Retinal Vessel Segmentation using UNet

In this proposed Paper a novel, simple lightweight structured Deep Learning method to solve the problem of Retinal Vessel Segmentation. Such kind of problem in the retinal vessel segmentation is very common in the field of medical image segmentation moreover which has present in the human eyes a com...

Full description

Saved in:

Bibliographic Details
Published in:	2023 2nd International Conference on Vision Towards Emerging Trends in Communication and Networking Technologies (ViTECoN) pp. 1 - 5
Main Authors:	Priyadarsini, M.Jasmin Pemeena, S, Sowmiya, Jabeena, A, Rajini, G.K., Subramanian, Ganesan, Clinton S, Ernest Bravin
Format:	Conference Proceeding
Language:	English
Published:	IEEE 05-05-2023
Subjects:	CNN Convolution Neural Network Deep learning Image resolution Image segmentation Pathology Retinal Vessel Segmentation Retinal vessels Solid modeling
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	In this proposed Paper a novel, simple lightweight structured Deep Learning method to solve the problem of Retinal Vessel Segmentation. Such kind of problem in the retinal vessel segmentation is very common in the field of medical image segmentation moreover which has present in the human eyes a computer-aided diagnosis (CAD) based solution to allow easier, quicker, and more effective diagnosis of pathological diseases. This problem will be solved through the analysis of the morphological properties of the blood vessels present in the human retina. There have been many approaches using Deep Learning to solve the problem of retinal vessel segmentation in the earlier few years and the performance of these models have kept increasing consistently. Our proposed model is a multiresolution pathway U-Net which is a modified U-Net with intermediate nodes which perform multi-resolution aggregation of features. Our design was find to achieve comparable results in the comparison of the state in the DRIVE and STARE datasets.
DOI:	10.1109/ViTECoN58111.2023.10157589