Robust parametric modeling of Alzheimer's disease progression

Comparative analysis of multimodal biomarkers for amyloid-beta positivity detection in Alzheimer's disease cohorts

Introduction

Efforts to develop cost-effective approaches for detecting amyloid pathology in Alzheimer's disease (AD) have gained significant momentum with a focus on biomarker classification. Recent research has explored non-invasive and readily accessible biomarkers, including magnetic resonance imaging (MRI) biomarkers and some AD risk factors.

Methods

In this comprehensive study, we leveraged a diverse dataset, encompassing participants with varying cognitive statuses from multiple sources, including cohorts from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and our in-house Dementia Disease Initiation (DDI) cohort. As brain amyloid plaques have been proposed as sufficient for AD diagnosis, our primary aim was to assess the effectiveness of multimodal biomarkers in identifying amyloid plaques, using deep machine learning methodologies.

Results

Our findings underscore the robustness of the utilized methods in detecting amyloid beta positivity across multiple cohorts. Additionally, we investigated the potential of demographic data to enhance MRI-based amyloid detection. Notably, the inclusion of demographic risk factors significantly improved our models' ability to detect amyloid-beta positivity, particularly in early-stage cases, exemplified by an average area under the ROC curve of 0.836 in the unimpaired DDI cohort.

Discussion

These promising, non-invasive, and cost-effective predictors of MRI biomarkers and demographic variables hold the potential for further refinement through considerations like APOE genotype and plasma markers.

Disease progression modelling of Alzheimer's disease using probabilistic principal components analysis

The recent biological redefinition of Alzheimer's Disease (AD) has spurred the development of statistical models that relate changes in biomarkers with neurodegeneration and worsening condition linked to AD. The ability to measure such changes may facilitate earlier diagnoses for affected individuals and help in monitoring the evolution of their condition. Amongst such statistical tools, disease progression models (DPMs) are quantitative, data-driven methods that specifically attempt to describe the temporal dynamics of biomarkers relevant to AD. Due to the heterogeneous nature of this disease, with patients of similar age experiencing different AD-related changes, a challenge facing longitudinal mixed-effects-based DPMs is the estimation of patient-realigning time-shifts. These time-shifts are indispensable for meaningful biomarker modelling, but may impact fitting time or vary with missing data in jointly estimated models. In this work, we estimate an individual's progression through Alzheimer's disease by combining multiple biomarkers into a single value using a probabilistic formulation of principal components analysis. Our results show that this variable, which summarises AD through observable biomarkers, is remarkably similar to jointly estimated time-shifts when we compute our scores for the baseline visit, on cross-sectional data from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Reproducing the expected properties of clinical datasets, we confirm that estimated scores are robust to missing data or unavailable biomarkers. In addition to cross-sectional insights, we can model the latent variable as an individual progression score by repeating estimations at follow-up examinations and refining long-term estimates as more data is gathered, which would be ideal in a clinical setting. Finally, we verify that our score can be used as a pseudo-temporal scale instead of age to ignore some patient heterogeneity in cohort data and highlight the general trend in expected biomarker evolution in affected individuals.

An ontology-based approach for harmonization and cross-cohort query of Alzheimer's disease data resources

BACKGROUND

In the United States, the National Alzheimer's Coordinating Center (NACC) and the Alzheimer's Disease Neuroimaging Initiative (ADNI) are two major data sharing resources for Alzheimer's Disease (AD) research. NACC and ADNI strive to make their data more FAIR (findable, interoperable, accessible and reusable) for the broader research community. However, there is limited work harmonizing and supporting cross-cohort interoperability of the two resources.

METHOD

In this paper, we leverage an ontology-based approach to harmonize data elements in the two resources and develop a web-based query system to search patient cohorts across the two resources. We first mapped data elements across NACC and ADNI, and performed value harmonization for the mapped data elements with inconsistent permissible values. Then we built an Alzheimer's Disease Data Element Ontology (ADEO) to model the mapped data elements in NACC and ADNI. We further developed a prototype cross-cohort query system to search patient cohorts across NACC and ADNI.

RESULTS

After manual review, we found 172 mappings between NACC and ADNI. These 172 mappings were further used to construct common concepts in ADEO. Our data element mapping and harmonization resulted in five files storing common concepts, variables in NACC and ADNI, mappings between variables and common concepts, permissible values of categorical type data elements, and coding inconsistency harmonization, respectively. Our cross-cohort query system consists of three core architectural elements: a web-based interface, an advanced query engine, and a backend MongoDB database.

CONCLUSIONS

In this work, ADEO has been specifically designed to facilitate data harmonization and cross-cohort query of NACC and ADNI data resources. Although our prototype cross-cohort query system was developed for exploring NACC and ADNI, its backend and frontend framework has been designed and implemented to be generally applicable to other domains for querying patient cohorts from multiple heterogeneous data sources.

Vascular endothelial-cadherin as a marker of endothelial injury in preclinical Alzheimer disease

OBJECTIVE

Endothelial dysfunction is an early and prevalent pathology in Alzheimer disease (AD). We here investigate the value of vascular endothelial-cadherin (VEC) as a cerebrospinal fluid (CSF) marker of endothelial injury in preclinical AD.

METHODS

Cognitively normal participants (Clinical Dementia Rating [CDR] 0) from the Knight Washington University-ADRC were included in this study (n = 700). Preclinical Alzheimer's Cognitive Composite (PACC) scores, CSF VEC, tau, p-tau181, Aβ42/Aβ40, neurofilament light-chain (NFL) levels, and magnetic resonance imaging (MRI) assessments of white matter injury (WMI) were obtained from all participants. A subset of participants underwent brain amyloid imaging using positron emission tomography (amyloid-PET) (n = 534). Linear regression examined associations of CSF VEC with PACC and individual cognitive scores in preclinical AD. Mediation analyses examined whether CSF VEC mediated effects of CSF amyloid and tau markers on cognition in preclinical AD.

RESULTS

CSF VEC levels significantly correlated with PACC and individual cognitive scores in participants with amyloid (A+T±N±; n = 558) or those with amyloid and tau pathologies (A+T+N±; n = 259), after adjusting for covariates. CSF VEC also correlated with CSF measures of amyloid, tau, and neurodegeneration and global amyloid burden on amyloid-PET scans in our cohort. Importantly, our findings suggest that CSF VEC mediates associations of CSF Aβ42/Aβ40, p-tau181, and global amyloid burden with cognitive outcomes in preclinical AD.

INTERPRETATION

Our results support the utility of CSF VEC as a marker of endothelial injury in AD and highlight the importance of endothelial injury as an early pathology that contributes to cognitive impairment in even the earliest preclinical stages.

CARRNN: A Continuous Autoregressive Recurrent Neural Network for Deep Representation Learning From Sporadic Temporal Data

Learning temporal patterns from multivariate longitudinal data is challenging especially in cases when data is sporadic, as often seen in, e.g., healthcare applications where the data can suffer from irregularity and asynchronicity as the time between consecutive data points can vary across features and samples, hindering the application of existing deep learning models that are constructed for complete, evenly spaced data with fixed sequence lengths. In this article, a novel deep learning-based model is developed for modeling multiple temporal features in sporadic data using an integrated deep learning architecture based on a recurrent neural network (RNN) unit and a continuous-time autoregressive (CAR) model. The proposed model, called CARRNN, uses a generalized discrete-time autoregressive (AR) model that is trainable end-to-end using neural networks modulated by time lags to describe the changes caused by the irregularity and asynchronicity. It is applied to time-series regression and classification tasks for Alzheimer's disease progression modeling, intensive care unit (ICU) mortality rate prediction, human activity recognition, and event-based digit recognition, where the proposed model based on a gated recurrent unit (GRU) in all cases achieves significantly better predictive performance than the state-of-the-art methods using RNNs, GRUs, and long short-term memory (LSTM) networks.

Deep Learning Model for Prediction of Progressive Mild Cognitive Impairment to Alzheimer’s Disease Using Structural MRI

Alzheimer’s disease (AD) is an irreversible neurological disorder that affects the vast majority of dementia cases, leading patients to experience gradual memory loss and cognitive function decline. Despite the lack of a cure, early detection of Alzheimer’s disease permits the provision of preventive medication to slow the disease’s progression. The objective of this project is to develop a computer-aided method based on a deep learning model to distinguish Alzheimer’s disease (AD) from cognitively normal and its early stage, mild cognitive impairment (MCI), by just using structural MRI (sMRI). To attain this purpose, we proposed a multiclass classification method based on 3D T1-weight brain sMRI images from the ADNI database. Axial brain images were extracted from 3D MRI and fed into the convolutional neural network (CNN) for multiclass classification. Three separate models were tested: a CNN built from scratch, VGG-16, and ResNet-50. As a feature extractor, the VGG-16 and ResNet-50 convolutional bases trained on the ImageNet dataset were employed. To achieve classification, a new densely connected classifier was implemented on top of the convolutional bases.