Scouting Report-Cleerly: Applying AI for Early Detection and Treatment of Coronary Disease
New York-based Cleerly secured $43 million in series B funding providing a cutting-edge digital health platform with machine learning capabilities for early detection and treatment of coronary disease and heart attacks. The Series B funding was led by Vensana Capital, with additional backing from LRVHealth, New Leaf Venture Partners, DigiTx Partners, the American College of Cardiology, and Cigna Ventures. Founded in 2017 by Dr. James K. Min, a cardiologist and director of the Dalio Institute for Cardiac Imaging at New York Presbyterian Hospital/Weill Cornell Medical College, Cleerly has raised $54M in total funding. Cleerly has two FDA approvals and will use the funding to commercially scale their company, obtain more FDA approvals, and invest in Research and Development (R&D) for their “precision prevention” technology.
Cleerly has spent five years building out its data science teams, perfecting its algorithms, and obtaining some approvals from the Food and Drug Administration in order to have recently emerged from “stealth” mode.
Cleerly believes that by applying their AI-based technology to analyze heart scans they could reduce costs of cardiovascular care by 60% including a 75% reduction in invasive cardiac catheterization tests.
Heart disease is largely asymptomatic and the first sign of heart disease is often a heart attack. 60% of those who have a heart attack have no prior symptoms.
The causes of heart disease are not thoroughly understood and healthcare providers rely on risk indicators that are insufficient to prevent heart attacks.
James K. Min, MD FACC, founder, and CEO of Cleerly is attempting to revolutionize the diagnosis and prevention of heart diseases by empowering primary care providers to “reach patients earlier [prior to hospitalization] where costs spiral out of control.” Cleerly is meant to allow for intervention by primary care providers by helping them to understand and interpret its advanced imaging without the need for a specialist. This will allow earlier diagnosis and treatment, before the patient complains of chest pain which is the “end-stage phenomenon” reflecting arteries that are already compromised with plaque or fatty deposits. For example, while risk factors such as cholesterol levels are currently used to determine the patient’s cardiac health, there is an 80% overlap “of cholesterol levels for people who do and don’t have heart attacks” according to Dr. Min in a recent Forbes article. In other words, risk factors are not enough to prevent heart attacks. Indeed, Dr. Min noted that often the first symptom of coronary artery disease can be a patient actually experiencing a heart attack itself, indicating a need for earlier diagnostics and preventative care plans.
According to the company, Cleerly has a database of 50,000 CT scans that Dr. Min and his team have cataloged and labeled to highlight certain characteristics. He and his team used AI and machine learning to train the algorithms to identify common patterns which characterize cardiovascular disease including the “presence, extent, severity and type” of disease. Unlike an invasive cardiac catheterization test, which is expensive, requires sedation and several hours in the hospital for recovery, Cleerly offers non-invasive medical interventions to conduct “comprehensive coronary artery phenotyping” which will help healthcare professionals to formulate a preventative healthcare plan. In addition, Cleerly’s platform is based upon analysis of what are called Cardiac Computer Tomography Angiogram (CTA) images, which are non-invasive but can still take detailed images of the heart illuminated by a dye injected into the patient. Although CTA is not yet first line therapy in the U.S., the American College of Cardiology has noted that the technology holds great promise and United Healthcare does reimburse for the procedure for lower risk patients with chest pain. CTA became first line therapy for those with chest pain in 2016. In addition, while Min initially theorized that “the more narrow and blocked a person’s arteries were, the more likely they would be to experience a heart attack”, however his team found it was not the amount of plaque but the thickness of the plaque that mattered. Patients who had so called fibrofatty or necrotic core plaque in their hearts, which was more likely to be softer and made of fat, cholesterol and other fatty compounds. According to the National Institutes of Health, understanding how the necrotic core develops is an urgent goal in heart-disease research. While the company is not positioning its product to replace trained doctors who can interpret scans, a June 2021 study commissioned by the company noted that the AI had a diagnostic accuracy of about 99.7 when assessing scans of patients whose tests indicated they had severe narrowing in their arteries.
The Big Picture:
According to the U.S. Centers for Disease Control (CDC) approximately 650,00 people will die from heart disease this year and heart disease will cost the healthcare system approximately $219 billion annually. As noted above, often the first indication of disease is severe chest pain or a heart attack, with about 60% of those who have heart attacks never having any prior symptoms of disease. Cleerly intends to change this dynamic by reaching patients earlier in the process and through their primary care providers instead of the ER when it can be too late to intervene. Considering the nature of heart disease, silent and asymptomatic, Cleerly will allow various healthcare professionals (PCPs, specialists, radiologists) to quickly identify the presence and nature of the heart disease without the need for invasive procedures like cardiac catheterization. As a result the company believes it can reduce expensive, invasive procedures by 75% and lower costs of cardiovascular treatment by 60%. By taking advantage of AI, Cleerly’s technology can analyze CT scans within minutes, compared to the 8 hours it would take humans to manually analyze the CT scans which makes the analysis feasible and cost effective. By deploying a non-invasive test such as this, Cleery’s technology would likely allow for broader scanning of the approximately one-half of Americans who currently display one of the three highest risk factors for the nation’s number one killer; high blood pressure, high cholesterol or smoking. As such it may be a more effective and efficient way to find some of the almost 20M people the CDC estimates currently have heart disease in the U.S.