Fundus Image Classification for the Early Detection of Issues in the DR for the Effective Disease Diagnosis
Kotte Vinay Kumar1,*, Narasimha Reddy Soora2 and N.C.Santoshkumar3
1,2,3Department of Computer Science and Engineering, Kakatiya Institute of Technology and Science, Warangal, Telangana,506004, India.
*Corresponding Author Name: Kotte Vinay Kumar .Email: kvk.cse@kitsw.ac.in
Journal of Computer Allied Intelligence(JCAI),31 December 2023, 1(1), 27-40
Received: 30 November 2023 Accepted: 22 December 2023 Published: 31 December 2023
Diabetic Retinopathy and fundus images, “DR” could also refer to Digital Retinography, which involves the digital imaging of the retina. Digital retinography allows for the storage, retrieval, and transmission of retinal images for analysis and diagnosis. It plays a crucial role in screening for diabetic retinopathy and other eye diseases. The use of artificial intelligence and machine learning algorithms in the analysis of fundus images has gained traction in recent years. These technologies can assist healthcare professionals in the early detection and monitoring of diabetic retinopathy by analyzing patterns and abnormalities in the retinal images. This paper introduces a novel approach to image analysis, combining Local Semantic Object Feature Extraction (LsOFE) and Gray Level Co-occurrence Matrix (GLCM) metrics for robust object recognition and classification. The LsOFE process captures intricate local features, enhancing the representation of objects in images. The subsequent GLCM-based feature extraction provides detailed insights into texture characteristics. The final classification step, utilizing LsOFE features, demonstrates high precision, recall, and F1-Score for most images, showcasing the effectiveness of the proposed methodology. However, challenges are identified in the classification of specific images, suggesting areas for improvement and refinement. Overall, this research contributes a valuable framework for image processing, with the potential for broader applications in fields such as computer vision and pattern recognition. The results presented underscore the promising performance of the proposed approach, offering a foundation for further exploration and optimization in the realm of image analysis.
Keywords : Fundus Image; diabetic retinopathy; feature extraction; classification, deep learning.
Citation : K.V. Kumar, S.N. Reddy and N.C. Santosh Kumar, “Fundus Image Classification for the Early Detection of Issues in the DR for the Effective Disease Diagnosis”, Journal of Computer Allied Intelligence (JCAI), vol.01, no.01, pp.27-40,2023.