Currently, up to one-third of all nuclear medicine studies are performed for cardiovascular disorders, and mostly are myocardial perfusion imaging (MPI) using SPECT acquisition. MPI aims at detection of acute myocardial ischemia and the scarred myocardium and it is increasingly used to plan myocardial revascularization and to assess the effectiveness of medical and surgical interventions. However, nonuniform gamma-ray attenuation in the thoracic region may severely impair the accuracy of SPECT cardiac imaging and frequently result in image artifacts. Therefore, attenuation corrected must be applied. We have derived an algorithm (called EM-IntraSPECT) that is able to reconstruct tomographic cross-sections of a patient from SPECT studies alone. The reconstructed cross-sections in turn, may be used for attenuation correction of cardiac SPECT images thus improving quality analysis based of these images. In this research we intend to continue evaluation of the new attenuation compensation algorithm and to assess the clinical usefulness of this method. These goals can be reached by comparing the patient's cross sectional anatomical data reconstructed by EM-IntraSPECT and obtained from MRI. It is proposed that 10 patients already undergoing nuclear cardiac imaging for clinical purpose, with 99mTc sestamibi or 99mTc radiolabeled red cells will be selected. These patients would be utilized to reconstruct transmission images and determination of internal anatomical structures in the thoracic region from SPECT projections alone. Further, direct and quantitative comparison of the patient's anatomy obtained by the MRI and the SPECT methods will be performed. The same algorithm would be used for reconstruction of attenuation compensated emission SPECT cardiac images. Finally, the accuracy and the clinical usefulness of this new, attenuation corrected cardiac SPECT image reconstruction would be quantitatively evaluated. All these procedures apply only to the image data and NOT the patient.