Please use this identifier to cite or link to this item:
Title: Investigation of noise removal algorithms for medical applications
Authors: Ng, Cherlyn Hui Ting
Keywords: DRNTU::Engineering::Electrical and electronic engineering
Issue Date: 2017
Abstract: There are various ways to review, study and diagnose new and existing undesired health symptoms. Two of the common used types are Laboratory diagnosis and Radiology diagnosis. Rather than getting the information from the physical examination of the patient, Laboratory diagnosis rely greatly on laboratory reports and test outcomes. On the other hand, Radiology diagnosis is based on the results of medical imaging. Radiology diagnosis will be the main focus for this project. Radiology diagnosis uses imaging technology to diagnose and treat diseases. By using this method of diagnosis, health professionals such as doctors, will be able to see the inner structure of the body and hence, review the cause of the undesired health symptoms. Imaging technology can be used to diagnose different types of illnesses like breast cancer, colon cancer, heart disease e.g. Imaging technology is also used to monitor the health conditions of the patient. One of the examples is to monitor how well the body system is responding to a certain treatment or medication that has been given to treat the disease. There are numerous types of diagnostic radiology. Computed Tomography (CT), also known as CAT (Computerized Axial Tomography), uses X-rays to produce images of the inner structure of the body. Ultrasound is a diagnostic radiology, which uses high frequency sound waves to capture images of the inner structure of the body. Magnetic resonance imaging (MRI) uses both magnetic field and radio wave signals to create images of the organs and structures in the body. The advantage of MRI is that it provides different information about the structure inside the body which can be seen with CT and Ultrasound scan. It also provides information which cannot be shown in both CT and Ultrasound scan. MRI will be the main focus for this project. However, unwanted signals like noise, often exist in the medical images. With the existing of noise in the radiology images, it creates challenges for health professionals to effectively analyze and diagnose the health symptoms that the patient is having. In diagnostic radiology such as medical imaging, to ensure that productive and effective diagnose can be made by the health professionals, it is important that the medical images contains as minimal amount of noise as possible. Noise removal is a process of which unwanted noise is being extracted from the signal. To prevent inaccurate diagnosis of the medical images, it is a challenging problem to make sure that apart from the undesired noise, the rest of information in the medical images are being preserved after performing noise removal practices. There are numerous types of noise removal algorithms for medical applications. This report will be focusing on the removal of undesirable noise using algorithms of appropriate filters that have been implemented by the author. This report will give the explanation of the performance of four types of filters. These four filters are then applied to the medical images that contain different types of unwanted noise. The medical images that were being used are MRI images. To analyze which filter is the most effective in removing the particular noise from the particular MRI image, performance measurements are performed, calculated and evaluated.
Schools: School of Electrical and Electronic Engineering 
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
Cherlyn Ng's FYP Report.pdf
  Restricted Access
Cherlyn Ng Hui Ting's Final Year Project Report2.67 MBAdobe PDFView/Open

Page view(s)

Updated on Sep 23, 2023

Download(s) 50

Updated on Sep 23, 2023

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.