Nuclear imaging is a commonly used tool in today's diagnostics and therapy planning. For interventional use however they suffer from drawbacks which limit their application. Freehand SPECT was developed to provide 3D functional imaging during an intervention. It combines a nuclear probe with an optical tracking system to obtain its position and orientation in space synchronized with its reading. This information can be used to compute a 3D tomographic reconstruction of an activity distribution. However as there is no fixed geometry the system matrix has to be computed on the fly, using ad hoc models of the detection process. The accuracy of the reconstructions is therefore highly dependent on both the used models and the acquisition geometry. Two different models were evaluated. The first model (M1) computes the amount of activity detected from a single point source over the solid angle subtended by the point source and the front of the probe. The second model (M2) computes the mean lengths that rays from a point source pass through the detector and shielding and uses these lengths to compute the effects of the shielding and the detection process. A set of experiments with six different acquisition geometries was performed using a rectangular phantom containing two hollow spheres filled with 0.5MBq of a Tc99m solution. The different acquisition geometries include scans from only one to all six sides of the phantom. The resulting data sets were reconstructed using both described models and MLEM. These reconstructions were then compared with a CT of the phantom. For the data sets acquired only from one or two sides, model M1 did not reconstruct activity in the correct position, while model M2 showed activity in the correct area but not resolving the two spheres. With the data sets acquired from at least three sides both models gave reasonable reconstructions with a mean error of about 10.04mm with M1 and 8.54mm with M2.
«
Login
BibTeX