A model that accounts for the interdependence of extent and severity in the automatic evaluation of myocardial...

The extent and severity are two important parameters when describing a regional defect in myocardial single-photon emission computed tomography (SPECT) perfusion imaging. Intuitively, these two parameters should be independent of each other, but we have shown in a previous study that there is an interdependence. This interdependence has been investigated in two commercially available analysis programs (AutoQUANT and 4D-MSPECT) using Monte Carlo-simulated SPECT data.

An anthropomorphic digital computer phantom and a Monte Carlo program were used to generate SPECT data. Several defects of different volumes and lesion activity uptake reductions were simulated and evaluated. Comparison of the measures of extent and severity with their corresponding known values led to the development of a correction model based on least-squares parameter estimation. This model was then applied to a test group consisting of 10 different computer-simulated patients.

Our results showed that the interdependence was reduced significantly for most of our test cases after applying the correction method. The application of the correction model to the test group demonstrated that the model was robust with respect to different patient geometries. A further test with projections that simulated a perfect SPECT system revealed that the interdependence between the extent and severity was not caused by the limited spatial resolution of the SPECT system, but rather the inherent design of the algorithms of the analysis programs.

A model has been developed to take into account the interdependence of the extent and severity.