Explainable Artificial Intelligence for Predicting Attention Deficit Hyperactivity Disorder in Children and Adults

Attention Deficit Hyperactivity Disorder (ADHD) is a disorder that starts in childhood, sometimes persisting into adulthood. It puts a strain on their social, professional, family, and environmental lives, which can exacerbate disorders such as anxiety, depression, and bipolar disorder. Background/Objectives: This paper aims to predict ADHD in children and adults and explain the main factors impacting this disorder. Methods: We start by introducing the main symptoms and challenges ADHD poses for children and adults such as epilepsy and depression. Then, we present the results of existing research on three ADHD comorbidities: anxiety, depression, and bipolar disorder, and their possible continuity in adulthood with therapeutic implications. After that, we explain the impact of this disorder and its relationship with these comorbidities on the affected patient's health and environment and list proposed treatments. We propose a methodology for predicting this impairment in children and adults by using Machine Learning algorithms (ML), Explainable Artificial Intelligence (XAI), and two datasets, the National Survey for Children's Health (NSCH) (2022) for the children and the ADHD|Mental Health for the adults. Results: Logistic Regression (LR) was the most suitable algorithm for children, with an accuracy of 99%. As for adults, the XGBoost (XGB) was the most performant ML method, with an accuracy of 100%. Conclusions: Lack of sleep and excessive smiling/laughing are among the factors having an impact on ADHD for children, whereas anxiety and depression affect ADHD adults.