Powerful Medical Receives €40 Million IPCEI Grant — read the full story

Powerful Medical
1. October 2024
1 min to read

Time for a Diagnostic Paradigm Shift From STEMI/​NSTEMI to OMI/​NOMI (DIFOCCULT-3)

Summary

DIFOCCULT-3 is a randomized controlled study actively enrolling patients across 18 sites in Turkey. It evaluates AI-assisted ECG interpretation in detecting high-risk heart attack patterns.

By comparing traditional STEMI/NSTEMI classification with an occlusion/non-occlusion (OMI/NOMI) model, the trial aims to improve acute coronary occlusion detection. The primary composite endpoint includes all-cause mortality and all-cause re-hospitalization at 1-year follow-up, assessing its impact on long-term patient outcomes.

Study DesignRandomized Controlled Trial (RCT)
StatusEnrolling patients
Countries / Sites1 (Turkey) / 18
Study StartOctober 1, 2024
Primary Completion (estimated)October 1, 2025
Enrollment (target)6,000
clinicaltrials.gov ID NCT06570759

Author-Logo_PM
Powerful Medical leads one of the most important shifts in modern medicine by augmenting human-made clinical decisions with artificial intelligence. Our primary focus is on cardiovascular diseases, the world’s leading cause of death.

About PMcardio

PMcardio is a CE-certified AI that reads ECGs and offers a complex assessment of 49 cardiac conditions. Clinically validated in 15+ studies and trusted by over 100,000 clinicians, it delivers rapid, expert‑level interpretations, empowering emergency physicians, GPs, nurses, paramedics, and cardiologists to act with confidence at the point of care. Available for Individuals and Organizations.

About Powerful Medical

Established in 2017, Powerful Medical has embarked on a mission to revolutionize the diagnosis and treatment of cardiovascular diseases. We are a medical company backed by 28 world-class cardiologists and led by our expert Scientific Board with decades of experience in daily patient care, clinical research, and medical devices. The results of our research are implemented, developed, certified, and brought to market by our 50+ strong interdisciplinary team of physicians, data scientists, AI experts, software engineers, regulatory specialists, and commercial teams.

Share this article

Relevant Publications

Bifascicular Block Associated With Myocardial Infarction: A Marker of Proximal Left Anterior Descending Artery Occlusion Confirmed by the Artificial Intelligence-Based Smartphone App Queen of Hearts

This single-patient case report describes an elderly man presenting with chest pain, hypotension, and bifascicular block (BFB)—a combination of right bundle branch block (RBBB) and left anterior fascicular block (LAFB)—whose ECG showed QRS‑concordant anterior and lateral ST‑segment elevation consistent with a STEMI‑equivalent / occlusive myocardial infarction (OMI) pattern. Urgent coronary angiography revealed a long, severely calcified, near‑occlusive proximal left anterior descending (LAD) artery lesion, successfully treated with primary PCI and drug‑eluting stent implantation, achieving TIMI 3 flow. The Queen of Hearts (PMcardio) AI‑based smartphone app correctly classified the ECG as STEMI‑equivalent, identified atrial flutter and BFB, and predicted reduced left ventricular ejection fraction, later confirmed by echocardiography (EF 38%). This case underscores BFB with concordant anterior ST elevation as a high‑risk marker of proximal LAD‑culprit OMI and provides anecdotal evidence that specialized AI‑enabled ECG interpretation can support rapid, accurate decision‑making in ACS. 

Artificial Intelligence Versus Human Expertise: ECG-Based Detection of Occlusive Myocardial Infarction After Cardiac Arrest

This single-centre study tested whether AI-based ECG analysis can detect occlusive myocardial infarction (OMI) after cardiac arrest using post-ROSC ECGs from 97 patients with subsequent coronary angiography. A dedicated deep neural network (Queen of Hearts, QoH) achieved the highest discrimination for acute coronary occlusion (AUC 0.85) and OMI (AUC 0.75), outperforming human experts, with a more balanced trade-off between sensitivity and specificity. In contrast, two large language model–based chatbots (ChatGPT and a GPT-based EKG Analyst) showed near-perfect sensitivity but almost no specificity, labelling nearly all ECGs as OMI and thus providing no meaningful diagnostic discrimination. These findings suggest that specialized ECG-trained AI, such as QoH, may serve as a useful adjunct in post-resuscitation decision-making. In contrast, general-purpose LLMs are currently unsuitable for critical ECG diagnosis.

Join over 100,000 healthcare professionals who are already taking advantage of AI