Robert M. Hirschfeld, M.D.; Richard D. Judge, M.D., F.A.C.P.
This content is PDF only. Please click on the PDF icon to access.
This study investigates the premise that medical diagnosis is a reasoning process which could be closely simulated by Bayes' theorem. To do this, the diagnosis of three university specialists was compared with that of a computer program in actual cases of congenital and rheumatic heart disease. A significant correlation was demonstrable in 45 cases studied.
The performance of the clinicians correlated closely with that of their corresponding computer model. (The physician generally outperformed the computer.) A test evaluation between computer and clinician showed individual values of 4.8 for Drs. A and B and 3.84 for Dr. C, which is a
Hirschfeld RM, Judge RD. Clinician Versus Computer: A Study of the Application of Bayes' Theorem to Clinical Diagnosis.. Ann Intern Med. 1970;72:777. doi: 10.7326/0003-4819-72-5-777_1
Download citation file:
Published: Ann Intern Med. 1970;72(5):777.
Copyright © 2018 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use