Bunkerhill Health has received US Food and Drug Administration (FDA) clearance for the Bunkerhill MAC artificial intelligence (AI) algorithm for the detection and quantification of mitral annular calcification (MAC). The algorithm runs on routine, non-gated chest computed tomography (CT) scans.
MAC is a chronic condition in which calcium deposits build up in the mitral valve ring. Although often discovered incidentally, studies have linked MAC with increased cardiovascular risk, including higher rates of mortality and complications in structural heart procedures. Despite its significance, MAC is often not quantified or consistently reported on routine chest CTs.
The algorithm was developed and tested using data from Bunkerhill’s multi-institutional research consortium of more than 25 academic medical centres. For this clearance, data from seven consortium sites were included, among them Emory University, Thomas Jefferson University and the University of California, San Francisco (UCSF). The clearance represents one of the broadest academic collaboration efforts yet applied to the development of an FDA-cleared AI algorithm.
Bunkerhill MAC is available within Carebricks, Bunkerhill’s AI platform for clinical reasoning and action. Carebricks allows health systems to apply AI-powered reasoning—including large language models (LLMs) and FDA-cleared algorithms—across the full scope of patient data and to customise workflows that automate appropriate next steps.
“FDA clearance of this algorithm is a landmark not just for Bunkerhill, but for how we use routine data to advance cardiac care,” said Nishith Khandwala, co-founder and CEO of Bunkerhill Health. “By making MAC quantification available at scale through routine CT scans, we’re giving clinicians a new tool to better understand risk and tailor patient care, all embedded seamlessly into the workflows they already use.”
