TY - GEN
T1 - A 3D coarse-to-fine framework for volumetric medical image segmentation
AU - Zhu, Zhuotun
AU - Xia, Yingda
AU - Shen, Wei
AU - Fishman, Elliot
AU - Yuille, Alan
N1 - Funding Information:
Though this work mainly focuses on segmentation for the pancreas, we can naturally apply the proposed idea to other small organs, e.g., spleen, duodenum and gallbladder, etc, In the future, we will target on error causes that lead to inaccurate segmentation to make our framework more stable, and extend our 3D coarse-to-fine framework to cyst segmentation which can cause cancerous tumors, and the very important tumor segmentation [36] task. Acknowledgements This work was supported by the Lust-garten Foundation for Pancreatic Cancer Research, and National Natural Science Foundation of China No. 61672336. We appreciate enormous help from Seyoun Park, Lingxi Xie, Yuyin Zhou, Yan Wang, Fengze Liu, and valuable discussions from Qing Liu, Yan Zheng, Chenxi Liu, Zhe Ren.
Publisher Copyright:
© 2018 IEEE.
PY - 2018/10/12
Y1 - 2018/10/12
N2 - In this paper, we adopt 3D Convolutional Neural Networks to segment volumetric medical images. Although deep neural networks have been proven to be very effective on many 2D vision tasks, it is still challenging to apply them to 3D tasks due to the limited amount of annotated 3D data and limited computational resources. We propose a novel 3D-based coarse-to-fine framework to effectively and efficiently tackle these challenges. The proposed 3D-based framework outperforms the 2D counterpart to a large margin since it can leverage the rich spatial information along all three axes. We conduct experiments on two datasets which include healthy and pathological pancreases respectively, and achieve the current state-of-the-art in terms of Dice-Sorensen Coefficient (DSC). On the NIH pancreas segmentation dataset, we outperform the previous best by an average of over 2%, and the worst case is improved by 7% to reach almost 70%, which indicates the reliability of our framework in clinical applications.
AB - In this paper, we adopt 3D Convolutional Neural Networks to segment volumetric medical images. Although deep neural networks have been proven to be very effective on many 2D vision tasks, it is still challenging to apply them to 3D tasks due to the limited amount of annotated 3D data and limited computational resources. We propose a novel 3D-based coarse-to-fine framework to effectively and efficiently tackle these challenges. The proposed 3D-based framework outperforms the 2D counterpart to a large margin since it can leverage the rich spatial information along all three axes. We conduct experiments on two datasets which include healthy and pathological pancreases respectively, and achieve the current state-of-the-art in terms of Dice-Sorensen Coefficient (DSC). On the NIH pancreas segmentation dataset, we outperform the previous best by an average of over 2%, and the worst case is improved by 7% to reach almost 70%, which indicates the reliability of our framework in clinical applications.
KW - 3D CNNs
KW - Pancreas Segmentation
UR - http://www.scopus.com/inward/record.url?scp=85056766181&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85056766181&partnerID=8YFLogxK
U2 - 10.1109/3DV.2018.00083
DO - 10.1109/3DV.2018.00083
M3 - Conference contribution
AN - SCOPUS:85056766181
T3 - Proceedings - 2018 International Conference on 3D Vision, 3DV 2018
SP - 682
EP - 690
BT - Proceedings - 2018 International Conference on 3D Vision, 3DV 2018
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 6th International Conference on 3D Vision, 3DV 2018
Y2 - 5 September 2018 through 8 September 2018
ER -