TY - GEN
T1 - Memory Network-Based Quality Normalization of Magnetic Resonance Images for Brain Segmentation
AU - Su, Yang
AU - Wei, Jie
AU - Ma, Benteng
AU - Xia, Yong
AU - Zhang, Yanning
N1 - Publisher Copyright:
© 2019, Springer Nature Switzerland AG.
PY - 2019
Y1 - 2019
N2 - Medical images of the same modality but acquired at different centers, with different machines, using different protocols, and by different operators may have highly variable quality. Due to its limited generalization ability, a deep learning model usually cannot achieve the same performance on another database as it has done on the database with which it was trained. In this paper, we use the segmentation of brain magnetic resonance (MR) images as a case study to investigate the possibility of improving the performance of medical image analysis via normalizing the quality of images. Specifically, we propose a memory network (MemNet)-based algorithm to normalize the quality of brain MR images and adopt the widely used 3D U-Net to segment the images before and after quality normalization. We evaluated the proposed algorithm on the benchmark IBSR V2.0 database. Our results suggest that the MemNet-based algorithm can not only normalize and improve the quality of brain MR images, but also enable the same 3D U-Net to produce substantially more accurate segmentation of major brain tissues.
AB - Medical images of the same modality but acquired at different centers, with different machines, using different protocols, and by different operators may have highly variable quality. Due to its limited generalization ability, a deep learning model usually cannot achieve the same performance on another database as it has done on the database with which it was trained. In this paper, we use the segmentation of brain magnetic resonance (MR) images as a case study to investigate the possibility of improving the performance of medical image analysis via normalizing the quality of images. Specifically, we propose a memory network (MemNet)-based algorithm to normalize the quality of brain MR images and adopt the widely used 3D U-Net to segment the images before and after quality normalization. We evaluated the proposed algorithm on the benchmark IBSR V2.0 database. Our results suggest that the MemNet-based algorithm can not only normalize and improve the quality of brain MR images, but also enable the same 3D U-Net to produce substantially more accurate segmentation of major brain tissues.
KW - Brain tissue segmentation
KW - Deep learning
KW - Magnetic resonance image
KW - Medical image quality normalization
UR - http://www.scopus.com/inward/record.url?scp=85077126384&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-36189-1_5
DO - 10.1007/978-3-030-36189-1_5
M3 - 会议稿件
AN - SCOPUS:85077126384
SN - 9783030361884
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 58
EP - 67
BT - Intelligence Science and Big Data Engineering. Visual Data Engineering - 9th International Conference, IScIDE 2019, Proceedings, Part 1
A2 - Cui, Zhen
A2 - Pan, Jinshan
A2 - Zhang, Shanshan
A2 - Xiao, Liang
A2 - Yang, Jian
PB - Springer
T2 - 9th International Conference on Intelligence Science and Big Data Engineering, IScIDE 2019
Y2 - 17 October 2019 through 20 October 2019
ER -