respectively), without any a priori assumptions or need for neuroradiology reads. The new test is superior to standard hippocampal atrophy (26% accuracy) and cerebrospinal fluid beta amyloid measure (62% accuracy). A multiparametric analysis compared DTI-MRI derived fractional anisotropy, whose readout of neuronal loss agrees with ADrp phenotype, and SNPrs2075650 is significantly altered in patients with ADrp-like phenotype.

Conclusions

This new data analytic method demonstrates potential for increasing accuracy of Alzheimer diagnosis.

Cerebrospinal Fluid Biomarkers, Brain Structural and Cognitive Performances Between Normotensive and Hypertensive Controlled, Uncontrolled and Untreated 70-Year-Old Adults

Background: Hypertension is an important risk factor for Alzheimer's disease (AD). The pathophysiological mechanisms underlying the relationship between AD and hypertension are not fully understood, but they most likely involve microvascular dysfunction and cerebrovascular pathology. Although previous studies have assessed the impact of hypertension on different markers of brain integrity, no study has yet provided a comprehensive comparison of cerebrospinal fluid (CSF) biomarkers and structural brain differences between normotensive and hypertensive groups in a single and large cohort of older adults in relationship to cognitive performances. Objective: The aim of the present work was to investigate the differences in cognitive performances, CSF biomarkers and magnetic resonance imaging (MRI) of brain structure between normotensive, controlled hypertensive, uncontrolled hypertensive, and untreated hypertensive older adults from the Gothenburg H70 Birth Cohort Studies. Methods: As an indicator of vascular brain pathology, we measured white matter hyperintensities (WMHs), lacunes, cerebral microbleeds, enlarged perivascular space (epvs), and fractional anisotropy (FA). To assess markers of AD pathology/neurodegeneration, we measured hippocampal volume, temporal cortical thickness on MRI, and amyloid-β42, phosphorylated tau, and neurofilament light protein (NfL) in cerebrospinal fluid. Various neuropsychological tests were used to assess performances in memory, attention/processing speed, executive function, verbal fluency, and visuospatial abilities. Results: We found more white matter pathology in hypertensive compared to normotensive participants, with the highest vascular burden in uncontrolled participants (e.g., lower FA, more WMHs, and epvs). No significant difference was found in any MRI or CSF markers of AD pathology/neurodegeneration when comparing normotensive and hypertensive participants, nor among hypertensive groups. No significant difference was found in most cognitive functions between groups. Conclusion: Our results suggest that good blood pressure control may help prevent cerebrovascular pathology. In addition, hypertension may contribute to cognitive decline through its effect on cerebrovascular pathology rather than AD-related pathology. These findings suggest that hypertension is associated with MRI markers of vascular pathology in the absence of a significant decline in cognitive functions.

Prediction of transition to psychosis from an at-risk mental state using structural neuroimaging, genetic, and environmental data

INTRODUCTION

Psychosis is usually preceded by a prodromal phase in which patients are clinically identified as being at in an "At Risk Mental State" (ARMS). A few studies have demonstrated the feasibility of predicting psychosis transition from an ARMS using structural magnetic resonance imaging (sMRI) data and machine learning (ML) methods. However, the reliability of these findings is unclear due to possible sampling bias. Moreover, the value of genetic and environmental data in predicting transition to psychosis from an ARMS is yet to be explored.

METHODS

In this study we aimed to predict transition to psychosis from an ARMS using a combination of ML, sMRI, genome-wide genotypes, and environmental risk factors as predictors, in a sample drawn from a pool of 246 ARMS subjects (60 of whom later transitioned to psychosis). First, the modality-specific values in predicting transition to psychosis were evaluated using several: (a) feature types; (b) feature manipulation strategies; (c) ML algorithms; (d) cross-validation strategies, as well as sample balancing and bootstrapping. Subsequently, the modalities whose at least 60% of the classification models showed an balanced accuracy (BAC) statistically better than chance level were included in a multimodal classification model.

RESULTS AND DISCUSSION

Results showed that none of the modalities alone, i.e., neuroimaging, genetic or environmental data, could predict psychosis from an ARMS statistically better than chance and, as such, no multimodal classification model was trained/tested. These results suggest that the value of structural MRI data and genome-wide genotypes in predicting psychosis from an ARMS, which has been fostered by previous evidence, should be reconsidered.

Identifying individuals with Alzheimer's disease-like brains based on structural imaging in the Human Connectome Project Aging cohort

Given the difficulty in factoring out typical age effects from subtle Alzheimer's disease (AD) effects on brain structure, identification of very early, as well as younger preclinical “at‐risk” individuals has unique challenges. We examined whether age‐correction procedures could be used to better identify individuals at very early potential risk from adults who did not have any existing cognitive diagnosis. First, we obtained cross‐sectional age effects for each structural feature using data from a selected portion of the Human Connectome Project Aging (HCP‐A) cohort. After age detrending, we weighted AD structural deterioration with patterns quantified from data of the Alzheimer's Disease Neuroimaging Initiative. Support vector machine was then used to classify individuals with brains that most resembled atrophy in AD across the entire HCP‐A sample. Additionally, we iteratively adjusted the pipeline by removing individuals classified as AD‐like from the HCP‐A cohort to minimize atypical brain structural contributions to the age detrending. The classifier had a mean cross‐validation accuracy of 94.0% for AD recognition. It also could identify mild cognitive impairment with more severe AD‐specific biomarkers and worse cognition. In an independent HCP‐A cohort, 8.8% were identified as AD‐like, and they trended toward worse cognition. An “AD risk” score derived from the machine learning models also significantly correlated with cognition. This work provides a proof of concept for the potential to use structural brain imaging to identify asymptomatic individuals at young ages who show structural brain patterns similar to AD and are potentially at risk for a future clinical disorder.

Comparing different approaches for operationalizing subjective cognitive decline: impact on syndromic and biomarker profiles

Subjective cognitive decline (SCD) has been proposed as a risk factor for future cognitive decline and dementia. Given the heterogeneity of SCD and the lack of consensus about how to classify this condition, different operationalization approaches still need to be compared. In this study, we used the same sample of individuals to compare  different SCD operationalization approaches. We included 399 cognitively healthy individuals from a community-based cohort. SCD was assessed through nine questions about memory and non-memory subjective complaints. We applied four approaches to operationalize SCD: two hypothesis-driven approaches and two data-driven approaches. We characterized the resulting groups from each operationalization approach using multivariate methods on comprehensive demographic, clinical, cognitive, and neuroimaging data. We identified two main phenotypes: an amnestic phenotype characterized by an Alzheimer's Disease (AD) signature pattern of brain atrophy; and an anomic phenotype, which was mainly related to cerebrovascular pathology. Furthermore, language complaints other than naming helped to identify a subgroup with subclinical cognitive impairment and difficulties in activities of daily living. This subgroup also showed an AD signature pattern of atrophy. The identification of SCD phenotypes, characterized by different syndromic and biomarker profiles, varies depending on the operationalization approach used. In this study we discuss how these findings may be used in clinical practice and research.

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.

Prediction of clinical and biomarker conformed Alzheimer's disease and mild cognitive impairment from multi-feature brain structural MRI using age-correction from a large independent lifespan sample

Structural neuroimaging has been applied to the identification of individuals with Alzheimer's disease (AD) and mild cognitive impairment (MCI). However, these methods are greatly impacted by age limiting their utility for detection of preclinical pathology. We built linear models for age based on multiple combined structural features using a large independent lifespan sample of 272 healthy adults across a wide age range from the Human Connectome Project Aging study. These models were then used to create a new support vector machine (SVM) training model in 136 AD and 268 control participants based on residues of fit from the expected age-effects relationship. Subsequent validation assessed the accuracy of the SVM model in new datasets. Finally, we applied the classifier to 276 individuals with MCI to evaluate prediction for early impairment and longitudinal cognitive change. The optimal 10-fold cross-validation accuracy was 93.07%, compared to 91.83% without age detrending. In the validation dataset, the classifier for AD obtained an accuracy of 84.85% (56/66), sensitivity of 85.36% (35/41) and specificity of 84% (21/25). Classification accuracy was improved when using the lifespan sample as opposed to the classification sample. Importantly, we observed cross-sectional greater AD specific biomarkers, as well as faster cognitive decline in MCI who were classified as more 'AD-like' (MCI-AD), and these effects were pronounced in individuals who were late MCI. The top five contributive features were volumes of left hippocampus, right hippocampus, left amygdala, the thickness of left and right middle temporal & parahippocampus gyrus. Linear detrending for age in SVM for combined structural features resulted in good performance for recognition of AD and AD-specific biomarkers, as well as prediction of MCI progression. Such procedures may be used in future work to enhance prediction in samples with atypical age distributions.

An attempt to identify reproducible high-density EEG markers of PTSD during sleep

STUDY OBJECTIVES

We examined electroencephalogram (EEG) spectral power to study abnormalities in regional brain activity in post-traumatic stress disorder (PTSD) during sleep. We aimed to identify sleep EEG markers of PTSD that were reproducible across nights and subsamples of our study population.

METHODS

Seventy-eight combat-exposed veteran men with (n = 31) and without (n = 47) PTSD completed two consecutive nights of high-density EEG recordings in a laboratory. We performed spectral-topographical EEG analyses on data from both nights. To assess reproducibility, we used the first 47 consecutive participants (18 with PTSD) for initial discovery and the remaining 31 participants (13 with PTSD) for replication.

RESULTS

In the discovery analysis, compared with non-PTSD participants, PTSD participants exhibited (1) reduced delta power (1-4 Hz) in the centro-parietal regions during nonrapid eye movement (NREM) sleep and (2) elevated high-frequency power, most prominent in the gamma band (30-40 Hz), in the antero-frontal regions during both NREM and rapid eye movement (REM) sleep. These findings were consistent across the two study nights, with reproducible trends in the replication analysis. We found no significant group differences in theta power (4-8 Hz) during REM sleep and sigma power (12-15 Hz) during N2 sleep.

CONCLUSIONS

The reduced centro-parietal NREM delta power, indicating reduced sleep depth, and the elevated antero-frontal NREM and REM gamma powers, indicating heightened central arousal, are potential objective sleep markers of PTSD. If independently validated, these putative EEG markers may offer new targets for the development of sleep-specific PTSD diagnostics and interventions.

Fully bayesian longitudinal unsupervised learning for the assessment and visualization of AD heterogeneity and progression

Tau pathology and brain atrophy are the closest correlate of cognitive decline in Alzheimer's disease (AD). Understanding heterogeneity and longitudinal progression of atrophy during the disease course will play a key role in understanding AD pathogenesis. We propose a framework for longitudinal clustering that simultaneously: 1) incorporates whole brain data, 2) leverages unequal visits per individual, 3) compares clusters with a control group, 4) allows for study confounding effects, 5) provides cluster visualization, 6) measures clustering uncertainty. We used amyloid-β positive AD and negative healthy subjects, three longitudinal structural magnetic resonance imaging scans (cortical thickness and subcortical volume) over two years. We found three distinct longitudinal AD brain atrophy patterns: one typical diffuse pattern (n=34, 47.2%), and two atypical patterns: minimal atrophy (n=23 31.9%) and hippocampal sparing (n=9, 12.5%). We also identified outliers (n=3, 4.2%) and observations with uncertain classification (n=3, 4.2%). The clusters differed not only in regional distributions of atrophy at baseline, but also longitudinal atrophy progression, age at AD onset, and cognitive decline. A framework for the longitudinal assessment of variability in cohorts with several neuroimaging measures was successfully developed. We believe this framework may aid in disentangling distinct subtypes of AD from disease staging.

The reliability of a deep learning model in clinical out-of-distribution MRI data: A multicohort study

Deep learning (DL) methods have in recent years yielded impressive results in medical imaging, with the potential to function as clinical aid to radiologists. However, DL models in medical imaging are often trained on public research cohorts with images acquired with a single scanner or with strict protocol harmonization, which is not representative of a clinical setting. The aim of this study was to investigate how well a DL model performs in unseen clinical datasets-collected with different scanners, protocols and disease populations-and whether more heterogeneous training data improves generalization. In total, 3117 MRI scans of brains from multiple dementia research cohorts and memory clinics, that had been visually rated by a neuroradiologist according to Scheltens' scale of medial temporal atrophy (MTA), were included in this study. By training multiple versions of a convolutional neural network on different subsets of this data to predict MTA ratings, we assessed the impact of including images from a wider distribution during training had on performance in external memory clinic data. Our results showed that our model generalized well to datasets acquired with similar protocols as the training data, but substantially worse in clinical cohorts with visibly different tissue contrasts in the images. This implies that future DL studies investigating performance in out-of-distribution (OOD) MRI data need to assess multiple external cohorts for reliable results. Further, by including data from a wider range of scanners and protocols the performance improved in OOD data, which suggests that more heterogeneous training data makes the model generalize better. To conclude, this is the most comprehensive study to date investigating the domain shift in deep learning on MRI data, and we advocate rigorous evaluation of DL models on clinical data prior to being certified for deployment.

Predicting Global Cognitive Decline in the General Population Using the Disease State Index

Background

Identifying persons at risk for cognitive decline may aid in early detection of persons at risk of dementia and to select those that would benefit most from therapeutic or preventive measures for dementia.

Objective

In this study we aimed to validate whether cognitive decline in the general population can be predicted with multivariate data using a previously proposed supervised classification method: Disease State Index (DSI).

Methods

We included 2,542 participants, non-demented and without mild cognitive impairment at baseline, from the population-based Rotterdam Study (mean age 60.9 ± 9.1 years). Participants with significant global cognitive decline were defined as the 5% of participants with the largest cognitive decline per year. We trained DSI to predict occurrence of significant global cognitive decline using a large variety of baseline features, including magnetic resonance imaging (MRI) features, cardiovascular risk factors, APOE-ε4 allele carriership, gait features, education, and baseline cognitive function as predictors. The prediction performance was assessed as area under the receiver operating characteristic curve (AUC), using 500 repetitions of 2-fold cross-validation experiments, in which (a randomly selected) half of the data was used for training and the other half for testing.

Results

A mean AUC (95% confidence interval) for DSI prediction was 0.78 (0.77–0.79) using only age as input feature. When using all available features, a mean AUC of 0.77 (0.75–0.78) was obtained. Without age, and with age-corrected features and feature selection on MRI features, a mean AUC of 0.70 (0.63–0.76) was obtained, showing the potential of other features besides age.

Conclusion

The best performance in the prediction of global cognitive decline in the general population by DSI was obtained using only age as input feature. Other features showed potential, but did not improve prediction. Future studies should evaluate whether the performance could be improved by new features, e.g., longitudinal features, and other prediction methods.

Machine learning in mental health: a scoping review of methods and applications

BACKGROUND

This paper aims to synthesise the literature on machine learning (ML) and big data applications for mental health, highlighting current research and applications in practice.

METHODS

We employed a scoping review methodology to rapidly map the field of ML in mental health. Eight health and information technology research databases were searched for papers covering this domain. Articles were assessed by two reviewers, and data were extracted on the article's mental health application, ML technique, data type, and study results. Articles were then synthesised via narrative review.

RESULTS

Three hundred papers focusing on the application of ML to mental health were identified. Four main application domains emerged in the literature, including: (i) detection and diagnosis; (ii) prognosis, treatment and support; (iii) public health, and; (iv) research and clinical administration. The most common mental health conditions addressed included depression, schizophrenia, and Alzheimer's disease. ML techniques used included support vector machines, decision trees, neural networks, latent Dirichlet allocation, and clustering.

CONCLUSIONS

Overall, the application of ML to mental health has demonstrated a range of benefits across the areas of diagnosis, treatment and support, research, and clinical administration. With the majority of studies identified focusing on the detection and diagnosis of mental health conditions, it is evident that there is significant room for the application of ML to other areas of psychology and mental health. The challenges of using ML techniques are discussed, as well as opportunities to improve and advance the field.

Alterations in the tyrosine and phenylalanine pathways revealed by biochemical profiling in cerebrospinal fluid of Huntington's disease subjects

Huntington’s disease (HD) is a severe neurological disease leading to psychiatric symptoms, motor impairment and cognitive decline. The disease is caused by a CAG expansion in the huntingtin (HTT) gene, but how this translates into the clinical phenotype of HD remains elusive. Using liquid chromatography mass spectrometry, we analyzed the metabolome of cerebrospinal fluid (CSF) from premanifest and manifest HD subjects as well as control subjects. Inter-group differences revealed that the tyrosine metabolism, including tyrosine, thyroxine, L-DOPA and dopamine, was significantly altered in manifest compared with premanifest HD. These metabolites demonstrated moderate to strong associations to measures of disease severity and symptoms. Thyroxine and dopamine also correlated with the five year risk of onset in premanifest HD subjects. The phenylalanine and the purine metabolisms were also significantly altered, but associated less to disease severity. Decreased levels of lumichrome were commonly found in mutated HTT carriers and the levels correlated with the five year risk of disease onset in premanifest carriers. These biochemical findings demonstrates that the CSF metabolome can be used to characterize molecular pathogenesis occurring in HD, which may be essential for future development of novel HD therapies.

Biochemical Differences in Cerebrospinal Fluid between Secondary Progressive and Relapsing⁻Remitting Multiple Sclerosis

To better understand the pathophysiological differences between secondary progressive multiple sclerosis (SPMS) and relapsing-remitting multiple sclerosis (RRMS), and to identify potential biomarkers of disease progression, we applied high-resolution mass spectrometry (HRMS) to investigate the metabolome of cerebrospinal fluid (CSF). The biochemical differences were determined using partial least squares discriminant analysis (PLS-DA) and connected to biochemical pathways as well as associated to clinical and radiological measures. Tryptophan metabolism was significantly altered, with perturbed levels of kynurenate, 5-hydroxytryptophan, 5-hydroxyindoleacetate, and N-acetylserotonin in SPMS patients compared with RRMS and controls. SPMS patients had altered kynurenine compared with RRMS patients, and altered indole-3-acetate compared with controls. Regarding the pyrimidine metabolism, SPMS patients had altered levels of uridine and deoxyuridine compared with RRMS and controls, and altered thymine and glutamine compared with RRMS patients. Metabolites from the pyrimidine metabolism were significantly associated with disability, disease activity and brain atrophy, making them of particular interest for understanding the disease mechanisms and as markers of disease progression. Overall, these findings are of importance for the characterization of the molecular pathogenesis of SPMS and support the hypothesis that the CSF metabolome may be used to explore changes that occur in the transition between the RRMS and SPMS pathologies.

Using deep autoencoders to identify abnormal brain structural patterns in neuropsychiatric disorders: A large-scale multi-sample study

Machine learning is becoming an increasingly popular approach for investigating spatially distributed and subtle neuroanatomical alterations in brain-based disorders. However, some machine learning models have been criticized for requiring a large number of cases in each experimental group, and for resembling a "black box" that provides little or no insight into the nature of the data. In this article, we propose an alternative conceptual and practical approach for investigating brain-based disorders which aim to overcome these limitations. We used an artificial neural network known as "deep autoencoder" to create a normative model using structural magnetic resonance imaging data from 1,113 healthy people. We then used this model to estimate total and regional neuroanatomical deviation in individual patients with schizophrenia and autism spectrum disorder using two independent data sets (n = 263). We report that the model was able to generate different values of total neuroanatomical deviation for each disease under investigation relative to their control group (p < .005). Furthermore, the model revealed distinct patterns of neuroanatomical deviations for the two diseases, consistent with the existing neuroimaging literature. We conclude that the deep autoencoder provides a flexible and promising framework for assessing total and regional neuroanatomical deviations in neuropsychiatric populations.

Integration of magnetic resonance imaging and protein and metabolite CSF measurements to enable early diagnosis of secondary progressive multiple sclerosis

Molecular networks in neurological diseases are complex. Despite this fact, contemporary biomarkers are in most cases interpreted in isolation, leading to a significant loss of information and power. We present an analytical approach to scrutinize and combine information from biomarkers originating from multiple sources with the aim of discovering a condensed set of biomarkers that in combination could distinguish the progressive degenerative phenotype of multiple sclerosis (SPMS) from the relapsing-remitting phenotype (RRMS). Methods: Clinical and magnetic resonance imaging (MRI) data were integrated with data from protein and metabolite measurements of cerebrospinal fluid, and a method was developed to sift through all the variables to establish a small set of highly informative measurements. This prospective study included 16 SPMS patients, 30 RRMS patients and 10 controls. Protein concentrations were quantitated with multiplexed fluorescent bead-based immunoassays and ELISA. The metabolome was recorded using liquid chromatography-mass spectrometry. Clinical follow-up data of the SPMS patients were used to assess disease progression and development of disability. Results: Eleven variables were in combination able to distinguish SPMS from RRMS patients with high confidence superior to any single measurement. The identified variables consisted of three MRI variables: the size of the spinal cord and the third ventricle and the total number of T1 hypointense lesions; six proteins: galectin-9, monocyte chemoattractant protein-1 (MCP-1), transforming growth factor alpha (TGF-α), tumor necrosis factor alpha (TNF-α), soluble CD40L (sCD40L) and platelet-derived growth factor AA (PDGF-AA); and two metabolites: 20β-dihydrocortisol (20β-DHF) and indolepyruvate. The proteins myelin basic protein (MBP) and macrophage-derived chemokine (MDC), as well as the metabolites 20β-DHF and 5,6-dihydroxyprostaglandin F1a (5,6-DH-PGF1), were identified as potential biomarkers of disability progression. Conclusion: Our study demonstrates, in a limited but well-defined and data-rich cohort, the importance and value of combining multiple biomarkers to aid diagnostics and track disease progression.

Stability of graph theoretical measures in structural brain networks in Alzheimer's disease

Graph analysis has become a popular approach to study structural brain networks in neurodegenerative disorders such as Alzheimer's disease (AD). However, reported results across similar studies are often not consistent. In this paper we investigated the stability of the graph analysis measures clustering, path length, global efficiency and transitivity in a cohort of AD (N = 293) and control subjects (N = 293). More specifically, we studied the effect that group size and composition, choice of neuroanatomical atlas, and choice of cortical measure (thickness or volume) have on binary and weighted network properties and relate them to the magnitude of the differences between groups of AD and control subjects. Our results showed that specific group composition heavily influenced the network properties, particularly for groups with less than 150 subjects. Weighted measures generally required fewer subjects to stabilize and all assessed measures showed robust significant differences, consistent across atlases and cortical measures. However, all these measures were driven by the average correlation strength, which implies a limitation of capturing more complex features in weighted networks. In binary graphs, significant differences were only found in the global efficiency and transitivity measures when using cortical thickness measures to define edges. The findings were consistent across the two atlases, but no differences were found when using cortical volumes. Our findings merits future investigations of weighted brain networks and suggest that cortical thickness measures should be preferred in future AD studies if using binary networks. Further, studying cortical networks in small cohorts should be complemented by analyzing smaller, subsampled groups to reduce the risk that findings are spurious.

Proposal for a hierarchical, multidimensional, and multivariate approach to investigate cognitive aging

Cognitive aging is highly complex. We applied a data-driven statistical method to investigate aging from a hierarchical, multidimensional, and multivariate approach. Orthogonal partial least squares to latent structures and hierarchical models were applied for the first time in a study of cognitive aging. The association between age and a total of 316 demographic, clinical, cognitive, and neuroimaging measures was simultaneously analyzed in 460 cognitively normal individuals (35-85 years). Age showed a strong association with brain structure, especially with cortical thickness in frontal and parietal association regions. Age also showed a fairly strong association with cognition. Although a strong association of age with executive functions and processing speed was captured as expected, the association of age with visual memory was stronger. Clinical measures were less strongly associated with age. Hierarchical and correlation analyses further showed these associations in a neuroimaging-cognitive-clinical order of importance. We conclude that orthogonal partial least square and hierarchical models are a promising approach to better understand the complexity in cognitive aging.

Factors influencing accuracy of cortical thickness in the diagnosis of Alzheimer's disease

There is great value to use of structural neuroimaging in the assessment of Alzheimer's disease (AD). However, to date, predictive value of structural imaging tend to range between 80% and 90% in accuracy and it is unclear why this is the case given that structural imaging should parallel the pathologic processes of AD. There is a possibility that clinical misdiagnosis relative to the gold standard pathologic diagnosis and/or additional brain pathologies are confounding factors contributing to reduced structural imaging classification accuracy. We examined potential factors contributing to misclassification of individuals with clinically diagnosed AD purely from cortical thickness measures. Correctly classified and incorrectly classified groups were compared across a range of demographic, biological, and neuropsychological data including cerebrospinal fluid biomarkers, amyloid imaging, white matter hyperintensity (WMH) volume, cognitive, and genetic factors. Individual subject analyses suggested that at least a portion of the control individuals misclassified as AD from structural imaging additionally harbor substantial AD biomarker pathology and risk, yet are relatively resistant to cognitive symptoms, likely due to "cognitive reserve," and therefore clinically unimpaired. In contrast, certain clinical control individuals misclassified as AD from cortical thickness had increased WMH volume relative to other controls in the sample, suggesting that vascular conditions may contribute to classification accuracy from cortical thickness measures. These results provide examples of factors that contribute to the accuracy of structural imaging in predicting a clinical diagnosis of AD, and provide important information about considerations for future work aimed at optimizing structural based diagnostic classifiers for AD.

Monitoring disease progression in mild cognitive impairment: Associations between atrophy patterns, cognition, APOE and amyloid

BACKGROUND

A disease severity index (SI) for Alzheimer's disease (AD) has been proposed that summarizes MRI-derived structural measures into a single score using multivariate data analysis.

OBJECTIVES

To longitudinally evaluate the use of the SI to monitor disease progression and predict future progression to AD in mild cognitive impairment (MCI). Further, to investigate the association between longitudinal change in the SI and cognitive impairment, Apolipoprotein E (APOE) genotype as well as the levels of cerebrospinal fluid amyloid-beta 1-42 (Aβ) peptide.

METHODS

The dataset included 195 AD, 145 MCI and 228 control subjects with annual follow-up for three years, where 70 MCI subjects progressed to AD (MCI-p). For each subject the SI was generated at baseline and follow-ups using 55 regional cortical thickness and subcortical volumes measures that extracted by the FreeSurfer longitudinal stream.

RESULTS

MCI-p subjects had a faster increase of the SI over time (p < 0.001). A higher SI at baseline in MCI-p was related to progression to AD at earlier follow-ups (p < 0.001) and worse cognitive impairment (p < 0.001). AD-like MCI patients with the APOE ε4 allele and abnormal Aβ levels had a faster increase of the SI, independently (p = 0.003 and p = 0.004).

CONCLUSIONS

Longitudinal changes in the SI reflect structural brain changes and can identify MCI patients at risk of progression to AD. Disease-related brain structural changes are influenced independently by APOE genotype and amyloid pathology. The SI has the potential to be used as a sensitive tool to predict future dementia, monitor disease progression as well as an outcome measure for clinical trials.

Recent publications from the Alzheimer's Disease Neuroimaging Initiative: Reviewing progress toward improved AD clinical trials

INTRODUCTION

The Alzheimer's Disease Neuroimaging Initiative (ADNI) has continued development and standardization of methodologies for biomarkers and has provided an increased depth and breadth of data available to qualified researchers. This review summarizes the over 400 publications using ADNI data during 2014 and 2015.

METHODS

We used standard searches to find publications using ADNI data.

RESULTS

(1) Structural and functional changes, including subtle changes to hippocampal shape and texture, atrophy in areas outside of hippocampus, and disruption to functional networks, are detectable in presymptomatic subjects before hippocampal atrophy; (2) In subjects with abnormal β-amyloid deposition (Aβ+), biomarkers become abnormal in the order predicted by the amyloid cascade hypothesis; (3) Cognitive decline is more closely linked to tau than Aβ deposition; (4) Cerebrovascular risk factors may interact with Aβ to increase white-matter (WM) abnormalities which may accelerate Alzheimer's disease (AD) progression in conjunction with tau abnormalities; (5) Different patterns of atrophy are associated with impairment of memory and executive function and may underlie psychiatric symptoms; (6) Structural, functional, and metabolic network connectivities are disrupted as AD progresses. Models of prion-like spreading of Aβ pathology along WM tracts predict known patterns of cortical Aβ deposition and declines in glucose metabolism; (7) New AD risk and protective gene loci have been identified using biologically informed approaches; (8) Cognitively normal and mild cognitive impairment (MCI) subjects are heterogeneous and include groups typified not only by "classic" AD pathology but also by normal biomarkers, accelerated decline, and suspected non-Alzheimer's pathology; (9) Selection of subjects at risk of imminent decline on the basis of one or more pathologies improves the power of clinical trials; (10) Sensitivity of cognitive outcome measures to early changes in cognition has been improved and surrogate outcome measures using longitudinal structural magnetic resonance imaging may further reduce clinical trial cost and duration; (11) Advances in machine learning techniques such as neural networks have improved diagnostic and prognostic accuracy especially in challenges involving MCI subjects; and (12) Network connectivity measures and genetic variants show promise in multimodal classification and some classifiers using single modalities are rivaling multimodal classifiers.

DISCUSSION

Taken together, these studies fundamentally deepen our understanding of AD progression and its underlying genetic basis, which in turn informs and improves clinical trial design.

A 'Disease Severity Index' to identify individuals with Subjective Memory Decline who will progress to mild cognitive impairment or dementia

Subjective memory decline (SMD) is a heterogeneous condition. While SMD might be the earliest sign of Alzheimer’s disease (AD), it also occurs in aging and various neurological, medical, and psychiatric conditions. Identifying those with higher risk to develop dementia is thus a major challenge. We tested a novel disease severity index generated by multivariate data analysis with numerous structural MRI measures as input. The index was used to identify SMD individuals with high risk of progression to mild cognitive impairment (MCI) or AD. A total of 69 healthy controls, 86 SMD, 45 MCI, and 38 AD patients were included. Subjects were followed up for 7.5 years. Clinical, cognitive, PET amyloid imaging and APOE ε4 data were used as outcome variables. The results showed that SMD evidenced cognitive performance intermediate between healthy controls and MCI. The disease severity index identified eleven (13%) SMD individuals with an AD-like pattern of brain atrophy. These individuals showed lower cognitive performance, increased CDR-SOB, higher amyloid burden and worse clinical progression (6.2 times higher likelihood to develop MCI, dementia or die than healthy controls). The current disease severity index may have relevance for clinical practice, as well as for selecting appropriate individuals for clinical trials.

Disrupted Network Topology in Patients with Stable and Progressive Mild Cognitive Impairment and Alzheimer's Disease

Recent findings suggest that Alzheimer's disease (AD) is a disconnection syndrome characterized by abnormalities in large-scale networks. However, the alterations that occur in network topology during the prodromal stages of AD, particularly in patients with stable mild cognitive impairment (MCI) and those that show a slow or faster progression to dementia, are still poorly understood. In this study, we used graph theory to assess the organization of structural MRI networks in stable MCI (sMCI) subjects, late MCI converters (lMCIc), early MCI converters (eMCIc), and AD patients from 2 large multicenter cohorts: ADNI and AddNeuroMed. Our findings showed an abnormal global network organization in all patient groups, as reflected by an increased path length, reduced transitivity, and increased modularity compared with controls. In addition, lMCIc, eMCIc, and AD patients showed a decreased path length and mean clustering compared with the sMCI group. At the local level, there were nodal clustering decreases mostly in AD patients, while the nodal closeness centrality detected abnormalities across all patient groups, showing overlapping changes in the hippocampi and amygdala and nonoverlapping changes in parietal, entorhinal, and orbitofrontal regions. These findings suggest that the prodromal and clinical stages of AD are associated with an abnormal network topology.