Retinal Image Segmentation Assisted Diagnosis System based on Django

To address the problem of limited application of eye diagnosis in the current market, we developed a web-based system using the Django framework to assist in retinal image segmentation for diagnosis. Doctors can use the segmented images of retinal vessels and optic disc and cup to assist in diagnosing patients. The front-end is mainly built using HTML, CSS, Javascript and other languages to construct clear and concise pages, and display the results of user interaction and retinal image segmentation. For the segmentation of retinal vessels and optic disc and cup, we used improved UNet model and posal model in the backend, respectively. The system has been tested and demonstrated good segmentation results using the retinal image segmentation model, and can be applied to medical fields to provide assistance.