It is possible to predict when a patient will die with artificial intelligence. According to a new research made by the scientists of the University of Adelaide in Australia, the artificial intelligence system and the developed algorithm is able to predict whether or not a patient will die soon.
Scientists at the University of Adelaide used artificial intelligence technology to analyze the medical imaging of the breast of 48 patients under the age of 60. The developed system succeeded in predicting that the patients they studied on would have died within the next 5 years with 69 percent accuracy. It has been noted that these estimations show great similarities with the estimations of human doctors.
Although researchers are not entirely sure what computers see in computerized tomography scans, they think that artificial intelligence can analyze the data in large quantities so that the doctors can detect shapes and anomalies that they cannot do. The developed artificial intelligence can examine organ changes such as air entrapment between system tissues and organs, cardiac growth, blood vessel clotting, and other vascular disorders.
The author of this research who is a radiologist from the University of Adelaide Public Health School in Australia, Luke Oakden-Rayner said that the assessment of biological age has been limited and the life expectancy of the patient has been limited by the fact that physicians could not look into the body until now and could not measure the health of each organ. He stated that their study has researched how the computer will understand the test results and how the “deep learning” which can be learnt will be used.