Transfer of literature data to virtual patient cases: Development and implementation of a method to generate categorical data using Bernoulli-distributed random values

Abstract

Abstract Background: Teaching medicine is a complex task because medical teachers are also heavily involved in clinical practice. Automatic creation of virtual patient cases would be a great benefit to them, saving them time and providing a wider choice of virtual patient cases for student training. Our aim was to develop and test a computer program which simulates clinical case scenarios based on information about the occurrence of specific symptoms in certain diagnoses from the literature.Methods: Medical literature was searched for suitable diagnoses with information on the respective probabilities of specific symptoms. We developed a computer program that generates virtual patient cases with symptons whose occurrence is randomly generated by Bernoulli experiments, according to probabilities reported in the literature. The number of runs and thus the number of patient cases generated is arbitrary.Results: We illustrate the function of our generator with the exemplary diagnosis “brain abscess” with the related symptoms and the respective probabilities from the literature. With a growing number of repetitions of the Bernoulli experiment, the relative frequencies of occurence increasingly converge with the probabilities from the literature. E.g., the relative frequency for headache after 10.000 repetitions was 0.7267 and, after rounding, equals the mean value of the probability range of 0.73 reported in the literature. The same applies to the other symptoms. Conclusion: The results suggest that an automated creation of virtual patient cases is possible, but with regard to the limitation to symptom constellations, it is not yet suitable for professional use. Based on additional information provided in the literatur, an extension of the generator can be implemented in further research.