Avtor/Urednik		Babić, Ankica; Krusinska, Ewa; Bodemar, Goran; Wigertz, Ove; Mathiesen, Ulrik; Franzen, Lennart
Naslov		Building a knowledge-base for non-specified liver diaeases
Tip		članek
Vir		In: Babić A, editor. Medical knowledge extraction: application of data analysis methods to support clinical decisions. Linkoping: Department of biomedical engineering, Medical informatics,
Leto izdaje		1993
Obseg		str. 137-50
Jezik		eng
Abstrakt		Discovering as yet unseen knowledge from clinical data is of importance in the field of unspecified liver diseases. Making a decision to confirm presence of disease is difficult even for an experienced clinician. Knowledge extraction by means of suitable data analysis can therefore benefit clinical decisions. Avoiding a liver biopsy, the final investigation for disease confirmation, by making a decision based on relevant laboratory findings only, would be considered an essential support. This might improve the cost-effectiveness of a standard medical examination, as wel. Thus, information technology is likely to contribute to health care procedures. The system based on Qinlan's ID3 algorithm was simple and efficient to use. The knowledge extracted was to be used togive a biopsy recommendation or not. Basic principles concerning the systems are described, as well as a medical evaluation after which some of the diagnostic rules were found to be useful as decision algorithms i.e.they could be directly applied in clinical work up, and become a part of knowledge-base of automated decision support system.
Deskriptorji		LIVER DISEASES DECISION SUPPORT TECHNIQUES DECISION TREES