This study explores the explainable detection of three diseases—pathological myopia, glaucoma, and diabetic retinopathy—using retinal fundus images. Both deep learning and feature-based methods are examined for each condition. The deep learning approaches employ transfer learning, while UNet-based models are utilised for feature segmentation. Feature maps are created from segmented features and passed through simple CNNs to detect diseases. Data augmentation techniques are applied across methods to enhance performance, and Grad-CAM/Grad-CAM++ are used to interpret and validate the insights gained from the deep learning models. Our results include accuracy, precision, and recall of 98% for pathological myopia, 97% for glaucoma, and 92% for diabetic retinopathy presence. For diabetic retinopathy grading, Cohen’s kappa scores of 0.83 (linear) and 0.90 (quadratic) were obtained.