Title :
Rough set techniques for medical diagnosis systems
Author :
Ilczuk, G. ; Mlynarski, R. ; Wakulicz-Deja, A. ; Drzewiecka, A. ; Kargul, W.
Author_Institution :
HEITEC AG Systemhaus fuer Automatisierung und Informationstechnol., Erlangen
Abstract :
The process of discovering natural phenomena or complex system was until recently limited to finding formulas that fit empirical data. This process used with success in science and engineering has its limits when the complexity of the natural processes increases. Therefore, to analyze data in the medical domain an alternative approach was needed. Several mathematical methods including neural nets, inductive learning fuzzy sets and rough sets were proposed to model data in the form of decision tables or rules. Pawlak´s rough sets theory for handling imprecision and uncertainty in data has a main advantage over the other techniques which is the possibility to analyze data without any preliminary information what favor its usage in medical decision systems. In this paper we present the results of rule generation from our implementation of the LEM2 algorithm on reduced sets of attributes calculated using the Wrapper method with different learning algorithms
Keywords :
data mining; learning (artificial intelligence); medical diagnostic computing; medical information systems; rough set theory; uncertainty handling; LEM2 algorithm; Pawlak´s rough sets theory; Wrapper method; hospital database; imprecision handling; knowledge discovery; machine learning algorithms; medical decision systems; medical diagnosis systems; rule generation; uncertainty handling; Biomedical engineering; Data analysis; Fuzzy sets; Information analysis; Mathematical model; Medical diagnosis; Medical diagnostic imaging; Neural networks; Rough sets; Uncertainty;
Conference_Titel :
Computers in Cardiology, 2005
Conference_Location :
Lyon
Print_ISBN :
0-7803-9337-6
DOI :
10.1109/CIC.2005.1588235
