Author/Authors :
Ahmad Bustami, Fadzlul Rahimi Universiti Kebangsaan Malaysia - MEMS-Automotive Research Group, Department of Mechanical and Materials Engineering, Malaysia , Md Saad, Mohd Hanif Universiti Kebangsaan Malaysia - MEMS-Automotive Research Group, Department of Mechanical and Materials Engineering, Malaysia , Mohd Nor, Mohd Jailani Universiti Kebangsaan Malaysia - MEMS-Automotive Research Group,Department of Mechanical and Materials Engineering, Malaysia , Aziz, Bilkis Banu Universiti Kebangsaan Malaysia - Faculty of Medicine - Pediatric Department, Malaysia
Abstract :
This paper describes heart abnormalities classification procedures utilising features obtained from time-frequency spectogram of ECG heart and image processing techniques. Enhanced spatial features of lime-frequency spectogram were extracted and fed into a forward chaining expert system and the corresponding abnormalities were identified. A confidence factor is calculated for every classification result indicating the degre of belief that the classification is true. It was observed that the classification method was able to give 100% correct classification based on features that was extracted from data sets which were included in the knowledge base and data sets which were not included in the knowledge base.
Keywords :
Heart abnormalities classification , expert system , simultaneous time , frequency analysis.
