Premium
Kabisa: an interactive computer‐assisted training program for tropical diseases
Author(s) -
Ende J,
Blot K,
Kestens L,
Gompel A,
Enden E
Publication year - 1997
Publication title -
medical education
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.776
H-Index - 138
eISSN - 1365-2923
pISSN - 0308-0110
DOI - 10.1111/j.1365-2923.1997.tb02568.x
Subject(s) - tropical disease , competence (human resources) , curriculum , computer science , bayes' theorem , disease , medical education , bayesian probability , artificial intelligence , medicine , data science , psychology , pathology , social psychology , pedagogy
SUMMARY In Europe, tropical pathology is usually taught in special short courses, intended for those planning to practise in developing countries. The theoretical knowledge to be assimilated during this short period is considerable, and turning such newly acquired knowledge into competence is difficult. Kabisa is a computer‐based training program for tropical diseases. Instead of concentrating on strictly tropical diseases, students are trained in recognizing diseases in patients presenting randomly in an imaginary reference hospital in a developing country. Databases are compiled by experts from experiences in various parts of Africa, Asia and tropical America. Seven languages and three levels of competence can be chosen by the student. Updating of all databases is possible by teachers who want to describe a particular setting. A ‘consistency checker'verifies the internal consistency of a new configuration. The logical engine is based upon both a ‘cluster'and a Bayesian logic, with built‐in corrections for related disease characteristics. This correction allows calculated probabilities to stay closer to real probabilities, and avoids the ‘probability overshoot'that is inherent to ‘idiot Bayes'calculations. The program provides training in diagnostic skills in an imaginary second‐line setting in a tropical country. It puts tropical and cosmopolitan diseases in perspective and combines applied clinical epidemiology and pattern recognition within varying sets of presenting symptoms. Students are guided in searching for the most relevant disease characteristics, in ranking disease probability, and in deciding when to stop investigating.