Integrating Cellular Automata and Deep Learning Architectures for Eye Fundus Disease Segmentation and Classification

View at publisher Summarise with VIRAT
Authors: Rena J. Jessie, R. Arulprakasam
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Diseases of the eye fundus are serious diseases capable of causing serious loss of vision and even permanent blindness in case of delayed identification and treatment. Manual identification is a time-consuming process heavily dependent on the experience of ophthalmologists. The aim of this research is to provide shape to an efficient and accurate eye fundus disease classification and segmentation system by combining Cellular Automata and deep learning techniques. The research combines deep learning models, such as Convolutional Neural Networks (CNNs) and U-Net, with cellular automata to improve the precision of image segmentation. Cellular automata, with the capability of simulating complex spatial patterns and local iterative interaction, improve the process of segmentation through iteratively enhanced boundary accuracy and elimination of noise within predicted masks. A large collection of eye fundus images is employed for the representation of diabetic retinopathy, age-related macular degeneration, and glaucoma diseases. The image in each case is linked to expert-annotated ground-truth segmentations by ophthalmologists. Experimental results confirm the efficacy and efficiency of the proposed system in classifying eye fundus diseases accurately and identifying the diseaseaffected regions in the images. The models perform well with high accuracy, reflecting valuable information regarding the nature and degree of a wide range of fundus diseases. The integration of cellular automata improves the accuracy and smoothness of the segmented regions, resulting in a good insight into the progression of the disease and new paradigms for next-generation diagnostic equipment in ophthalmology.

Fields of Study & Subjects

Retinal Imaging and Analysis Medical Image Segmentation Techniques Advanced Neural Network Applications

Publication details

Journal / source
CRC Press eBooks
Pages
234–241
Publication date
26 March 2026
Language
EN
Document type
Book chapter