Parametric image formation using clustering for dynamic cardiac SPECT

Dynamic cardiac SPECT imaging can provide quantitative and possibly even absolute measures of physiological parameters. However, a dynamic cardiac SPECT study involves a number of steps to obtain estimates of physiological parameters of interest. One of the key steps involves the selection of regions of interest. In the past, this has been done manually or by using a semi-automatic method. We propose to use cluster analysis to segment the data to obtain improved parameter estimates. The algorithm consists of using a standard k-means clustering followed by a blood input fine-tuning procedure using fuzzy k-means performed to obtain a more accurate blood input function. Computer simulations were used to test the algorithm and to compute bias in kinetic rate parameters with and without the use of blood input fine-tuning. This was followed by performing eight studies in three canines and three studies in two patients with a dynamic cardiac perfusion SPECT protocol. The short-axis slice image data were used as input for the cluster analysis program as well as for a previously validated semi-automatic method. All of the time activity curves were fit to a two-compartment model. Parametric images of the wash-in rate parameter were obtained after cluster analysis. The wash-in rate estimates from the selected regions of interest with both of the methods were compared using microsphere derived flows as a gold standard in the case of canine studies. Our results suggest that in regions with low noise, cluster analysis provides parameter estimates comparable to the semi-automatic method in addition to providing improved visual defect localization and contrast. Moreover, the clustered curves have less noise and yield reasonable fits where with the semi-automatic method the fitting routine sometimes failed to converge. The use of clustering also required less manual intervention than the semi-automatic method. These results indicate that use of clustering may bring dynamic cardiac SPECT closer to clinical feasibility.