Establishment of a Corneal Ulcer Prognostic Model Based on Machine Learning
Friday, 12 July 2024, 16:50
Corneal Ulcer Prognostic Model Using Machine Learning
Introduction
Corneal infections are a major concern globally, causing corneal blindness. Analyzing the risk of perforation and impairment in ulcer patients is crucial for early treatment.
Methodology
- Dataset: Includes 4973 slit lamp images and clinical data of 240 patients.
- Modeling: Developed a deep learning model for accurate segmentation and classification of corneal lesions.
- Machine Learning Algorithms: Implemented XGBoost and LightGBM for prognostic model creation.
Results
- High accuracy rates were achieved in lesion segmentation and classification.
- XGBoost Model: Excelled in predicting patient prognoses within 1 and 3 months.
- ROC curve analysis showed promising performance of the model.
This article was prepared using information from open sources in accordance with the principles of Ethical Policy. The editorial team is not responsible for absolute accuracy, as it relies on data from the sources referenced.