CorVista Health Unveils Machine Learning–Driven Noninvasive Approach to Detect Cardiac Ischemia at ACC.26
CorVista Health has announced promising new clinical data supporting a novel, noninvasive diagnostic approach for detecting cardiac ischemia using physiologic signal analysis combined with advanced machine learning. The findings were presented at the American College of Cardiology Annual Scientific Session 2026, held from March 28–30, 2026, in New Orleans.
The study, titled “Noninvasive Ischemia Detection in Symptomatic Patients: A Physiologic Feature Machine-Learned Model,” introduces an innovative methodology designed to improve the identification of ischemia in patients presenting with symptoms suggestive of coronary artery disease. By leveraging physiologic signals rather than traditional imaging, the approach aims to provide clinicians with a more accessible, efficient, and potentially more inclusive diagnostic tool.
A New Paradigm in Ischemia Detection
Cardiac ischemia, a condition caused by reduced blood flow to the heart muscle, remains a major contributor to cardiovascular morbidity and mortality worldwide. Early and accurate detection is critical for guiding treatment decisions and preventing adverse outcomes such as myocardial infarction.
Traditional diagnostic pathways often rely on imaging modalities or invasive procedures, including stress testing, coronary angiography, or advanced imaging techniques. While these approaches have proven clinical utility, they can be costly, resource-intensive, and, in some cases, limited in their ability to detect certain forms of ischemia.
CorVista’s physiologic signal–based model represents a shift toward a more patient-friendly and scalable solution. By analyzing subtle physiologic patterns captured noninvasively, the system applies machine learning algorithms to identify signatures associated with ischemia, potentially enabling earlier and more accurate detection across a broader patient population.
Strong Diagnostic Performance Across Key Metrics
The study results demonstrated robust diagnostic performance for the machine learning model, highlighting its potential as a reliable tool in clinical practice. Key findings included:
- An area under the curve (AUC) of 0.86, indicating strong overall accuracy in distinguishing patients with and without ischemia
- Sensitivity of 90%, reflecting a high ability to correctly identify patients with the condition
- Specificity of 59%, indicating moderate ability to correctly identify those without ischemia
- A negative predictive value (NPV) of 99%, suggesting that the model is highly effective at ruling out disease when the test result is negative
The high sensitivity and NPV are particularly significant, as they support the model’s potential use as a rule-out tool. In clinical settings, such tools can help physicians confidently exclude ischemia in low- to intermediate-risk patients, reducing the need for further testing.
Consistent Performance Across Diverse Patient Groups
One of the most notable aspects of the study was the model’s consistent performance across a range of patient subgroups. The analysis showed no significant variation in diagnostic accuracy based on sex, age, or body mass index (BMI), suggesting broad applicability in real-world clinical populations.
This level of consistency is especially important given the known variability in cardiovascular disease presentation across different demographic groups. A diagnostic tool that performs reliably across diverse populations can help address disparities in care and improve overall patient outcomes.
Detecting Both Obstructive and Nonobstructive Ischemia
The model also demonstrated strong sensitivity in detecting two distinct forms of ischemia:
- Epicardial coronary artery disease (eCAD), which is caused by obstructive plaque in the major coronary arteries
- Ischemia with nonobstructive coronary arteries (INOCA), a condition in which ischemia occurs despite the absence of visible large-vessel blockages
Sensitivity was reported at 91% for eCAD and 86% for INOCA, underscoring the model’s ability to identify both traditional and less visible forms of coronary disease.
INOCA, in particular, represents a significant diagnostic challenge in current clinical practice. Unlike obstructive coronary artery disease, it is not easily detected באמצעות standard imaging techniques, as it often involves dysfunction in smaller vessels or abnormalities in coronary blood flow regulation.
The ability of CorVista’s model to detect INOCA noninvasively could represent a major advancement, as this condition is frequently underdiagnosed and undertreated despite its clinical significance.
Addressing a Critical Gap in Women’s Cardiovascular Health
The findings are particularly relevant for women, who often experience different patterns of coronary disease compared to men. Research has shown that a substantial proportion of women referred for coronary angiography do not exhibit obstructive coronary artery disease, even when they present with persistent symptoms.
In fact, up to two-thirds of women undergoing angiography may receive “normal” results, a scenario that can lead to frustration, delayed diagnosis, and inadequate treatment. In contrast, the majority of men undergoing the same procedure are more likely to have obstructive disease.
This discrepancy is partly explained by the higher prevalence of INOCA among women. Studies indicate that INOCA occurs at rates similar to eCAD in women and is nearly twice as common in women as in men.
Despite its prevalence, INOCA remains underrecognized, largely due to limitations in current diagnostic approaches. As a result, many women undergo invasive procedures that fail to identify the underlying cause of their symptoms, leaving them without clear answers or appropriate management strategies.
By offering a noninvasive method capable of detecting both eCAD and INOCA, CorVista’s technology has the potential to improve diagnostic accuracy for women and address a longstanding gap in cardiovascular care.
Limitations of Current Diagnostic Modalities
Current noninvasive options for detecting INOCA are limited. Positron Emission Computed Tomography is one of the few techniques capable of identifying this condition, but it is not widely available, with access limited to a small fraction of patients.
Invasive procedures such as left heart catheterization can provide definitive diagnosis, but they come with inherent risks and are not routinely used to evaluate INOCA. Moreover, the specialized testing required to identify microvascular dysfunction during catheterization is often not performed in standard practice.
These limitations highlight the need for more accessible, noninvasive diagnostic tools that can accurately identify all forms of ischemia without exposing patients to unnecessary risks or costs.
Potential to Reduce Unnecessary Procedures
The high negative predictive value of CorVista’s model suggests that it could play a valuable role in reducing unnecessary diagnostic procedures. By reliably ruling out ischemia in many patients, the technology could help clinicians avoid unnecessary imaging or invasive testing.
This has important implications not only for patient experience but also for healthcare systems, where reducing the use of costly procedures can lead to significant savings. At the same time, improved diagnostic accuracy can ensure that patients who do require further evaluation are identified promptly and receive appropriate care.
Enhancing Clinical Decision-Making
CorVista’s leadership emphasized the potential of the technology to support more informed clinical decision-making. By providing a noninvasive, data-driven assessment of ischemia risk, the system can help clinicians determine the most appropriate next steps for each patient.
This could lead to more efficient care pathways, with fewer delays and more targeted use of diagnostic resources. For patients, this translates into faster answers, reduced uncertainty, and a clearer path to treatment.
Broader Implications for Cardiovascular Care
The introduction of machine learning–based diagnostics represents a broader trend in cardiovascular medicine, where advanced analytics are increasingly being used to enhance disease detection and management.
By integrating physiologic data with artificial intelligence, tools like CorVista’s model can uncover patterns that may not be visible through traditional methods. This approach has the potential to complement existing diagnostic strategies and expand the capabilities of clinicians in identifying complex conditions.
While further validation and real-world implementation will be important next steps, the data presented at ACC.26 provide strong evidence supporting the potential of physiologic signal–based diagnostics in cardiovascular care.
As healthcare systems continue to seek more efficient and patient-centered approaches, technologies that combine accuracy, accessibility, and noninvasiveness are likely to play an increasingly important role.
The findings from CorVista Health’s latest study highlight a promising new direction in the detection of cardiac ischemia. With strong diagnostic performance, consistent results across diverse populations, and the ability to identify both obstructive and nonobstructive disease, the machine learning model offers a compelling alternative to traditional diagnostic pathways.
By addressing key limitations in current approaches and providing a highly effective rule-out tool, this technology has the potential to improve patient outcomes, reduce unnecessary procedures, and close critical gaps in cardiovascular care—particularly for populations that have historically been underserved.
As innovation continues to reshape the field, CorVista’s physiologic signal–based approach may represent a significant step forward in delivering more precise, accessible, and equitable cardiovascular diagnostics.
About CorVista System®
The CorVista System is an FDA-cleared, non-invasive cardiovascular diagnostics platform designed to analyze cardiac and hemodynamic signals using machine-learned algorithms. The system synchronously collects physiological signals during a brief point-of-care test and applies advanced analytics to identify patterns associated with pulmonary hypertension (PH) and coronary artery disease (CAD). The CorVista System delivers actionable diagnostic insights without the use of radiation, contrast agents, injections, fasting, or exercise. The CorVista System is developed and manufactured by Analytics For Life, Inc. and licensed to CorVista Health, Inc.
About CorVista Health
CorVista Health is dedicated to transforming cardiovascular care through innovative diagnostics that shorten the path from symptoms to diagnosis. By enabling earlier detection of complex cardiovascular conditions, CorVista aims to empower clinicians with actionable insights and improve patient outcomes across diverse care settings.
Source Link:https://www.businesswire.com/
