Document Type

Article

Publication Date

5-14-2021

Publication Title

Alzheimers Dement (Amst)

Keywords

JMG, JGM

JAX Source

Alzheimers Dement (Amst) 2021 May 14; 13(1):e12140

Volume

13

Issue

1

First Page

12140

Last Page

12140

ISSN

2352-8729

PMID

34027015

DOI

https://doi.org/10.1002/dad2.12140

Grant

AG054354

Abstract

Introduction: Genome-wide association studies (GWAS) for late onset Alzheimer's disease (AD) may miss genetic variants relevant for delineating disease stages when using clinically defined case/control as a phenotype due to its loose definition and heterogeneity.

Methods: We use a transfer learning technique to train three-dimensional convolutional neural network (CNN) models based on structural magnetic resonance imaging (MRI) from the screening stage in the Alzheimer's Disease Neuroimaging Initiative consortium to derive image features that reflect AD progression.

Results: CNN-derived image phenotypes are significantly associated with fasting metabolites related to early lipid metabolic changes as well as insulin resistance and with genetic variants mapped to candidate genes enriched for amyloid beta degradation, tau phosphorylation, calcium ion binding-dependent synaptic loss,

Discussion: This is the first attempt to show that non-invasive MRI biomarkers are linked to AD progression characteristics, reinforcing their use in early AD diagnosis and monitoring.

Comments

We would like to thank Jim Peterson for his C++ script of cropping MRIs, and Nikhil Milind for his help in the AMP-AD gene expression submodule analysis.We are grateful to Kwangsik Nho for his advice on MRI pre-processing, and to Andrew J. Saykin for his insightful analysis suggestions.

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License.

Download

Share

COinS