Graph Neural Networks for ADHD Identification in Children
Runxuan Yu1, Xinmeng Zhang1, You Chen1,2
1Vanderbilt University, Nashville, TN, USA.
Studies in Health Technology and Informatics
|August 8, 2025
Summary
Graph Neural Networks, specifically GraphSAGE, show promise for diagnosing Attention-Deficit/Hyperactivity Disorder (ADHD) in children. This AI approach outperformed traditional methods in a large-scale study, enabling earlier intervention for ADHD.
Area of Science:
- Neuroscience
- Artificial Intelligence
- Pediatric Health
Background:
- Attention-Deficit/Hyperactivity Disorder (ADHD) is a prevalent neurodevelopmental disorder.
- Early and accurate diagnosis is crucial for effective intervention in children with ADHD.
Purpose of the Study:
- To compare the diagnostic performance of GraphSAGE, a Graph Neural Network, against XGBoost.
- To evaluate the utility of machine learning models for ADHD diagnosis using real-world data.
Main Methods:
- Utilized data from 42,041 children from the 2022 National Survey of Children's Health.
- Trained and evaluated GraphSAGE (a Graph Neural Network) and XGBoost models for ADHD classification.
Main Results:
- GraphSAGE achieved a higher F1 score (0.8331) compared to XGBoost (0.8241).
- The Graph Neural Network demonstrated superior performance in identifying ADHD in the study cohort.
Conclusions:
- GraphSAGE presents a more effective machine learning approach for ADHD diagnosis than XGBoost.
- These findings support the potential of advanced AI in improving pediatric neurodevelopmental disorder diagnostics.
More Related Videos
13:09Using Brain Activation nir-HEG/Q-EEG and Execution Measures CPTs in a ADHD Assessment Protocol
Published on: April 1, 2018
10.4K
05:48The Adventures of Fundi Intervention Based on the Cognitive and Emotional Processing in Attention Deficit Hyperactive Disorder Patients
Published on: June 12, 2020
5.8K
