ARTICLE SUMMARY:
The paradigm for screening, diagnosing, and staging cardiac diseases ought to be more like that of cancer—based not on risk factors alone, but on the nature of the disease present in a particular patient. AI is making that happen. We discuss the predictive technologies of Cleerly, Heartflow, and Corvista for CAD; AccurKardia in aortic stenosis; and Cardiosense in heart failure.
Screening for cancer involves imaging modalities and tests that look at the tissue inside a patient, that is, mammography for breast cancer, colonoscopy for gastrointestinal cancer, and low-dose CT scans for the screening of patients at risk of lung cancer. But the process for identifying cardiac conditions early relies heavily on risk factors: age, sex, family history, comorbidities, lifestyle, race, zip code, and other social determinants of health.
Statistics from the CDC clearly demonstrate that it’s not an effective way to manage patients, with 605,000 people in the US having their first heart attack each year. Autopsies have also shown that 34% of people who died of sudden cardiac death showed evidence of a previous heart attack that went undetected. And heart failure, which accounts for more than 425,000 deaths in the US annually, is largely undiagnosed until later stages, when structural damage has already occurred.
We also know that many should-be cardiac patients fall through the cracks because of incidental findings of cardiac disease in electronic records when imaging is done for other purposes, although that’s the gap that Bunkerhill Health is solving, with a coronary artery calcium algorithm that finds cardiovascular disease in CT scans. (See “Cleerly and Bunkerhill Health Partner to Find Cardiac Patients at Risk,” MedTech Strategist, March 12, 2025.)
