Stereoscopic full aperture imaging in nuclear medicine

Abstract

Images of planar scintigraphy and single photon emission computerized tomography (SPECT) used in nuclear medicine are often low quality. They usually appear to be blurred and noisy. This problem is due to the low spatial resolution and poor sensitivity of the acquisition technique with the gamma camera (GC). Other techniques, such as coded aperture imaging (CAI) reach higher spatial resolutions than GC. However, CAI is not frequently used for imaging in nuclear medicine, due to the decoding complexity of some images and the difficulty in controlling the noise magnitude. Summing up, the images obtained through GC are low quality and it is still difficult to implement CAI technique. A novel technique, full aperture Imaging (FAI), also uses gamma ray-encoding to obtain images, but the coding system and the method of images reconstruction are simpler than those used in CAI. In addition, FAI also reaches higher spatial resolution than GC. In this work, the principles of FAI technique and the method of images reconstruction are explained in detail. The FAI technique is tested by means of Monte Carlo simulations with filiform and spherical sources. Spatial resolution tests of GC versus FAI were performed using two different source-detector distances. First, simulations were made without interposing any material between the sources and the detector. Then, other more realistic simulations were made. In these, the sources were placed in the centre of a rectangular prismatic region, filled with water. A rigorous comparison was made between GC and FAI images of the linear filiform sources, by means of two methods: mean fluence profile graphs and correlation tests. Finally, three-dimensional capacity of FAI was tested with two spherical sources. The results show that FAI technique has greater sensitivity (>100 times) and greater spatial resolution (>2.6 times) than that of GC with LEHR collimator, in both cases, with and without attenuating material and long and short-distance configurations. The FAI decoding algorithm reconstructs simultaneously four different projections which are located in separate image fields on the detector plane, while GC produces only one projection per acquisition. Simulations have allowed comparison of both techniques under ideal identical conditions. Our results show it is possible to apply an extremely simple encoded imaging technique, and get three-dimensional radioactivity information for simplistic geometry sources. The results are promising enough to evaluate the possibility of future research with more complex sources typical of nuclear medicine imaging.