eduzhai > Applied Sciences > Engineering >

Multi-Task Driven Explainable Diagnosis of COVID-19 using Chest X-ray Images

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 11 pages

Abstract: With increasing number of COVID-19 cases globally, all the countries areramping up the testing numbers. While the RT-PCR kits are available insufficient quantity in several countries, others are facing challenges withlimited availability of testing kits and processing centers in remote areas.This has motivated researchers to find alternate methods of testing which arereliable, easily accessible and faster. Chest X-Ray is one of the modalitiesthat is gaining acceptance as a screening modality. Towards this direction, thepaper has two primary contributions. Firstly, we present the COVID-19Multi-Task Network which is an automated end-to-end network for COVID-19screening. The proposed network not only predicts whether the CXR has COVID-19features present or not, it also performs semantic segmentation of the regionsof interest to make the model explainable. Secondly, with the help of medicalprofessionals, we manually annotate the lung regions of 9000 frontal chestradiographs taken from ChestXray-14, CheXpert and a consolidated COVID-19dataset. Further, 200 chest radiographs pertaining to COVID-19 patients arealso annotated for semantic segmentation. This database will be released to theresearch community.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×