Related Experiment Video
Updated: Jun 26, 2026
08:39
Assessing Cortical Cerebral Microinfarcts on High Resolution MR Images
Published on: November 20, 2015
Image based diagnosis of cortical cataract
Huiqi Li1, Liling Ko, Joo Hwee Lim
1Institute for Infocomm Research, A*STAR, 21 Heng Mui Keng Terrace, Singapore. huiqili@i2r.a-star.edu.sg
Summary
This study introduces an automated method for detecting and grading cortical cataract using retro-illumination images. The approach accurately identifies opacities, showing promise for clinical diagnosis.
Area of Science:
- Ophthalmology
- Medical Imaging
- Computer Vision
Background:
- Cataract detection and grading are crucial for diagnosing vision impairment.
- Current methods may be subjective or time-consuming.
- Automated analysis of retro-illumination images offers potential for objective assessment.
Purpose of the Study:
- To develop an automatic approach for detecting cortical opacities.
- To grade the severity of cortical cataract using retro-illumination images.
- To evaluate the accuracy and reliability of the proposed automated method.
Main Methods:
- Utilized spoke-like features to distinguish cortical opacities from other types.
- Developed algorithms for region of interest (ROI) detection and opacity area measurement.
- Tested algorithms on a dataset of retro-illumination images from a community study.
Main Results:
- Achieved a 98.2% success rate in ROI detection for 611 images.
- Demonstrated a mean error of 3.15% in opacity area detection compared to human graders for 466 images.
- Obtained 85.6% accuracy in exact cortical cataract grading.
Conclusions:
- The proposed automatic approach shows high accuracy in detecting and grading cortical opacities.
- The method is a promising tool for supporting clinical diagnosis of cortical cataract.
- Automated analysis of retro-illumination images can improve efficiency and objectivity in ophthalmology.
