Semi-supervised Cervical Segmentation on Ultrasound by A Dual Framework for Neural Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Fangyijie, Curran, Kathleen M., Silvestre, Guénolé
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916659062636544
author Wang, Fangyijie
Curran, Kathleen M.
Silvestre, Guénolé
author_facet Wang, Fangyijie
Curran, Kathleen M.
Silvestre, Guénolé
contents Accurate segmentation of ultrasound (US) images of the cervical muscles is crucial for precision healthcare. The demand for automatic computer-assisted methods is high. However, the scarcity of labeled data hinders the development of these methods. Advanced semi-supervised learning approaches have displayed promise in overcoming this challenge by utilizing labeled and unlabeled data. This study introduces a novel semi-supervised learning (SSL) framework that integrates dual neural networks. This SSL framework utilizes both networks to generate pseudo-labels and cross-supervise each other at the pixel level. Additionally, a self-supervised contrastive learning strategy is introduced, which employs a pair of deep representations to enhance feature learning capabilities, particularly on unlabeled data. Our framework demonstrates competitive performance in cervical segmentation tasks. Our codes are publicly available on https://github.com/13204942/SSL\_Cervical\_Segmentation.
format Preprint
id arxiv_https___arxiv_org_abs_2503_17057
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Semi-supervised Cervical Segmentation on Ultrasound by A Dual Framework for Neural Networks
Wang, Fangyijie
Curran, Kathleen M.
Silvestre, Guénolé
Image and Video Processing
Computer Vision and Pattern Recognition
Accurate segmentation of ultrasound (US) images of the cervical muscles is crucial for precision healthcare. The demand for automatic computer-assisted methods is high. However, the scarcity of labeled data hinders the development of these methods. Advanced semi-supervised learning approaches have displayed promise in overcoming this challenge by utilizing labeled and unlabeled data. This study introduces a novel semi-supervised learning (SSL) framework that integrates dual neural networks. This SSL framework utilizes both networks to generate pseudo-labels and cross-supervise each other at the pixel level. Additionally, a self-supervised contrastive learning strategy is introduced, which employs a pair of deep representations to enhance feature learning capabilities, particularly on unlabeled data. Our framework demonstrates competitive performance in cervical segmentation tasks. Our codes are publicly available on https://github.com/13204942/SSL\_Cervical\_Segmentation.
title Semi-supervised Cervical Segmentation on Ultrasound by A Dual Framework for Neural Networks
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.17057