Paper Image

Multiview learning for medical image segmentation

Published on:

2 November 2023

Primary Category:

Computer Vision and Pattern Recognition

Paper Authors:

Yanming Guo


Key Details

Proposes multiview learning framework combining augmentation and contrastive learning

Integrates CNN and attention blocks for efficiency and performance

Achieves state-of-the-art results for retinal vessel segmentation

Can be trained in 30 minutes with under 3GB GPU RAM

Has potential to generalize to other medical segmentation tasks

AI generated summary

Multiview learning for medical image segmentation

This paper proposes a novel multiview learning framework for medical image segmentation that addresses key challenges like limited data and high computational cost. It integrates data augmentation and contrastive learning to extract robust features across augmented views of the input image. A hybrid CNN-attention network captures both local and global context. When tested on retinal vessel segmentation, this approach achieved high performance while being efficient to train.

Answers from this paper


No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up