Zebra Medical is hoofing its way through the FDA gauntlet, racking up eight 510(k) clearances for its artificial intelligence software systems in barely three years.
The latest of these gives the go-ahead to HealthCCSng, an AI program that measures the amount of calcium built up in the heart’s main arteries, helping physicians not only diagnose cardiovascular disease but also assess its severity.
The software is intended to catch heart disease and get patients started on a treatment plan well before a coronary event occurs, reversing the current trend in which up to half of patients don’t know they have cardiovascular disease until they experience their first heart attack.
Coronary artery calcium scoring is a major indicator of heart disease, according to Zebra Medical, with studies showing a strong correlation between the amount of calcium in the arteries and the severity of the disease. Patients with the highest levels of calcium buildup may be more than 20 times more likely to experience a cardiac event than other groups.
Calcium scoring is usually measured using cardiac-gated CT scans, where an electrocardiogram monitors heart rate throughout the scanning process. HealthCCSng, meanwhile, analyzes existing non-gated CT scans, which are less expensive and require less radiation exposure than gated scans.
After calculating the amount of calcium buildup in the imaging data, the software sorts patients into three categories, indicating the severity of their cardiovascular disease.
In a retrospective analysis (PDF) of the low-dose CT scans collected from more than 12,000 people in the National Lung Screening Trial, Zebra Medical’s machine learning algorithm was proven to be a stronger predictor of cardiovascular disease incidence and mortality than the current standard of manually calculating that risk. The 2020 study concluded that the software further outperformed manual predictors because it’s fully automated, allowing it to save time and also assess more potential risk factors at once.
Zebra Medical’s first-ever FDA clearance also concerned a calcium scoring tool. That AI algorithm, cleared in July 2018, was able to automatically calculate the amount of buildup from gated CT scans.
Between then and now, the Israeli company has reeled in regulatory nods for AI systems designed to calculate bone measurements from standard X-rays, flag potentially cancerous lesions in mammograms, spot pneumothorax and pleural effusion in chest scans and more.
All those wins led to digital X-ray machine maker Nanox’s decision last month to acquire Zebra Medical. The deal carries a total value of $200 million, with Zebra Medical getting half up front and the other half paid out as it hits certain milestones.
Zebra Medical’s AI algorithms will be embedded into Nanox’s digital X-ray system, creating a system that collects, processes and analyzes imaging data all in the same platform.