TY - JOUR
T1 - Inf-Net
T2 - Automatic COVID-19 Lung Infection Segmentation from CT Images
AU - Fan, Deng Ping
AU - Zhou, Tao
AU - Ji, Ge Peng
AU - Zhou, Yi
AU - Chen, Geng
AU - Fu, Huazhu
AU - Shen, Jianbing
AU - Shao, Ling
N1 - Publisher Copyright:
© 1982-2012 IEEE.
PY - 2020/8
Y1 - 2020/8
N2 - Coronavirus Disease 2019 (COVID-19) spread globally in early 2020, causing the world to face an existential health crisis. Automated detection of lung infections from computed tomography (CT) images offers a great potential to augment the traditional healthcare strategy for tackling COVID-19. However, segmenting infected regions from CT slices faces several challenges, including high variation in infection characteristics, and low intensity contrast between infections and normal tissues. Further, collecting a large amount of data is impractical within a short time period, inhibiting the training of a deep model. To address these challenges, a novel COVID-19 Lung Infection Segmentation Deep Network (Inf-Net) is proposed to automatically identify infected regions from chest CT slices. In our Inf-Net, a parallel partial decoder is used to aggregate the high-level features and generate a global map. Then, the implicit reverse attention and explicit edge-attention are utilized to model the boundaries and enhance the representations. Moreover, to alleviate the shortage of labeled data, we present a semi-supervised segmentation framework based on a randomly selected propagation strategy, which only requires a few labeled images and leverages primarily unlabeled data. Our semi-supervised framework can improve the learning ability and achieve a higher performance. Extensive experiments on our COVID-SemiSeg and real CT volumes demonstrate that the proposed Inf-Net outperforms most cutting-edge segmentation models and advances the state-of-the-art performance.
AB - Coronavirus Disease 2019 (COVID-19) spread globally in early 2020, causing the world to face an existential health crisis. Automated detection of lung infections from computed tomography (CT) images offers a great potential to augment the traditional healthcare strategy for tackling COVID-19. However, segmenting infected regions from CT slices faces several challenges, including high variation in infection characteristics, and low intensity contrast between infections and normal tissues. Further, collecting a large amount of data is impractical within a short time period, inhibiting the training of a deep model. To address these challenges, a novel COVID-19 Lung Infection Segmentation Deep Network (Inf-Net) is proposed to automatically identify infected regions from chest CT slices. In our Inf-Net, a parallel partial decoder is used to aggregate the high-level features and generate a global map. Then, the implicit reverse attention and explicit edge-attention are utilized to model the boundaries and enhance the representations. Moreover, to alleviate the shortage of labeled data, we present a semi-supervised segmentation framework based on a randomly selected propagation strategy, which only requires a few labeled images and leverages primarily unlabeled data. Our semi-supervised framework can improve the learning ability and achieve a higher performance. Extensive experiments on our COVID-SemiSeg and real CT volumes demonstrate that the proposed Inf-Net outperforms most cutting-edge segmentation models and advances the state-of-the-art performance.
KW - COVID-19
KW - CT image
KW - infection segmentation
KW - semi-supervised learning
UR - http://www.scopus.com/inward/record.url?scp=85088907242&partnerID=8YFLogxK
U2 - 10.1109/TMI.2020.2996645
DO - 10.1109/TMI.2020.2996645
M3 - 文章
C2 - 32730213
AN - SCOPUS:85088907242
SN - 0278-0062
VL - 39
SP - 2626
EP - 2637
JO - IEEE Transactions on Medical Imaging
JF - IEEE Transactions on Medical Imaging
IS - 8
M1 - 9098956
ER -