Calibrated Methods for Quantitative PET/CT Imaging

DESCRIPTION (provided by applicant): The overall goal of this STTR project is to develop calibrated 3D quantitative analysis tools that directly support improved quantitative accuracy in clinical trials using PET/CT imaging. Pocket phantom technology willbe extended to characterize, perform quality assurance, and improve measurement of PET/CT scans on a per-patient scan basis. We will extend a 2nd generation CT 'pocket phantom', developed by Kitware, to include a PET scanner calibration process, developedby the University of Washington (UW) that is based on National Institute of Standards and Technology (NIST) traceable calibration sources. We will also extend automated algorithms to detect and measure the phantom, and calculate key PET image characteristics. During the Phase I of this SBIR we will demonstrate the feasibility of our approach and identify the components needed to further develop the PET/CT quality assurance system. The Phase II project will conduct a large clinical study and complete all necessary work to market and sell PET/CT calibration and measurement phantoms and analysis services. This proposal is a partnership between Kitware, Inc. and the Imaging Research Laboratory at the University of Washington. Kitware is a leader in quantitativemedical imaging algorithm development and the creation and support of open-source imaging software (e.g. VTK, ITK, ParaView). Kitware has successfully developed and deployed CT pocket phantoms to characterize the fundamental imaging characteristics of CT acquisitions. The University of Washington team has extensive experience in PET/CT imaging and participated in the development of the research prototype PET/CT scanner as well as establishment of the PET scanner calibration method using NIST-traceable sources. The specific aims of the research are to: 1. Develop a PET/CT pocket phantom based on a long-lived 68-Ge source with cross calibration to an NIST-traceable dose calibrator source. 2. Develop automated pocket phantom detection and 68-Ge source measurement algorithms. The latter will estimate fundamental PET/CT scanner calibration and resolution properties that can be used to both measure and reduce the variability of PET/CT measurements. 3. Evaluate the performance of the PET/CT pocket phantom in conjunction with the phantom detection and measurement methods in realistic imaging scenarios. PUBLIC HEALTH RELEVANCE: The goal of this project is to develop a PET/CT pocket phantom and automated analysis software for scan specific PET/CT imaging characterization, quality assurance, and to improve quantitative measurements. The proposed methods have the potential to significantly improve image quality and quantitative measurements in PET/CT clinical trials.