November Journal of Cardiac Failure Focus Issue on Artificial Intelligence: AI Echo Raises HFpEF Detection to 78%

November Journal of Cardiac Failure Focus Issue on Artificial Intelligence: AI Echo Raises HFpEF Detection to 78%

PR Newswire

The November Journal of Cardiac Failure focus issue on artificial intelligence in heart failure also features a large language model study of why patients with heart failure go without SGLT2 inhibitors, plus a review of AI-enabled care from detection to chronic management

WASHINGTON, Nov. 17, 2026 /PRNewswire/ — Adding an artificial intelligence (AI) analysis of a single routine echocardiogram clip to the H₂FPEF clinical score identified 78% of patients with hemodynamically confirmed heart failure with preserved ejection fraction (HFpEF), compared with 48% using the score alone. That’s according to research published in the November 2026 issue of the Journal of Cardiac Failure (JCF), an issue focused on artificial intelligence in heart failure. JCF is the official journal of the Heart Failure Society of America (HFSA).

Journal of Cardiac Failure

Researchers applied an AI model (EchoGo HF) to resting echocardiograms from 280 outpatients referred for suspected HFpEF. Each patient also underwent exercise hemodynamic testing, the diagnostic gold standard. The AI model added the most value for patients with low or intermediate H₂FPEF scores, the group where diagnosis is hardest. The authors therefore tested a two-step strategy: calculate H₂FPEF first, then apply AI only when the score is inconclusive. An AI-detected HFpEF result also independently predicted elevated filling pressures and reduced exercise capacity. That suggests the model reflects disease severity as well as diagnosis.

Why a Focus Issue on AI in Heart Failure

In their Editor’s Page, Realizing the Potential of Artificial Intelligence in Heart Failure, guest editors Jonathan W. Cunningham, MD (Mass General Brigham/Harvard Medical School), Elena Donald (Columbia University Irving Medical Center), Nir Uriel, MD (Columbia University Irving Medical Center) and Ankeet S. Bhatt, MD (Kaiser Permanente San Francisco Medical Center) explain their reasoning. Heart failure’s population burden is growing and outcomes are worsening, while the number of heart failure physicians is flat and training spots go unfilled. AI could help close that gap.

“Promise alone is insufficient,” the editors write. “Like any innovation intended to change clinical care, AI must be rigorously studied to understand where it adds value, where it does not and how it can be implemented safely and effectively.”

The editors call on the heart failure community to move past retrospective and in silico demonstrations of model performance. Instead, they want prospective studies, including randomized trials where feasible, that test whether AI tools improve clinical decisions, treatment delivery, outcomes, equity and clinician workload. “AI researchers will continue to advance these tools, but clinicians and patients must lead their translation into care,” they write.

The November issue also includes the following articles:

Large Language Models Reveal Why Patients With Heart Failure Aren’t Prescribed SGLT2 Inhibitors

Many patients with heart failure who qualify for an SGLT2 inhibitor under guidelines never get a prescription. Using clinical notes from 599 patients seen in cardiology clinics, investigators optimized a large language model (OpenAI’s o1) that classified the documented reason for non-prescription with 84.7% accuracy. Physician-attributed factors were the most common theme (28.7% of notes). The leading specific reasons were deferring the decision to a future visit (15.4%) and the physician judging the patient low-risk or unlikely to benefit (7.4%). Those findings point to concrete targets for improving guideline-directed prescribing.

Review: Embedding AI in Heart Failure Care From Early Detection to Chronic Management

This review maps current AI applications across the heart failure care continuum:

  • AI-enabled ECG and imaging for opportunistic detection of low ejection fraction, HFpEF and structural heart disease
  • Risk stratification based on electronic health records
  • Clinician decision support
  • Remote medication titration
  • Wearable monitoring

The authors conclude that the strongest evidence favors AI built into active clinical workflows that guide targeted testing, treatment optimization and follow-up, rather than AI used only to send passive alerts.

Full Issue Available Online

The complete November 2026 issue of the Journal of Cardiac Failure includes these featured articles along with additional original investigations, reviews and clinical research. It is available online now. For interviews with authors, please contact Laura Poko at lpoko@hfsa.org.  

About the Journal of Cardiac Failure

The Journal of Cardiac Failure (JCF) publishes the highest quality science in the field of heart failure with a focus on diversity, equity, and inclusion, mentorship, multidisciplinary partnerships, and patient-centeredness. Published papers span original investigator-initiated work to state-of-the-art reviews, guidelines and scientific statements, expert perspectives, early career and trainee spotlight pieces, patient and patient-partner narratives. JCF also emphasizes the power of language and prioritizes innovative approaches to dissemination of published work to reach and impact the broader heart failure community.

About the Heart Failure Society of America

The Heart Failure Society of America, Inc. (HFSA) represents the first organized effort by heart failure experts from the Americas to provide a forum for all those interested in heart function, heart failure, and congestive heart failure (CHF) research and patient care. The mission of HFSA is to provide a platform to improve and expand heart failure care through collaboration, education, innovation, research, and advocacy. HFSA members include physicians, scientists, nurses, nurse practitioners, pharmacists, trainees, other healthcare workers, and patients. For more information, visit hfsa.org.

Media Contact: Laura Poko, 301-798-4493, ext. 226, lpoko@hfsa.org 

Heart Failure Society of America

Cision View original content to download multimedia:https://www.prnewswire.com/news-releases/november-journal-of-cardiac-failure-focus-issue-on-artificial-intelligence-ai-echo-raises-hfpef-detection-to-78-302942797.html

SOURCE Heart Failure Society of America