Classifying Alzheimer's disease with brain imaging and genetic data using a neural network framework

Abstract

A long-standing question is how to best use brain morphometric and genetic data to distinguish AD patients from cognitively normal (CN) subjects and to predict those who will progress from mild cognitive impairment (MCI) to AD. Here we use a neural network (NN) framework on both magnetic resonance imaging-derived quantitative structural brain measures and genetic data to address this question. We tested the effectiveness of NN models in classifying and predicting AD. We further performed a novel analysis of the NN model to gain insight into the most predictive imaging and genetics features, and to identify possible interactions between features that affect AD risk. Data was obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort and included baseline structural MRI data and single nucleotide polymorphism (SNP) data for 138 AD patients, 225 CN subjects, and 358 MCI patients. We found that NN models with both brain and SNP features as predictors perform significantly better than models with either alone in classifying AD and CN subjects, with an area under the receiver operating characteristic curve (AUC) of 0.992, and in predicting the progression from MCI to AD (AUC=0.835). The most important predictors in the NN model were the left middle temporal gyrus volume, the left hippocampus volume, the right entorhinal cortex volume, and the APOE ε4 risk allele. Further, we identified interactions between the right parahippocampal gyrus and the right lateral occipital gyrus, the right banks of the superior temporal sulcus and the left posterior cingulate, and SNP rs10838725 and the left lateral occipital gyrus. Our work shows the ability of NN models to not only classify and predict AD occurrence, but also to identify important AD risk factors and interactions among them.

A Correlation Analysis between SNPs and ROIs of Alzheimer’s Disease Based on Deep Learning

“…Multimodal neuroimaging data can provide complementary information and theoretically improve the accuracy of classification results. In order to systematically understand the formation mechanism of AD, multiscale, multimode, and heterogeneous data should be fused to mine the interaction between cross-omics variables [ 21 , 22 ]. Some methods of data fusion based on ensemble classifier, dimension-based, and multicore learning have been proposed to establish a fusion predictor for other complex diseases.…”

An update on magnetic resonance imaging markers in AD

“…In addition, the functional connectivity features of the regions of interest also predicted the severity of the neuropsychiatric symptoms, as well as the AD pathology (indexed by baseline and change of Aβ/phosphorylated tau ratio) with above 70% accuracy rate. Furthermore, measures of MRI at baseline (GM volumes of specific regions of interest, including the hippocampus) in combination with genetic [apolipoprotein E (APOE) ε4 allele and 19 single nucleotide polymorphisms (SNPs) significantly associated with AD] 35 or hypometabolism (qualitatively investigated with 18F-fluoro-2-deoxyglucose positron emission tomography – FDG-PET) 36 information could predict the progression to AD in MCI cases clinically followed up for 24 months, reaching accuracies higher than 80%.…”

“…Using different classification approaches, several studies demonstrated that brain structural MRI (particularly the medial temporal lobe), amyloid and FDG-PET, and genetic features (in particular the APOEε4 allele) better than other imaging and genetic features predicted the distinction between AD and healthy controls and the conversion from MCI to AD. 35 , 36 , 37 , 96 …”

Classification of Mild Cognitive Impairment With Multimodal Data Using Both Labeled and Unlabeled Samples

“…Single nucleotide polymorphism (SNP) can be used as features for classification. Ning et al reported the ability of AD risk loci in AD/NC classification [24]. Rodrí guez et al [25] selected 8 known AD-related loci and found there was no good discrimination for the classification of MCI.…”