Israeli Hospital Completes Study Using AI to Detect Cardiovascular Issues

Photo Credit: Piron Guillaume / Unsplash.jpg

Beilinson Hospital recently completed a study testing Nanox.AI’s HealthCCSng medical software that found 58 percent of patients unknowingly had moderate to severe levels of coronary artery calcium (CAC) or plaque.

CAC is the strongest predictor of future cardiac events, and measuring it typically subjects patients to an additional costly scan that is not normally covered by insurance companies.

The study was conducted by the Cardiology and Imaging departments at Beilinson Medical Center in Petah Tikva, one of the largest multidisciplinary medical institutions in Israel.

The Beilinson team utilized Nanox.AI’s HealthCCSng, an FDA-cleared and CE Marked tool designed specifically for cardiac health assessment.

HealthCCSng uses medical imaging data from routine chest CT scans to automatically quantify and analyze CAC, the strongest predictor of future cardiac events, with patients in the highest category being more than 20 times more likely to suffer a cardiac event.

Many studies show a clear correlation between coronary artery calcium detected in the coronary arteries and severity of disease.

“The patients in this study received routine CT scans that had nothing to do with cardiac concern. Nanox’s AI technology can enable physicians to route these unsuspecting individuals with high CAC levels to the appropriate care pathways and treatment,” said Professor Ran Kornowski, Director of the Cardiology Center at Beilinson Hospital.

Kornowski led the study with Prof. Ashraf Hamdan, director of the Cardiovascular Imaging unit.

Among the 326 eligible patients who participated in the study from January to July 2023, 101 (31 percent) exhibited Severe CAC, 88 (27 percent) had Moderate CAC, and 137 (42 percent) showed Low CAC.

Patients with severe CAC levels were referred to specialized preventive cardiology clinics for in-depth evaluation and treatment. Those categorized with Low and Moderate CAC were directed to primary care physicians for further assessment and medical optimization.

“While the study’s findings were staggering, we are encouraged by the important role AI can play in early risk identification and prevention of cardiovascular events.”

In the study, the software was employed to assess Coronary Artery Calcification (CAC) levels from non-gated, non-contrast chest CT scans. The software’s categorizations were subjected to qualitative evaluation by two radiologists who reviewed each case.

Patient categorization included Low (CAC 0-99), Moderate (CAC 100-399), and Severe (CAC > 400), with exclusions applied to individuals with certain medical histories or artifacts.

“We have integrated Nanox.AI’s HealthCCS solution within our HMO-owned (Clalit) teaching hospital, and the outcomes have been highly promising in terms of detecting previously undiagnosed patients with elevated calcium levels from routine chest CT scans.

“Even within our health-conscious population, we were able to identify a significant number of individuals with hidden cardiovascular disease and guide them onto the appropriate care trajectory,” emphasized Prof. Ashraf Hamdan, Head of Cardiovascular Imaging at Beilinson Hospital and chief investigator.

“Given the global prominence of cardiovascular disease as the leading cause of mortality and the fact that nearly half of patients realize their condition only after an initial heart attack, leveraging such technologies for general population screening and early detection is absolutely imperative,” noted Dr. Orit Wimpfheimer, Chief Medical Officer of Nanox.AI.

Nanox.AI’s portfolio also includes an FDA-cleared imaging solution in bone health (HealthOST) and one in development for fatty liver disease (HealthFLD), which also use routine medical CT imaging to help physicians identify early signs of diseases, such as osteoporosis and non-alcoholic fatty liver disease. The technology enables further work-up and possible treatment, helping prevent potentially major, life-changing health events.