Predicting heart disease death risk from lung cancer screening

22 April, 2021
Predicting heart disease death risk from lung cancer screening
  • A recently available study developed a method to incorporate data from lung cancer screening with artificial intelligence (AI) to greatly help predict the risk of death from cardiovascular disease.
  • The study’s calcium scoring method predicted cardiovascular mortality risk comparably to standard methods.
  • In line with the authors, their method is “automatic, fast, and only cause[s] minimal extra workload.”
Heart diseaseTrusted Source may be the number one reason behind death of people in the usa, with about 25%Trusted Way to obtain deaths from coronary disease (CVD) due to smoking.

Smoking doubles to quadruplesTrusted Source the chance for coronary heart disease and stroke.

Chemicals in tobacco smoke damage the liner of the arteries, resulting in inflammation. Smoking can also donate to atherosclerosis or the accumulation of plaque (composed of fat, calcium, and other substances), creating arterial narrowing and stiffness.

A coronary artery calcium scanTrusted Source is an X-ray that uses computerized tomography (CT) to create multiple images of calcium-containing plaques in the arteries. From these images, medical researchers can create a coronary artery calcium (CAC) score to determine a patient’s heart disease risk and guide treatment.

The 2021 United States Preventative Services Task Force (USPSTF) currently recommends total annual lung cancer screening with low-dose CT in adults aged 50-80 years who have a 20 pack-yearTrusted Source smoking history and currently smoke or quit within days gone by 15 years.

The authors of the recent study examined whether doctors could use these recommended lung screenings to predict the 5-year risk for heart disease-related deaths.

Deep learning method
The study findings come in the journal Radiology: Cardiothoracic Imaging. The analysis used low-dose CT lung scan data from participants signed up for the National Lung Screening Trial between August 2002 and April 2004.

The analysis used an programmed deep learning solution to quantify arterial calcium from the low-dose CT lung scans. Deep learning is a kind of AI that runs on the multi-layered structure of algorithms or “neural networks” to investigate data.

The first neural network aligns and crops the CT image to give attention to the heart. The second neural network measures calcium levels in the images extracted from the first neural network.

The deep learning network determined blood vessel calcification in six different parts of the heart, like the thoracic aorta, aortic valve, mitral valve, and the proper coronary artery.

Using this information, the authors predicted each participant’s 5-year CVD mortality. Then they compared the analysis prediction model with three other models:

  • self-reported participant characteristics only
  • self-reported characteristics and total CAC
  • self-reported characteristics and the study prediction model
The self-reported participant characteristics indicating increased CV risk included age, a brief history of smoking, and a brief history of other diseases, such as diabetes, stroke, high blood pressure, and heart disease.

Researchers trained the prediction model with data from 4,451 participants and tested it with data from 1,113 participants. Approximately 62% of the testing group were male, with a median age of 61 and a smoking history of about 50 pack-years.

The authors discovered that their programmed calcium scoring method performed equally at predicting 5-year CV mortality risk as the model using self-reported CV risk factors and total CAC.

The study prediction model performed much better than the self-reported patient characteristics alone. The most accurate predictions originated from the scientists’ new method in combo with self-reported participant characteristics.