Australian biologists have created an artificial intelligence system capable of predicting the life expectancy of a person from one photo of his organs with 69% accuracy, according to an article published in the journal Scientific Reports.
"The prediction of what awaits the patient is extremely important for the medical profession, as it will help them choose the treatment that is best suited for saving and prolonging the patient's life.While such predictions were limited to the experience of doctors and their eyes.We checked whether it is possible to use deep neural Network to solve this problem, "- says Luke Oakden-Rayner (Luke Oakden-Rayner) from the University of Adelaide (Australia).
The Magic of Artificial Intelligence
For this training, Oakden-Reiner and his colleagues used a collection of several thousand photographs of the chest and abdominal cavity, obtained with a tomograph during the health observations of 40 patients. This modest set of images, according to scientists, was enough to ensure that their offspring could reach the level of predictions, which are usually demonstrated by physicians trying "by sight" to determine the life span of their patients.
Convinced that the system they created correctly correctly predicts the lifespan of photographs of the bodies of already dead patients, the scientists checked how it will cope with work in "combat" conditions. To do this, they recruited a group of eight young and elderly patients, enlightened their chest with a tomograph and observed their lives for the next few years.
