Written by Sergio Sancho Azcoitia, Security Research en ElevenPaths
Since its appearance a few years ago, expert systems have been applied in different fields. Due to their effectiveness they have inmproved projects that need their application in the clinical field. We will briefly remind you what expert systems consist of and some of their possible applications, in this case associated with the field of medicine.
An Expert System (ES) is a program that uses a series of acquired human knowledge to solve problems or perform tasks that would normally be solved by expert humans. In many cases, these programs work better than any human expert because of their effectiveness and speed when making decisions.
At present, the expert systems constitute the area of application of AI with greater success in the world of medicine. Expert systems allow the storage and utilisation of knowledge of one or several human experts in a specific applied domain. The use of advanced tools such as expert systems increases productivity and decision-making efficiency, which is essential for solving problems when experts have doubts or are not present.
Below, we will briefly introduce you to some of the expert systems that have triumphed in the world of medicine:
- MYCIN: Designed at Stanford by Edward Shortliffe, it is capable of diagnosing infectious diseases in the blood and prescribing the appropriate antibiotics.
- PUFF: Designed in the late 70’s with the collaboration of Robert Fallat, (specialist in pulmonary diseases), able to diagnose lung diseases.
- CADUCEUS: Originally from the University of Pittsburgh, it is used for the diagnosis of internal medicine.
These expert systems are very useful and can greatly facilitate the work of professionals. Despite this, many experts in the world of medicine have expressed their concern about the idea that in the future the entire decision-making process could be in the hands of machines, rendering the role of doctors irrelevant.
However, it must be clarified that expert systems in medicine are not designed to replace doctors, but to complement their work. At the moment there is no machine capable of simulating the behavior of a doctor or his clinical eye (series of knowledge acquired by a professional throughout his practice).