Internal radionuclide treatment planning using tomographic data: The extended ULMDOS software

Abstract

1813

Objectives: Before therapy with unsealed radionuclides an accurate dosimetry must be performed for each patient. Planar dosimetric calculations are still state of the art and can be supported by the software tool ULMDOS, which facilitates traceability and documentation as a prerequisite for good clinical practice. Now an extended version of ULMDOS is available to process also tomographic data to overcome drawbacks of the planar method, e.g., organ overlay and background correction.

Methods: ULMDOS is implemented with IDL 6.1 (Interactive Data Language) for a Windows XP/2000 platform. Serial tomographic scans can be loaded in an ECAT7- or DICOM-format. Each image is corrected for decay and scaled to the administered activity. The definition of volumes of interest (VOIs) is supported by different drawing options, e.g., a freehand-, isocontour-, polygon-tool, a region growing algorithm or by a cluster analysis. Image navigation is possible in coronal, sagittal or transaxial view through single slices resp. stacked slices, presented as maximum intensity projection for simultaneous drawing of regions of interest. In a 3-dimensional viewer the resulting VOIs can be checked. Residence times can be calculated interactively by fits of the time activity curves of particular organs to exponential functions.

Results: Quantitative 3-dimensional data allow for an exact individual dosimetry. Problems like organ overlay, insufficient attenuation and scatter correction going along with the planar approach can be avoided. Additional information about inhomogeneous uptake within organs are gathered and displayed as histogram or as tabular statistics. To support traceability, documentation, retrospective examination and later processing, all generated data can be saved in binary or ASCII format.

Conclusions: Compared to the planar approach the extended ULMDOS software tool allows for a more accurate dosimetry. Additional organ specific information is obtained. Documentation features - due to shortend loading times - account for a reduced processing time and assure for reproducibility as a prerequisite for good clinical practice.

Research Support (if any): Deutsche Forschungsgemeinschaft (DFG, German Research Foundation), grant no. KFO120 GL 236/6-1.