Importance of phenotypic information in ADHD diagnosis

Attention Deficit Hyperactivity Disorder (ADHD) is one of the widely researched neuro-developmental disorder. This paper highlights the importance of phenotypic information in the diagnosis of ADHD, in addition to Magnetic Resonance Image (MRI) based features. In this study, features from amygdala region of the brain is extracted using region of interest based feature extraction technique. These features are used in combination with phenotypic information for effective discrimination of ADHD from typically developing controls. Further, to improve the effectiveness of the detection system, SBMLR based feature selection technique is used to select the most discriminative features. Performance of the proposed ADHD diagnosis is evaluated on a benchmark ADHD-200 consortium database. The performance evaluation on two state-of-the-art classification techniques clearly highlight the importance of phenotypic information in the detection of ADHD.