TY - GEN
T1 - Efficient Soft-Constrained Clustering for Group-Based Labeling
AU - Bise, Ryoma
AU - Abe, Kentaro
AU - Hayashi, Hideaki
AU - Tanaka, Kiyohito
AU - Uchida, Seiichi
N1 - Funding Information:
Acknowledgments. This work was supported by JSPS KAKENHI Grant JP19K22895 and AMED Grant Number JP18lk1010028.
Publisher Copyright:
© 2019, Springer Nature Switzerland AG.
PY - 2019
Y1 - 2019
N2 - We propose a soft-constrained clustering method for group-based labeling of medical images. Since the idea of group-based labeling is to attach the label to a group of samples at once, we need to have groups (i.e., clusters) with high purity. The proposed method is formulated to achieve high purity even for difficult clustering tasks such as medical image clustering, where image samples of the same class are often very distant in their feature space. In fact, those images degrade the performance of conventional constrained clustering methods. Experiments with an endoscopy image dataset demonstrated that our method outperformed various state-of-the-art methods.
AB - We propose a soft-constrained clustering method for group-based labeling of medical images. Since the idea of group-based labeling is to attach the label to a group of samples at once, we need to have groups (i.e., clusters) with high purity. The proposed method is formulated to achieve high purity even for difficult clustering tasks such as medical image clustering, where image samples of the same class are often very distant in their feature space. In fact, those images degrade the performance of conventional constrained clustering methods. Experiments with an endoscopy image dataset demonstrated that our method outperformed various state-of-the-art methods.
UR - http://www.scopus.com/inward/record.url?scp=85075680763&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85075680763&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-32254-0_47
DO - 10.1007/978-3-030-32254-0_47
M3 - Conference contribution
AN - SCOPUS:85075680763
SN - 9783030322533
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 421
EP - 430
BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2019 - 22nd International Conference, Proceedings
A2 - Shen, Dinggang
A2 - Yap, Pew-Thian
A2 - Liu, Tianming
A2 - Peters, Terry M.
A2 - Khan, Ali
A2 - Staib, Lawrence H.
A2 - Essert, Caroline
A2 - Zhou, Sean
PB - Springer
T2 - 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019
Y2 - 13 October 2019 through 17 October 2019
ER -