Utilizing the coincidence and anti-coincidence counting of crosstalk events to enhance measurement accuracy in BNCT-SPECT

Single Photon Emission Computed Tomography (SPECT) in Boron Neutron Capture Therapy (BNCT) plays a vital role in real-time monitoring of treatment effects. However, image quality can be degraded by background noise. We propose a novel noise-reduction technique that utilizes both coincidence and anti-coincidence counting methods, leveraging the crosstalk phenomenon between adjacent scintillators. To evaluate the feasibility of this approach, we measured γ-ray spectra from a 137 Cs standard source using two GAGG scintillators. The results showed noise reduction improvements of 15.9% with coincidence and 3.8% with anti-coincidence methods. Additionally, theoretical analysis demonstrated that this approach is effective in multiple scintillators arranged by the forward nature of Compton Scattering. This approach offers a promising path toward improving BNCT-SPECT imaging quality by effectively discriminating target signals from noise in high-density detector arrays. • BNCT-SPECT plays an important role in evaluating the treatment effect of Boron Neutron Capture Therapy in real-time. • Since GAGG Scintillator has high density and energy resolution, it is a promising detector for the BNCT-SPECT. • The Crosstalk Phenomenon can contribute to enhancing the noise reduction processed by coincidence/anti-coincidence methods simultaneously. • Simulation results using T-deposit 2 in PHITS were successfully benchmarked by the experiment with two scintillators, which demonstrated that the coincidence events can be analyzed in this tally. • The proposed method could significantly reduce the background noise, suggesting that this approach becomes a promising way to evaluate the real-time treatment effect in BNCT.