Quantum chimp-enanced SqueezeNet for precise diabetic retinopathy classification

Diabetic retinopathy remains a prominent cause of blindness throughout the world, and it is a product of prolonged high blood sugar levels that degrade the retinal blood vessels. This makes early detection of DR critical to stop irreversible blindness from occurring. This paper presents an advanced hybrid approach that utilizes the Quantum Chimp Optimization Algorithm (QCOA) integrated with SqueezeNet to enhance the accuracy and efficiency of DR classification significantly. The novel methodology was divided into two main stages: feature extraction and classification. Firstly, SqueezeNet enables efficient feature extraction from segmented fundus images with minimal computational complexity, ensuring that critical retinal features are captured effectively. The classification process, QCOA optimizes the Support Vector Machine (SVM) parameters and performs feature selection. The hybrid system effectively refines the performance of the SVM, consequently increasing the classification accuracies and optimizing the model's performance. By leveraging QCOA's capability to tune SVM parameters precisely, the proposed approach achieves remarkable classification accuracy, sensitivity, and specificity rates of 99.80%, 99.90%, and 100%, respectively. Fundamentally, the results display the efficiency and practical utility of the proposed approach, and its implementation in real-world clinical settings is likely to significantly improve the rates of early DR reflection and accurate classification for improved patient outcomes.