Diabetic Retinopathy (DR) is a leading cause of vision impairment globally, underscoring the need for accurate and early diagnosis to prevent disease progression. Although fundus imaging serves as a cornerstone of Computer-Aided Diagnosis (CAD) systems, several challenges persist, including lesion scale variability, blurry morphological patterns, inter-class imbalance, limited labeled datasets, and computational inefficiencies. To address these issues, this study proposes an end-to-end diagnostic framework that integrates an enhanced preprocessing pipeline with a novel deep learning architecture, Hierarchical-Inception-Residual-Dense Network (HIRD-Net). The preprocessing stage combines Contrast Limited Adaptive Histogram Equalization (CLAHE) with Dilated Difference of Gaussian (D-DoG) filtering to improve image contrast and highlight fine-grained retinal structures. HIRD-Net features a hierarchical feature fusion stem alongside multiscale, multilevel inception-residual-dense blocks for robust representation learning. The Squeeze-and-Excitation Channel Attention (SECA) is introduced before each Global Average Pooling (GAP) layer to refine the Feature Maps (FMs). It further incorporates four GAP layers for multi-scale semantic aggregation, employs the Hard-Swish activation to enhance gradient flow, and utilizes the Focal Loss function to mitigate class imbalance issues. Experimental results on the IDRiD-APTOS2019, DDR, and EyePACS datasets demonstrate that the proposed framework achieves 93.46%, 82.45% and 79.94% overall classification accuracy using only 4.8 million parameters, highlighting its strong generalization capability and computational efficiency. Furthermore, to ensure transparent predictions, an Explainable AI (XAI) approach known as Gradient-weighted Class Activation Mapping (Grad-CAM) is employed to visualize HIRD-Net’s decision-making process.

Diabetic Retinopathy (DR) has become a leading cause of blindness among diabetic patients. Accurate and timely diagnosis of DR is critical to slowing disease progression. This research proposes a Hybrid Convolutional Neural Network (CNN)-based model, named Diabetic Retinopathy Detection Network (DRD-Net). The proposed DRD-Net designed to enhance diagnostic accuracy by addressing key challenges such as gradient vanishing and lesion scale variability in fundus images. Contrast-Limited Adaptive Histogram Equalization (CLAHE) was used to enhance contrast and highlight lesions in fundus images. To increase the diversity of training samples, the proposed framework employs geometric data augmentation techniques. DRD-Net incorporates the Swish activation function along with densely connected blocks to mitigate gradient vanishing and enhancing feature propagation within the network. Additionally, the model integrates two Inception blocks to facilitate multiscale feature extraction, which is essential for detecting small Regions of Interest (RoI) in fundus images. Experimental results demonstrate that DRD-Net achieves a precision of 84.4%, recall of 84.5%, F1-score of 84.1%, and accuracy of 85.1%, outperforming several state-of-the-art models on the IDRiD dataset. These results highlight DRD-Net’s potential as an effective solution for automated DR diagnosis, contributing to more efficient and accurate DR screening.