AI researchers in Sweden and cardiac specialists Brazil have collaborated to build an AI tool that can interpret ECG and detect diseases affecting the heart. The tool can today diagnose cardiac fibrillation and five other common ECG abnormalities.
The AI system has been trained on more than 2 million collected ECGs that heart specialists have manually diagnosed. The result is a system that can diagnose the six abnormalities just as well as a cardiac specialist.
The system is not in use in hospitals yet, but when completed, it can be an important tool to relieve cardiac specialists. This is especially important in low- and middle-income countries with few specialist doctors in relation to the population.
The researchers also see this as another sign of the opportunities that exist for collaboration between AI researchers and medical practitioners.
“This is the first result of a collaboration that we have built up over the past two years. I dare to believe in investing in this type of deep collaboration between researchers in AI and researchers in medicine in order to create new knowledge that can help us people to a better life, "says Thomas Schön, professor of regulatory technology and researchers in machine learning / AI and the person responsible for the technical part of the study.