A team of scientists led by Sebastian Massaro of the University of Surrey (Britain) taught the neural network to recognize signs of heart failure in patients' ECGs. The accuracy and speed of the diagnostic system is amazing - it makes a 100% accurate diagnosis in just one session. Moreover, AI can also determine morphological features indicating the severity of the disease.

It should be noted that statements about 100% accuracy should be taken critically. At the current stage, AI trained thousands of ECGs of healthy people and patients with severe insufficiency. And therefore, the neural network will definitely distinguish a healthy person from a sick person, but what about those who have just begun to develop a disease and have no symptoms? The problem is that it is much more difficult to find such patients and get their ECG, and therefore there is simply nothing to train the neural network for.

Massaro himself believes that hundreds and thousands of tests are needed before an AI diagnostician can be involved in real work. But he, strictly speaking, is not intended for this. Neurity is optimized specifically for high-speed analysis using small amounts of data that even a small wearable gadget can collect. This is necessary to save people from having to undergo a long and tedious examination in a clinical setting. Instead, compact electronics will constantly monitor their health.

The devices for collecting bio-data are already there and are available to a wide mass of users. But AI for the analysis of the information collected is still in short supply, which is the bottleneck of digital diagnostics. The development of Massaro, as well as the projects of his colleagues from other countries, will help solve this problem.

techcult.ru