Related Experiment Video
Updated: May 12, 2026
Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
Published on: July 11, 2025
Design and application of a web-based intelligent ophthalmic image analysis teaching platform
Junyi Chen1, Yang Yang1, Yuanzhuo Song2
1School of Medical Information and Engineering, Southwest Medical University, Luzhou, China.
Background:
The rapid integration of artificial intelligence (AI) into ophthalmology has created new demands for interdisciplinary education that combines clinical image interpretation with intelligent analytical methods. However, structured teaching platforms that support end-to-end ophthalmic image analysis training remain limited. This study aimed to design and evaluate a web-based intelligent ophthalmic image analysis teaching platform and to assess its usability and educational effectiveness.
Methods:
A modular web-based platform was developed to support fundus photography and optical coherence tomography (OCT) analysis workflows, including image preprocessing, model training, performance evaluation, and AI-assisted interpretative guidance. A questionnaire survey (n = 121) was conducted to assess usability using the System Usability Scale (SUS) and competency ratings. In addition, a controlled teaching experiment involving 64 third-year undergraduate students compared learning outcomes between students receiving traditional instruction alone and those using the platform as a supplementary tool.
Results:
The platform successfully implemented a complete instructional workflow integrating clinical image cognition and AI-driven analysis. The median SUS score was 80.0 (70.0, 87.5), significantly exceeding the benchmark value of 68 (p < 0.001). Five self-assessed competency dimensions demonstrated mean scores ranging from 4.31 to 4.53 (maximum score = 5), all exceeding 4.0. In the controlled experiment, the experimental group achieved significantly higher scores in theoretical competence (78.62 ± 12.62 vs. 71.41 ± 13.03), practical competence (92.70 ± 4.18 vs. 86.45 ± 10.57), and comprehensive competence (87.34 ± 7.45 vs. 79.77 ± 10.94) compared with the control group (p < 0.05).
Conclusion:
The proposed web-based intelligent ophthalmic image analysis teaching platform demonstrated high usability and measurable educational benefits. Integrating clinically contextualized AI workflows into ophthalmic education may support competency development in intelligent ophthalmic education and training.
