Abstract
One thousand and forty-six patients, referred for barium meal examination, were interviewed and subsequently examined by one radiologist. Attempts were made to correlate the data obtained from the patient with radiologic diagnosis, using discriminant function analysis and Bayesian conditional probability calculations. Each patient was classified into one of eight diagnostic categories on the basis of these calculations. The accuracy of this method was compared with that of the clinical diagnoses made by the radiologist before examination. Other studies evaluating statistical methods of medical diagnoses have been confined to situations in which clinical diagnostic accuracy is high. As in these, the present study, which is directed to patients who challenge the acumen of the practising physician, yields a diagnostic accuracy similar to that achieved by the clinician.
Similar content being viewed by others
References
Ledley RS, Lusted LB: The role of computers in medical diagnosis. Med DoKum 5:70–78, 1961
Lipkir M, Engle RL Jr, Davis BJ et al: Digital computer as aid to differential diagnosis. Arch Int Med 108:124–140, 1961
Overall JE, Williams CM: Models for Medical Diagnosis Factor Analysis, Part Two, Experimental. Med DoKum 5:78–84, 1961
Warner HR, Toronto AF, Veasey LG, et al: A mathematical approach to medical diagnosis-application to congenital heart disease. JAMA 177:177–183, 1961
Rinaldo JA, Scheinok P, Rupe CE: Symptom diagnosis. A mathematical analysis of epigastric pain. Ann Intern Med 59:145–154, 1963
Lodwick GS, Keats TE, Dorst JP: The coding of Roentgen images for computer analysis as applied to lung cancer. Radiology 81:185–200, Aug 1963
Lodwick GS, Cosmo LH, Smith WE, et al: Computer diagnosis of primary bone tumors. Radiology 80:273–275, 1963
Wilson WJ, Templeton AW, Turner AH, et al: The computer analysis and diagnosis of gastric ulcers. Radiology 85:1064–1073, 1965
Johnstone AS: Diffuse spasm and diffuse muscle hypertrophy of lower oesophagus. Br J Radiol 33:723–735, 1960
Bock OAA: Radiological diagnosis of gastric atrophy. Br J Radiol 36:578–582, 1963
Fraser GM, Pitman RG, Lawrie JM, et al: The significance of the radiological finding of coarse mucosal folds in the duodenum. Lancet 2:979, 1964
Ostrow JD, Resnick RH: Hyperchlorhydria, duodenitis and duodenal ulcer. A clinical study of their inter-relationships. Ann Intern Med 51:1303, 1959
Rhodes J, Evans KT, Lawrie JH, et al: Coarse mucosal folds in the duodenum. Q J Med 37:151–169, 1968
Ross P: The accuracy of radiological examinations. Med J Aust 1:534, 1964
Bailey NTJ: Probability methods of diagnosis based on small samples, Mathematics and Computer Science in Biology and Medicine. London, Her Majesty's Stationery Office, 1965, 103–110
Vanderplas JM: A method for determining probabilities for correct use of Bayes' theorem in medical diagnosis. Comput Biomed Res 1:215–220, 1967
Anderson TW: Introduction to Multivariate Statistical Analysis. John Wiley and Sons, 1958, pp 147–152
Cornfield J: Discriminant functions, Proceedings of the Sixth IBM Medical Symposium, Oct 1964, pp 1–22
Biomedical Computer Programs. Edited by WJ Dixon. Health Sciences Computing Facility. Department of Preventive Medicine and Public Health, School of Medicine University of California, Los Angeles, Sept 1965
IBM, System/360 Scientific Subroutine Package (360-A-CM-03X), Version 2, 1967
Ross P: Computers in medical diagnosis. Crit Rev in Radiol Sci. To be published
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Ross, P., Dutton, A.M. Computer analysis of symptom complexes in patients having upper gastrointestinal examinations. Digest Dis Sci 17, 248–254 (1972). https://doi.org/10.1007/BF02232297
Issue Date:
DOI: https://doi.org/10.1007/BF02232297