The US Food and Drug Administration (FDA) has approved the first artificial intelligence algorithm to detect valvular heart disease using a digital stethoscope. The Register writes about it.
EMAS software, developed by American startup Eko, is designed to analyze patients’ pulses and help healthcare professionals detect heart murmurs.
The data is transferred to the cloud, where the system determines in seconds what type of heart disease the patient may have. EMAS then produces the result in JSON format, which is passed to the application and displayed in human-readable form.
“The software uses signal processing, such as filtering, as well as machine learning-based algorithms, to analyze the information received and generate output to support clinical decisions for clinicians,” the company said.
The developers evaluated the performance of the algorithm according to two criteria typical for cardiology: sensitivity and specificity. According to the first parameter, the accuracy of diagnosis was 85.6%, according to the second – 84.4%. For comparison, general practitioners for the same criteria showed a result of 44% and 69%, respectively.
“Cardiologists typically detect murmurs with greater accuracy than general practitioners, but most patients are seen in primary care,” the company said.
Eko spokesmen called the FDA approval an important step “on the road to commercialization” of the product.
Recall that in April, the manufacturer of the standalone AI X-ray analyzer Oxipit announced its intention to submit an application to the FDA for certification of the device.
In March 2021, the regulator approved the first AI device for diagnosing COVID-19.
Subscribe to ForkLog news on Telegram: ForkLog AI – all the news from the world of AI!
Found a mistake in the text? Select it and press CTRL+ENTER