Deep learning predicts cardiovascular disease risks from lung cancer screening low dose computed tomography

Hanqing Chao; Hongming Shan; Fatemeh Homayounieh; Ramandeep Singh; Ruhani Doda Khera; Hengtao Guo; Timothy Su; Ge Wang; Mannudeep K Kalra; Pingkun Yan

doi:10.1038/s41467-021-23235-4

Back

Deep learning predicts cardiovascular disease risks from lung cancer screening low dose computed tomography

Journal article

Open access

Peer reviewed

Deep learning predicts cardiovascular disease risks from lung cancer screening low dose computed tomography

Hanqing Chao, Hongming Shan, Fatemeh Homayounieh, Ramandeep Singh, Ruhani Doda Khera, Hengtao Guo, Timothy Su, Ge Wang, Mannudeep K Kalra and Pingkun Yan

Nature communications, Vol.12(1), pp.2963-2963

05/20/2021

DOI: 10.1038/s41467-021-23235-4

PMCID: PMC8137697

PMID: 34017001

Files and links (1)

url

https://doi.org/10.1038/s41467-021-23235-4View

Published (Version of record) Open Access

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, achieves an area under the curve (AUC) of 0.871 on a separate test set of 2,085 subjects and identifies patients with high CVD mortality risks (AUC of 0.768). We validate 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.

Adult

Aged

Aged, 80 and over

Cardiovascular Diseases - diagnosis

Cardiovascular Diseases - epidemiology

Cardiovascular Diseases - etiology

Clinical Trials as Topic

Coronary Vessels - diagnostic imaging

Datasets as Topic

Deep Learning

Electrocardiography

Female

Follow-Up Studies

Humans

Image Processing, Computer-Assisted - methods

Lung - diagnostic imaging

Lung Neoplasms - complications

Lung Neoplasms - diagnosis

Male

Mass Screening - methods

Mass Screening - statistics & numerical data

Middle Aged

Retrospective Studies

Risk Assessment - methods

Risk Assessment - statistics & numerical data

Risk Factors

ROC Curve

Tomography, X-Ray Computed - statistics & numerical data

Details

Title: Subtitle: Deep learning predicts cardiovascular disease risks from lung cancer screening low dose computed tomography
Creators: Hanqing Chao - Rensselaer Polytechnic Institute
Hongming Shan - Rensselaer Polytechnic Institute
Fatemeh Homayounieh - Massachusetts General Hospital
Ramandeep Singh - Massachusetts General Hospital
Ruhani Doda Khera - Harvard University
Hengtao Guo - Rensselaer Polytechnic Institute
Timothy Su - Niskayuna High School, Niskayuna, NY, USA
Ge Wang - Rensselaer Polytechnic Institute
Mannudeep K Kalra - Harvard University
Pingkun Yan - Rensselaer Polytechnic Institute
Resource Type: Journal article
Publication Details: Nature communications, Vol.12(1), pp.2963-2963
DOI: 10.1038/s41467-021-23235-4
PMID: 34017001
PMCID: PMC8137697
ISSN: 2041-1723
eISSN: 2041-1723
Grant note: R56 HL145172 / NHLBI NIH HHS
Language: English
Date published: 05/20/2021
Academic Unit: Radiology
Record Identifier: 9984697625602771

Metrics

5 Record Views

56 Times Cited - Web of Science

See more details