Assessement of the Scintigraphic Image Quality in Single Photon Emission Computed Tomography (SPECT) using Jaszczak Phantom

ABSTRACT Single Photon Emission Computed Tomography (SPECT) has advantages over other methods in that true 3-D images are easily obtained, appropriate radiopharmaceuticals are relatively less costly and more easily used, and the technology is particularly well-suited for imaging certain types of tissue. However, current methods for deriving SPECT images require design tradeoffs between sensitivity and spatial resolution. Failure to correct for image degradation could lead to low quality images and hence wrong diagnosis. This project aims at improving Scintigraphic images using a written MatLab code and filters. This study was conducted using the Single Head Single Photon Emission Computed Tomography machine at the Nuclear Medicine Department of the Korle-Bu Teaching Hospital. The e.cam® SPECT machine, Data Spectrum phantom, Technetium-99m and the ImageJ software were the materials used. A scintigraphic image was obtained with the e.cam® SPECT machine using a Low Energy High Resolution (LEHR) collimator. Tomographic images were acquired at 128 projections round the Jaszczak phantom. A MatLab algorithm code was applied to improve the contrast of the image and then five filters were applied to the images and comparative studies made. The contrast of the MatLab enhanced image was significantly better than that of the acquired image and the filters applied greatly increased the resolution of the acquired image. The acquired image had improved in quality by applying the MatLab algorithm code and filters. Improvement in quality is expected if the MatLab code is applied to acquired diagnostic Nuclear Medicine images which would help Nuclear Medicine Physicians make more accurate diagnostic decisions