Ultromics, a leader in AI-driven cardiology tools, has shared new research on how artificial intelligence can help in echocardiography at the 2025 American Society of Echocardiography Scientific Sessions in Nashville. The study, published as an abstract in the Journal of the American Society of Echocardiography, shows how AI can help detect cardiac amyloidosis earlier, a condition that often isn’t diagnosed until after years of unexplained heart failure symptoms. Early detection is important because new treatments are most effective when the disease is caught early.
Health Technology Insights: ConcertAI Appoints Casey Graves to Lead Clinical Solutions
The research looked at data from 4,815 patients across 17 hospitals in the US and UK. They used Ultromics’ EchoGo Amyloidosis AI system to see how it could improve referral decisions in real hospital settings. The AI was able to find cardiac amyloidosis more accurately and faster than traditional methods. This helped doctors find patients who might have been missed and reduced the number of unnecessary tests. The results were consistent whether the patient group had a low or high number of cases, showing how useful AI can be in everyday medical practice.
Health Technology Insights: Biocon Opens First U.S. Manufacturing Plant in New Jersey
In areas where there are fewer cases, relying only on wall thickness correctly referred about 65% of patients, but adding AI brought that up to between 76% and 80%, which means quicker diagnoses and fewer unnecessary referrals. In places with more cases, AI reduced unnecessary referrals by up to 18% without affecting detection rates. These findings were the same across hospitals in both countries, showing the technology’s wide potential.
Cardiac amyloidosis is becoming more recognized as a common cause of heart failure. New drugs like tafamidis and acoramidis can slow the disease and lower death rates, but they only work well if diagnosed early. Unfortunately, current medical practices miss about two-thirds of cases. Dr. Ashley Akerman, Director of Clinical Sciences at Ultromics and lead author of the study, said patients often have years of unexplained symptoms and damage before they are diagnosed. She said the EchoGo Amyloidosis AI tool can improve standard heart ultrasound exams to better spot those at risk, reduce unnecessary procedures, and ensure patients who need more testing and treatment get it sooner.
EchoGo Amyloidosis works by looking at echocardiograms at the pixel level to find subtle signs that even experienced doctors might miss. It was developed and tested using data from over 7,000 patients and nearly 10,000 echo videos from 15 international centers, and was also tested on 2,700 patients at 18 sites. The model has excellent accuracy with an AUC of 0.93 across different real-world groups. This AI tool provides consistent, automated evaluations that help doctors spot high-risk patients early, make better decisions about further testing, and connect more people with life-saving treatments.
This study adds to the growing body of clinical evidence supporting Ultromics’ EchoGo platform, which is the first AI system for echocardiography approved by the FDA and covered by Medicare. The platform has been featured in over 25 peer-reviewed studies and is currently used in top US hospitals such as UChicago Medicine, Northwestern, and City of Hope. There, it helps doctors detect complicated heart conditions earlier and manage patients more effectively.
Health Technology Insights: Corstasis Names Dr. Amin Medjamia VP of Medical Affairs
To participate in our interviews, please write to our HealthTech Media Room at sudipto@intentamplify.com
