An algorithm applied to scanners in lung cancer screening makes it possible to measure the calcium score in the arteries, a reliable marker of cardiovascular risk.
Screening for lung cancer could have other virtues than the prevention of a particularly aggressive disease diagnosed in more than 37,000 people each year in France. Scientists from the Cardiovascular Imaging Research Center at Massachusetts General Hospital in Boston (USA) have just shown, thanks to an artificial intelligence system, that this scanner examination also makes it possible to assess the risks of cardiovascular disease by measuring the deposition of calcium in the arteries.
They have created an algorithm capable of measuring calcium deposition in the arteries on chest CT images taken as part of lung cancer screening, which allows them to assess the risks of heart disease linked to this deposition. Their work was presented at the RSNA (American Society of Radiology) Congress on December 3.
The calcium score helps in the prescription of statins
“The new cholesterol guidelines encourage the use of the calcium score to help doctors and patients decide whether to take a statin. For some patients at intermediate risk of heart disease, if the calcium score is 0, statin therapy may be delayed. If the score is high, taking the statin can be postponed,” explains Michaël T. Lu, director of the Boston IARC.
Hence the search for this rate using an algorithm whose accuracy has been tested on thousands of heavy smokers aged 55 to 74 who are part of a pulmonary screening trial. The results showed that the coronary artery calcium scores obtained by this technique corresponded to those of human readers but, more importantly, they validated the link between calcium scores and cardiovascular mortality during the follow-up period. who was 7 years old.
A tool that can be used on almost all chest scanners
“If our tool detects a lot of calcium in a patient’s coronary artery, we can refer them to a specialist for follow-up to facilitate their access to appropriate treatment,” said Roman Zeleznik, lead author of the study. who adds: “We now have a tool that can be used on almost any chest CT scan to provide very clinically relevant information for a large number of patients.”
.
