S-Space College of Medicine/School of Medicine (의과대학/대학원) Dept. of Radiation Applied Life Science (대학원 협동과정 방사선응용생명과학전공) Theses (Ph.D. / Sc.D._협동과정 방사선응용생명과학전공)
Enhancement of 3-dimensional positioning accuracy in PET detector for high resolution PET system
고해상도 PET 시스템을 위한 PET 검출기의 3차원 위치 정보 정확도 향상 기술 개발
- 의과대학 협동과정 방사선응용생명과학전공
- Issue Date
- 서울대학교 대학원
- Positron Emission Tomography; PET detector; High resolution; Depth-of-interaction; Inter-crystal scattering
- 학위논문 (박사)-- 서울대학교 대학원 : 의과대학 협동과정 방사선응용생명과학전공, 2018. 2. 이재성.
- The positron emission tomography (PET) is a widely used imaging modality that provides biological information at the molecular level. The biological information in molecular and cellular level enables a new discovery in both pre-clinical studies and clinical cases. However, due to the fundamental limits of the spatial resolution in the PET system, the effectiveness of PET is limited when diagnosing small-sized lesions. Hence, improving the spatial resolution in PET is important for the maximization of the diagnosing power of the PET system.
In this thesis, studies on enhancing 3-dimensional (3D) positioning accuracy in PET detector for the high resolution PET system were conducted and presented. The depth-of-interaction (DOI) encoding/decoding and inter-crystal scattering (ICS) event identification technologies were developed and evaluated in the PET detector and system level.
Firstly, the DOI encoding PET detector was developed and detector performances were evaluated. Maximum-likelihood estimation based DOI decoding methodology was developed and optimization studies in several aspects were conducted to achieve the high z-axis positioning accuracy. Secondly, based on the developed DOI encoding/decoding technologies, a prototype DOI PET system was developed and system-level performances were evaluated. Phantom and animal imaging studies were conducted to evaluate imaging performances of the prototype DOI PET system. The proposed DOI encoding/decoding technology was successfully demonstrated at the system level showing its feasibility for the high resolution PET application. Thirdly, a new ICS event identification method was proposed: a new technology of classifying and identifying ICS events in PET detectors with light sharing design, which was not feasible with existing technologies. The proposed method was validated by conducting simulation and experimental studies. By recovering identified ICS events, which is improving x- and y-direction positioning accuracy in the PET detector, improvement in the PET intrinsic spatial resolution was observed. In conclusion, the technologies developed in this thesis enhanced the spatial resolution of the PET system.