Author ORCID Identifier
Semester
Fall
Date of Graduation
2025
Document Type
Problem/Project Report
Degree Type
MS
College
Statler College of Engineering and Mineral Resources
Department
Lane Department of Computer Science and Electrical Engineering
Committee Chair
Yenumula Reddy
Committee Member
Raymond Raylman
Committee Member
David Graham
Committee Member
Donald Adjeroh
Abstract
Monte Carlo (MC) simulations are often used to provide insight into complex physical phenomenon. They can generate realistic results without the necessity to construct costly and complex apparatus. The ultimate value of these simulations depend on how accurately the physics can be modeled and if a sufficient volume can be produced to account for statistical variance. In medical imaging device development, MC simulations of Positron Emission Tomography (PET) scanners have typically been used to model the performance of new systems. The imaging group in the Department of Radiology have developed a unique detector design for a pre- clinical PET scanner that utilizes a single, continuous annulus of scintillator (AnnPET). PET is used to map positron-emitting concentration in the small animal in the scanner. Fourteen facets were machined on the outer surface of the annulus to facilitate mounting of arrays of photodetectors that are used to detect the optical photons produced by the interaction of radiation with the scintillator. The challenge is to identify where in scintillator the interaction occurred. This process is complicated by use of the continuous scintillator. Typically, PET detectors utilize pixelated scintillator where the optical photons are constrained to well defined regions, simplifying the localization task. To localize interaction points in the continuous scintillator, we adapted a convolutional neural network (CNN) to learn where these locations are based on the optical photon distributions detected by the photodetectors. The goal of this project was to create a computer model of annular pre-clinical PET scanner, including the physics of the system, to provide the data set to train the CNN to localize event positioning in the annular scintillator, which is necessary to create images of the radiotracer activity object scanned. This problem report details the design, implementation and deployment of the MC simulation of AnnPET used to create CNN training data.
Recommended Citation
Martone, Pete, "Simulation of AnnPET for Neural Network Training and System Performance Benchmarking" (2025). Graduate Theses, Dissertations, and Problem Reports. 13058.
https://researchrepository.wvu.edu/etd/13058