A deep-learning-based framework for severity assessment of COVID-19 with CT images

作者：

Highlights：

• A novel prior-knowledge-based model for severity assessment of COVID-19.

• A new input strategy based on multi-view slices for 3D model of COVID-19.

• Sufficient CT images were collected for the fine-grained severity assessment.

• Potential values for accelerating triage, following up the treatment response, etc.

摘要

•A novel prior-knowledge-based model for severity assessment of COVID-19.•A new input strategy based on multi-view slices for 3D model of COVID-19.•Sufficient CT images were collected for the fine-grained severity assessment.•Potential values for accelerating triage, following up the treatment response, etc.

论文关键词：COVID-19,Deep learning,Severity assessment,Multi-view lesion,Dual-Siamese channels,Clinical metadata

论文评审过程：Received 6 March 2021, Revised 3 June 2021, Accepted 12 July 2021, Available online 27 July 2021, Version of Record 30 July 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115616