The Feasibility of Quad-Modal PET/SPECT/Spectral-CT/CBCT On-Board Imaging in a Small-Animal Radiation Therapy Platform
H. Wang,X. Li,L. Xu,Y. Ren,W. Deng,H. Feng,Z. Yang,S. Ma,Q. Ni,Y. Kuang
DOI: https://doi.org/10.1016/j.ijrobp.2022.07.557
IF: 8.013
2022-01-01
International Journal of Radiation Oncology*Biology*Physics
Abstract:Purpose/Objective(s)While radiation therapy (RT) paradigms have evolved rapidly, preclinical RT techniques, especially molecular/functional imaging guidance techniques, are still far behind, which hampers successful translation of novel RT techniques to bedside practices. In this study, we firstly investigated the feasibility of a highly integrated quad-modal on-board imaging configuration, including positron emission tomography (PET), single-photon emission computed tomography (SPECT), spectral CT, and cone-beam computed tomography (CBCT), in a small animal radiation therapy (SART) platform, which presents a near ideal one-stop comprehensive platform for preclinical RT studies.Materials/MethodsThe quad-modal imaging configuration in a SART platform was modeled by using the GATE Monte Carlo code as a proof-of-concept. A partial-ring on-board PET imaging subsystem was designed using a semiconductor thallium bromide detector. A single cadmium zinc telluride flat panel detector was utilized to obtain on-board SPECT, spectral-CT, and CBCT images of small animals. A preclinical SART workflow based on the quad-modal imaging guidance was designed, with image-guided RT and emission-guided radiation therapy (EGRT) incorporated as biologically guided RT modalities. The spatial resolution, sensitivity, and scatter fraction of the PET subsystem were measured using the simulated phantoms. To demonstrate the quad-model imaging performance of the platform, a simulated phantom with multiple imaging probes, including iodine and radioisotopes of 18F and 99mTc, was imaged. In spectral CT imaging of the phantom, the virtual non-contrast (VNC) electron density and iodine uptake fractions in Kidney1 inserts, mixed with an iodine contrast agent at electron fractions of 0.02 and 0.03, were decomposed and imaged quantitatively by using the Bayesian eigentissue decomposition (ETD) method.ResultsFor the PET subsystem, the spatial resolution was better than 1.2 mm, the absolute peak sensitivity at the platform center with an energy window of 175–560 KeV was 18.5%, and the measured scatter fraction was 3.5% for the mouse phantom described in the NEMA NU-4 standard with a default energy window of 480–540 KeV. In quad-model imaging of the phantom, all imaging probes, including iodine uptake fractions, were imaged clearly within the phantom. The PET and SPECT images of the phantom agreed well with the actual spatial distributions of the tracers within the phantom. The quantitative spectral-CT images of iodine uptake fraction and VNC electron density matched well with the ground truth values within a reasonable error range.ConclusionThe results demonstrated the feasibility of the proposed quad-modal imaging configuration in a SART platform. With the anatomical, molecular, and functional information of tumors being accounted for, this novel design can improve the accuracy of preclinical RT studies, thereby facilitating the successful translation of preclinical studies to clinical practices.