Original title: Artificial Intelligence, Machine Learning, and Deep Learning Approaches for Attention-Deficit/Hyperactivity Disorder Diagnosis: Protocol for a Scoping Review
ADHD diagnosis often relies on interviews and behavior ratings, which can differ between clinicians and settings. This paper is a plan for a review, not a test of a new diagnostic tool. The reviewers will search published studies that use AI to classify ADHD, then compare the data, methods, checks, and explanations those studies report. Their goal is to map how reliable and ready for clinical use this research appears to be.
Imagine sorting a pile of recipes to see what people cook and which ingredients they use. The pile is large, and each recipe describes its ingredients and results in a different way, so comparing them fairly is hard. In this paper, the recipes are published ADHD studies, and the reviewers will sort and compare their data, methods, checks, and explanations to map what the research shows.
ADHD diagnosis can vary across practitioners and settings. The planned review will map how studies use different kinds of data and test AI-based classification, helping show what is known—and what still needs to be established—before these tools can be considered for clinical use.
This is a review protocol, and the screening and analysis are still in progress. It reports no completed synthesis or findings about which AI approaches perform best. Its planned scope is limited to peer-reviewed English-language empirical studies published from January 2019 to April 2026, with more than 100 participants and a non-ADHD control group.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
Artificial Intelligence, Machine Learning, and Deep Learning Approaches for Attention-Deficit/Hyperactivity Disorder Diagnosis: Protocol for a Scoping Review — B C Ibe, Amir Aly, Shaymaa Al-Juboori et al. — JMIR Research Protocols — doi:10.2196/99551
Abstract licence: CC BY
Explanation generated by Synthesis from the abstract — it is not the original paper.