Artificial intelligence techniques for automatic screening of amblyogenic factors

Area of Science:

• Ophthalmology
• Computer Science
• Artificial Intelligence

Background:

• Amblyogenic factors in children require early detection for effective treatment.
• Traditional screening methods can be resource-intensive and time-consuming.
• Automated systems offer potential for improved efficiency and accessibility.

Purpose of the Study:

• To develop a low-cost automated video system for screening amblyogenic factors in children aged 6 months to 6 years.
• To leverage computer vision and artificial intelligence for automated analysis of eye images.
• To evaluate the performance of AI algorithms in identifying vision abnormalities.

Main Methods:

• Automated capture of digital video frames and pupil images.
• Application of computer vision and artificial intelligence (AI) for image analysis.
• Evaluation of AI systems using a tenfold testing method, comparing to specialist examination.

Main Results:

• The decision tree learning approach achieved 77% accuracy in identifying amblyogenic factors.
• The system correctly identified 82% of strabismic individuals and 90% of high refractive errors/anisometropia.
• Moderate refractive errors were less accurately identified by the automated system.

Conclusions:

• The automated video system provides acceptable cost benefits for detecting amblyogenic factors in young children.
• Further research is ongoing to improve the accuracy of the automated analysis.
• The system shows promise for widespread use in pediatric vision screening.