3738

3532

كارشناسي ارشد

مخابرات

اصفهان :دانشگاه صنعتي اصفهان ، دانشكده برق و كامپيوتر

1386

يازده ،100،[II]ص. : مصور ،جدول،نمودار

ص.ع. به فارسي و انگليسي

رسول امير فتاحي

سعيد صدري

29/08/86

مهندسي برق و كامپيوتر

ID3532

به فارسي و انگليسي :قابل رؤيت در نسخه ديجيتال

AbstractCoronary heart disease is now the leading cause of death worldwide 3 8 million men and3 4 million women die each year from coronary heart disease Many elderly people sufferfrom coronary artery disease Early detection is important to prevent the risks associatedwith this disease For these reasons approaches for early detection of CAD are veryimportant The commonly used method for detection of CAD is angiography which isinvasive time consuming and costly Noninvasive clinical detecting methods such asECG echocardiography and heart scan can not detect the coronary artery disease in anearlier stage In this thesis we use phonocardiography which is noninvasive method andcan detect CAD in initial stages We decompose normal and abnormal heart sounds using wavelet and wavelet packettransforms Then we compute several factors on resulted coefficients Some of thesefactors are used providing a feature vector as the input to the neural network Neuralnetwork is used for classification of normal patients from abnormal ones

رسول امير فتاحي

سعيد صدري