Investigation of Image Quality for Quantitative Lu-177 in SPECT imaging: A Phantom Study

Abstract

A lutetium 177 (177Lu) radiopharmaceutical has been used for a theragnostic agent in molecular radiotherapies. This study aimed to investigate the image quality of SPECT image from 177Lu from Jaszczak Cylindrical Phantom based on tomographic uniformity, local-sphere uniformity, and signal-to-noise ratio (SNR). Data acquisitions were conducted using a SPECT/CT unit. For contrast measurement, six hollow sphere inserts with diameters of 9.9, 12.4, 15.6, 19.7, 24.8, and 31.2 mm were filled by 177Lu with radioactivity concentration 10 times higher than the warm background. Images were reconstructed using three different iterative reconstruction algorithms, including Flash3D, OSEM2D, and Wallis. All reconstructions were carried out with the same iteration number of 4, subset number of 4, and Gaussian Filter. Results for tomographic uniformity measurement were (112.47±8.40%), (114.30±9.59%), and (105.94±17.49%) counts for Flash3D, OSEM2D, and Wallis algorithms, respectively. Flash3D algorithm provided better tomographic uniformity than others, while Wallis algorithm yielded the highest noisy image with low SNR. Local-sphere uniformity and SNR tended to significantly increase for sphere diameters larger than 19.70 mm. It was concluded that the reconstruction method significantly affects the image quality in 177Lu quantification. Then, it seems that Flash3D is the best reconstruction method for 177Lu SPECT image acquisition.