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

TL;DR

This study develops a deep learning model that predicts cardiovascular disease risk from low dose CT scans used for lung cancer screening, enabling dual screening from a single imaging modality.

Contribution

The paper introduces a novel deep learning approach that estimates CVD risk directly from LDCT scans, validated against established cardiac markers and risk scores.

Findings

Deep learning model achieved AUC of 0.871 for CVD risk prediction.

Model identified high CVD risk patients with an AUC of 0.768.

Validated against ECG-gated cardiac CT markers and risk scores.

Abstract

Cancer patients have a higher risk of cardiovascular disease (CVD) mortality than the general population. Low dose computed tomography (LDCT) for lung cancer screening offers an opportunity for simultaneous CVD risk estimation in at-risk patients. Our deep learning CVD risk prediction model, trained with 30,286 LDCTs from the National Lung Cancer Screening Trial, achieved an area under the curve (AUC) of 0.871 on a separate test set of 2,085 subjects and identified patients with high CVD mortality risks (AUC of 0.768). We validated our model against ECG-gated cardiac CT based markers, including coronary artery calcification (CAC) score, CAD-RADS score, and MESA 10-year risk score from an independent dataset of 335 subjects. Our work shows that, in high-risk patients, deep learning can convert LDCT for lung cancer screening into a dual-screening quantitative tool for CVD risk estimation.