Skip to content

A deep learning project for accurate retinal vein segmentation using U-Net model. This repository includes detailed steps for data preprocessing, model training, and evaluation on the DRIVE dataset.

Notifications You must be signed in to change notification settings

numaanfarooq/RetinaVeinSegmentation-U-Net

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 

Repository files navigation

RetinaVeinSegmentation-U-Net

A deep learning project for accurate retinal vein segmentation using U-Net model. This repository includes detailed steps for data preprocessing, model training, and evaluation on the DRIVE dataset.

01_test_0

Overview

This project focuses on segmenting retinal veins from medical images using a UNET-based deep learning model implemented in PyTorch. The goal is to assist in early diagnosis and monitoring of retinal diseases, which could potentially help healthcare professionals in detecting conditions like diabetic retinopathy, glaucoma, and hypertension.

The model is trained using the **DRIVE (Digital Retinal Images for Vessel Extraction) dataset consisting of retina vessel images along with their annotated binary mask. The dataset is consist of 20 images for both training and testing set.

#UNET architecture

U-Net is a U-shaped encoder-decoder network consisting of four encoder blocks and four decoder blocks connected via a bridge, where the encoder network halves spatial dimensions and doubles feature channels, and the decoder network doubles spatial dimensions and halves feature channels, with skip connections between encoder and decoder blocks, making it a symmetric architecture efficient for image segmentation tasks, particularly in medical imaging and computer vision applications.

##Dataset

Name: DRIVE (Digital Retinal Images for Vessel Extraction) Structure:

About

A deep learning project for accurate retinal vein segmentation using U-Net model. This repository includes detailed steps for data preprocessing, model training, and evaluation on the DRIVE dataset.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages