This project involves the development of a novel method for the in-vivo quantitation of pure beta-emitting labeled antibodies proposed for antibody therapy (32P and 90Y). The use of a conventional high energy collimated gamma camera is proposed, operated in the planar, tomographic and planar whole body imaging modes for serial localized, relative and absolute measurements, and whole body retention respectively. A new class of image restoration filters is proposed, an order statistic-neural network hybrid (OSNNH) filter, which may partially compensate for the enhanced image degradation in bremsstrahlung detection and yield an effective linear attenuation coefficient. Conjugate counting geometry is therefore proposed for both planar modes to achieve relative and approximate absolute measurements. Compared to a Wiener restoration filter, the OSNNH technique demonstrates better noise suppression, stability, and image restoration. It also preserves image content and measured counts and should therefore prove useful as a pre-reconstruction filter for the tomographic detection mode increasing the accuracy of the proposed absolute measurements. Extensive preliminary data using 32P already demonstrate the feasibility of the proposed research plan for each detection mode and the advantages of the OSNNH filter. The physical characteristics of 90Y should yield comparable results with enhanced sensitivity due to the more energetic beta particles but without significant resolution loss because of the restoration properties of the OSNNH filter. The anticipated errors in the proposed measurements, although larger than those observed for single-photon detection, will not affect the validity of the results which should prove valuable for antibody therapy management.
| Qian, W; Clarke, L P (1996) A restoration algorithm for P-32 and Y-90 bremsstrahlung emission nuclear imaging: a wavelet-neural network approach. Med Phys 23:1309-23 |
