Landmark Study Shows Ultromics’ EchoGo® Amyloidosis Significantly Improves Detection of Cardiac Amyloidosis with Echocardiography

• Cardiac amyloidosis is an often-missed, life-threatening cause of heart failure. As new disease-modifying therapies emerge, timely and earlier diagnosis is critical.
  
  
• EchoGo® Amyloidosis is the first FDA-cleared AI screening tool for cardiac amyloidosis using an echocardiogram and has received Breakthrough Device designation from the FDA.
  
  
• Study results demonstrate accuracy of EchoGo® Amyloidosis in detecting cardiac amyloidosis, with strong performance across AL, ATTRwt, and ATTRv subtypes.
  
  Full study: https://academic.oup.com/eurheartj/article-lookup/doi/10.1093/eurheartj/ehaf387 

Oxford, UK, July 09, 2025 /PRNewswire/ - A large-scale, multi-center international study published in the European Heart Journal has shown that EchoGo® Amyloidosis, an AI-powered tool developed by Ultromics, significantly improves the screening of cardiac amyloidosis from a standard echocardiogram. It is the first FDA-cleared AI tool for this condition and has also received Breakthrough Device designation from the FDA.

Researchers from Ultromics and Mayo Clinic, with investigators at The University of Chicago Medicine and collaborators around the world, validated and tested the model in a large and multiethnic patient population, and compared its performance to conventional diagnostic methods.

The findings demonstrate that EchoGo® Amyloidosis is highly accurate, achieving 85% sensitivity and 93% specificity.¹ The model performed consistently well across all major cardiac amyloidosis subtypes, and crucially distinguished the disease from phenotypically similar conditions such as hypertensive heart disease, HFpEF, and hypertrophic cardiomyopathy, conditions that often contribute to missed or delayed diagnosis.¹

Geographic representation of the separate training, tuning, and international, multi-ethnic external validation cohorts which included patients withcardiac amyloidosis (CA) and controls referred for transthoracic echocardiography. The artificial intelligence model, based on a single apical four chamber echocardiographic videoclip, was validated in the entire external validation cohort of 2719 patients, in which prevalence of CA was 22%,with AUC 0.93. The model’s accuracy was maintained in testing in various subgroups.

“EchoGo® Amyloidosis is a breakthrough tool that can help us identify at-risk patients so they can receive the treatment they need,” said Ross Upton, CEO and Founder of Ultromics. “It uses deep learning to analyze a single routine echocardiography videoclip to deliver insights, helping clinicians decide when further investigation is needed. Early identification is critical in guiding access to therapies that are most effective when initiated at earlier stages of disease.”

Cardiac Amyloidosis: An Urgent Clinical 

Cardiac amyloidosis is a serious and often underdiagnosed cause of heart failure, driven by abnormal amyloid protein deposits, either light chain (AL) or transthyretin-derived (ATTRwt and ATTRv), that stiffen the heart and impair its function. Symptoms often mimic those of other cardiac conditions, making diagnosis challenging. As many as two-thirds of cases may be missed clinically. ^2-4 Early diagnosis is crucial, as new drug therapies such as Tafamidis are now available that can slow or halt disease progression.⁵

The condition is especially difficult to identify in patients with heart failure with preserved ejection fraction (HFpEF), a common but diagnostically complex subtype of heart failure. Studies suggest that an estimated 15% of HFpEF patients may have underlying cardiac amyloidosis,⁶ a hidden burden that often goes unrecognized.

Study Design and Key Findings

The clinical study evaluating EchoGo® Amyloidosis followed a rigorous two-phase process:

• Development & Optimization Phase: Conducted at Mayo Clinic using 9,786 patients, including 1,349 biopsy-confirmed cardiac amyloidosis (CA) cases and 1,263 matched controls, to train and refine the deep learning model.
• External Validation Phase: Conducted across 2,719 patients at 18 global centers, where the AI was independently tested against gold-standard diagnostic criteria

The external validation cohort included a broad range of institutions including The University of Chicago Medicine, Columbia University Irving Medical Center, Brigham and Women’s Hospital, University of Pennsylvania, The Ohio State University Wexner Medical Center, University of Washington, Hospital of the University of Occupational and Environmental Health (Japan), Instituto do Coração – INCOR (Brazil), ICBA and Centro Privado de Cardiología (Argentina), The University of Texas MD Anderson Cancer Center, NorthShore University HealthSystem, University of Virginia Medical Center, Boston University, MedStar Health Research Institute, University of Leicester (UK), and Beth Israel Deaconess Medical Center.

EchoGo® Amyloidosis was trained using apical 4-chamber echocardiographic video clips and validated against established diagnostic benchmarks, including biopsy and Tc-PYP imaging.

The AI demonstrated high diagnostic performance, achieving 85% sensitivity and 93% specificity, indicating its ability to detect cardiac amyloidosis accurately from a single routine echocardiogram.¹

EchoGo® Amyloidosis demonstrated strong performance across all major subtypes of cardiac amyloidosis, with sensitivities of 84% for AL, 85% for ATTRwt, and 86% for ATTRv.

In a high-risk subgroup of HFpEF patients with increased wall thickness, EchoGo® Amyloidosis maintained strong diagnostic performance, demonstrating potential utility in one of the most diagnostically challenging settings in cardiovascular care.¹

In comparative analysis, EchoGo® Amyloidosis outperformed two validated clinical scoring systems, the Transthyretin Cardiac Amyloidosis Score (TCAS) and the Increased Wall Thickness Score (IWT). The AI model demonstrated an AUC of 0.921, significantly exceeding TCAS (0.74) and IWT (0.80) in diagnostic accuracy.¹

Decision curve analysis showed EchoGo® Amyloidosis identified 36.4% more true positive cases and reduced unnecessary referrals by 6.9% compared to the next best method.¹

Discrimination and classification of the artificial intelligence model, transthyretin cardiac amyloidosis score, and increased wall thickness score. (A) Receiver operating characteristic curves for the artificial intelligence model, transthyretin cardiac amyloidosis score, and increased wall thickness score. Decision thresholds for the artificial intelligence model, the transthyretin cardiac amyloidosis score, and the increased wall thickness score
were 0.06, 6, and 8, respectively. The reported statistics and confidence intervals represent the median, 2.5th and 97.5th percentiles from bootstrapping. (B) Positive predictive value of the artificial intelligence model, transthyretin cardiac amyloidosis score, and increased wall thickness score according to modelled disease prevalence. AI, artificial intelligence; AUC, area under the curve; IWT, increased wall thickness; NPV, negative predictive value; PPV, positive predictive value; Sens, sensitivity; Spec, specificity; TCAS, transthyretin cardiac amyloidosis score

“EchoGo® Amyloidosis achieved high diagnostic accuracy across a broad spectrum of patients and clinical environments,” said Patricia A. Pellikka, MD, Vice Chair, Department of Cardiovascular Medicine, Mayo Clinic, and senior author of the study.

“In our subgroup analysis of older adults with HFpEF, where diagnosis is particularly challenging, the model not only maintained strong performance but also significantly outperformed traditional clinical and transthoracic echo-based screening methods. These results highlight its potential to improve early detection, reduce diagnostic uncertainty, and enhance patient care.”

EchoGo® Amyloidosis is FDA-cleared and currently in use across multiple U.S. centers. It is part of Ultromics’ growing AI portfolio, which also includes EchoGo® Heart Failure, an FDA-cleared device designed to aid in the detection of HFpEF, reimbursable under Medicare and commercial payer pathways, including Category III CPT Code 0932T for outpatient use and NTAP (XXE2X19) coverage for inpatient settings. Both tools operate through the EchoGo® platform, delivering diagnostic and clinical decision support from standard echocardiographic video clips, while integrating seamlessly into existing workflows to enable timely, informed care.

“Current approaches to detecting cardiac amyloidosis on echocardiography often rely on markers that are either unreliable in contemporary clinical settings or time-consuming to implement consistently across high-volume echocardiography laboratories,” said Jeremy A. Slivnick, MD, co-author and Assistant Professor at The University of Chicago Medicine.

“With its ability to provide fully automated detection of cardiac amyloidosis using a single apical 4-chamber view, EchoGo® Amyloidosis offers a practical alternative that can be seamlessly integrated into routine workflows without compromising diagnostic performance.”

Ultromics continues to advance the field of cardiovascular imaging by integrating AI and deep learning into everyday practice. Its mission is to support earlier detection, smarter triage, and broader access to therapies that are most effective when introduced at earlier stages of disease.

Full study: https://academic.oup.com/eurheartj/article-lookup/doi/10.1093/eurheartj/ehaf387 

References:

1. Slivnick, Hawkes et al., Eur Heart J (in press).
2. González-López E, et al., Eur Heart J. 2015;36:2585–94.
3. Hahn VS, et al., JACC Heart Fail. 2020;8:712–24.
4. AbouEzzeddine OF, et al., JAMA Cardiol. 2021;6:1267–74.
5. Maurer MS, et al., N Engl J Med 2018;379:1007–16.
6. Hahn VS et al, JACC Heart Fail. 2020;8:712–724.