TY - GEN
T1 - Semi-supervised 3D abdominal multi-organ segmentation via deep multi-planar co-training
AU - Zhou, Yuyin
AU - Wang, Yan
AU - Tang, Peng
AU - Bai, Song
AU - Shen, Wei
AU - Fishman, Elliot K.
AU - Yuille, Alan
N1 - Funding Information:
In this paper, we designed a systematic framework DM-PCT for multi-organ segmentation in abdominal CT scans, which is motivated by the traditional co-training strategy to incorporate multi-planar information for the unlabeled data during training. The pseudo-labels are iteratively updated by inferencing comprehensively on multiple configurations of unlabeled data with a multi-planar fusion module. We evaluate our approach on our own large newly collected high-quality dataset. The results show that 1) our method outperforms the fully supervised learning approach by a large margin; 2) it outperforms the single planar method, which further demonstrates the benefit of multi-planar fusion; 3) it can learn better if more unlabeled data provided especially when the scale of labeled data is small. Our framework can be practical in assisting radiologists for clinical applications since the annotation of multiple organs in 3D volumes requires massive labor from radiologists. Our framework is not specific to a certain structure, but shows robust results in multiple complex anatomical structures within efficient computational time. We believe that our algorithm may achieve even higher accuracy if a more powerful backbone network or an advanced fusion algorithm is employed, which we leave as the future work. Acknowledgement. This work was supported by the Lust-garten Foundation for Pancreatic Cancer Research and also supported by NSFC No. 61672336. We thank Prof. Seyoun Park, Dr. Lingxi Xie, Cihang Xie, Zhishuai Zhang, Fengze Liu, Zhuotun Zhu and Yingda Xia for instructive discussions.
Publisher Copyright:
© 2019 IEEE
PY - 2019/3/4
Y1 - 2019/3/4
N2 - In multi-organ segmentation of abdominal CT scans, most existing fully supervised deep learning algorithms require lots of voxel-wise annotations, which are usually difficult, expensive, and slow to obtain. In comparison, massive unlabeled 3D CT volumes are usually easily accessible. Current mainstream works to address semi-supervised biomedical image segmentation problem are mostly graph-based. By contrast, deep network based semi-supervised learning methods have not drawn much attention in this field. In this work, we propose Deep Multi-Planar Co-Training (DMPCT), whose contributions can be divided into two folds: 1) The deep model is learned in a co-training style which can mine consensus information from multiple planes like the sagittal, coronal, and axial planes; 2) Multi-planar fusion is applied to generate more reliable pseudo-labels, which alleviates the errors occurring in the pseudo-labels and thus can help to train better segmentation networks. Experiments are done on our newly collected large dataset with 100 unlabeled cases as well as 210 labeled cases where 16 anatomical structures are manually annotated by four radiologists and confirmed by a senior expert. The results suggest that DMPCT significantly outperforms the fully supervised method by more than 4% especially when only a small set of annotations is used.
AB - In multi-organ segmentation of abdominal CT scans, most existing fully supervised deep learning algorithms require lots of voxel-wise annotations, which are usually difficult, expensive, and slow to obtain. In comparison, massive unlabeled 3D CT volumes are usually easily accessible. Current mainstream works to address semi-supervised biomedical image segmentation problem are mostly graph-based. By contrast, deep network based semi-supervised learning methods have not drawn much attention in this field. In this work, we propose Deep Multi-Planar Co-Training (DMPCT), whose contributions can be divided into two folds: 1) The deep model is learned in a co-training style which can mine consensus information from multiple planes like the sagittal, coronal, and axial planes; 2) Multi-planar fusion is applied to generate more reliable pseudo-labels, which alleviates the errors occurring in the pseudo-labels and thus can help to train better segmentation networks. Experiments are done on our newly collected large dataset with 100 unlabeled cases as well as 210 labeled cases where 16 anatomical structures are manually annotated by four radiologists and confirmed by a senior expert. The results suggest that DMPCT significantly outperforms the fully supervised method by more than 4% especially when only a small set of annotations is used.
UR - http://www.scopus.com/inward/record.url?scp=85063592457&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85063592457&partnerID=8YFLogxK
U2 - 10.1109/WACV.2019.00020
DO - 10.1109/WACV.2019.00020
M3 - Conference contribution
AN - SCOPUS:85063592457
T3 - Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019
SP - 121
EP - 140
BT - Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 19th IEEE Winter Conference on Applications of Computer Vision, WACV 2019
Y2 - 7 January 2019 through 11 January 2019
ER -