TY - GEN
T1 - Training Multi-organ Segmentation Networks with Sample Selection by Relaxed Upper Confident Bound
AU - Wang, Yan
AU - Zhou, Yuyin
AU - Tang, Peng
AU - Shen, Wei
AU - Fishman, Elliot K.
AU - Yuille, Alan L.
N1 - Publisher Copyright:
© 2018, Springer Nature Switzerland AG.
PY - 2018
Y1 - 2018
N2 - Convolutional neural networks (CNNs), especially fully convolutional networks, have been widely applied to automatic medical image segmentation problems, e.g., multi-organ segmentation. Existing CNN-based segmentation methods mainly focus on looking for increasingly powerful network architectures, but pay less attention to data sampling strategies for training networks more effectively. In this paper, we present a simple but effective sample selection method for training multi-organ segmentation networks. Sample selection exhibits an exploitation-exploration strategy, i.e., exploiting hard samples and exploring less frequently visited samples. Based on the fact that very hard samples might have annotation errors, we propose a new sample selection policy, named Relaxed Upper Confident Bound (RUCB). Compared with other sample selection policies, e.g., Upper Confident Bound (UCB), it exploits a range of hard samples rather than being stuck with a small set of very hard ones, which mitigates the influence of annotation errors during training. We apply this new sample selection policy to training a multi-organ segmentation network on a dataset containing 120 abdominal CT scans and show that it boosts segmentation performance significantly.
AB - Convolutional neural networks (CNNs), especially fully convolutional networks, have been widely applied to automatic medical image segmentation problems, e.g., multi-organ segmentation. Existing CNN-based segmentation methods mainly focus on looking for increasingly powerful network architectures, but pay less attention to data sampling strategies for training networks more effectively. In this paper, we present a simple but effective sample selection method for training multi-organ segmentation networks. Sample selection exhibits an exploitation-exploration strategy, i.e., exploiting hard samples and exploring less frequently visited samples. Based on the fact that very hard samples might have annotation errors, we propose a new sample selection policy, named Relaxed Upper Confident Bound (RUCB). Compared with other sample selection policies, e.g., Upper Confident Bound (UCB), it exploits a range of hard samples rather than being stuck with a small set of very hard ones, which mitigates the influence of annotation errors during training. We apply this new sample selection policy to training a multi-organ segmentation network on a dataset containing 120 abdominal CT scans and show that it boosts segmentation performance significantly.
UR - https://www.scopus.com/pages/publications/85053843728
U2 - 10.1007/978-3-030-00937-3_50
DO - 10.1007/978-3-030-00937-3_50
M3 - 会议稿件
AN - SCOPUS:85053843728
SN - 9783030009366
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 434
EP - 442
BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings
A2 - Frangi, Alejandro F.
A2 - Fichtinger, Gabor
A2 - Schnabel, Julia A.
A2 - Alberola-López, Carlos
A2 - Davatzikos, Christos
PB - Springer Verlag
T2 - 21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018
Y2 - 16 September 2018 through 20 September 2018
ER -