Robust Detection of Impaired Resting State Functional Connectivity Networks in Alzheimer's Disease Using Elastic Net Regularized Regression.

Stefan J Teipel,Michel J Grothe,Nicolai Franzmeier,Christian Sorg,Timo Grimmer,Michael Ewers,Martin Walter,Stefan Klöppel,Coraline D Metzger,Martin Dyrba,Viola Borchardt,Eva Meisenzahl

doi:10.3389/fnagi.2016.00318

Abstract

The large number of multicollinear regional features that are provided by resting state (rs) fMRI data requires robust feature selection to uncover consistent networks of functional disconnection in Alzheimer's disease (AD). Here, we compared elastic net regularized and classical stepwise logistic regression in respect to consistency of feature selection and diagnostic accuracy using rs-fMRI data from four centers of the “German resting-state initiative for diagnostic biomarkers” (psymri.org), comprising 53 AD patients and 118 age and sex matched healthy controls. Using all possible pairs of correlations between the time series of rs-fMRI signal from 84 functionally defined brain regions as the initial set of predictor variables, we calculated accuracy of group discrimination and consistency of feature selection with bootstrap cross-validation. Mean areas under the receiver operating characteristic curves as measure of diagnostic accuracy were 0.70 in unregularized and 0.80 in regularized regression. Elastic net regression was insensitive to scanner effects and recovered a consistent network of functional connectivity decline in AD that encompassed parts of the dorsal default mode as well as brain regions involved in attention, executive control, and language processing. Stepwise logistic regression found no consistent network of AD related functional connectivity decline. Regularized regression has high potential to increase diagnostic accuracy and consistency of feature selection from multicollinear functional neuroimaging data in AD. Our findings suggest an extended network of functional alterations in AD, but the diagnostic accuracy of rs-fMRI in this multicenter setting did not reach the benchmark defined for a useful biomarker of AD.

Highlights

Many studies have identified altered functional connectivity networks in resting state examinations of Alzheimer’s disease (AD) patients compared to controls using functional imaging techniques such as FDG-PET or resting state functional MRI
In accordance with our hypothesis, we found more accurate group discrimination between AD dementia cases and controls and more homogeneous feature selection from resting state fMRI data when using regularized logistic regression with an elastic net penalty compared with a classical stepwise logistic regression
These findings support the notion that regularized regression is superior to classical stepwise feature selection for dealing with highly collinear multidimensional functional imaging data

Summary

Introduction

Many studies have identified altered functional connectivity networks in resting state examinations of Alzheimer’s disease (AD) patients compared to controls using functional imaging techniques such as FDG-PET or resting state functional MRI (rsfMRI) (for a recent review see Teipel et al, 2016). To identify the network characteristics of AD-related changes in functional imaging data, most studies have employed stepwise or multiple linear regression approaches (Agosta et al, 2012; Koch et al, 2012; Sheline and Raichle, 2013). Features from rs-fMRI and other functional imaging data are often highly collinear across regions, and linear regression approaches are known to be highly sensitive toward collinearity (James et al, 2013; Section 3.3.6). J. et al, 2015; Schouten et al, 2016; de Vos et al, 2016)

Methods

Results

Conclusion