Towards quantifying psychiatric diagnosis using machine learning algorithms and big fMRI data

Abstract

Abstract Attention Deficit Hyperactivity Disorder (ADHD) is one of the most common brain disorders among children and is very difficult to diagnose using current methods. Similarly other mental disorders are subject to the same systematic errors with sufficient evidence of diagnostic errors as well as over-prescribing of drugs due to misdiagnosis. For most mental health disorders there is no quantitative method that will inform the presence or absence of a given mental disorder. We argue that definitive and quantitative diagnostic tests are necessary for ADHD and other mental disorders. To this end, big data Functional Magnetic Resonance Imaging (fMRI) and machine learning algorithms can be instrumental in changing the way psychiatric disorders are diagnosed and treated. We briefly discuss our recent research efforts and future directions for a quantitative gold standard tests for psychiatric diagnosis.

Keywords

• Machine learning and data mining
• ADHD
• fMRI
• Big data
• High performance computing