Key points:

  • Heartflow is the first FDA-approved AI tool for diagnosing cardiac disease.
  • As the leading cause of death in the US, coronary artery disease can now be assessed more quickly and accurately with Heartflow.
  • The AI platform achieves accuracy close to invasive methods and is backed by numerous studies.
  • Heartflow’s scalable, per-report model has fueled rapid adoption and a successful IPO.
  • AI-driven early detection and prevention is a major investment theme for our digital health investment strategy.

Coronary artery disease affects 1 in 14 US adults and remains the leading cause of death, accounting for 1 in 5 heart-related fatalities, or around 400,000 deaths per year. Its most severe manifestation is a heart attack.

Angina, or chest pain, is often the earliest warning sign. Every year, 9.5 million Americans seek medical care for chest pain,which underscores the critical importance of accurate diagnosis and prevention in improving population health.

What is CAD?

Coronary artery disease (CAD) occurs when the arteries that supply blood to the heart become narrowed or blocked by plaque. This condition often develops silently over many years before symptoms occur. CAD is strongly linked to age, family history, and male sex, as well as lifestyle factors and chronic conditions such as high blood pressure (hypertension), high cholesterol, diabetes, smoking, and obesity.

Visualizing coronary anatomy

To assess CAD severity, doctors typically order a coronary computed tomography angiography (CTA) when a patient presents with chest pain. This imaging technique provides detailed anatomical views of coronary arteries, revealing areas of narrowing or blockage.

Radiologists then analyze thousands of CTA images to identify plaque, measure narrowing, and evaluate overall artery health. This process is highly labor-intensive and can take hours, even for experienced specialists, and often results in subjective conclusions.

By contrast, automatic AI tools like Heartflow can quantify plaque in just 5-10 minutes after image upload, delivering objective quantitative results powered by artificial intelligence and machine learning. 

Heartflow: faster, more accurate functional assessment

Heartflow provides a quicker and more precise evaluation than traditional human interpretation.

Using raw CTA images generated by computed tomography scanners – often thousands to tens of thousands – Heartflow constructs a detailed 3D model of the coronary arteries. It then applies computational fluid dynamics (CFD) to simulate blood flow and pressure throughout the coronary tree in just minutes.

The system calculates a fractional flow reserve (FFRCT) value at every point in the arteries – a virtual pressure measurement that predicts how much blood flow is reduced downstream of a stenosis, or narrowing of the vessel (See Image 1). This insight helps physicians determine whether blockages are functionally significant, meaning whether they truly restrict blood flow enough to cause ischemia – a condition where the heart muscle does not receive sufficient oxygen. Not all narrowings require immediate intervention.

Image 1: Heartflow’s AI-powered fractional flow reserve mapping visualization

A color-coded 3D model of coronary arteries generated by Heartflow’s AI, with annotated fractional flow reserve values at different artery segments, visually indicating areas of normal and reduced blood flow to help assess the functional significance of coronary blockages.
RCA: right coronary artery, LCX: left circumflex artery, LAD: left anterior descending artery

Color-coded 3D coronary arteries with FFR values, showing normal and reduced blood flow.

How accurate is AI interpretation?

Heartflow’s FFRCT analysis can achieve accuracy comparable to invasive fractional flow reserve (FFR), provided image quality is high. Clinical trials and real world data show sensitivity of around 85% to 90% and specificity of around 80%.3 Physicians value it as a triage tool, following CTA, particularly for intermediate stenosis (40% to 70%), where anatomical imaging alone is inconclusive. The solution is also cost-effective – around USD 1,000 per case, reimbursed by Medicare and most commercial insurers in the US.

Its credibility is reinforced by 3,000 peer-reviewed studies, which supported successful commercialization. As of October 2025, approximately 1,400 of 2,700 US hospitals and outpatient facilities performing CCTA have already adopted the Heartflow platform.4

Since launch, Heartflow has improved the accuracy of its algorithms, using a vast library of 110 million annotated CCTA images,5 a critical resource for machine learning and large language model training. Each new analysis further enhances performance.

The first FDA-approved AI algorithm

Heartflow received US FDA approval for its FFRCT analysis in 2014 and Plaque analysis in 2022, and went public on Nasdaq in August 2025. This marked the first IPO of a successfully commercialized SaaS-like AI algorithm – scalable and with clear visibility to operational profitability. Its per-reporting pricing model validates the clinical value it provides, helping physicians avoid risky and costly interventions such as stents or bypass surgery in acute settings where rapid decision-making is critical.

Competition is intensifying. In 2024, Cleerly’s ISCHEMIA – a rival to Heartflow’s FFRCT analysis – also received FDA clearance, while Elucid is advancing in the same space.6

The opportunity in CAD prevention

While FFRCT analysis provides actionable insights to treat symptomatic CAD patients, the greatest clinical impact lies in early detection and slowing disease progression. Heart attacks often occur as the first major event of CAD, even in individuals who show no symptoms.

Heartflow’s plaque analysis addresses this need by providing quantitative assessment of coronary plaque in seemingly healthy patients. This capability could offer predictive insights, enabling preventive measures to avert life-threatening events.

More transformative AI applications in healthcare

Heartflow’s IPO is more than a single success story – we believe it signals a broader transformation in healthcare driven by artificial intelligence. As diagnostic algorithms, predictive analytics, and large language models mature, we expect AI to redefine clinical workflows, improve patient outcomes, and unlock new efficiencies across the healthcare ecosystem.

For investors, this represents a long-term opportunity. Our Digital Health strategy is designed to capture these trends, focusing on companies that leverage AI to deliver scalable, cost-effective solutions with proven clinical impact. We believe the convergence of technology and medicine will remain one of the most powerful investment themes of the decade.

S-11/25 M-002660

About the author
  • Fang Liu

    Fang Liu

    CFA, Portfolio manager, Thematic Equities

    Fang Liu is a senior portfolio manager for the Digital Health Equity strategy on the Thematic Equity team at UBS Asset Management. Before joining the team in February 2020, Fang worked for 3 years in the equity investment team at Calibrium AG, managing a few global all-sector concentrated high-conviction strategies. Prior to that, she worked for Lombard Odier as an equity analyst in the thematic team since 2015. Fang spent 4 years as an academic researcher at IMD business school, where she acquired comprehensive research skills and broad industries and sectors knowledge. Fang holds a master’s degree in Management from the University of Lausanne (HEC) and is a CFA Charterholder, a member of the CFA Institute and the CFA Society of Zurich.

Related insights

We’re here to help

Contact us

For general inquiries with UBS Asset Management, fill in a form with your details and we’ll be back in touch.

Our leadership team

Our global leadership team is deep, diverse, and dedicated to our ethos of delivering investment excellence.

Find your local UBS office

As your expert global partner, we're closer than you think. Discover UBS's locations in your region.