Abstract
This article traces the history of the development of artificial intelligence as a science that constantly responds to current problems that arise in medical practice. Attention is drawn to the fact that almost all modern neural systems of medical diagnostics are static. This means that they do not have a time axis, and therefore, can only make diagnoses of diseases at the current time. As a result, doctors have to make prescriptions for prevention and treatment courses without checking on computer models what this may lead to in the future. Thus, consciously or unconsciously, doctors have to experiment on patients, which is an ethical problem. This article shows that this centuriesold ethical problem can be solved by further development and application of modern methods of artificial intelligence. Optimal selection of prevention and treatment courses can be made by virtual predictive experimentation on dynamic computer models of patients.
Keywords: Artificial intelligence, neural network, diagnostics, forecasting, virtual computer experiments, bioethics.
Graphical Abstract
35