02/18/2025
Amna Zaheer, a volunteer at Darul Qalb, has published an article titled 'AI as a Support to Diagnose ADHD: An Insight of Unorthodox Approaches: A Scoping Review' in the journal Child Neuropsychology, under the mentorship of Dr. Ahmad Akhtar.
Dr. Ahmad Akhtar advised his mentee to first learn about AI before working on the project, demonstrating his commitment to ensuring that his mentees not only write but also fully understand their subject matter. Following his advice, Amna did a short course on AI then wrote this article under the mentorship of Dr. Akhtar.
This scoping review systematically analyzes 54 studies from the past two decades to evaluate AI's role in diagnosing ADHD. It explores AI applications in brain imaging (MRI), brain activity monitoring (EEG, ECG), behavioral assessments, virtual reality-based testing, and motion-tracking sensors. Machine learning and deep learning models, such as convolutional neural networks (CNNs) and support vector machines (SVMs), show promising diagnostic accuracy ranging from 70% to 95%. Despite these advancements, challenges like algorithmic bias, data inconsistency, and the need for diverse datasets hinder widespread clinical implementation.
