Characterization of Heart Disease in Single Photon Emission Tomography Images Using Imaging Processing

Abstract

Left ventricle functions, ejection fraction, end systolic volume and end diastolic volume, are measured using several noninvasive imaging techniques such as two and three dimensional echocardiography, cardiac magnetic resonance imaging, and different radionuclide methods such as planar multi-gated radionuclide angiography and gated myocardial perfusion single-photon emission computed tomography, these techniques vary considerably regarding precision, ease of use, availability, and costs. This study aim to assess myocardial perfusion at rest and stress condition using image processing concerning the ejection Fraction, end systolic volume and end diastolic volume in myocardial perfusion imaging (SPECT) using a function written in the Interactive Data Language (IDL) software to normalize the image to range of count that classify the muscles of the heart into four classes normal, mild, moderate and severe, and recognize each part of the images as one of the classes. This analytical study consisted of 60 random patients with myocardium disorder, given 8 to 30 mci of TC 99m-MIBI intravenously. The result of this study showed that it is possible to estimate the ejection fraction with an accuracy of 90%, end systolic volume with accuracy 95% and end diastolic volume with and accuracy of 83% of the heart at stress status test from rest status test without stressing the patients. In conclusion this study prove that; it is possible to estimate the classes of stress test from rest test without stressing the patients, and hence overcome the potential danger of the stress condition for unfit heart.