Artificial intelligence will examine the eyes and calculate the risk of a heart attack
At the University of Leeds a system with artificial intelligence (AI) that analyzes eye scans taken during routine visits to the optometrist or optometrist and identifies those who are at high risk of heart attack. The system analyzes changes in the miniature blood vessels of the retina, which we know to be on a broader cardiovascular problem . indicate
The Leeds specialists used deep learning techniques to train the AI to automatically read the scans and identify the people most likely to have a heart attack within the next year.
The system that Nature Machine Intelligence is characterized by an accuracy of 70-80 percent and, according to the developers, could be used to diagnose Cardiovascular diseases be used.
Cardiovascular diseases, including heart attacks, is the leading cause of death worldwide and the second leading cause of death in the UK. These are chronic diseases that affect the quality of life. This technique has the potential to revolutionize diagnostics. Retinal scanning is an inexpensive and routine procedure used by many eye exams used, says Professor Alex Frangi, who led the development of the new system. People who are examined by an ophthalmologist or optometrist can find out if they have a cardiovascular disease Suffer. This allows treatment to be started earlier, before other symptoms appear.
The system of artificial intelligence was trained on the ophthalmological and cardiological data of more than 5.000 people. It has learned to distinguish between pathological and normal conditions. Once learned, it was capable of the size and power of the left ventricle determined solely from retinal scans. an enlarged one ventricle is associated with a higher risk of heart disease. The AI can then, by combining data about the condition of the left ventricle with information about the patient's age and gender, predict the risk of a heart attack over the next 12 months.
Currently we can measure the size and function of the left ventricle using the Echocardiography or the MRT determine. These are specialized and expensive tests that are far less accessible than the tests performed by ophthalmologists or optometrists. Not only will the new system reduce costs and increase the availability of early heart diagnosis, but it can also play a major role in countries with less developed health systems where access to specialized tests is very difficult.
This system of artificial intelligence is a great tool for recognizing patterns found in nature. And that's exactly what it does, it links patterns of change in the retina changes in the heart, enthuses Professor Sven Plein, one of the authors of the study.