ADHD diagnosis from multiple data sources with batch effects

Emanuele Olivetti,Susanne Greiner,Paolo Avesani

doi:10.3389/fnsys.2012.00070

Abstract

The Attention Deficit Hyperactivity Disorder (ADHD) affects the school-age population and has large social costs. The scientific community is still lacking a pathophysiological model of the disorder and there are no objective biomarkers to support the diagnosis. In 2011 the ADHD-200 Consortium provided a rich, heterogeneous neuroimaging dataset aimed at studying neural correlates of ADHD and to promote the development of systems for automated diagnosis. Concurrently a competition was set up with the goal of addressing the wide range of different types of data for the accurate prediction of the presence of ADHD. Phenotypic information, structural magnetic resonance imaging (MRI) scans and resting state fMRI recordings were provided for nearly 1000 typical and non-typical young individuals. Data were collected by eight different research centers in the consortium. This work is not concerned with the main task of the contest, i.e., achieving a high prediction accuracy on the competition dataset, but we rather address the proper handling of such a heterogeneous dataset when performing classification-based analysis. Our interest lies in the clustered structure of the data causing the so-called batch effects which have strong impact when assessing the performance of classifiers built on the ADHD-200 dataset. We propose a method to eliminate the biases introduced by such batch effects. Its application on the ADHD-200 dataset generates such a significant drop in prediction accuracy that most of the conclusions from a standard analysis had to be revised. In addition we propose to adopt the dissimilarity representation to set up effective representation spaces for the heterogeneous ADHD-200 dataset. Moreover we propose to evaluate the quality of predictions through a recently proposed test of independence in order to cope with the unbalancedness of the dataset.

Highlights

The advance of computational methods for data analysis is opening new perspectives for exploiting of structural and functional magnetic resonance imaging data in the field of neuroscience
Our interest lies in the clustered structure of the data causing the so-called batch effects which have strong impact when assessing the performance of classifiers built on the Attention Deficit Hyperactivity Disorder (ADHD)-200 dataset
We propose the use of the dissimilarity representation (Pekalska et al, 2002; Balcan et al, 2008; Chen et al, 2009) as a mean to construct a common representation space for all the heterogeneous types of data available in the ADHD-200 dataset

Summary

Introduction

The advance of computational methods for data analysis is opening new perspectives for exploiting of structural and functional magnetic resonance imaging (fMRI) data in the field of neuroscience. The statistical learning framework (Hastie et al, 2009), and multivariate pattern analysis, is a prominent example of these methods. In this framework the approaches are data-driven, i.e., they do not require the complete and explicit modeling of the underlying physiology of the brain. For this reason these methods are referred to as model free or nonparametric. The challenge is to achieve accurate prediction on future subjects. Since this approach is data-driven, a successful detection of the disease does not always correspond to a deeper understanding of the pathology. The classifier acts as an information extractor and the basic inference that is derived from an accurate classifier is that the data carry information about the condition of interest

Objectives

Methods

Results

Discussion

Conclusion