Team Unicorn

ADHD is the most common pediatric psychiatric condition and has an estimated $42 billion annual cost. However, due to the complex nature of ADHD and its overlapping symptomology with other psychiatric diseases, it can be difficult to diagnose–taking providers up to 8 hours to diagnose one patient. Today there is no single conclusive test for identifying ADHD. Multi-modal evaluation methods, including clinical evaluation of the child’s motor skills, verbal skills, and diagnostic questionnaires, are commonly used.

Our team has utilized features derived from these diagnostic questionnaires and performed statistical modeling to generate individual patient risk scores for ADHD. Using machine learning methods, we have begun to identify the features most significantly associated with ADHD with the aim of aiding clinicians in diagnosing ADHD more efficiently during their clinical assessments and enabling a more personalized patient evaluation.