AI Detects Hidden Heart Disease That Doctors Missed
A 45-year-old man who arrived at a Queens emergency room coughing up blood and struggling to breathe was sent home with an asthma diagnosis and an inhaler. But an artificial intelligence program analyzing his routine electrocardiogram (ECG) had already flagged a far more serious problem: structural heart disease so advanced that he would ultimately need a heart transplant. The case, published in Nature Medicine on June 22, represents the world’s first heart transplant attributable to AI-based disease detection.
A Missed Diagnosis, an AI Catch
Louie Quiros, a caregiver and security guard, arrived at a Queens emergency room in February 2025 with alarming symptoms: coughing up blood, difficulty breathing, and a rapid heartbeat. A chest X-ray showed no abnormalities. Doctors diagnosed asthma and sent him home with an inhaler.
But EchoNext, an AI program developed at NewYork-Presbyterian Hospital and Columbia University, had analyzed Quiros’s routine ECG less than 10 minutes after it was performed. The AI flagged structural heart disease — a condition that human doctors had missed. According to ColumbiaDoctors, the detection led to a heart transplant for Quiros, described as a world first for AI-driven disease detection.
How EchoNext Works
EchoNext is a deep learning model trained on more than 1.2 million ECG-echocardiogram pairs from 230,000 patients across the NewYork-Presbyterian healthcare system. The AI analyzes routine 12-lead ECG data to identify patterns indicative of structural heart disease that are imperceptible to the human eye. When it flags a patient as high-risk, the clinician is prompted to order an echocardiogram for definitive diagnosis.
“EchoNext basically uses the cheaper test to figure out who needs the more expensive ultrasound,” said Dr. Pierre Elias, founder and CEO of Pathway Labs, the company commercializing the technology, and medical director for AI at NewYork-Presbyterian. “It detects diseases cardiologists can’t from an ECG. We think that ECG plus AI has the potential to create an entirely new screening paradigm.”
Outperforming Cardiologists
In a head-to-head comparison with 13 cardiologists on 3,200 ECGs, EchoNext accurately identified 77% of structural heart problems versus 64% for cardiologists, according to the original study published in Nature in July 2025. When deployed in nearly 85,000 patients who had not previously had an echocardiogram, EchoNext identified 9% as high-risk. Among those who subsequently received an echocardiogram, nearly three-quarters were diagnosed with structural heart disease — twice the rate of positivity compared to patients without AI screening.
“You can’t treat the patient you don’t know about,” Dr. Elias said. “Using our technology, we may be able to turn the estimated 400 million ECGs that will be performed worldwide this year into 400 million chances to screen for structural heart disease.”
FDA Clearance and Widespread Availability
Pathway Labs recently received FDA clearance for EchoNext covering six indications: right-sided heart failure, left-sided heart failure, valve disease, severe hypertrophy compatible with infiltrative cardiomyopathy, and pulmonary hypertension. The company described it as the world’s first multicondition AI clearance in cardiology.
Crucially, EchoNext will be made available through a partnership with OpenEvidence, a clinical decision platform used by over 500,000 US physicians, as reported by STAT News. This integration bypasses the slow, hospital-by-hospital adoption model and makes the technology immediately accessible at the point of care.
“FDA-approved AI shouldn’t sit siloed in the ivory tower while patients wait years for it to reach them,” said Dr. Travis Zack, Chief Medical Officer of OpenEvidence. “Putting EchoNext on OpenEvidence means a breakthrough in heart disease detection is available everywhere care happens, from major hospitals to community practices.”
A New Screening Paradigm
Structural heart disease — including valve disease, cardiomyopathy, and other conditions that impair heart function — affects millions worldwide. Unlike cancer, which has routine screening tools like mammograms and colonoscopies, there has been no equivalent affordable screening test for structural heart disease. Many cases go undetected until significant function has been lost or a major cardiac event occurs.
“While we have mammograms and colonoscopies for cancer, we have never had an equivalent form of early detection for the most common cause of death in the world — heart disease,” Dr. Elias said. “Through EchoNext, we are able to detect high-risk conditions that the human eye can’t and may otherwise be missed.”
Funding and Future Steps
Pathway Labs announced an $8.5 million seed round led by AlleyCorp and Breyer Capital to expand deployment across health systems, grow its clinical and commercial teams, and support ongoing research and development, according to the official announcement.
“One of the most compelling opportunities in medical AI is uncovering clinically meaningful signals from data we already collect,” said Dr. Morgan Cheatham, Partner at Breyer Capital. “Pathway Labs is redefining what can be discerned from one of the most widely ordered tests in medicine, the ECG.”
A randomized controlled trial is planned to further validate the technology’s impact on patient outcomes. If successful, EchoNext could fundamentally change how heart disease is detected — turning every routine ECG into a potential life-saving screening opportunity.
This article was compiled from reporting by the New York Times, STAT News, Nature, Nature Medicine, Columbia University Irving Medical Center, and official announcements from Pathway Labs.