Saved in:
Bibliographic Details
Main Authors: Le, Tran Quoc Khanh, Vu, Nguyen Lan Vi, Pham, Ha-Hieu, Huynh, Xuan-Loc, Nguyen, Tien-Huy, Le, Minh Huu Nhat, Nguyen, Quan, Nguyen, Hien D.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.09876
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916693487386624
author Le, Tran Quoc Khanh
Vu, Nguyen Lan Vi
Pham, Ha-Hieu
Huynh, Xuan-Loc
Nguyen, Tien-Huy
Le, Minh Huu Nhat
Nguyen, Quan
Nguyen, Hien D.
author_facet Le, Tran Quoc Khanh
Vu, Nguyen Lan Vi
Pham, Ha-Hieu
Huynh, Xuan-Loc
Nguyen, Tien-Huy
Le, Minh Huu Nhat
Nguyen, Quan
Nguyen, Hien D.
contents Transvaginal ultrasound is a critical imaging modality for evaluating cervical anatomy and detecting physiological changes. However, accurate segmentation of cervical structures remains challenging due to low contrast, shadow artifacts, and indistinct boundaries. While convolutional neural networks (CNNs) have demonstrated efficacy in medical image segmentation, their reliance on large-scale annotated datasets presents a significant limitation in clinical ultrasound imaging. Semi-supervised learning (SSL) offers a potential solution by utilizing unlabeled data, yet existing teacher-student frameworks often encounter confirmation bias and high computational costs. In this paper, a novel semi-supervised segmentation framework, called HDC, is proposed incorporating adaptive consistency learning with a single-teacher architecture. The framework introduces a hierarchical distillation mechanism with two objectives: Correlation Guidance Loss for aligning feature representations and Mutual Information Loss for stabilizing noisy student learning. The proposed approach reduces model complexity while enhancing generalization. Experiments on fetal ultrasound datasets, FUGC and PSFH, demonstrate competitive performance with reduced computational overhead compared to multi-teacher models.
format Preprint
id arxiv_https___arxiv_org_abs_2504_09876
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle HDC: Hierarchical Distillation for Multi-level Noisy Consistency in Semi-Supervised Fetal Ultrasound Segmentation
Le, Tran Quoc Khanh
Vu, Nguyen Lan Vi
Pham, Ha-Hieu
Huynh, Xuan-Loc
Nguyen, Tien-Huy
Le, Minh Huu Nhat
Nguyen, Quan
Nguyen, Hien D.
Computer Vision and Pattern Recognition
Artificial Intelligence
Transvaginal ultrasound is a critical imaging modality for evaluating cervical anatomy and detecting physiological changes. However, accurate segmentation of cervical structures remains challenging due to low contrast, shadow artifacts, and indistinct boundaries. While convolutional neural networks (CNNs) have demonstrated efficacy in medical image segmentation, their reliance on large-scale annotated datasets presents a significant limitation in clinical ultrasound imaging. Semi-supervised learning (SSL) offers a potential solution by utilizing unlabeled data, yet existing teacher-student frameworks often encounter confirmation bias and high computational costs. In this paper, a novel semi-supervised segmentation framework, called HDC, is proposed incorporating adaptive consistency learning with a single-teacher architecture. The framework introduces a hierarchical distillation mechanism with two objectives: Correlation Guidance Loss for aligning feature representations and Mutual Information Loss for stabilizing noisy student learning. The proposed approach reduces model complexity while enhancing generalization. Experiments on fetal ultrasound datasets, FUGC and PSFH, demonstrate competitive performance with reduced computational overhead compared to multi-teacher models.
title HDC: Hierarchical Distillation for Multi-level Noisy Consistency in Semi-Supervised Fetal Ultrasound Segmentation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2504.09876