Machine learning methods to predict amyloid positivity using domain scores from cognitive tests

Abstract

Amyloid-$$\beta$$ β (A$$\beta$$ β ) is the target in many clinical trials for Alzheimer’s disease (AD). Preclinical AD patients are heterogeneous with regards to different backgrounds and diagnosis. Accurately predicting A$$\beta$$ β status of participants by using machine learning (ML) models based on easily accessible data, could improve the effectiveness of AD clinical trials. We will develop optimal ML models for each subpopulation stratified by sex and disease stages using sub scores from screening neurological tests. Data from the AD Neuroimaging Initiative (ADNI) were used to build the ML models, for three groups: individuals with significant memory concern, early mild cognitive impairment (MCI), and late MCI. Data were further separated into 6 groups by disease stage (3 levels) and sex (2 categories). The outcome was defined as the A$$\beta$$ β status confirmed by the PET imaging, and the features include demographic data, newly identified risk factors, screening tests, and the domain scores from screening tests. Monte Carlo simulation studies were used together with k-fold cross-validation technique to compute model performance metric. We also develop a new feature selection method based on the stochastic ordering to avoiding searching all possible combinations of features. Accuracy of the identified optimal model for SMC male was over 90% by using domain scores, and accuracy for LMCI female was above 86%. Domain scores can improve the ML model prediction as compared to the total scores. Accurate ML prediction models can identify the proper population for AD clinical trials.

Predicting Amyloid-β Positivity in Alzheimer’s Disease: Comprehensive Analysis of Feature Selection and Machine Learning Models for Accurate Identification

“…Women have demonstrated an advantage in verbal memory, suggesting that memory-based measures may be less effective in screening women for early changes associated with AD. Furthermore, the strong memory performance exhibited by women may mask early signs of the disease, making it less effective to predict the presence of brain amyloid using memory test scores, especially in women without detectable memory deficits [31]. Therefore, it is crucial to develop separate statistical prediction models for each subpopulation stratified by sex and APOE ε4 status.…”

Potential Applications of Artificial Intelligence in Clinical Trials for Alzheimer’s Disease

“…AI prescreening algorithms can reduce challenges of PET and CSF, such as high costs and participants' fear of radiation exposure, by selecting a subset of individuals who are likely to be Aβ or tau positive. Therefore, in AI research for AD clinical trials, many aimed to predict amyloidosis in subjects with MCI [23][24][25], while others focused on preclinical stages before neurodegeneration is too substantial [26,27] (Table 2).…”

Comparison of deep learning architectures for predicting amyloid positivity in Alzheimer’s disease, mild cognitive impairment, and healthy aging, from T1-weighted brain structural MRI

“…In Shan et al (2021) , Shan et al used Monte Carlo simulations with k -fold cross validation to predict Aβ positivity using domain scores from cognitive tests, obtaining an accuracy of 0.90 and 0.86 on men and women, respectively, with subjective memory complaints. In Ezzati et al (2020) , Ezzati et al used an ensemble linear discriminant model to predict Aβ positivity using demographic information, ApoE4 genotype (as this is the major risk gene for late onset AD), MRI volumetrics and CSF biomarkers, yielding AUCs between 0.89 and 0.92 in participants with amnestic mild cognitive impairment (aMCI).…”