Related Experiment Video
Updated: Oct 7, 2025
05:33
Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
Published on: July 11, 2025
353
Automated histopathological evaluation of pterygium using artificial intelligence
Jong Hoon Kim1, Young Jae Kim1, Yeon Jeong Lee2
1Department of Biomedical Engineering, Gachon University of Medicine and Science, Incheon, Korea (the Republic of).
The British Journal of Ophthalmology
|January 12, 2022
Summary
A new artificial intelligence (AI) automated method accurately grades pterygium histopathology images. This AI tool shows promise as a reliable approach for quantitative analysis in pterygium evaluation.
Area of Science:
- Ophthalmology
- Digital Pathology
- Artificial Intelligence
Background:
- Pterygium diagnosis relies on histopathological evaluation.
- Accurate grading of pterygium histopathology is crucial for treatment and prognosis.
- Current manual grading can be subjective and time-consuming.
Purpose of the Study:
- To develop and evaluate an automated artificial intelligence (AI) method for grading pterygium histopathological images.
- To assess the efficacy and reliability of AI in quantitative histopathological analysis of pterygium.
Main Methods:
- Developed in-house software utilizing AI for automated histopathological image grading.
- Trained four-grade classification models using 400 images from 40 patients, with manual grading as ground truth.
- Employed region of interest segmentation and combined expectation-maximisation with k-nearest neighbours for grade classification, analyzing 55 radiomic features.
Main Results:
- The bagging tree classifier demonstrated the best performance with 75.9% true positive rate (TPR) and 75.8% positive predictive value (PPV) in internal validation.
- External validation confirmed the method's reproducibility, achieving an average of 81.3% TPR and 82.0% PPV across four classification grades.
Conclusions:
- The developed automated AI method offers a reliable approach for the quantitative grading of pterygium histopathological images.
- This AI-driven tool may enhance the objectivity and efficiency of histopathological evaluations in pterygium research and clinical practice.
