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.
View on arXiv@article{le2025_2504.09876,
title={ HDC: Hierarchical Distillation for Multi-level Noisy Consistency in Semi-Supervised Fetal Ultrasound Segmentation },
author={ Tran Quoc Khanh Le and Nguyen Lan Vi Vu and Ha-Hieu Pham and Xuan-Loc Huynh and Tien-Huy Nguyen and Minh Huu Nhat Le and Quan Nguyen and Hien D. Nguyen },
journal={arXiv preprint arXiv:2504.09876},
year={ 2025 }
}