The rapid and uncontrollable spread of COVID-19 has seriously threatened global public health. Rapid and accurate diagnosis of COVID-19 is the main key to control and manage the epidemic. COVID-19 segmentation can provide great insights to accelerate clinical decisions. COVID-19 segmentation based on deep learning method has attracted extensive research in the field of medical image analysis. However, most existing networks are heavyweight networks, which causes structural redundancy and expensive computational cost. Moreover, the segmentation problem of low-contrast COVID-19 image and the obscure boundary between the infected area and normal tissues affects accuracy of segmentation. To solve the problems, we propose a lightweight contextual information fusion network, LCFNet, for COVID-19 segmentation. We introduce a contextual information fusion strategy combining multiple global pyramid guidance (GPG) with scale-aware pyramid fusion (SAPF) module and deep supervision (DS) module, which can capture more fine-grained image features. We conduct experiments on two COVID-19 datasets. We perform the ablation studies, demonstrating the effectiveness of key components of the proposed method. Compared with traditional segmentation methods, LCFNet model match more consistently with the ground-truth boundary, which shows the superiority of the proposed method. Moreove, LCFNet has 1.68M parameters, which demonstrates the robustness of the proposed method. Our proposed method can further induce the number of model parameters. Compared with the state-of-the-art methods, the proposed approach has achieved significant improvements and is also superior to other segmentation methods.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.