Artificial Intelligence Detects Signs of Heart Disease on Lung Cancer Screenings

The use of artificial intelligence (AI) can provide an automated and accurate tool to measure a common marker of heart disease in patients undergoing lung cancer screening, according to a study presented today at the annual meeting of the Radiological Society of North America (RSNA).

“The new cholesterol guidelines encourage using the calcium score to help physicians and patients decide whether to take a statin,” said study co-senior author Michael T. Lu, M.D., M.P.H., director of AI in the Cardiovascular Imaging Research Center (CIRC) at Massachusetts General Hospital (MGH) in Boston in a press release about the findings. “For select patients at intermediate risk of heart disease, if the calcium score is 0, statin can be deferred. If the calcium score is high, then those patients should be on a statin.”

In this study, researchers trained a deep-learning system on cardiac CTs and chest CTs in which the coronary artery calcium had been measured manually. Subsequently, they tested the system on the CT scans of thousands of heavy smokers, age 55-74, who were part of the National Lung Screening Trial (NLST).

According to the results of the study, the deep learning system coronary artery calcium scores corresponded to those of human readers. Moreover, the researchers observed a notable link between deep learning calcium scores and cardiovascular death over follow-up period of almost seven years.

“There’s information about cardiovascular health on these CT scans,” Dr. Lu continued. “This is an automated way to extract that information, which can help patients and physicians make decisions about preventative therapy.”

“If our tool detects a lot of coronary artery calcium in a patient, then maybe we can send that patient to a specialist for follow up,” said lead author Roman Zeleznik, M.Sc., B.Sc., from the Artificial Intelligence in Medicine (AIM) Program at Boston’s Brigham and Women’s Hospital (BWH) and Dana-Farber Cancer Institute. “This would make it easier for patients to get appropriate treatment.”