Related Experiment Video
Updated: May 15, 2026
09:17
Using Retinal Imaging to Study Dementia
Published on: November 6, 2017
Automatic detection of optic disc based on PCA and mathematical morphology
Sandra Morales1, Valery Naranjo, Us Angulo
1Instituto Interuniversitario de Investigación en Bioingeniería y Tecnología Orientada al Ser Humano, Universitat Politècnica de València, 46022 Valencia, Spain. smorales@labhuman.i3bh.es
IEEE Transactions on Medical Imaging
|January 15, 2013
Summary
This study introduces an automated algorithm for optic disc segmentation in fundus images, enhancing early disease detection. The method combines mathematical morphology and principal component analysis (PCA) for robust and reliable results.
Area of Science:
- Ophthalmology
- Medical Imaging
- Computer Vision
Background:
- Accurate optic disc segmentation is crucial for diagnosing various retinal pathologies.
- Automating this process can reduce the need for specialist intervention and improve efficiency.
Purpose of the Study:
- To develop and validate an automated algorithm for optic disc segmentation from fundus images.
- To improve the early detection of eye diseases through enhanced image analysis.
Main Methods:
- The algorithm utilizes mathematical morphology operations, including generalized distance function (GDF), watershed transformation variants, and geodesic transformations.
- Principal component analysis (PCA) is employed to obtain an optimal grayscale representation from RGB fundus images, serving as input for segmentation.
Main Results:
- The algorithm achieved high performance metrics, with average Jaccard's and Dice's coefficients of 0.8200 and 0.8932, respectively.
- An overall accuracy of 0.9947 was recorded, along with a true positive fraction of 0.9275 and a false positive fraction of 0.0036.
- Validation on five public databases demonstrated robustness across varied datasets.
Conclusions:
- The proposed method offers a robust and reliable tool for automatic optic disc segmentation.
- This automated approach has the potential to significantly aid in the early diagnosis of ocular conditions.
- The algorithm demonstrates superior performance compared to existing state-of-the-art methods.
