TY - CHAP
T1 - Volumetric Medical Image Segmentation
T2 - A 3D Deep Coarse-to-Fine Framework and Its Adversarial Examples
AU - Li, Yingwei
AU - Zhu, Zhuotun
AU - Zhou, Yuyin
AU - Xia, Yingda
AU - Shen, Wei
AU - Fishman, Elliot K.
AU - Yuille, Alan L.
N1 - Publisher Copyright:
© 2019, Springer Nature Switzerland AG.
PY - 2019
Y1 - 2019
N2 - Although deep neural networks have been a dominant method for many 2D vision tasks, it is still challenging to apply them to 3D tasks, such as medical image segmentation, due to the limited amount of annotated 3D data and limited computational resources. In this chapter, by rethinking the strategy to apply 3D Convolutional Neural Networks to segment medical images, we propose a novel 3D-based coarse-to-fine framework to efficiently tackle these challenges. The proposed 3D-based framework outperforms their 2D counterparts by a large margin since it can leverage the rich spatial information along all three axes. We further analyze the threat of adversarial attacks on the proposed framework and show how to defend against the attack. We conduct experiments on three datasets, the NIH pancreas dataset, the JHMI pancreas dataset and the JHMI pathological cyst dataset, where the first two and the last one contain healthy and pathological pancreases, respectively, and achieve the current state of the art in terms of Dice-Sørensen Coefficient (DSC) on all of them. Especially, on the NIH pancreas 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 - Although deep neural networks have been a dominant method for many 2D vision tasks, it is still challenging to apply them to 3D tasks, such as medical image segmentation, due to the limited amount of annotated 3D data and limited computational resources. In this chapter, by rethinking the strategy to apply 3D Convolutional Neural Networks to segment medical images, we propose a novel 3D-based coarse-to-fine framework to efficiently tackle these challenges. The proposed 3D-based framework outperforms their 2D counterparts by a large margin since it can leverage the rich spatial information along all three axes. We further analyze the threat of adversarial attacks on the proposed framework and show how to defend against the attack. We conduct experiments on three datasets, the NIH pancreas dataset, the JHMI pancreas dataset and the JHMI pathological cyst dataset, where the first two and the last one contain healthy and pathological pancreases, respectively, and achieve the current state of the art in terms of Dice-Sørensen Coefficient (DSC) on all of them. Especially, on the NIH pancreas 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.
UR - http://www.scopus.com/inward/record.url?scp=85073234652&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85073234652&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-13969-8_4
DO - 10.1007/978-3-030-13969-8_4
M3 - Chapter
AN - SCOPUS:85073234652
T3 - Advances in Computer Vision and Pattern Recognition
SP - 69
EP - 91
BT - Advances in Computer Vision and Pattern Recognition
PB - Springer London
ER -