Related Experiment Video
Updated: Jun 9, 2025
05:32
Comparing Eye-tracking Data of Children with High-functioning ASD, Comorbid ADHD, and of a Control Watching Social Videos
Published on: December 7, 2018
8.9K
Auxiliary Diagnosis of Children With Attention-Deficit/Hyperactivity Disorder Using Eye-Tracking and Digital
Zhongling Liu1, Jinkai Li2, Yuanyuan Zhang1
1Child Health Care Medical Division, Shanghai Children's Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China.
JMIR Mhealth and Uhealth
|October 30, 2024
Summary
Eye-tracking technology offers a new way to diagnose Attention-deficit/hyperactivity disorder (ADHD) in children. This study developed an objective system using eye movement biomarkers and machine learning, achieving high accuracy in identifying ADHD.
Area of Science:
- Neuroscience
- Developmental Psychology
- Biomedical Engineering
Background:
- Attention-deficit/hyperactivity disorder (ADHD) is a prevalent neurodevelopmental disorder in school-aged children.
- Current diagnostic challenges stem from a lack of objective biomarkers, leading to potential misdiagnoses and delayed interventions.
- Eye-tracking technology presents a promising objective method for assessing neuropsychological behaviors in children.
Purpose of the Study:
- To develop a reliable and objective auxiliary diagnostic system for ADHD utilizing eye-tracking technology.
- To establish the system's utility for ADHD screening in school and community settings.
- To identify potential objective biomarkers for the clinical diagnosis of ADHD.
Main Methods:
- A case-control study comparing children with ADHD and typically developing (TD) children.
- Design of an eye-tracking assessment paradigm targeting core ADHD cognitive deficits.
- Extraction and analysis of digital biomarkers (e.g., pupil diameter, gaze regularity) and developmental patterns.
- Application of machine learning (ML) models for ADHD prediction, validated via 5-fold cross-validation.
Main Results:
- 216 participants (94 ADHD, 122 TD) were recruited.
- The ADHD group exhibited significantly poorer performance in saccade tasks and distinct digital biomarker patterns compared to the TD group.
- ML models demonstrated high efficacy in discriminating between groups, achieving an AUC of 0.965 and accuracy of 0.908.
Conclusions:
- Eye-tracking biomarkers effectively differentiate ADHD and TD groups based on eye movement patterns.
- The developed ML model provides an accurate and reliable method for ADHD identification.
- The system can aid in early ADHD screening and offer objective biomarkers for clinical reference.
