A NOVEL HYBRID CLASSIFICATION METHOD WITH PARTICLE SWARM OPTIMIZATION AND K-NEAREST NEIGHBOR ALGORITHM FOR DIAGNOSIS OF CORONARY ARTERY DISEASE USING EXERCISE STRESS TEST DATA
Küçük Resim Yok
Tarih
2012
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
ICIC INTERNATIONAL
Erişim Hakkı
info:eu-repo/semantics/closedAccess
Özet
The aim of this study is to investigate the effectiveness of a novel hybrid method, particle swarm optimization with k-nearest neighbor classifier (PSOkNN), on determination of coronary artery disease (CAD) existence upon exercise stress testing (EST) data. The PSOkNN method is composed of two steps. At the first step, one particle which demonstrates the whole samples optimally in training dataset is generated for both healthy and unhealthy patients. Then, at the second one, the class of the test sample is determined according to the distance of the test sample to the generated particles utilizing k-nearest neighbor algorithm. To demonstrate the effectiveness of this novel method, the results of PSOkNN are compared with the classification results of the artificial immune recognition system and k-nearest neighbor algorithm. Besides, reliability of the proposed method on determination of CAD existence upon EST data is examined by using classification accuracy, k-fold cross-validation method and Cohen's kappa coefficient.
Açıklama
Anahtar Kelimeler
Coronary artery disease, Exercise stress testing, Particle swarm optimization, Artificial immune recognition system
Kaynak
INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL
WoS Q Değeri
N/A
Scopus Q Değeri
Q3
Cilt
8
Sayı
5B
