Medicalgorithmics’ AI Algorithms Described in Nature Medicine Magazine

Medicalgorithmics’ DeepRhythmAI (DRAI) is revolutionizing heart diagnostics. Independent research has shown 14 times fewer missed diagnoses of critical arrhythmias. A recent publication in Nature Medicine demonstrates the potential and benefits of direct reporting of ambulatory EKG results to doctors.

Medicalgorithmics, a pioneer in cardiac diagnostics based on artificial intelligence, proudly announces the publication of groundbreaking, independent studies in the journal Nature Medicine, demonstrating the excellent performance of the DeepRhythmAI (DRAI) model in diagnosing heart arrhythmias.

The development of DeepRhythmAI represents a paradigm shift in EKG analysis, promising faster diagnoses, better patient outcomes, and healthcare system efficiency, which should translate into shorter waiting times for doctors. The global shortage of EKG technicians and patients’ increasing waiting times create critical demand for innovative solutions in heart care. Analysis of human heartbeats for potentially life-threatening arrhythmias is a complex and often error-prone process. DRAI faces this challenge, offering a solution that significantly outperforms traditional technician-led analysis.

The “DRAI MARTINI” study was conducted by Dr. Linda S. Johnson, Associate Professor at Lund University in Sweden, and Dr. Jeffrey S. Healey, Cardiology Professor at McMaster University in Canada. The study tested DRAI on records previously analyzed by licensed EKG technicians in U.S. clinical practice. This is the largest artificial intelligence study in arrhythmia diagnostics worldwide, involving over 200,000 days of EKG data from more than 14,000 adult patients. Both AI and technician analyses were compared to diagnoses made by panels of three cardiologists who independently reviewed 5,000 randomly selected arrhythmias to establish the final “gold standard” of diagnosis, ensuring impartial evaluation. In total, 17 panels of three cardiologists from leading institutions worldwide participated in the study, providing exceptionally high-quality “beat-to-beat” diagnoses for over 5,000 rhythmic events.

“We tested what would happen if technicians were replaced by DeepRhythmAI for diagnosing arrhythmias on outpatient EKGs. The study was rigorous, employing over 200,000 EKG days, with 50 independent experts involved in providing a remarkably high-quality reference point to which both artificial intelligence and technicians were compared. Quite frankly, the results surprised us. The AI tool has a significant advantage in terms of patient safety, with 14-times fewer missed critical arrhythmia diagnoses – a result of significantly better sensitivity. The implications for patient care are substantial. DeepRhythmAI is good enough to be used for direct physician reporting, improving care access, shortening diagnosis times and, hopefully, improving patient outcomes too,” says Dr. Linda S. Johnson, Associate Professor at Lund University.

DRAI demonstrated significantly higher sensitivity in detecting critical arrhythmias, including atrial fibrillation, complete heart block, pauses, and supraventricular and ventricular tachycardias. The false-negative rate for DRAI was low at 3.2 per 1000 patients compared to 44.3 per 1000 patients for technician analysis – reducing the missed diagnosis rate by 14-times compared to technician analysis. This significantly better accuracy underlines the superiority of DRAI technology in terms of patient safety. It translates into an impressive 99.9% negative predictive value for DRAI as a direct physician reporting tool for excluding the presence of critical arrhythmias in ambulatory EKGs.

The exceptional performance of DRAI is powered by both CNN and transformer models and an unmatched training dataset comprising 250 billion examined heartbeats. This vast data library allows artificial intelligence to learn and identify subtle patterns indicating arrhythmias with exceptional performance. DRAI, as a 100% cloud-based solution, has the ability to directly report ambulatory EKG results to the doctor and can streamline clinical workflows by as much as 65%, which leads to lower healthcare costs, increased access to monitoring, and faster diagnoses, regardless of the number of subsequent sessions.

“The results of this independent study confirm what we have been working on at Medicalgorithmics – using artificial intelligence to transform cardiac diagnostics. DeepRhythmAI’s ability to reduce the number of missed arrhythmia diagnoses by 14-times compared to human analysis is a breakthrough that directly impacts patient safety and clinical efficiency. Thanks to the level of accuracy, which gives doctors the certainty of excluding critical arrhythmias, DRAI sets a new standard for direct EKG reporting to the doctor, reducing the workload on health service staff while providing faster and more reliable diagnoses. This is the future of cardiac care – more precise, scalable, and accessible to patients worldwide,” comments Przemysław Tadla, CTO of Medicalgorithmics S.A.

Utilizing the power of the only FDA-approved AI-based integrative solution with third-party software, DRAI offers a solution that exceeds traditional health care standards, significantly reducing the risk of missed diagnosis and improving the speed and accuracy of cardiac care. This breakthrough technology represents a significant advance in the fight against heart disease.