<?xml version="1.0" encoding="UTF-8"?><?xml-stylesheet type="text/xsl" href="static/style.xsl"?><OAI-PMH xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/ http://www.openarchives.org/OAI/2.0/OAI-PMH.xsd"><responseDate>2026-11-21T01:51:27Z</responseDate><request verb="GetRecord" identifier="oai:macsphere.mcmaster.ca:11375/24862" metadataPrefix="dim">https://prod-ms-be.lib.mcmaster.ca/server/oai/request</request><GetRecord><record><header><identifier>oai:macsphere.mcmaster.ca:11375/24862</identifier><datestamp>2026-09-24T06:31:32Z</datestamp><setSpec>com_11375_271</setSpec><setSpec>col_11375_272</setSpec></header><metadata><dim:dim xmlns:dim="http://www.dspace.org/xmlns/dspace/dim" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:doc="http://www.lyncode.com/xoai" xsi:schemaLocation="http://www.dspace.org/xmlns/dspace/dim http://www.dspace.org/schema/dim.xsd">
   <dim:field mdschema="dc" element="contributor" qualifier="advisor">Shirani, Shahram</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="author">Kargar Samani, Paria</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="department" lang="en_US">Electrical and Computer Engineering</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="accessioned">2019-10-01T02:27:16Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="available">2019-10-01T02:27:16Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="issued">2019</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">http://hdl.handle.net/11375/24862</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">https://doi.org/10.71548/9336</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="abstract" lang="en_US">High-intensity X-ray radiation through the human body can cause damage to cells,&#xd;
increasing the chance of complications such as cancer. A possible solution is lowering&#xd;
the dosage of radiation. In low dose CT (LDCT) images, fundamental structures are&#xd;
still easily identifiable. However, noise and other additional artifacts are introduced.&#xd;
Removing the visual effects of artifacts caused by lowering radiation dose has been an&#xd;
active area of research in the last few years. Recently, deep learning approaches have&#xd;
demonstrated impressive performance for LDCT denoising. In this thesis, we propose a&#xd;
new machine learning-based approach for LDCT noise reduction that outperforms other&#xd;
methods.&#xd;
Deep Learning is based on the idea of stacking many layers of neurons together. Over&#xd;
the past years, deep learning researchers have successfully optimized the performance&#xd;
of neural networks by stacking more layers. With the growing availability of high performance GPUs as well as more massive datasets, deep learning technology has proven&#xd;
very useful. However, deeper networks are more challenging to train, not because of their&#xd;
computational cost, but due to the difficulty of propagating gradients through so many&#xd;
layers.&#xd;
Deeper neural networks are more complex to train. We present a residual framework&#xd;
to ease the training of networks that are substantially deeper than the others. Residual&#xd;
learning means each subsequent layer in a deep neural network is only responsible for,&#xd;
fine-tuning the output from a previous layer, which is possible only by adding a learned&#xd;
"residual" to the input. This method differs from a more traditional approach where&#xd;
each layer had to generate the total desired output. By using residual learning in LDCT&#xd;
denoising, we prevent degradation of training accuracy once traversing the network and,&#xd;
increase the training pace. Another aspect of our work is using Generative Adversarial Network (GAN), which&#xd;
is a framework for estimating generative models via an adversarial process. We simultaneously train two models a generative model G, that we are using for generating Normal Dose CT (NDCT) and a discriminative model D that estimates the probability that a&#xd;
sample came from the training data rather than G. GANs’ potential is enormous since&#xd;
they can learn to mimic any distribution of data.&#xd;
The novelty of our approach is in combining Residual learning and GAN. For training&#xd;
a convolutional neural network (CNN), a large amount of data is needed. We address this&#xd;
problem by using patch coding. Inspired by the idea of deep learning, we combine the&#xd;
autoencoder, deconvolutional network, and skip connections into residual learning. One&#xd;
motivation for skipping over the layers is to avoid the problem of gradient vanishing. Our&#xd;
experiments show that our method outperforms recent works on LDCT image denoising&#xd;
in terms of Peak SNR (PSNR), and SSIM.</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="degreetype" lang="en_US">Thesis</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="degree" lang="en_US">Master of Electrical Engineering (MEE)</dim:field>
   <dim:field mdschema="dc" element="language" qualifier="iso" lang="en_US">en</dim:field>
   <dim:field mdschema="dc" element="title" lang="en_US">Towards Safer X-rays</dim:field>
   <dim:field mdschema="dc" element="type" lang="en_US">Thesis</dim:field>
   <dim:field mdschema="others" element="access-status">open.access</dim:field>
</dim:dim></metadata></record></GetRecord></OAI-PMH>