Title :
Diagnosis of cardiac problems from SPECT images by feedforward networks
Author :
Bakirci, O. ; Yildirim, Tulay
Author_Institution :
Yildiz Teknik Universitesi, Istanbul, Turkey
Abstract :
In this work, the performance of the feedforward networks on myocardial diagnosis using SPECT (single photon emission computed tomography) images is investigated. Results of the work shows that feedforward artificial neural networks reached 100% accuracy on diagnosis while machine learning algorithms have achieved maximum 90.4% accuracy up to now.
Keywords :
cardiology; feedforward neural nets; medical image processing; single photon emission computed tomography; SPECT images; cardiac problems; feedforward artificial neural networks; myocardial diagnosis; performance; single photon emission computed tomography; Feedforward neural networks; Ischemic pain; Machine learning algorithms; Multi-layer neural network; Myocardium; Neural networks; Positron emission tomography; Single photon emission computed tomography; Testing; Virtual colonoscopy;
Conference_Titel :
Signal Processing and Communications Applications Conference, 2004. Proceedings of the IEEE 12th
Print_ISBN :
0-7803-8318-4
DOI :
10.1109/SIU.2004.1338268
