TY - JOUR
T1 - ψ-Net
T2 - Stacking Densely Convolutional LSTMs for Sub-Cortical Brain Structure Segmentation
AU - Liu, Lihao
AU - Hu, Xiaowei
AU - Zhu, Lei
AU - Fu, Chi Wing
AU - Qin, Jing
AU - Heng, Pheng Ann
N1 - Funding Information:
Manuscript received November 29, 2019; revised February 3, 2020; accepted February 8, 2020. Date of publication February 24, 2020; date of current version August 31, 2020. This work was supported in part by a grant from the Hong Kong Research Grants Council under Project CUHK 14225616, in part by a grant from the National Natural Science Foundation of China under Project U1813204, in part by a grant from the National Natural Science Foundation of China under Grant 61902275, and in part by the Research Grants Council of the Hong Kong Special Administrative Region (Project no. CUHK 14201717). (Lihao Liu and Xiaowei Hu contributed equally to this work.) (Corresponding author: Lei Zhu.) Lihao Liu, Xiaowei Hu, and Lei Zhu are with the Department of Computer Science and Engineering, The Chinese University of Hong Kong, Hong Kong (e-mail: [email protected]).
Publisher Copyright:
© 1982-2012 IEEE.
PY - 2020/9
Y1 - 2020/9
N2 - Sub-cortical brain structure segmentation is of great importance for diagnosing neuropsychiatric disorders. However, developing an automatic approach to segmenting sub-cortical brain structures remains very challenging due to the ambiguous boundaries, complex anatomical structures, and large variance of shapes. This paper presents a novel deep network architecture, namely \Psi -Net, for sub-cortical brain structure segmentation, aiming at selectively aggregating features and boosting the information propagation in a deep convolutional neural network (CNN). To achieve this, we first formulate a densely convolutional LSTM module (DC-LSTM) to selectively aggregate the convolutional features with the same spatial resolution at the same stage of a CNN. This helps to promote the discriminativeness of features at each CNN stage. Second, we stack multiple DC-LSTMs from the deepest stage to the shallowest stage to progressively enrich low-level feature maps with high-level context. We employ two benchmark datasets on sub-cortical brain structure segmentation, and perform various experiments to evaluate the proposed \Psi -Net. The experimental results show that our network performs favorably against the state-of-the-art methods on both benchmark datasets.
AB - Sub-cortical brain structure segmentation is of great importance for diagnosing neuropsychiatric disorders. However, developing an automatic approach to segmenting sub-cortical brain structures remains very challenging due to the ambiguous boundaries, complex anatomical structures, and large variance of shapes. This paper presents a novel deep network architecture, namely \Psi -Net, for sub-cortical brain structure segmentation, aiming at selectively aggregating features and boosting the information propagation in a deep convolutional neural network (CNN). To achieve this, we first formulate a densely convolutional LSTM module (DC-LSTM) to selectively aggregate the convolutional features with the same spatial resolution at the same stage of a CNN. This helps to promote the discriminativeness of features at each CNN stage. Second, we stack multiple DC-LSTMs from the deepest stage to the shallowest stage to progressively enrich low-level feature maps with high-level context. We employ two benchmark datasets on sub-cortical brain structure segmentation, and perform various experiments to evaluate the proposed \Psi -Net. The experimental results show that our network performs favorably against the state-of-the-art methods on both benchmark datasets.
KW - densely convolutional LSTM
KW - feature integration
KW - Sub-cortical brain structure segmentation
UR - http://www.scopus.com/inward/record.url?scp=85090178245&partnerID=8YFLogxK
U2 - 10.1109/TMI.2020.2975642
DO - 10.1109/TMI.2020.2975642
M3 - Journal article
C2 - 32091996
AN - SCOPUS:85090178245
SN - 0278-0062
VL - 39
SP - 2806
EP - 2817
JO - IEEE Transactions on Medical Imaging
JF - IEEE Transactions on Medical Imaging
IS - 9
M1 - 9007625
ER -