English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Coronavirus Disease 2019 (COVID-19) has caused great casualties and becomesalmost the most urgent public health events worldwide. Computed tomography (CT)is a significant screening tool for COVID-19 infection, and automatedsegmentation of lung infection in COVID-19 CT images will greatly assistdiagnosis and health care of patients. However, accurate and automaticsegmentation of COVID-19 lung infections remains to be challenging. In thispaper we propose a dilated dual attention U-Net (D2A U-Net) for COVID-19 lesionsegmentation in CT slices based on dilated convolution and a novel dualattention mechanism to address the issues above. We introduce a dilatedconvolution module in model decoder to achieve large receptive field, whichrefines decoding process and contributes to segmentation accuracy. Also, wepresent a dual attention mechanism composed of two attention modules which areinserted to skip connection and model decoder respectively. The dual attentionmechanism is utilized to refine feature maps and reduce semantic gap betweendifferent levels of the model. The proposed method has been evaluated onopen-source dataset and outperforms cutting edges methods in semanticsegmentation. Our proposed D2A U-Net with pretrained encoder achieves a Dicescore of 0.7298 and recall score of 0.7071. Besides, we also build a simplifiedD2A U-Net without pretrained encoder to provide a fair comparison with othermodels trained from scratch, which still outperforms popular U-Net familymodels with a Dice score of 0.7047 and recall score of 0.6626. Our experimentresults have shown that by introducing dilated convolution and dual attentionmechanism, the number of false positives is significantly reduced, whichimproves sensitivity to COVID-19 lesions and subsequently brings significantincrease to Dice score.

D2A U-Net: Automatic Segmentation of COVID-19 Lesions from CT Slices with Dilated Convolution and Dual Attention Mechanism