Cardiac Image Reconstruction From Low-dosage SPECT Scans via Deep Learning (CIRFLDSSvDL)

Authors

• Evan Gunderson, Florida Institute of Technology
• Timothy Shane, Florida Institute of Technology
• Alexander Thoms, Florida Institute of Technology

Document Type

Poster

Publication Title

Northrop Grumman Engineering & Science Student Design Showcase

Abstract

● Single-Photon Emission Computed Tomogrophy (SPECT) scans are a type of medical scan ● A patient is injected with a radiotracer, which emits gamma rays that are detected by gamma cameras that rotate around the patient ● Using a series of images collectively referred to as a sinogram, a 3D reconstruction of the patient is created using techniques such as Maximum Likelihood Expectation Maximization (MLEM) ●MLEM takes a long time, but using a neural network can speed up the reconstruction process and allow for a lower dosage of radiotracer to be used

Advisor

Marius C. Silaghi

Publication Date

4-24-2026

Recommended Citation

Gunderson, Evan; Shane, Timothy; and Thoms, Alexander, "Cardiac Image Reconstruction From Low-dosage SPECT Scans via Deep Learning (CIRFLDSSvDL)" (2026). Electrical Engineering and Computer Science Student Publications. 81.
https://repository.fit.edu/ces_student/81