The Spike-and-Slab Elastic Net as a Classification Tool in Alzheimer's Disease

• Activity Number: 26 - Imaging Speed Session
• Type: Contributed
• Date/Time: Sunday, August 8, 2021 : 1:30 PM to 3:20 PM
• Sponsor: Section on Statistics in Imaging
• Author(s): Justin Leach* and Lloyd Edwards and Rajesh Kana and Kristina Visscher and Nengjun Yi and Inmaculada Aban
• Companies: University of Alabama at Birmingham and University of Alabama at Birmingham and University of Alabama and University of Alabama at Birmingham and University of Alabama at Birmingham and University of Alabama at Birmingham
• Keywords: Spike-and-slab; Bayesian variable selection; Penalized likelihood; Generalized linear models; Elastic net; Alzheimer's Disease

Abstract

Alzheimer’s disease (AD) is the leading cause of dementia and has received considerable research attention, including using neuroimaging biomarkers to classify patients and/or predict disease progression. Generalized linear models, e.g., logistic regression, can be used as classifiers, but spatial measurements are correlated and often outnumber subjects, in which case penalized and/or Bayesian models, but not classical models, will usually be identifiable. Many useful models, e.g., the elastic net and spike-and-slab lasso, perform automatic variable selection, removing extraneous predictors and reducing model variance, but neither model exploits spatial information in selecting variables. Spatial information can be incorporated into variable selection by placing intrinsic autoregressive priors on the logit probabilities of inclusion within a spike-and-slab elastic net framework. We use cortical thickness and tau-PET images from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) study for binary classification of subjects who are cognitively normal, mildly cognitively impaired, or diagnosed with dementia to demonstrate that this framework can improve classification performance.