scholarly journals IMPROVING BRAIN AGE PREDICTION MODELS: INCORPORATION OF AMYLOID STATUS IN ALZHEIMER’S DISEASE

2019 ◽  
Vol 3 (Supplement_1) ◽  
pp. S91-S91
Author(s):  
Maria Ly ◽  
Nishita Muppidi ◽  
Helmet Karim ◽  
Gary Yu ◽  
Akiko Mizuno ◽  
...  

Abstract Brain age prediction may serve as a promising, individualized biomarker of brain health and may help us understand the heterogeneous biological changes that occur in aging. Brain age prediction is a machine learning method that estimates an individual’s chronological age from their neuroimaging scans. If predicted brain age is greater than chronological age, that individual may have an “older” brain than expected, suggesting that they may have experienced a higher cumulative exposure to brain insults or were more impacted by those pathological insults. However, contemporary brain age models include older participants with amyloid pathology in their training sets and thus may be confounded when studying Alzheimer’s disease (AD). We showed that amyloid status is a critical feature for brain age prediction models by training a model on 808 individuals without significant amyloid pathology from the ADNI, OASIS-3, and IXI cohorts. Our model accurately predicted brain age in the training and independent test sets, comparable to previous published models: [r(807) = 0.94, R2 = 0.88, p=0.001, MAE = 4.9 years, p=0.001], [r(39) = 0.67, R2 = 0.45, and MAE = 4.6 years]. We demonstrated significant differences between AD diagnostic groups [F(3,431)=30.7, p<0.001], and our model was the first to delineate significant differences in brain age relative to chronological age between cognitively normal individuals with and without amyloid [mean difference, 95% CI; CN-Aβ(-) (-3.4, -4.9:-1.8), CN-Aβ(+) (-0.7, -1.9:0.5)]. Ultimately, incorporation of amyloid status in brain age prediction models improves the utility of brain age as a biomarker for aging and AD.

2020 ◽  
Vol 87 ◽  
pp. 44-48 ◽  
Author(s):  
Maria Ly ◽  
Gary Z. Yu ◽  
Helmet T. Karim ◽  
Nishita R. Muppidi ◽  
Akiko Mizuno ◽  
...  

2021 ◽  
Vol 12 ◽  
Author(s):  
Matthias S. Treder ◽  
Jonathan P. Shock ◽  
Dan J. Stein ◽  
Stéfan du Plessis ◽  
Soraya Seedat ◽  
...  

In neuroimaging, the difference between chronological age and predicted brain age, also known as brain age delta, has been proposed as a pathology marker linked to a range of phenotypes. Brain age delta is estimated using regression, which involves a frequently observed bias due to a negative correlation between chronological age and brain age delta. In brain age prediction models, this correlation can manifest as an overprediction of the age of young brains and an underprediction for elderly ones. We show that this bias can be controlled for by adding correlation constraints to the model training procedure. We develop an analytical solution to this constrained optimization problem for Linear, Ridge, and Kernel Ridge regression. The solution is optimal in the least-squares sense i.e., there is no other model that satisfies the correlation constraints and has a better fit. Analyses on the PAC2019 competition data demonstrate that this approach produces optimal unbiased predictive models with a number of advantages over existing approaches. Finally, we introduce regression toolboxes for Python and MATLAB that implement our algorithm.


2020 ◽  
Author(s):  
Julie Gonneaud ◽  
Alex T. Baria ◽  
Alexa Pichet Binette ◽  
Brian A. Gordon ◽  
Jasmeer P. Chhatwal ◽  
...  

AbstractWe aimed at developing a model able to predict brain aging from resting state functional connectivity (rs-fMRI) and assessing whether genetic risk/determinants of Alzheimer’s disease (AD) and amyloid (Aβ) pathology contributes to accelerated brain aging. Using data collected in 1340 cognitively unimpaired participants from 18 to 94 years old selected across multi-site cohorts, we showed that chronological age can be predicted across the whole lifespan from topological properties of graphs constructed from rs-fMRI. We subsequently used the difference between the model-predicted age and the chronological age in pre-symptomatic autosomal dominant AD (ADAD) mutation carriers and asymptomatic individuals at risk of sporadic AD and assessed the influence of genetics and Aβ pathology on brain age. Applying our predictive model in the context of preclinical AD revealed that the pre-symptomatic phase of ADAD is characterized by accelerated functional brain aging. This phenomenon is independent from, and might precede, detectable fibrillar Aβ deposition.


2019 ◽  
Vol 2 (3) ◽  
pp. e201900303 ◽  
Author(s):  
Christin A Glorioso ◽  
Andreas R Pfenning ◽  
Sam S Lee ◽  
David A Bennett ◽  
Etienne L Sibille ◽  
...  

Advanced age and the APOE ε4 allele are the two biggest risk factors for Alzheimer’s disease (AD) and declining cognitive function. We describe a universal gauge to measure molecular brain age using transcriptome analysis of four human postmortem cohorts (n = 673, ages 25–97) free of neurological disease. In a fifth cohort of older subjects with or without neurological disease (n = 438, ages 67–108), we show that subjects with brains deviating in the older direction from what would be expected based on chronological age show an increase in AD, Parkinson’s disease, and cognitive decline. Strikingly, a younger molecular age (−5 yr than chronological age) protects against AD even in the presence of APOE ε4. An established DNA methylation gauge for age correlates well with the transcriptome gauge for determination of molecular age and assigning deviations from the expected. Our results suggest that rapid brain aging and APOE ε4 are synergistic risk factors, and interventions that slow aging may substantially reduce risk of neurological disease and decline even in the presence of APOE ε4.


2021 ◽  
Vol 79 (4) ◽  
pp. 1533-1546
Author(s):  
Mithilesh Prakash ◽  
Mahmoud Abdelaziz ◽  
Linda Zhang ◽  
Bryan A. Strange ◽  
Jussi Tohka ◽  
...  

Background: Quantitatively predicting the progression of Alzheimer’s disease (AD) in an individual on a continuous scale, such as the Alzheimer’s Disease Assessment Scale-cognitive (ADAS-cog) scores, is informative for a personalized approach as opposed to qualitatively classifying the individual into a broad disease category. Objective: To evaluate the hypothesis that the multi-modal data and predictive learning models can be employed for future predicting ADAS-cog scores. Methods: Unimodal and multi-modal regression models were trained on baseline data comprised of demographics, neuroimaging, and cerebrospinal fluid based markers, and genetic factors to predict future ADAS-cog scores for 12, 24, and 36 months. We subjected the prediction models to repeated cross-validation and assessed the resulting mean absolute error (MAE) and cross-validated correlation (ρ) of the model. Results: Prediction models trained on multi-modal data outperformed the models trained on single modal data in predicting future ADAS-cog scores (MAE12, 24 & 36 months= 4.1, 4.5, and 5.0, ρ12, 24 & 36 months= 0.88, 0.82, and 0.75). Including baseline ADAS-cog scores to prediction models improved predictive performance (MAE12, 24 & 36 months= 3.5, 3.7, and 4.6, ρ12, 24 & 36 months= 0.89, 0.87, and 0.80). Conclusion: Future ADAS-cog scores were predicted which could aid clinicians in identifying those at greater risk of decline and apply interventions at an earlier disease stage and inform likely future disease progression in individuals enrolled in AD clinical trials.


Aging Cell ◽  
2021 ◽  
Vol 20 (3) ◽  
Author(s):  
Jiangong Wang ◽  
Bin Liu ◽  
Yong Xu ◽  
Meizi Yang ◽  
Chaoyun Wang ◽  
...  

2016 ◽  
Vol 68 (22) ◽  
pp. 2395-2407 ◽  
Author(s):  
Luca Troncone ◽  
Marco Luciani ◽  
Matthew Coggins ◽  
Elissa H. Wilker ◽  
Cheng-Ying Ho ◽  
...  

2012 ◽  
Vol 7 (Suppl 1) ◽  
pp. O7 ◽  
Author(s):  
Wenfeng Yu ◽  
Naguib Mechawar ◽  
Slavica Krantic ◽  
Jean-Guy Chabot ◽  
Rémi Quirion

Sign in / Sign up

Export Citation Format

Share Document