Lung Cancer Screening Images Used to Identify Heart Conditions

RSNA 2020 research highlights importance of screening images for multiple conditions


ML Assesses Calcium Buildup in Arteries on CT Lung Scans


Machine learning (ML) can be used to assess coronary artery calcification (CAC) on CT lung screening scans, according to research presented at RSNA 2020.

CAC — calcium buildup within the walls of the arteries of the heart — suggests the presence of atherosclerotic coronary artery disease (CAD) and the potential for serious cardiac events such as heart attack or stroke. The severity of CAD is an independent indicator of a patient’s risk of such an event. The patient’s risk is established through the CAC score derived through imaging and which typically measures CAC volume and density.

“Usually these are assessed on cardiac-gated CT scans,” said presenter Jordan Fuhrman, a PhD candidate, Department of Radiology, University of Chicago. “However, recent efforts have attempted to quantify CAC on low-dose CT scans obtained from lung cancer screening. A complete evaluation of these scans can be time-consuming and inconsistent.”

Fuhrman and colleagues used ML to develop an automatic, complete assessment of CAC. Using thoracic CT scans obtained from the International Early Lung Cancer Action Program, researchers divided 814 scans into four categories of CAC severity – none, mild, moderate and severe. Fuhrman explained that these images could also include other sources of calcium not within the coronary arteries. Efforts to identify CAC from an input image have utilized convolutional neural networks (CNNS).

Some studies have attempted to reduce these false positive detections through the use of multiple CNNs.

“The use of these additional networks is, in my opinion, not necessary, and we can achieve equivalent performance with reduced computation using only one convolution neural network,” he said.

Fuhrman and colleagues used a revised U-Net architecture, commonly used for biomedical image segmentation, that simultaneously identifies CAC and assigns ordinal scores between 0 and 3 to each of the four coronary artery branches — the left main artery, left anterior descending artery, right coronary artery, and the circumflex — which corresponds to the volume of CAC in each. The total sum is the CAC volume score (0-12).

Researchers determined whether a low-dose lung screening CT scan contained CAC with an area under the curve (AUC) of 0.96, establishing a baseline metric that allows them to specify whether a case has CAC. Fuhrman and colleagues further tested this approach by investigating each of the four coronary artery branches.

The left anterior descending artery performed the best (AUC=0.95), and the left main artery performed the worst (AUC=0.84), results show.


Coronary Artery Calcification in Lung Cancer Screening


According to data from the National Lung Screening Trial, atherosclerotic cardiovascular disease (ASCVD) is the leading cause of death for individuals receiving lung cancer screening (LCS).

“This speaks to the significant need for cardiovascular risk stratification and reduction in the lung cancer screening population,” said Tina Tailor, MD, an assistant professor of radiology at Duke University Medical Center, during an RSNA 2020 session.

Coronary artery calcification (CAC) is an imaging biomarker of ASCVD that is commonly detected by LCS examinations.

“It’s not unusual to find CAC in individuals screened for lung cancer,” Dr. Tailor said. “However, the role of CAC in the lung cancer screening population is unclear.”

“We sought to determine the proportion of lung cancer screening individuals eligible for statin therapy for primary ASCVD prevention by American Heart Association (AHA) guidelines, assess the rate of statin prescription amongst statin-eligible patients, and determine the impact of lung cancer screening-reported CAC on downstream statin initiation,” Dr. Tailor said.

The study was a retrospective review of electronic health record data of patients undergoing baseline lung cancer screening at three academic institutions and their affiliated hospitals. A total of 5,495 individuals received lung cancer screening during the span of the study.

After excluding those with pre-existing clinical ASCVD, 73.6% of patients were determined to be statin-eligible according to the AHA guidelines.

“Yet most statin-eligible individuals were not on a statin at the time of the lung cancer screening, which shows a significant disconnect between those eligible for statin and those actually receiving it,” Dr. Tailor said.

Researchers found a significant increase in statin prescription downstream of the LCS exam for individuals with LCS-reported CAC.

“Additionally, the likelihood of downstream statin prescription was higher in patients with increasing CAC severity. This suggests a positive trend towards statin prescription with the reporting of CAC by radiologists,” she said.

The study shows that statins are grossly under-prescribed in the LCS population, a group at high risk for ASCVD. The reporting of CAC by radiologists at LCS examination may prompt downstream preventive statin prescription.

“Results suggest that radiologists should report both the presence and severity of CAC at lung cancer screening exams,” Dr. Tailor said.

For More Information

Access these RSNA 2020 sessions at