Detection of retinal diseases from OCT images using a VGG16 and transfer learning

Abstract

The field of ophthalmology relies on digital image processing techniques, such as Optical Coherence Tomography (OCT), for diagnosing retinal diseases. However, manual interpretation of OCT images is time-consuming and prone to human error. This study developed a deep learning-based model to assist in diagnosing retinal pathologies from OCT images. A modified VGG16 architecture was trained on a dataset of OCT images to classify four retinal conditions: choroidal neovascularization, diabetic macular edema, drusen, and normal. Rigorous evaluation, including cross-validation and independent testing, demonstrated the model’s ability to achieve an accuracy of 95.19% and high precision (95.29%), recall (95.19%), and F1-score (95.20%). In addition, gradient-weighted class activation mapping was employed to visualize network decisions, and a graphical user interface was developed to enhance user interaction with the diagnostic tool. The developed model can potentially improve the early detection and diagnosis of retinal diseases, ultimately enhancing patient care.