eduzhai > Applied Sciences > Engineering >

Deep Learning Predicts Cardiovascular Disease Risks from Lung Cancer Screening Low Dose Computed Tomography

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 29 pages

Abstract: Cancer patients have a higher risk of cardiovascular disease (CVD) mortalitythan the general population. Low dose computed tomography (LDCT) for lungcancer screening offers an opportunity for simultaneous CVD risk estimation inat-risk patients. Our deep learning CVD risk prediction model, trained with30,286 LDCTs from the National Lung Cancer Screening Trial, achieved an areaunder the curve (AUC) of 0.871 on a separate test set of 2,085 subjects andidentified patients with high CVD mortality risks (AUC of 0.768). We validatedour model against ECG-gated cardiac CT based markers, including coronary arterycalcification (CAC) score, CAD-RADS score, and MESA 10-year risk score from anindependent dataset of 335 subjects. Our work shows that, in high-riskpatients, deep learning can convert LDCT for lung cancer screening into adual-screening quantitative tool for CVD risk estimation.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...