Abstract
The sentinel lymph node (SLN) hypothesis is applied as part of the standard procedure for identifying early-stage breast cancer. Thus, an imaging system that can locate SLNs in operating rooms is desired. Several 2-D probe imaging systems and a freehand single-photon emission-computed tomography (fhSPECT) system have been proposed. However, 2-D probe imaging systems are affected by shine-through and shadowing effects. Here, we proposed an alternative to 3D imaging systems, i.e., a dual-photon emission computed tomography (DuPECT) system, which integrates both preoperative and intraoperative information to locate SLNs using cascade photons emitted isotopes such as Se-75 and In-111. The system consists of a LaBr3-based probe and planar head, a collimation system, and a coincidence circuit. When two photons from each disintegration were detected simultaneously, the slat and parallel-hole collimator define a plane and a line, respectively, which represent the possible flight paths of each photon. SLNs can be located using the line-plane intersection. In this study, Se-75 was used to evaluate the DuPECT concept, performance, and optimization of collimator configurations using Monte Carlo software developed in our laboratory. The result of the performance evaluation indicates that the randoms rate increases with increased initial activities, while the scatter rate is lower than 1.2 count/s for various activities. The sensitivity is 0.23±0.01 cps/MBq, which is significantly lower than that of most 2-D probe imaging systems (6.5–2,200 cps/MBq). In a simulated imaging study, four injection sites and two LNs placed at various depths are minimally distinguishable. However, the LNs are clearly identifiable in the absence of injection sites, indicating that photons emitted from the injection sites seriously deteriorate the image quality. To reduce the influence of injection sites, a pinhole-slat collimation system was proposed. Preliminary results show that the pinhole-slat collimation system succeeds in eliminating photons emitted from injection sites. In addition, a feasibility study of In-111 was conducted with a delay-time-window technique. In-111 was another potential cascade isotope for its appropriategamma energies (171 and 245 keV), short half-life (2.8 days), and relative low dose equivalent. Preliminary result indicates that In-111 is not appropriate for the DuPECT system due to its relative long half-time (85 ns) of the 245 keV gamma-ray. The number of random events increases significantly, leading to failed SLNs identification, as a wide coincidence time window is needed to accommodate the long life-lived 245 keV gamma. The proposed three-dimensional imaging system has the potential to identify injection sites and SLNs. However, difficulties with the low sensitivity for LN detection and in the choice of appropriate radioisotope must be overcome before its clinical usage.