A Fuzzy MCDM Framework Using AHP–COPRAS to Prioritize Design Criteria and Performance Challenges in Retinal Image Classification

The accurate classification of retinal images is critical for the early detection and treatment of ophthalmic conditions such as diabetic retinopathy and age-related macular degeneration. However, effective classification remains challenged by several factors, including image quality variability, complex feature extraction, and computational limitations. In this paper, we propose a Fuzzy Hybrid Multi-Criteria Decision-Making approach to prioritize these challenges systematically. By using Fuzzy AHP for criteria weighting and Fuzzy COPRAS for ranking, we provide a robust framework to address the key issues hindering classification performance. Expert surveys were conducted to assess multiple criteria, including impact on classification accuracy, computational cost, and generalizability. The results highlight the most critical challenges, providing insights for future improvements in retinal image classification models and systems. This study serves as a roadmap for researchers and practitioners looking to enhance retinal image analysis in clinical settings.