Temporal Trajectory and Progression Score Estimation from Voxelwise Longitudinal Imaging Measures: Application to Amyloid Imaging

Machine learning for modeling the progression of Alzheimer disease dementia using clinical data: a systematic literature review

OBJECTIVE

Alzheimer disease (AD) is the most common cause of dementia, a syndrome characterized by cognitive impairment severe enough to interfere with activities of daily life. We aimed to conduct a systematic literature review (SLR) of studies that applied machine learning (ML) methods to clinical data derived from electronic health records in order to model risk for progression of AD dementia.

MATERIALS AND METHODS

We searched for articles published between January 1, 2010, and May 31, 2020, in PubMed, Scopus, ScienceDirect, IEEE Explore Digital Library, Association for Computing Machinery Digital Library, and arXiv. We used predefined criteria to select relevant articles and summarized them according to key components of ML analysis such as data characteristics, computational algorithms, and research focus.

RESULTS

There has been a considerable rise over the past 5 years in the number of research papers using ML-based analysis for AD dementia modeling. We reviewed 64 relevant articles in our SLR. The results suggest that majority of existing research has focused on predicting progression of AD dementia using publicly available datasets containing both neuroimaging and clinical data (neurobehavioral status exam scores, patient demographics, neuroimaging data, and laboratory test values).

DISCUSSION

Identifying individuals at risk for progression of AD dementia could potentially help to personalize disease management to plan future care. Clinical data consisting of both structured data tables and clinical notes can be effectively used in ML-based approaches to model risk for AD dementia progression. Data sharing and reproducibility of results can enhance the impact, adaptation, and generalizability of this research.

Simulating the outcome of amyloid treatments in Alzheimer's disease from imaging and clinical data

In this study, we investigate SimulAD, a novel quantitative instrument for the development of intervention strategies for disease-modifying drugs in Alzheimer's disease. SimulAD is based on the modeling of the spatio-temporal dynamics governing the joint evolution of imaging and clinical biomarkers along the history of the disease, and allows the simulation of the effect of intervention time and drug dosage on the biomarkers' progression. When applied to multi-modal imaging and clinical data from the Alzheimer's Disease Neuroimaging Initiative the method enables to generate hypothetical scenarios of amyloid lowering interventions. The results quantify the crucial role of intervention time, and provide a theoretical justification for testing amyloid modifying drugs in the pre-clinical stage. Our experimental simulations are compatible with the outcomes observed in past clinical trials, and suggest that anti-amyloid treatments should be administered at least 7 years earlier than what is currently being done in order to obtain statistically powered improvement of clinical endpoints.

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.

The effect of rare variants in TREM2 and PLD3 on longitudinal cognitive function in the Wisconsin Registry for Alzheimer's Prevention

Recent studies have found an association between functional variants in TREM2 and PLD3 and Alzheimer's disease (AD), but their effect on cognitive function is unknown. We examined the effect of these variants on cognitive function in 1449 participants from the Wisconsin Registry for Alzheimer's Prevention, a longitudinal study of initially asymptomatic adults, aged 36-73 years at baseline, enriched for a parental history of AD. A comprehensive cognitive test battery was performed at up to 5 visits. A factor analysis resulted in 6 cognitive factors that were standardized into z scores (∼N [0, 1]); the mean of these z scores was also calculated. In linear mixed models adjusted for age, gender, practice effects, and self-reported race/ethnicity, PLD3 V232M carriers had significantly lower mean z scores (p = 0.02) and lower z scores for story recall (p = 0.04), visual learning and memory (p = 0.049), and speed and flexibility (p = 0.02) than noncarriers. TREM2 R47H carriers had marginally lower z scores for speed and flexibility (p = 0.06). In conclusion, a functional variant in PLD3 was associated with significantly lower cognitive function in individuals carrying the variant than in noncarriers.

Associations between Performance on an Abbreviated CogState Battery, Other Measures of Cognitive Function, and Biomarkers in People at Risk for Alzheimer's Disease

It is not known whether computerized cognitive assessments, like the CogState battery, are sensitive to preclinical cognitive changes or pathology in people at risk for Alzheimer's disease(AD). In 469 late middle-aged participants from the Wisconsin Registry for Alzheimer's Prevention(mean age 63.8±7 years at testing; 67% female; 39% APOE4+), we examined relationships between a CogState abbreviated battery(CAB) of seven tests and demographic characteristics, traditional paper-based neuropsychological tests as well as a composite cognitive impairment index, cognitive impairment status(determined by consensus review), and biomarkers for amyloid and tau(CSF phosphorylated-tau/Aβ42 and global PET-PiB burden) and neural injury(CSF neurofilament light protein). CSF and PET-PiB were collected in n = 71 and n = 91 participants, respectively, approximately four years prior to CAB testing. For comparison, we examined three traditional tests of delayed memory in parallel. Similar to studies in older samples, the CAB was less influenced by demographic factors than traditional tests. CAB tests were generally correlated with most paper-based cognitive tests examined and mapped onto the same cognitive domains. Greater composite cognitive impairment index was associated with worse performance on all CAB tests. Cognitively impaired participants performed significantly worse compared to normal controls on all but one CAB test. Poorer One Card Learning test performance was associated with higher levels of CSF phosphorylated-tau/Aβ42. These results support the use of the CogState battery as measures of early cognitive impairment in studies of people at risk for AD.

Computerized Cognitive Tests Are Associated with Biomarkers of Alzheimer's Disease in Cognitively Normal Individuals 10 Years Prior

OBJECTIVES

Evidence suggests that Alzheimer's disease (AD) biomarkers become abnormal many years before the emergence of clinical symptoms of AD, raising the possibility that biomarker levels measured in cognitively normal individuals would be associated with cognitive performance many years later. This study examined whether performance on computerized cognitive tests is associated with levels of cerebrospinal fluid (CSF) biomarkers of amyloid, tau, and phosphorylated tau (p-tau) obtained approximately 10 years earlier, when individuals were cognitively normal and primarily middle-aged.

METHODS

Individuals from the BIOCARD cohort (mean age at testing=69 years) were tested on two computerized tasks hypothesized to rely on brain regions affected by the early accumulation of AD pathology: (1) a Paired Associates Learning (PAL) task (n=67) and (2) a visual search task (n=86).

RESULTS

In regression analyses, poorer performance on the PAL task was associated with higher levels of CSF p-tau obtained years earlier, whereas worse performance in the visual search task was associated with lower levels of CSF Aβ1-42.

CONCLUSIONS

These findings suggest that AD biomarker levels may be differentially predictive of specific cognitive functions many years later. In line with the pattern of early accumulation of AD pathology, the PAL task, hypothesized to rely on medial temporal lobe function, was associated with CSF p-tau, whereas the visual search task, hypothesized to rely on frontoparietal function, was associated with CSF amyloid. Studies using amyloid and tau PET imaging will be useful in examining these hypothesized relationships further. (JINS, 2016, 22, 968-977).

A multivariate nonlinear mixed effects model for longitudinal image analysis: Application to amyloid imaging

It is important to characterize the temporal trajectories of disease-related biomarkers in order to monitor progression and identify potential points of intervention. These are especially important for neurodegenerative diseases, as therapeutic intervention is most likely to be effective in the preclinical disease stages prior to significant neuronal damage. Neuroimaging allows for the measurement of structural, functional, and metabolic integrity of the brain at the level of voxels, whose volumes are on the order of mm(3). These voxelwise measurements provide a rich collection of disease indicators. Longitudinal neuroimaging studies enable the analysis of changes in these voxelwise measures. However, commonly used longitudinal analysis approaches, such as linear mixed effects models, do not account for the fact that individuals enter a study at various disease stages and progress at different rates, and generally consider each voxelwise measure independently. We propose a multivariate nonlinear mixed effects model for estimating the trajectories of voxelwise neuroimaging biomarkers from longitudinal data that accounts for such differences across individuals. The method involves the prediction of a progression score for each visit based on a collective analysis of voxelwise biomarker data within an expectation-maximization framework that efficiently handles large amounts of measurements and variable number of visits per individual, and accounts for spatial correlations among voxels. This score allows individuals with similar progressions to be aligned and analyzed together, which enables the construction of a trajectory of brain changes as a function of an underlying progression or disease stage. We apply our method to studying cortical β-amyloid deposition, a hallmark of preclinical Alzheimer's disease, as measured using positron emission tomography. Results on 104 individuals with a total of 300 visits suggest that precuneus is the earliest cortical region to accumulate amyloid, closely followed by the cingulate and frontal cortices, then by the lateral parietal cortex. The extracted progression scores reveal a pattern similar to mean cortical distribution volume ratio (DVR), an index of global brain amyloid levels. The proposed method can be applied to other types of longitudinal imaging data, including metabolism, blood flow, tau, and structural imaging-derived measures, to extract individualized summary scores indicating disease progression and to provide voxelwise trajectories that can be compared between brain regions.

Dynamics of biomarkers in relation to aging and mortality

Contemporary longitudinal studies collect repeated measurements of biomarkers allowing one to analyze their dynamics in relation to mortality, morbidity, or other health-related outcomes. Rich and diverse data collected in such studies provide opportunities to investigate how various socio-economic, demographic, behavioral and other variables can interact with biological and genetic factors to produce differential rates of aging in individuals. In this paper, we review some recent publications investigating dynamics of biomarkers in relation to mortality, which use single biomarkers as well as cumulative measures combining information from multiple biomarkers. We also discuss the analytical approach, the stochastic process models, which conceptualizes several aging-related mechanisms in the structure of the model and allows evaluating "hidden" characteristics of aging-related changes indirectly from available longitudinal data on biomarkers and follow-up on mortality or onset of diseases taking into account other relevant factors (both genetic and non-genetic). We also discuss an extension of the approach, which considers ranges of "optimal values" of biomarkers rather than a single optimal value as in the original model. We discuss practical applications of the approach to single biomarkers and cumulative measures highlighting that the potential of applications to cumulative measures is still largely underused.