Medical expert systems are a type of artificial intelligence accessed through computer software that helps medical practitioners, such as hospital doctors, nurses and general practitioners, make informed decisions about patient care. These systems help doctors evaluate, diagnose and, ultimately, treat patients, and they are particularly useful for solving problems in areas in which the doctor is not an expert. For example, a patient might display signs of an infection to the doctor, who might suspect that it is blood-related but requires more information from a trusted knowledge base — the online medical expert system. Among the uses for medical expert systems could be to determine whether a patient should leave the general practitioner's office and go to the hospital, to prioritize in a triage situation and to prevent doctors from becoming stuck in diagnostic habits.
When a doctor uses a medical expert system, he typically enters his patients' symptoms directly into the program's user interface and receives potential diagnoses in return. Those diagnoses are created in the development stages of the software, in consultation with experts on various medical ailments, then stored by the software developer in the program's massive database. The crux of medical expert systems is the advice and information they store from elite medical practitioners, whose expertise provides the foundation for the software's knowledge base. Through these systems, those leading experts in their fields are able to share their insights regarding patient treatment 24 hours a day, seven days a week. Like other expert system applications, medical expert systems might be available online, with password protections, or through a private network or intranet, such as a hospital might have.
Medical expert systems can help hospitals and doctors' offices improve quality of care, reduce costs and help with compliance issues mandated by the government or insurance companies. Traditionally, doctors have treated patients by coming up with a diagnosis and then determining whether the symptoms fit that diagnosis. Medical expert systems handle the situation in reverse: given the symptoms presented, they determine what diseases the patient might have. Proponents suggest that it is a more intuitive way to make an accurate diagnosis, sometimes referring to the traditional way as "backward."
One limitation of medical expert systems is that they seem to work most effectively when diagnosing patients who have one primary symptom. Another is that they do often depend on a physical examination. Therefore, for the most accurate results, they should be used in conjunction with or by an attending physician or nurse.