Autism Detection in Children by Combined Use of Gaze Preference and the M-CHAT-R in a Resource-Scarce Setting

Area of Science:

• Developmental Psychology
• Clinical Psychology
• Biomedical Engineering

Background:

• Autism spectrum disorder (ASD) diagnosis in resource-limited settings (RLS) often occurs after age four.
• Timely diagnosis is critical for effective early intervention and improved outcomes.
• Current screening tools may have limitations in accessibility and accuracy in RLS.

Purpose of the Study:

• To evaluate the effectiveness of a novel screening tool, GP-MCHAT-R, for early identification of ASD in young children.
• To compare the diagnostic accuracy of GP-MCHAT-R with existing methods like M-CHAT-R and gaze preference alone.
• To assess the feasibility of using eye tracking for ASD screening in RLS.

Main Methods:

• The study involved 73 typically developing (TD) children and 28 children with ASD, aged 36-99 months.
• A combined approach, GP-MCHAT-R, was evaluated, integrating the initial 15 seconds of gaze preference (GP) video coding with M-CHAT-R results.
• Eye tracking data from the initial 15 seconds of video was analyzed for its discriminative ability.

Main Results:

• The GP-MCHAT-R demonstrated high diagnostic accuracy with an AUC of 0.89 (95% CI: 0.82-0.95).
• GP-MCHAT-R significantly outperformed M-CHAT-R alone (AUC = 0.78) and gaze preference alone (AUC = 0.76).
• The initial 15 seconds of eye tracking proved as effective as the full video for discrimination.

Conclusions:

• The GP-MCHAT-R tool shows significant potential for enabling early ASD screening in resource-limited settings.
• Combining gaze preference analysis with M-CHAT-R enhances diagnostic accuracy for ASD.
• This approach may facilitate earlier identification and intervention for children with ASD in underserved regions.