Related Experiment Video
Updated: Aug 16, 2025
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
2.9K
Automated Detection of Corneal Ulcer Using Combination Image Processing and Deep Learning.
Isam Abu Qasmieh1, Hiam Alquran1, Ala'a Zyout1
1Biomedical Systems and Medical Informatics Engineering, Yarmouk University, Irbid 21163, Jordan.
Diagnostics (Basel, Switzerland)
|December 23, 2022
Summary
Automated systems accurately detect corneal ulcers using image processing and deep learning. Deep learning achieved 98.9% accuracy, aiding early diagnosis and preventing vision loss.
Area of Science:
- Ophthalmology
- Medical Imaging
- Computer Vision
Background:
- Corneal ulcers are a common cause of visual impairment.
- Early detection of corneal ulcers is crucial to prevent vision loss.
- Slit-lamp imaging is a primary method for corneal ulcer screening.
Purpose of the Study:
- To develop and compare two automated systems for corneal ulcer localization.
- To evaluate the accuracy and efficiency of image processing and deep learning approaches.
- To provide tools for improved corneal ulcer assessment and treatment.
Main Methods:
- Image processing techniques utilizing the Hough transform.
- Deep learning approaches for automated region localization.
- Validation on the publicly available SUSTech-SYSU database.
Main Results:
- Both systems achieved over 90% accuracy in corneal ulcer detection.
- The deep learning approach reached 98.9% accuracy and 99.3% Dice similarity.
- The image processing method requires no explicit training model optimization.
Conclusions:
- Automated systems show high accuracy for corneal ulcer detection.
- Deep learning offers superior accuracy but requires large datasets.
- Image processing methods offer a viable alternative, especially with limited data, aiding clinical assessment.
