Synergistic learning of lung lobe segmentation and hierarchical multi-instance classification for automated severity assessment of COVID-19 in CT images

作者：

Highlights：

• A multi-task multi-instance learning framework is proposed to jointly assess the severity of the COVID-19 patients and segment lung lobe.

• A unique hierarchical multi-instance learning strategy is developed to predict the severity of patients in a weakly supervised manner for 3D CT images.

• An embedding-level multi-instance learning method is proposed to predict labels beyond the instance-level.

• The proposed method achieving promising results in severity assessment in a real-world COVID-19 dataset compared to several state-of-the-art methods.

摘要

•A multi-task multi-instance learning framework is proposed to jointly assess the severity of the COVID-19 patients and segment lung lobe.•A unique hierarchical multi-instance learning strategy is developed to predict the severity of patients in a weakly supervised manner for 3D CT images.•An embedding-level multi-instance learning method is proposed to predict labels beyond the instance-level.•The proposed method achieving promising results in severity assessment in a real-world COVID-19 dataset compared to several state-of-the-art methods.

论文关键词：COVID-19,CT,Severity assessment,Lung lobe segmentation,Multi-instance learning

论文评审过程：Received 16 June 2020, Revised 10 December 2020, Accepted 22 December 2020, Available online 16 January 2021, Version of Record 19 January 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.107828