Use of a deep-learning convolutional neural network (CNN) — a form of artificial intelligence — can help reveal patterns on chest computed tomography (CT) scans that identify smokers at high long-term risk for lung cancer well beyond the Centers for Medicare & Medicaid Services (CMS) criteria for lung screening eligibility, according to the results of an analysis published in the Annals of Internal Medicine.

Investigators sought to create and validate a CNN — that is, the CXR-LC model — with the ability to predict long-term incident lung cancer via the use of data typically available in a patient’s electronic medical record, including chest radiographs, sex, age, and current smoking status. The CXR-LC model was developed in the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial, which included to total of 41,856 patients. The final CXR-LC model was validated in additional smokers from the PLCO study (n=5615; 12-year follow-up) and National Lung Screening Trial (NLST) heavy smokers (n=5493; 6-year follow-up).

There were more current smokers (50.4% vs 20.2%, respectively) and higher mean pack-years (55.7 vs 35.4, respectively) in the NLST data set than in the PLCO data set. Further, although the follow-up times differed (12 years for PLCO and 6 years for NLST), the occurrence of incident lung cancer was similar between PLCO (3.7% [207 of 5615]) and NLST (3.8% [206 of 5493]), which is linked to the greater smoking burden in the NLST data set despite the shorter follow-up time.

Continue Reading

In the PLCO validation data set, the CXR-LC model had better discrimination for incident lung cancer — that is, a significantly higher area under the receiver-operating characteristic curve (AUC) of 0.755 — compared with an AUC of 0.634 in the CMS eligibility data set (P <.001) and a positive radiograph screen (AUC, 0.550; P <.001). Additionally, CXR-LC eligibility was significantly more sensitive than CMS eligibility (74.9% [95% CI, 69.0-80.8] vs 63.8% [95% CI, 57.2-70.3], respectively; P =.012) in predicting the development of lung cancer.

The investigators concluded that based on the findings from this study, high-risk individuals may benefit from the use of lung cancer screening CT scans.

Disclosure: Several study authors declared affiliations with the pharmaceutical industry. Please see the original reference for a full list of authors’ disclosures.


Lu MT, Raghu VK, Mayrhofer T, Aerts HJWL, Hoffmann U. Deep learning using chest radiographs to identify high-risk smokers for lung cancer screening computed tomography: development and validation of a prediction model. Ann Intern Med. Published online September 1, 2020. doi:10.7326/M20-1868