Design of a PG-SPECT for boron concentration monitoring in BNCT at XJTU

Abstract

Boron Neutron Capture Therapy (BNCT) has garnered significant attention in cancer treatment in recent years. As the efficacy of BNCT relies on the radiation dose deposited in the tumor region through the reaction of ¹⁰B(n, α) ⁷Li, the boron concentration monitoring within the cancer tissue is extremely important. In this study, we have proposed a Single Photon Emission Computed Tomography (SPECT) system by using 478 keV prompt gamma rays emitted from the reaction above for X-BNCT, which mainly consists of three groups of collimator and CZT detector arrays. Simultaneous Algebraic Reconstruction Technique (SART) and Maximum Likelihood-Expectation Maximization (ML-EM), as well as the convex set Adaptive Steepest Descent Projection Onto Convex Sets algorithm (ASD-POCS) were utilized to image reconstruction with projection data simulated by PHITS code for different types of tumor phantoms. The following conclusions were achieved: the highest reconstruction accuracy was obtained by ASD-POCS algorithm under 12 or 18 views, while ML-EM algorithm took much shorter reconstruction time; the tumors located at the depth of 7 cm under subcutaneous tissue can be satisfactorily reconstructed; an acquisition time of 240s was recommended to achieve compromise results in considering of image quality and treatment time under beam current of 20 mA.