Evaluation of Pattern Recognition and Feature Extraction Methods in ADHD Prediction

João Ricardo Sato,Marcelo Queiroz Hoexter,Luis Augusto Rohde,André Fujita

doi:10.3389/fnsys.2012.00068

Abstract

Attention Deficit/Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder, being one of the most prevalent psychiatric disorders in childhood. The neural substrates associated with this condition, both from structural and functional perspectives, are not yet well established. Recent studies have highlighted the relevance of neuroimaging not only to provide a more solid understanding about the disorder but also for possible clinical support. The ADHD-200 Consortium organized the ADHD-200 global competition making publicly available, hundreds of structural magnetic resonance imaging (MRI) and functional MRI (fMRI) datasets of both ADHD patients and typically developing (TD) controls for research use. In the current study, we evaluate the predictive power of a set of three different feature extraction methods and 10 different pattern recognition methods. The features tested were regional homogeneity (ReHo), amplitude of low frequency fluctuations (ALFF), and independent components analysis maps (resting state networks; RSN). Our findings suggest that the combination ALFF+ReHo maps contain relevant information to discriminate ADHD patients from TD controls, but with limited accuracy. All classifiers provided almost the same performance in this case. In addition, the combination ALFF+ReHo+RSN was relevant in combined vs. inattentive ADHD classification, achieving a score accuracy of 67%. In this latter case, the performances of the classifiers were not equivalent and L2-regularized logistic regression (both in primal and dual space) provided the most accurate predictions. The analysis of brain regions containing most discriminative information suggested that in both classifications (ADHD vs. TD controls and combined vs. inattentive), the relevant information is not confined only to a small set of regions but it is spatially distributed across the whole brain.

Highlights

Attention Deficit/Hyperactivity Disorder (ADHD) is a worldwide prevalent disorder (Polanczyk et al, 2007) and is characterized by excessive childhood onset inattention, hyperactivity, and impulsivity (American Psychiatric Association, 1994) that usually persists into adulthood (Mannuzza and Klein, 2000).In recent years, structural magnetic resonance imaging (MRI) and functional MRI techniques have been extensively used in the quantitative analysis of the brain in healthy individuals and patients with psychiatric disorders in an attempt to increase our understanding of human brain structural and functional networks (Bassett and Bullmore, 2009; Biswal et al, 2010)
Boxplots showing the classifiers performance for controls vs. ADHD and combined vs. inattentive ADHD predictions are presented in Figures 2 and 3, respectively
In the case of controls vs. ADHD predictions, it was found that the regional homogeneity (ReHo) and fALFF can separately and jointly (Figure 2) provide some information about the class of the subjects

Summary

Introduction

Attention Deficit/Hyperactivity Disorder (ADHD) is a worldwide prevalent disorder (Polanczyk et al, 2007) and is characterized by excessive childhood onset inattention, hyperactivity, and impulsivity (American Psychiatric Association, 1994) that usually persists into adulthood (Mannuzza and Klein, 2000).In recent years, structural magnetic resonance imaging (MRI) and functional MRI (fMRI) techniques have been extensively used in the quantitative analysis of the brain in healthy individuals and patients with psychiatric disorders in an attempt to increase our understanding of human brain structural and functional networks (Bassett and Bullmore, 2009; Biswal et al, 2010). In comparison to typically developing (TD) individuals, structural neuroimaging studies have shown that ADHD patients present abnormalities in several regions including the frontal, parietal, and occipital lobes, the basal ganglia and the cerebellum (Castellanos et al, 1996; Sowell et al, 2003; Seidman et al, 2006). Sato et al (2011) investigated the potential of structural MRI as a clinical tool to identify individuals with psychopathy. These methods have the potential to translate objective biological information extracted from neuroimaging data to clinical practice (Marquand et al, 2008; Zhu et al, 2008; Dosenbach et al, 2010)

Objectives

Methods

Results

Conclusion