The Application of a Case-Based Reasoning System to Attention-Deficit Hyperactivity Disorder

摘要

Attention-deficit hyperactivity disorder (ADHD) is a prevalent neuropsychiatric disorder. Diagnosis is currently made using a collection of information from multiple sources, many of which are subjective and not always correlated. This highlights the need for more objective tests of ADHD. We address this need with the development of a system for differentiation based on altered control of saccadic eye movements. Our hypothesis is that there is sufficient predictive information contained in eye movement data to allow for the application of a case-based reasoning (CBR) system capable of identifying meaningful groups of ADHD subjects. An iterative refinement methodology was used to incrementally improve a CBR system, resulting in a tool that could distinguish ADHD from control subjects with over 70% accuracy. Moreover, the incorrectly classified ADHD subjects demonstrated a decreased benefit from medication when compared to correctly classified subjects.