AI technology detects heart disease from ECGs in under two seconds

artificial intelligence electrocardiogram heart disease detection cardiology medical diagnostics

Researchers have developed an artificial intelligence tool capable of detecting heart disease, including heart failure and valvular disease, from an electrocardiogram (ECG) in less than two seconds. The technology, which was presented at the annual European Society of Cardiology congress in Munich, was trained on data from millions of patients.

In a US-based trial involving 67,000 patients, the AI tool successfully identified 81% of heart failure cases and 90% of valvular disease cases. This capability is significant because while standard ECGs are widely used, they often cannot detect structural heart diseases without follow-up echocardiograms, which can involve long wait times.

A separate AI model, known as ECG-CLIP, developed by Scripps Research, utilizes a different approach by training on ECG recordings paired with cardiologists' written notes. This model has demonstrated the ability to predict risks beyond cardiac issues, such as the likelihood of death following surgery or emergency visits, as well as the future onset of type 2 diabetes and chronic kidney disease.

Entities

Ahmed El-Medany · British Heart Foundation · Dr. Sonya Babu-Narayan · ECG-CLIP · European Parliament Research Service · European Society of Cardiology · Imperial College London · Munich · Scripps Research · United States

Claims

What the coverage asserts, and how many sources carry each claim.

[● 7 SOURCES] The findings were presented at the annual ESC congress in Munich. kirkinews.gr · www.efsyn.gr · www.haberaktuel.com · www.diken.com.tr · www.cna.gr · +2 more
[● 7 SOURCES] The technology was trained on data from millions of patients. kirkinews.gr · www.efsyn.gr · www.haberaktuel.com · www.diken.com.tr · www.cna.gr · +2 more
[● 7 SOURCES] The AI tool identifies signs of heart failure and valvular disease from an electrocardiogram (ECG). kirkinews.gr · www.efsyn.gr · www.haberaktuel.com · www.diken.com.tr · www.cna.gr · +2 more
[● 7 SOURCES] A research team has developed an AI tool capable of detecting heart disease in less than two seconds. kirkinews.gr · www.efsyn.gr · www.haberaktuel.com · www.diken.com.tr · www.cna.gr · +2 more
[● 6 SOURCES] In a trial involving 67,000 patients in the US, the AI tool identified up to 81% of cases. www.efsyn.gr · www.cna.gr · www.haberaktuel.com · www.diken.com.tr · www.egeszsegkalauz.hu · +1 more
[● 2 SOURCES] Researchers from Imperial College London developed an AI tool to detect heart disease. www.haberaktuel.com · www.diken.com.tr
[● 5 SOURCES] In a US trial of 67,000 patients, the AI identified 81% of heart failure cases and 90% of valvular disease cases. www.haberaktuel.com · www.diken.com.tr · www.efsyn.gr · www.egeszsegkalauz.hu · hvg.hu
[○ 1 SOURCE] The ECG-CLIP system was trained on electrocardiograms paired with cardiologists' written notes. studyfinds.com
[○ 1 SOURCE] The ECG-CLIP AI model can predict the risk of chronic kidney disease and type 2 diabetes. studyfinds.com