Instantiated mixed effects modeling of Alzheimer's disease markers

Data-driven modelling of neurodegenerative disease progression: thinking outside the black box

Data-driven disease progression models are an emerging set of computational tools that reconstruct disease timelines for long-term chronic diseases, providing unique insights into disease processes and their underlying mechanisms. Such methods combine a priori human knowledge and assumptions with large-scale data processing and parameter estimation to infer long-term disease trajectories from short-term data. In contrast to 'black box' machine learning tools, data-driven disease progression models typically require fewer data and are inherently interpretable, thereby aiding disease understanding in addition to enabling classification, prediction and stratification. In this Review, we place the current landscape of data-driven disease progression models in a general framework and discuss their enhanced utility for constructing a disease timeline compared with wider machine learning tools that construct static disease profiles. We review the insights they have enabled across multiple neurodegenerative diseases, notably Alzheimer disease, for applications such as determining temporal trajectories of disease biomarkers, testing hypotheses about disease mechanisms and uncovering disease subtypes. We outline key areas for technological development and translation to a broader range of neuroscience and non-neuroscience applications. Finally, we discuss potential pathways and barriers to integrating disease progression models into clinical practice and trial settings.

Exploring cognitive progression subtypes in the Framingham Heart Study

INTRODUCTION

Alzheimer's disease (AD) is a heterogeneous disorder characterized by complex underlying neuropathology that is not fully understood. This study aimed to identify cognitive progression subtypes and examine their correlation with clinical outcomes.

METHODS

Participants of this study were recruited from the Framingham Heart Study. The Subtype and Stage Inference (SuStaIn) method was used to identify cognitive progression subtypes based on eight cognitive domains.

RESULTS

Three cognitive progression subtypes were identified, including verbal learning (Subtype 1), abstract reasoning (Subtype 2), and visual memory (Subtype 3). These subtypes represent different domains of cognitive decline during the progression of AD. Significant differences in age of onset among the different subtypes were also observed. A higher SuStaIn stage was significantly associated with increased mortality risk.

DISCUSSION

This study provides a characterization of AD heterogeneity in cognitive progression, emphasizing the importance of developing personalized approaches for risk stratification and intervention.

Highlights

We used the Subtype and Stage Inference (SuStaIn) method to identify three cognitive progression subtypes.Different subtypes have significant variations in age of onset.Higher stages of progression are associated with increased mortality risk.

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.

Multimodal attention-based deep learning for Alzheimer's disease diagnosis

OBJECTIVE

Alzheimer's disease (AD) is the most common neurodegenerative disorder with one of the most complex pathogeneses, making effective and clinically actionable decision support difficult. The objective of this study was to develop a novel multimodal deep learning framework to aid medical professionals in AD diagnosis.

MATERIALS AND METHODS

We present a Multimodal Alzheimer's Disease Diagnosis framework (MADDi) to accurately detect the presence of AD and mild cognitive impairment (MCI) from imaging, genetic, and clinical data. MADDi is novel in that we use cross-modal attention, which captures interactions between modalities-a method not previously explored in this domain. We perform multi-class classification, a challenging task considering the strong similarities between MCI and AD. We compare with previous state-of-the-art models, evaluate the importance of attention, and examine the contribution of each modality to the model's performance.

RESULTS

MADDi classifies MCI, AD, and controls with 96.88% accuracy on a held-out test set. When examining the contribution of different attention schemes, we found that the combination of cross-modal attention with self-attention performed the best, and no attention layers in the model performed the worst, with a 7.9% difference in F1-scores.

DISCUSSION

Our experiments underlined the importance of structured clinical data to help machine learning models contextualize and interpret the remaining modalities. Extensive ablation studies showed that any multimodal mixture of input features without access to structured clinical information suffered marked performance losses.

CONCLUSION

This study demonstrates the merit of combining multiple input modalities via cross-modal attention to deliver highly accurate AD diagnostic decision support.

Evaluating the performance of Bayesian and frequentist approaches for longitudinal modeling: application to Alzheimer's disease

Linear mixed effects (LME) modelling under both frequentist and Bayesian frameworks can be used to study longitudinal trajectories. We studied the performance of both frameworks on different dataset configurations using hippocampal volumes from longitudinal MRI data across groups—healthy controls (HC), mild cognitive impairment (MCI) and Alzheimer’s disease (AD) patients, including subjects that converted from MCI to AD. We started from a big database of 1250 subjects from the Alzheimer’s disease neuroimaging initiative (ADNI), and we created different reduced datasets simulating real-life situations using a random-removal permutation-based approach. The number of subjects needed to differentiate groups and to detect conversion to AD was 147 and 115 respectively. The Bayesian approach allowed estimating the LME model even with very sparse databases, with high number of missing points, which was not possible with the frequentist approach. Our results indicate that the frequentist approach is computationally simpler, but it fails in modelling data with high number of missing values.

Categorical predictive and disease progression modeling in the early stage of Alzheimer's disease

BACKGROUND

A preclinical stage of Alzheimer's disease (AD) precedes the symptomatic phases of mild cognitive impairment (MCI) and dementia, which constitutes a window of opportunities for preventive therapies or delaying dementia onset.

NEW METHOD

We propose to use categorical predictive models based on survival analysis with longitudinal data which are capable of determining subsets of markers to classify cognitively unimpaired (CU) subjects who progress into MCI/dementia or not. Subsequently, the proposed combination of markers was used to construct disease progression models (DPMs), which reveal long-term pathological trajectories from short-term clinical data. The proposed methodology was applied to a population recruited by the ADNI.

RESULTS

A very small subset of standard MRI-based data, CSF markers and cognitive measures was used to predict CU-to-MCI/dementia progression. The longitudinal data of these selected markers were used to construct DPMs using the algorithms of growth models by alternating conditional expectation (GRACE) and the latent time joint mixed effects model (LTJMM). The results show that the natural history of the proposed cognitive decline classifies the subjects well according to the clinical groups and shows a moderate correlation between the conversion times and their estimates by the algorithms.

COMPARISON WITH EXISTING METHODS

Unlike the training of the DPM algorithms without preselection of the markers, here, it is proposed to construct and evaluate the DPMs using the subsets of markers defined by the categorical predictive models.

CONCLUSIONS

The estimates of the natural history of the proposed cognitive decline from GRACE were more robust than those using LTJMM. The transition from normal to cognitive decline is mostly associated with an increase in temporal atrophy, worsening of clinical scores and pTAU/Aβ. Furthermore, pTAU/Aβ, Everyday Cognition score and the normalized volume of the entorhinal cortex show alterations of more than 20% fifteen years before the onset of cognitive decline.

In-depth insights into Alzheimer's disease by using explainable machine learning approach

Alzheimer's disease is still a field of research with lots of open questions. The complexity of the disease prevents the early diagnosis before visible symptoms regarding the individual's cognitive capabilities occur. This research presents an in-depth analysis of a huge data set encompassing medical, cognitive and lifestyle's measurements from more than 12,000 individuals. Several hypothesis were established whose validity has been questioned considering the obtained results. The importance of appropriate experimental design is highly stressed in the research. Thus, a sequence of methods for handling missing data, redundancy, data imbalance, and correlation analysis have been applied for appropriate preprocessing of the data set, and consequently XGBoost model has been trained and evaluated with special attention to the hyperparameters tuning. The model was explained by using the Shapley values produced by the SHAP method. XGBoost produced a f1-score of 0.84 and as such is considered to be highly competitive among those published in the literature. This achievement, however, was not the main contribution of this paper. This research's goal was to perform global and local interpretability of the intelligent model and derive valuable conclusions over the established hypothesis. Those methods led to a single scheme which presents either positive, or, negative influence of the values of each of the features whose importance has been confirmed by means of Shapley values. This scheme might be considered as additional source of knowledge for the physicians and other experts whose concern is the exact diagnosis of early stage of Alzheimer's disease. The conclusions derived from the intelligent model's data-driven interpretability confronted all the established hypotheses. This research clearly showed the importance of explainable Machine learning approach that opens the black box and clearly unveils the relationships among the features and the diagnoses.

Mapping cerebral atrophic trajectory from amnestic mild cognitive impairment to Alzheimer's disease

Alzheimer's disease (AD) patients suffer progressive cerebral atrophy before dementia onset. However, the region-specific atrophic processes and the influences of age and apolipoprotein E (APOE) on atrophic trajectory are still unclear. By mapping the region-specific nonlinear atrophic trajectory of whole cerebrum from amnestic mild cognitive impairment (aMCI) to AD based on longitudinal structural magnetic resonance imaging data from Alzheimer's disease Neuroimaging Initiative (ADNI) database, we unraveled a quadratic accelerated atrophic trajectory of 68 cerebral regions from aMCI to AD, especially in the superior temporal pole, caudate, and hippocampus. Besides, interaction analyses demonstrated that APOE ε4 carriers had faster atrophic rates than noncarriers in 8 regions, including the caudate, hippocampus, insula, etc.; younger patients progressed faster than older patients in 32 regions, especially for the superior temporal pole, hippocampus, and superior temporal gyrus; and 15 regions demonstrated complex interaction among age, APOE, and disease progression, including the caudate, hippocampus, etc. (P < 0.05/68, Bonferroni correction). Finally, Cox proportional hazards regression model based on the identified region-specific biomarkers could effectively predict the time to AD conversion within 10 years. In summary, cerebral atrophic trajectory mapping could help a comprehensive understanding of AD development and offer potential biomarkers for predicting AD conversion.

Deep recurrent model for individualized prediction of Alzheimer's disease progression

Alzheimer's disease (AD) is known as one of the major causes of dementia and is characterized by slow progression over several years, with no treatments or available medicines. In this regard, there have been efforts to identify the risk of developing AD in its earliest time. While many of the previous works considered cross-sectional analysis, more recent studies have focused on the diagnosis and prognosis of AD with longitudinal or time series data in a way of disease progression modeling. Under the same problem settings, in this work, we propose a novel computational framework that can predict the phenotypic measurements of MRI biomarkers and trajectories of clinical status along with cognitive scores at multiple future time points. However, in handling time series data, it generally faces many unexpected missing observations. In regard to such an unfavorable situation, we define a secondary problem of estimating those missing values and tackle it in a systematic way by taking account of temporal and multivariate relations inherent in time series data. Concretely, we propose a deep recurrent network that jointly tackles the four problems of (i) missing value imputation, (ii) phenotypic measurements forecasting, (iii) trajectory estimation of a cognitive score, and (iv) clinical status prediction of a subject based on his/her longitudinal imaging biomarkers. Notably, the learnable parameters of all the modules in our predictive models are trained in an end-to-end manner by taking the morphological features and cognitive scores as input, with our circumspectly defined loss function. In our experiments over The Alzheimers Disease Prediction Of Longitudinal Evolution (TADPOLE) challenge cohort, we measured performance for various metrics and compared our method to competing methods in the literature. Exhaustive analyses and ablation studies were also conducted to better confirm the effectiveness of our method.

A predictive framework based on brain volume trajectories enabling early detection of Alzheimer's disease

We present a framework for constructing predictive models of cognitive decline from longitudinal MRI examinations, based on mixed effects models and machine learning. We apply the framework to detect conversion from cognitively normal (CN) to mild cognitive impairment (MCI) and from MCI to Alzheimer's disease (AD), using a large collection of subjects sourced from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and the Australian Imaging, Biomarkers and Lifestyle Flagship Study of Aging (AIBL). We extract subcortical segmentation and cortical parcellation from corresponding T1-weighted images using FreeSurfer v.6.0, select bilateral 3D regions of interest relevant to neurodegeneration/dementia, and fit their longitudinal volume trajectories using linear mixed effects models. Features describing these model-based trajectories are then used to train an ensemble of machine learning classifiers to distinguish stable CN from converters to MCI, and stable MCI from converters to AD. On separate test sets the models achieved an average of accuracy/precision/recall score of 69/73/60% for converted to MCI and 75/74/77% for converted to AD, illustrating the framework's ability to extract predictive imaging-based biomarkers from routine T1-weighted MRI acquisitions.

Robust parametric modeling of Alzheimer's disease progression

Quantitative characterization of disease progression using longitudinal data can provide long-term predictions for the pathological stages of individuals. This work studies the robust modeling of Alzheimer's disease progression using parametric methods. The proposed method linearly maps the individual's age to a disease progression score (DPS) and jointly fits constrained generalized logistic functions to the longitudinal dynamics of biomarkers as functions of the DPS using M-estimation. Robustness of the estimates is quantified using bootstrapping via Monte Carlo resampling, and the estimated inflection points of the fitted functions are used to temporally order the modeled biomarkers in the disease course. Kernel density estimation is applied to the obtained DPSs for clinical status classification using a Bayesian classifier. Different M-estimators and logistic functions, including a novel type proposed in this study, called modified Stannard, are evaluated on the data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) for robust modeling of volumetric magnetic resonance imaging (MRI) and positron emission tomography (PET) biomarkers, cerebrospinal fluid (CSF) measurements, as well as cognitive tests. The results show that the modified Stannard function fitted using the logistic loss achieves the best modeling performance with an average normalized mean absolute error (NMAE) of 0.991 across all biomarkers and bootstraps. Applied to the ADNI test set, this model achieves a multiclass area under the ROC curve (AUC) of 0.934 in clinical status classification. The obtained results for the proposed model outperform almost all state-of-the-art results in predicting biomarker values and classifying clinical status. Finally, the experiments show that the proposed model, trained using abundant ADNI data, generalizes well to data from the National Alzheimer's Coordinating Center (NACC) with an average NMAE of 1.182 and a multiclass AUC of 0.929.

Predicting Alzheimer's conversion in mild cognitive impairment patients using longitudinal neuroimaging and clinical markers

Patients with mild cognitive impairment (MCI) have a high risk for conversion to Alzheimer's disease (AD). Early diagnose of AD in MCI subjects could help to slow or halt the disease progression. Selecting a set of relevant markers from multimodal data to predict conversion from MCI to probable AD has become a challenging task. The aim of this paper is to quantify the impact of longitudinal predictive models with single- or multisource data for predicting MCI-to-AD conversion and identifying a very small subset of features that are highly predictive of conversion. We developed predictive models of MCI-to-AD progression that combine magnetic resonance imaging (MRI)-based markers (cortical thickness and volume of subcortical structures) with neuropsychological tests. These models were built with longitudinal data and validated using baseline values. By using a linear mixed effects approach, we modeled the longitudinal trajectories of the markers. A set of longitudinal features potentially discriminating between MCI subjects who convert to dementia and those who remain stable over a period of 3 years was obtained. Classifier were trained using the marginal longitudinal trajectory residues from the selected features. Our best models predicted conversion with 77% accuracy at baseline (AUC = 0.855, 84% sensitivity, 70% specificity). As more visits were available, longitudinal predictive models improved their predictions with 84% accuracy (AUC = 0.912, 83% sensitivity, 84% specificity). The proposed approach was developed, trained and evaluated using the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset with a total of 2491 visits from 610 subjects.

Longitudinal survival analysis and two-group comparison for predicting the progression of mild cognitive impairment to Alzheimer's disease

BACKGROUND

Longitudinal studies using structural magnetic resonance imaging (MRI) and neuropsychological measurements (NMs) allow a noninvasive means of following the subtle anatomical changes occurring during the evolution of AD.

NEW METHOD

This paper compared two approaches for the construction of longitudinal predictive models: a) two-group comparison between converter and nonconverter MCI subjects and b) longitudinal survival analysis. Predictive models combined MRI-based markers with NMs and included demographic and clinical information as covariates. Both approaches employed linear mixed effects modeling to capture the longitudinal trajectories of the markers. The two-group comparison approaches used linear discriminant analysis and the survival analysis used risk ratios obtained from the extended Cox model and logistic regression.

RESULTS

The proposed approaches were developed and evaluated using the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset with a total of 1330 visits from 321 subjects. With both approaches, a very small number of features were selected. These markers are easily interpretable, generating robust, verifiable and reliable predictive models. Our best models predicted conversion with 78% accuracy at baseline (AUC = 0.860, 79% sensitivity, 76% specificity). As more visits were made, longitudinal predictive models improved their predictions with 85% accuracy (AUC = 0.944, 86% sensitivity, 85% specificity).

COMPARISON WITH EXISTING METHOD

Unlike the recently published models, there was also an improvement in the prediction accuracy of the conversion to AD when considering the longitudinal trajectory of the patients.

CONCLUSIONS

The survival-based predictive models showed a better balance between sensitivity and specificity with respect to the models based on the two-group comparison approach.

A survey on machine and statistical learning for longitudinal analysis of neuroimaging data in Alzheimer's disease

BACKGROUND AND OBJECTIVES

Recently, longitudinal studies of Alzheimer's disease have gathered a substantial amount of neuroimaging data. New methods are needed to successfully leverage and distill meaningful information on the progression of the disease from the deluge of available data. Machine learning has been used successfully for many different tasks, including neuroimaging related problems. In this paper, we review recent statistical and machine learning applications in Alzheimer's disease using longitudinal neuroimaging.

METHODS

We search for papers using longitudinal imaging data, focused on Alzheimer's Disease and published between 2007 and 2019 on four different search engines.

RESULTS

After the search, we obtain 104 relevant papers. We analyze their approach to typical challenges in longitudinal data analysis, such as missing data and variability in the number and extent of acquisitions.

CONCLUSIONS

Reviewed works show that machine learning methods using longitudinal data have potential for disease progression modelling and computer-aided diagnosis. We compare results and models, and propose future research directions in the field.

Predicting time to dementia using a quantitative template of disease progression

INTRODUCTION

Characterization of longitudinal trajectories of biomarkers implicated in sporadic Alzheimer's disease (AD) in decades before clinical diagnosis is important for disease prevention and monitoring.

METHODS

We used a multivariate Bayesian model to temporally align 1369 Alzheimer's disease Neuroimaging Initiative participants based on the similarity of their longitudinal biomarker measures and estimated a quantitative template of the temporal evolution of cerebrospinal fluid Aβ 1 - 42 , p- ta u 181 p , and t-tau and hippocampal volume, brain glucose metabolism, and cognitive measurements. We computed biomarker trajectories as a function of time to AD dementia and predicted AD dementia onset age in a disjoint sample.

RESULTS

Quantitative template showed early changes in verbal memory, cerebrospinal fluid Aβ1-42 and p-tau181p, and hippocampal volume. Mean error in predicted AD dementia onset age was < 1.5 years.

DISCUSSION

Our method provides a quantitative approach for characterizing the natural history of AD starting at preclinical stages despite the lack of individual-level longitudinal data spanning the entire disease timeline.

TADPOLE Challenge: Accurate Alzheimer's disease prediction through crowdsourced forecasting of future data

The Alzheimer's Disease Prediction Of Longitudinal Evolution (TADPOLE) Challenge compares the performance of algorithms at predicting the future evolution of individuals at risk of Alzheimer's disease. TADPOLE Challenge participants train their models and algorithms on historical data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study. Participants are then required to make forecasts of three key outcomes for ADNI-3 rollover participants: clinical diagnosis, Alzheimer's Disease Assessment Scale Cognitive Subdomain (ADAS-Cog 13), and total volume of the ventricles - which are then compared with future measurements. Strong points of the challenge are that the test data did not exist at the time of forecasting (it was acquired afterwards), and that it focuses on the challenging problem of cohort selection for clinical trials by identifying fast progressors. The submission phase of TADPOLE was open until 15 November 2017; since then data has been acquired until April 2019 from 219 subjects with 223 clinical visits and 150 Magnetic Resonance Imaging (MRI) scans, which was used for the evaluation of the participants' predictions. Thirty-three teams participated with a total of 92 submissions. No single submission was best at predicting all three outcomes. For diagnosis prediction, the best forecast (team Frog), which was based on gradient boosting, obtained a multiclass area under the receiver-operating curve (MAUC) of 0.931, while for ventricle prediction the best forecast (team EMC1 ), which was based on disease progression modelling and spline regression, obtained mean absolute error of 0.41% of total intracranial volume (ICV). For ADAS-Cog 13, no forecast was considerably better than the benchmark mixed effects model (BenchmarkME ), provided to participants before the submission deadline. Further analysis can help understand which input features and algorithms are most suitable for Alzheimer's disease prediction and for aiding patient stratification in clinical trials. The submission system remains open via the website: https://tadpole.grand-challenge.org/.

Computational Causal Modeling of the Dynamic Biomarker Cascade in Alzheimer's Disease

Background

Alzheimer's disease (AD) is a major public health concern, and there is an urgent need to better understand its complex biology and develop effective therapies. AD progression can be tracked in patients through validated imaging and spinal fluid biomarkers of pathology and neuronal loss. We still, however, lack a coherent quantitative model that explains how these biomarkers interact and evolve over time. Such a model could potentially help identify the major drivers of disease in individual patients and simulate response to therapy prior to entry in clinical trials. A current theory of AD biomarker progression, known as the dynamic biomarker cascade model, hypothesizes AD biomarkers evolve in a sequential but temporally overlapping manner. A computational model incorporating assumptions about the underlying biology of this theory and its variations would be useful to test and refine its accuracy with longitudinal biomarker data from clinical trials.

Methods

We implemented a causal model to simulate time-dependent biomarker data under the descriptive assumptions of the dynamic biomarker cascade theory. We modeled pathologic biomarkers (beta-amyloid and tau), neuronal loss biomarkers, and cognitive impairment as nonlinear first-order ordinary differential equations (ODEs) to include amyloid-dependent and nondependent neurodegenerative cascades. We tested the feasibility of the model by adjusting its parameters to simulate three specific natural history scenarios in early-onset autosomal dominant AD and late-onset AD and determine whether computed biomarker trajectories agreed with current assumptions of AD biomarker progression. We also simulated the effects of antiamyloid therapy in late-onset AD.

Results

The computational model of early-onset AD demonstrated the initial appearance of amyloid, followed by biomarkers of tau and neurodegeneration and the onset of cognitive decline based on cognitive reserve, as predicted by the prior literature. Similarly, the late-onset AD computational models demonstrated the first appearance of amyloid or nonamyloid-related tauopathy, depending on the magnitude of comorbid pathology, and also closely matched the biomarker cascades predicted by the prior literature. Forward simulation of antiamyloid therapy in symptomatic late-onset AD failed to demonstrate any slowing in progression of cognitive decline, consistent with prior failed clinical trials in symptomatic patients.

Conclusions

We have developed and computationally implemented a mathematical causal model of the dynamic biomarker cascade theory in AD. We demonstrate the feasibility of this model by simulating biomarker evolution and cognitive decline in early- and late-onset natural history scenarios, as well as in a treatment scenario targeted at core AD pathology. Models resulting from this causal approach can be further developed and refined using patient data from longitudinal biomarker studies and may in the future play a key role in personalizing approaches to treatment.

Convolutional Neural Networks-Based MRI Image Analysis for the Alzheimer's Disease Prediction From Mild Cognitive Impairment

Mild cognitive impairment (MCI) is the prodromal stage of Alzheimer's disease (AD). Identifying MCI subjects who are at high risk of converting to AD is crucial for effective treatments. In this study, a deep learning approach based on convolutional neural networks (CNN), is designed to accurately predict MCI-to-AD conversion with magnetic resonance imaging (MRI) data. First, MRI images are prepared with age-correction and other processing. Second, local patches, which are assembled into 2.5 dimensions, are extracted from these images. Then, the patches from AD and normal controls (NC) are used to train a CNN to identify deep learning features of MCI subjects. After that, structural brain image features are mined with FreeSurfer to assist CNN. Finally, both types of features are fed into an extreme learning machine classifier to predict the AD conversion. The proposed approach is validated on the standardized MRI datasets from the Alzheimer's Disease Neuroimaging Initiative (ADNI) project. This approach achieves an accuracy of 79.9% and an area under the receiver operating characteristic curve (AUC) of 86.1% in leave-one-out cross validations. Compared with other state-of-the-art methods, the proposed one outperforms others with higher accuracy and AUC, while keeping a good balance between the sensitivity and specificity. Results demonstrate great potentials of the proposed CNN-based approach for the prediction of MCI-to-AD conversion with solely MRI data. Age correction and assisted structural brain image features can boost the prediction performance of CNN.

Uncovering the heterogeneity and temporal complexity of neurodegenerative diseases with Subtype and Stage Inference

The heterogeneity of neurodegenerative diseases is a key confound to disease understanding and treatment development, as study cohorts typically include multiple phenotypes on distinct disease trajectories. Here we introduce a machine-learning technique—Subtype and Stage Inference (SuStaIn)—able to uncover data-driven disease phenotypes with distinct temporal progression patterns, from widely available cross-sectional patient studies. Results from imaging studies in two neurodegenerative diseases reveal subgroups and their distinct trajectories of regional neurodegeneration. In genetic frontotemporal dementia, SuStaIn identifies genotypes from imaging alone, validating its ability to identify subtypes; further the technique reveals within-genotype heterogeneity. In Alzheimer’s disease, SuStaIn uncovers three subtypes, uniquely characterising their temporal complexity. SuStaIn provides fine-grained patient stratification, which substantially enhances the ability to predict conversion between diagnostic categories over standard models that ignore subtype (p = 7.18 × 10⁻⁴) or temporal stage (p = 3.96 × 10⁻⁵). SuStaIn offers new promise for enabling disease subtype discovery and precision medicine.

Progressive diseases tend to be heterogeneous in their underlying aetiology mechanism, disease manifestation, and disease time course. Here, Young and colleagues devise a computational method to account for both phenotypic heterogeneity and temporal heterogeneity, and demonstrate it using two neurodegenerative disease cohorts.

Estimating the Evolution of Disease in the Parkinson's Progression Markers Initiative

Parkinson's disease is the second most common neurological disease and affects about 1% of persons over the age of 60 years. Due to the lack of approved surrogate markers, confirmation of the disease still requires postmortem examination. Identifying and validating biomarkers are essential steps toward improving clinical diagnosis and accelerating the search for therapeutic drugs to ameliorate disease symptoms. Until recently, statistical analysis of multicohort longitudinal studies of neurodegenerative diseases has usually been restricted to a single analysis per outcome with simple comparisons between diagnostic groups. However, an important methodological consideration is to allow the modeling framework to handle multiple outcomes simultaneously and consider the transitions between diagnostic groups. This enables researchers to monitor multiple trajectories, correctly account for the correlation among biomarkers, and assess how these associations may jointly change over the long-term course of disease. In this study, we apply a latent time joint mixed-effects model to study biomarker progression and disease dynamics in the Parkinson's Progression Markers Initiative (PPMI) and examine which markers might be most informative in the earliest phases of disease. The results reveal that, even though diagnostic category was not included in the model, it seems to accurately reflect the temporal ordering of the disease state consistent with diagnosis categorization at baseline. In addition, results indicated that the specific binding ratio on striatum and the total Unified Parkinson's Disease Rating Scale (UPDRS) show high discriminability between disease stages. An extended latent time joint mixed-effects model with heterogeneous latent time variance also showed improvement in model fit in a simulation study and when applied to real data.

Structural brain imaging in Alzheimer's disease and mild cognitive impairment: biomarker analysis and shared morphometry database

Magnetic resonance (MR) imaging is a powerful technique for non-invasive in-vivo imaging of the human brain. We employed a recently validated method for robust cross-sectional and longitudinal segmentation of MR brain images from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort. Specifically, we segmented 5074 MR brain images into 138 anatomical regions and extracted time-point specific structural volumes and volume change during follow-up intervals of 12 or 24 months. We assessed the extracted biomarkers by determining their power to predict diagnostic classification and by comparing atrophy rates to published meta-studies. The approach enables comprehensive analysis of structural changes within the whole brain. The discriminative power of individual biomarkers (volumes/atrophy rates) is on par with results published by other groups. We publish all quality-checked brain masks, structural segmentations, and extracted biomarkers along with this article. We further share the methodology for brain extraction (pincram) and segmentation (MALPEM, MALPEM4D) as open source projects with the community. The identified biomarkers hold great potential for deeper analysis, and the validated methodology can readily be applied to other imaging cohorts.

Personalized Gaussian Processes for Forecasting of Alzheimer's Disease Assessment Scale-Cognition Sub-Scale (ADAS-Cog13)

In this paper, we introduce the use of a personalized Gaussian Process pGP model to predict per-patient changes in ADAS-Cog13-a significant predictor of Alzheimer's Disease (AD) in the cognitive domain - using data from each patient's previous visits, and testing on future (held-out) data. We start by learning a population-level model using multi- modal data from previously seen patients using a base Gaussian Process (GP) regression. The pGP is then formed by adapting the base GP sequentially over time to a new (target) patient using domain adaptive GPs [1]. We extend this personalized approach to predict the values of ADAS-Cog13 over the future 6, 12, 18, and 24 months. We compare this approach to a GP model trained only on past data of the target patients tGP, as well as to a new approach that combines pGP with tGP. We find that this new approach (pGP+tGP) leads to significant improvements in accurately forecasting future ADAS-Cog13 scores.