A research team in Japan has unveiled an innovative system based on artificial intelligence, which can detect high blood pressure and diabetes through a short video of no more than 30 seconds. This system, which is innocent, is patent, a video analysis of the skin of the face and the comfort of the hand, which allows it to discover health indicators with accuracy similar to those provided by traditional blood pressure measurement devices. In addition to the ability to determine blood pressure, the initial study has shown that the system can detect the presence of diabetes in its first and second species, making it a possible tool for the early detection of these chronic diseases. The research team strives to develop the system to be available for practical use in medical environments, which can provide a quick and without contact with a health examination, especially in light of the increasing need for techniques that can avoid contact. The system combines high -speed video and an artificial intelligence algorithm, which provides fast and non -contact tests to detect high blood pressure and diabetes of the first and second types, without performing blood tests or using traditional blood pressure measurements. This innovation is shown in the American Heart Association sessions held in Chicago, as this system works through a high speed camera; Video clips of the facial skin and the palm of the hand are recorded at 150 photos per second. The system analyzes wavelength data to detect pulse waves that reflect blood flow in the skin as it uses an artificial intelligence algorithm to discover high blood pressure and diabetes, through the properties derived from the blood flow. The resolution of the results and the system achieved a 94% resolution in the discovery of high blood pressure from the first phase (according to the prescriptions of the American Heart Association for Blood Pressure 130/80 mm HG or higher); Compared to a continuous blood pressure benchmark, it has also shown 86% accuracy in tracking high blood pressure when using a 30 -second video, and 81% when using a video for 5 seconds. The system also showed an accuracy of 75% in identifying diabetics, compared to the results of the cumulative sugar test measuring the blood sugar, and the researchers say they were surprised by the possibility of applying the blood flow algorithm to detect diabetes. It contains some most important complications of diabetes; Surround neuropathy, weakness, pain, numbness in the hands and feet, and other diseases associated with the damage of blood vessels, and the researchers believe it makes sense that the changes in blood flow are a characteristic of diabetes. To detect high blood pressure, the system must include an algorithm that takes into account the disturbance of the heart system or irregular heartbeat. In the future, the typical camera used by researchers must develop an economic sensor that uses only basic wavelengths, and only needs to collect data only a few seconds; The sensor can be added to smartphones, or even installed on a mirror in which a person sits for a few moments, producing it at a low cost.
Creating an artificial intelligence system that discovers chronic diseases without contact
