ADVERTISEMENT
DRAI MARTINI: Artificial Intelligence Beats Human Readers at Arrhythmia Detection
European Society of Cardiology Press Release
WARSAW, Poland, Sept. 2, 2024 -- New research presented at the ESC Congress 2024 reveals that a novel artificial intelligence (AI) algorithm has significantly higher sensitivity than human specialists in detecting heart rhythm disorders on long ECG recordings. The DRAI MARTINI study tested the AI, finding that conventional methods using human technicians to read ECG led to missed diagnosis 14 times more often than when using artificial intelligence.
The DRAI MARTINI study evaluated whether the DeepRhythmAI (DRAI), an advanced AI technology developed by Medicalgorithmics, could replace human technicians for reporting of ECG data, using data from more than 14,600 patients. The patients were monitored on average for 14 days due to symptoms such as palpitations, fainting, and dizziness. The AI system's performance was compared with traditional ECG analysis conducted by technicians, in collaboration with 50 leading cardiologists from the USA and Europe.
"Long term ECG analysis is time consuming and when human technicians read ECG there is potential for error, leading to missed diagnoses of clinically meaningful arrhythmias. Our study tests a state-of-the-art AI algorithm and finds that the AI had a 14 times lower risk of missing a critical arrhythmia diagnosis" stated Associate Professor Linda Johnson of Lund University, Sweden, lead researcher of the study.
DRAI substantially reduced false negative findings compared to traditional methods. False negative findings occurred in only 3.2 per 1,000 patients using AI, compared to 44.3 per 1,000 patients for technicians. With DRAI analyses reported directly to physicians, critical arrhythmia diagnoses would be missed in 14 times fewer patients. This was the result of a superior sensitivity for detecting critical arrhythmias, at 98.6% (95% CI 97.7-99.4), compared to 80.3% sensitivity by ECG technicians. The high AI sensitivity translated into a negative predictive value of 99.9%, ensuring nearly all patients without arrhythmias were correctly identified.
"The DRAI MARTINI study underscores the value of AI in augmenting the capabilities of ECG technicians and cardiologists. We anticipate that AI will become a standard tool in cardiac diagnostics, supported by research such as the DRAI MARTINI study," commented Przemysław Tadla, CTO at Medicalgorithmics.
Medicalgorithmics is a global provider of software for non-invasive cardiac diagnostics, using proprietary artificial intelligence algorithms to analyze ECG data and heart CT data. For nearly 20 years, the company has been conducting R&D on ECG arrhythmia diagnostic systems.
It is the world’s largest gathering of cardiovascular professionals, disseminating ground-breaking science both onsite in London and online – from 30 August to 2 September. Explore the scientific programme. More information is available from the ESC Press Office at press@escardio.org.
About the European Society of Cardiology
The ESC brings together health care professionals from more than 150 countries, working to advance cardiovascular medicine and help people to live longer, healthier lives.
