PA-Seg: Learning From Point Annotations for 3D Medical Image Segmentation Using Contextual Regularization and Cross Knowledge Distillation
IEEE Transactions on Medical Imaging · 2023-08 · 42 (8) · 2235-2246
DOI: 10.1109/tmi.2023.3245068 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Medical Imaging, 2023-08. Verfasst von Shuwei Zhai und 5 weiteren Autor:innen. Hauptthemen laut MEDLINE: Brain Neoplasms. Die Arbeit wurde bisher 40-mal zitiert (Crossref) und nennt 58 Literaturangaben. Gefördert von: National Natural Science Foundation of China, National Key Research and Development Program of China, 1.3.5 Project for disciplines of excellence, West China Hospital, Sichuan University.
Abstract (Originalsprache)
The success of Convolutional Neural Networks (CNNs) in 3D medical image segmentation relies on massive fully annotated 3D volumes for training that are time-consuming and labor-intensive to acquire. In this paper, we propose to annotate a segmentation target with only seven points in 3D medical images, and design a two-stage weakly supervised learning framework PA-Seg. In the first stage, we employ geodesic distance transform to expand the seed points to provide more supervision signal. To further deal with unannotated image regions during training, we propose two contextual regularization strategies, i.e., multi-view Conditional Random Field (mCRF) loss and Variance Minimization (VM) loss, where the first one encourages pixels with similar features to have consistent labels, and the second one minimizes the intensity variance for the segmented foreground and background, respectively. In the second stage, we use predictions obtained by the model pre-trained in the first stage as pseudo labels. To overcome noises in the pseudo labels, we introduce a Self and Cross Monitoring (SCM) strategy, which combines self-training with Cross Knowledge Distillation (CKD) between a primary model and an auxiliary model that learn from soft labels generated by each other. Experiments on public datasets for Vestibular Schwannoma (VS) segmentation and Brain Tumor Segmentation (BraTS) demonstrated that our model trained in the first stage outperformed existing state-of-the-art weakly supervised approaches by a large margin, and after using SCM for additional training, the model's performance was close to its fully supervised counterpart on the BraTS dataset.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Förderung
- National Natural Science Foundation of China
- National Key Research and Development Program of China
- 1.3.5 Project for disciplines of excellence, West China Hospital, Sichuan University
- Fundamental Research Funds for the Central Universities
Zitieren
APA
Zhai, S., Wang, G., Luo, X., Yue, Q., Li, K., & Zhang, S. (2023). PA-Seg: Learning From Point Annotations for 3D Medical Image Segmentation Using Contextual Regularization and Cross Knowledge Distillation. IEEE Transactions on Medical Imaging, 42(8), 2235-2246. https://doi.org/10.1109/tmi.2023.3245068
Vancouver
Zhai S, Wang G, Luo X, Yue Q, Li K, Zhang S. PA-Seg: Learning From Point Annotations for 3D Medical Image Segmentation Using Contextual Regularization and Cross Knowledge Distillation. IEEE Transactions on Medical Imaging. 2023;42(8):2235-2246. doi:10.1109/tmi.2023.3245068
BibTeX
@article{Zhai2023,
author = {Zhai, Shuwei and Wang, Guotai and Luo, Xiangde and Yue, Qiang and Li, Kang and Zhang, Shaoting},
title = {PA-Seg: Learning From Point Annotations for 3D Medical Image Segmentation Using Contextual Regularization and Cross Knowledge Distillation},
journal = {IEEE Transactions on Medical Imaging},
year = {2023},
volume = {42},
number = {8},
pages = {2235--2246},
doi = {10.1109/tmi.2023.3245068}
}
EUVIMED