Integrative analysis of multi-dimensional imaging genomics data for Alzheimer's disease prediction.

Ziming Zhang,Dinggang Shen,Heng Huang

doi:10.3389/fnagi.2014.00260

Abstract

In this paper, we explore the effects of integrating multi-dimensional imaging genomics data for Alzheimer's disease (AD) prediction using machine learning approaches. Precisely, we compare our three recent proposed feature selection methods [i.e., multiple kernel learning (MKL), high-order graph matching based feature selection (HGM-FS), sparse multimodal learning (SMML)] using four widely-used modalities [i.e., magnetic resonance imaging (MRI), positron emission tomography (PET), cerebrospinal fluid (CSF), and genetic modality single-nucleotide polymorphism (SNP)]. This study demonstrates the performance of each method using these modalities individually or integratively, and may be valuable to clinical tests in practice. Our experimental results suggest that for AD prediction, in general, (1) in terms of accuracy, PET is the best modality; (2) Even though the discriminant power of genetic SNP features is weak, adding this modality to other modalities does help improve the classification accuracy; (3) HGM-FS works best among the three feature selection methods; (4) Some of the selected features are shared by all the feature selection methods, which may have high correlation with the disease. Using all the modalities on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, the best accuracies, described as (mean ± standard deviation)%, among the three methods are (76.2 ± 11.3)% for AD vs. MCI, (94.8 ± 7.3)% for AD vs. HC, (76.5 ± 11.1)% for MCI vs. HC, and (71.0 ± 8.4)% for AD vs. MCI vs. HC, respectively.

Highlights

Alzheimer’s disease (AD) is a complex chronically progressive neurodegenerative disease and the most common form of dementia in elderly people worldwide
Different combinations result in selecting different features, and we briefly summarize the details of multiple kernel learning (MKL), high-order graph matching based feature selection (HGM-FS), and sparse multimodal learning (SMML) to show how they developed for feature selection
In order to verify the effect of integration of both imaging and genetic information on the AD prediction accuracy, we perform two classification tasks separately: (1) binary classification, i.e., AD vs. Mild Cognitive Impairment (MCI), AD vs. Healthy Control (HC), and MCI vs. HC; (2) multiclass classification, i.e., AD vs. MCI vs. HC

Summary

Introduction

Alzheimer’s disease (AD) is a complex chronically progressive neurodegenerative disease and the most common form of dementia in elderly people worldwide. As reported in Wimo et al (1997), the prevalence of clinically manifest AD is about 2% at the age of 65 years but increases to about 30% at the age of 85 years. With the increase of human’s life expectancy, more and more elderly people will suffer from AD, and it will cause a heavy socioeconomic burden. There is no treatment to cure or even slow the progression of this disorder currently (Weiner et al, 2012). Huge effort has been put on the better understanding of the disease for more effective treatment (Hardy and Selkoe, 2002; Jack et al, 2010; Weiner et al, 2010, 2012)

Methods

Results

Conclusion