Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Computer Engineering and Sciences

First Advisor

Debasis Mitra

Second Advisor

Samuel P. Kozaitis

Third Advisor

Eraldo Ribeiro

Fourth Advisor

Marius C. Silaghi


Medical imaging plays a vital role in modern healthcare, enabling clinicians to diagnose and treat a range of conditions. However, image acquisition and processing can be challenging because they can often be hindered by motion blurring, leading to inaccurate results. To address that, this dissertation proposes a novel approach based on nonlinear mathematical transformations and artificial neural networks (ANN). The dissertation begins with an introduction to Nuclear Medicine and the problem of motion blur in image reconstruction. A background on Medical Imaging techniques, including the Radon transform and Image Reconstruction methods such as Filtered Back Projection and Iterative Reconstruction methods are presented. The ANN is introduced, including the Fully Connected Neural Network (FCN) and Convolutional Neural Network (CNN). Then related work is presented, including previous studies on Deep Learning for Medical Imaging and motion correction techniques, such as image denoising and motion correction. Preliminary experiments are conducted to test the viability of using Deep Learning techniques for parameter prediction, synthetic object reconstruction, and motion blur handling. The main contribution of this dissertation is the proposed ANN model for reconstruction from motion blurred sinogram data, and moreover using zero-shot learning to reconstruct the motion-free image. The methodology is described in detail, including the use of a CNN with the Self-Attention mechanism. Experimental results demonstrate the effectiveness of the proposed method in producing accurate image reconstructions, with improved image quality and reduced motion blur. Overall, this dissertation presents a novel approach to image reconstruction for Nuclear Medicine Imaging, using Deep Learning techniques to address the problem of motion blur in sinogram data. The proposed method has the potential to improve diagnostic accuracy and enhance patient care in clinical settings.


Copyright held by author.