Professor Paul Leeson
Professor of Cardiovascular Medicine at the University of Oxford, and Founder and CMO at Ultromics.
Preliminary results of a large-scale study demonstrating the clinical benefits of EchoGo® Heart Failure were shared at the American College of Cardiology (ACC 2023) conference last month.
Using only an apical four-chamber ultrasound view, this new AI-enabled platform can identify heart failure with preserved ejection fraction (HFpEF) and has been shown to minimize the indeterminate classification of HFpEF in both undefined and well-defined cohorts. 
In this post, Professor Paul Leeson, Professor of Cardiovascular Medicine at the University of Oxford and Founder and Chief Medical Officer at Ultromics, discusses the clinical and operational benefits of integrating the company’s HFpEF platform in cardiac imaging to support diagnostic confidence, early preventative care, and clinical efficiency.
AI’s progress in reducing the crisis of indeterminates
Due to the heterogeneity of HFpEF, diagnosis remains a challenge, and optimal selection for clinical trials and potential targeted therapy remains difficult for clinicians to obtain.
To help clinicians improve HFpEF classification, a proposed solution came with the 2016 European Society of Cardiology (ESC) recommendation for a diagnostic algorithm for this syndrome. Since then, scoring systems have been published to purportedly improve diagnosis in patients with suspected HFpEF. These scoring systems, however, are not always accurate. According to literature, at least 30% of cases are classified as indeterminate, which means that a significant number of patients are likely to require additional invasive hemodynamic testing that is technically complex, costly, and increases risk for the patient.
With this background, we worked with Mayo Clinic to develop and study the generalizability of EchoGo® Heart Failure, a novel AI device that automates HFpEF classification, against current clinical scoring systems.
Preliminary results shared during a poster session at ACC 2023 showed that while using EchoGo® Heart Failure, 74% of patient cases were correctly re-classified that were otherwise classed as indeterminate by clinical scoring systems, including HFA-PEFF and H2FPEF.
While using EchoGo® Heart Failure, clinicians were able to identify more HFpEF cases that may have otherwise been missed. The echocardiography solution can improve accuracy, time to detection and diagnosis, benefiting overall clinical daily work.
In the 1,284 patient cases studied, diastolic assessment was indeterminate in 820 patients (64%) using the HFA-PEFF score, and 776 of cases were Indeterminate (60%) using the H2FPEF score.
The platform enhances clinicians’ ability to follow the 2022 AHA/ACC/HFSA Guideline for the Management of Heart Failure, which highlights the prominence of echocardiography and emphasizes the need for prevention, which may help mitigate the health and economic burden associated with heart failure.  EchoGo® Heart Failure can ultimately bring patients closer to a definitive diagnosis, accelerating personalised treatment plans and enabling earlier diagnosis and prevention, improving quality of life.
How EchoGo® Heart Failure integrates with workflows
EchoGo® Heart Failure uses AI to directly analyse a patient’s scan, analysing thousands of pixels which go unnoticed by eye, identifying the presence of HFpEF, and sending a report through the PACS system to support diagnosis. The platform is based on a three-dimensional convolutional neural network and is trained to detect HFpEF using only a single apical four-chamber view.
This replaces complex conventional scoring practices with a single platform that enables care teams to identify more HFpEF cases, improve efficiency, and helps improve the quality of lives for their patients, using a unique software-as-a-service model to provide the best experience for providers.
The platform has a proven ability to produce fewer indeterminates in comparison to the HFA-PEFF score and the H2FPEF score. This allows patients to avoid invasive hemodynamic testing whilst reducing technical complexity and unnecessary costs for clinicians.
The cloud native connection enables flexible clinical workflows and easy access to data to support collaboration and drive confident decision-making, all while simplifying the pathways to diagnosis.
As a full, vendor neutral solution, it is accessible in any care location and integrates with existing ultrasound systems and hospital IT infrastructure.
The combination of components of Ultromics’ echocardiography solution to help drive clinical benefits include:
Exceptional precision in the detection of HFpEF, with high accuracy using only a single apical four-chamber view.
Proven ability to produce fewer indeterminates in comparison to the HFA-PEFF score and the H2FPEF score.
Supports earlier intervention; non-urgent HFPEF cases can be detected and monitored in outpatient clinics and caught early before hospital admission.
Streamlined workflow and improved operational performance throughout the cardiovascular care pathway.
Simple integration into both outpatient and hospital environments with scalable, vendor-neutral report viewing.
Akerman AP, Porumb M, Beqiri A, et al. Comparison of clinical algorithms and artificial intelligence applied to an echocardiogram to categorize risk of heart failure with preserved ejection fraction. Journal of the American College of Cardiology. ACC. 2023;81 (8_Supplement) 360.
Nielsen OW, Valeur N, Sajadieh A, et al. Open Heart Journal. Echocardiographic subtypes of heart failure in consecutive hospitalised patients with dyspnoea. 2019. 20;6:e000928
Heidenreich PA, Bozkurt B, Aguilar D, et al. 2022 AHA/ACC/HFSA Guideline for the Management of Heart Failure: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Journal of the American College of Cardiology. 2022;79:263-421