Semester

Fall

Date of Graduation

2020

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

Gianfranco Doretto

Committee Member

Raymond Raylman

Committee Member

Alexander Stolin

Committee Member

Donald Adjeroh

Committee Member

Powsiri Klinkhachorn

Abstract

Biochemical processes are chemical processes that occur in living organisms. They can be studied with nuclear medicine through the help of radioactive tracers. Based on the radioisotope used, the photons that are emitted from the body tissue are either detected by single-photon emission computed tomography (SPECT) or by positron emission tomography (PET) scanners. SPECT uses gamma rays as tracer but gives a weaker contrast and spatial resolution compared to a PET scanner which uses positrons as tracer. PET scans show the metabolic changes occurring at the cellular level in an organ or a tissue. This detection is important because diseases begin at the cellular level and PET scans can detect these changes at a very early stage. To detect these changes machine learning plays an important role, mainly in sensing the position of positron emissions from the tissue. \tab In this work, we first generated a dataset of images representing the photon distribution on the PET photodetectors, using the Geant4 Application for Tomographic Emission (GATE) simulation software package and the DETECT2000 software environment. Second, we designed and developed a fully connected and two different convolutional neural network models to correlate the depth of interaction (DOI) with the shape of the photon distribution on the photodetectors, and to detect the position of the annihilation photon interactions inside the PET scanner. Experimental results show that the top performing network can detect the annihilation photon interactions or event positions with higher accuracy than current methods, on a very large dataset.

Share

COinS