An integrated framework with machine learning and radiomics for accurate and rapid early diagnosis of COVID-19 from Chest X-ray

作者：

Highlights：

• A new diagnostic tool to detect COVID-19 from chest X-ray (CXR) at early stage.

• The new tool can detect COVID-19 from CXR with 0 severity (scored by radiologists).

• 88% sensitivity and 97% specificity against many different lung infections.

• 90% accuracy with completely independent test dataset and different disease stages.

• Requires less than 2 minutes to provide the diagnosis result.

摘要

•A new diagnostic tool to detect COVID-19 from chest X-ray (CXR) at early stage.•The new tool can detect COVID-19 from CXR with 0 severity (scored by radiologists).•88% sensitivity and 97% specificity against many different lung infections.•90% accuracy with completely independent test dataset and different disease stages.•Requires less than 2 minutes to provide the diagnosis result.

论文关键词：COVID-19,Early diagnosis,Chest X-ray,Radiomics,Machine learning

论文评审过程：Received 23 November 2020, Revised 1 May 2021, Accepted 1 May 2021, Available online 4 May 2021, Version of Record 8 May 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115152