A novel method for the estimation of infarct size in a reperfused rat model for pinhole SPECT

Cardiac SPECT imaging using pinhole collimators can be used to perform in-vivo rat studies in order to evaluate cardiovascular diseases. The estimation of infarct size in a rat model of coronary artery disease, with [^99mTc]sestamibi, is typically done using a single threshold for all the studies. However, the presence of high statistical noise in the data can result in inaccuracies in the infarct size estimates. In this work, a novel method of obtaining infarct sizes using a variation-dependant threshold and spatial information is proposed. First, the mouse body (MOBY) phantom was used to simulate a wide range of infarct sizes for high and low count statistics. In addition, experimental studies consisting of five control and thirteen reperfused rat studies with [^99mTc]sestamibi were performed, by scanning each rat for 1 hr on the pinhole SPECT system. The projection data were reconstructed with OS-EM with penetration compensation and reoriented into short-axis slices. Circumferential profiles were obtained for each study. A relationship was obtained between the threshold and the variation in circumferential profiles for all the control studies. Using this relationship, a threshold was computed for each reperfused study. Since the site of occlusion was known, a region growing technique along with the selected threshold were used to obtain an estimate of the infarct size. The infarct sizes obtained with the proposed method were closer to the corresponding true infarct sizes, in the case of simulated data and triphenyltetrazolium chloride (TTC) staining derived infarct sizes, in the case of experimental rat studies, than those with a constant threshold. The proposed method has the potential to provide accurate estimates of infarct sizes for in-vivo rat studies performed with pinhole SPECT.