Room: AAPM ePoster Library
Purpose: Recently, multi-modality imaging such as PET and CT has been developed for image guidance in Radiotherapy. In this work, we present a novel compact architecture of COplanar Transmission and Emission Guidance in RadioTherapy (Co-TeGRT), which has great potential of multi-modality imaging with benefits in data alignment and registration, and simplified workflow. To assess its feasibility, CT image reconstruction from Co-TeGRT is explored.
Methods: The Co-TeGRT is enabled by leveraging a common detector developed to detect both X-ray and gamma ray photons, and by combining the common detector with a small-sized KVCT detector to form a hybrid radiation sensor (Fig. 1), where PET and CT imaging and MV treatment can all be assembled in the same scan plane on a rotating gantry. Such a compact architecture design balances both PET and KVCT performance. For KVCT, a normal scan field of view KVCT imaging is realized with an interior high spatial resolution (HSR) region from KVCT detector and two-sided low spatial resolution (LSR) regions from the common detector. CT image reconstruction therefore needs careful considerations (Fig. 2), where data completion by projection interpolations (can be improved by neural network method) for HSR and LSR data allows a multi-pass filtering, followed by data fusion and backprojection to generate CT images.
Results: Preliminary results from a numerical simulation of CT imaging in Co-TeGRT are shown in Fig. 3, where CT images reconstructed from the small-sized KVCT only data and from common detector for all data (no KVCT detector inserted) are provided as comparison. Better imaging performance is clearly observed from our proposed Co-TeGRT.
Conclusion: CT image reconstruction from Co-TeGRT using multi-pass filtering is developed and validate the feasibility of our proposed concept of Co-TeGRT. Monte-Carlo simulation using GATE is undergoing and neural network based super-resolution reconstruction is also under development.
Funding Support, Disclosures, and Conflict of Interest: This project was supported in part by the New Faculty Startup Funding of Tsinghua University, Beijing, China (53331100120).