Optimizing the quality of Positron Emission Tomography (PET) image using filtering methods

Abdul Shukor, Lyana Syaheerah (2022) Optimizing the quality of Positron Emission Tomography (PET) image using filtering methods. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

	Text (24 Pages) Optimizing the quality of Positron Emission Tomography (PET) image using filtering methods.pdf - Submitted Version Download (176kB)
	Text (Full text) Optimizing the quality of Positron Emission Tomography (PET) image using filtering methods.pdf - Submitted Version Restricted to Repository staff only Download (4MB)

Abstract

Positron Emission Tomography (PET) is a medical imaging machine that assists in discovering the metabolic or biochemical activity of tissues and organs. A radioactive substance (tracer) is used in the PET scan to show normal and abnormal metabolic activity. However, the image produced by the PET machine seems to be low quality, making it challenging for doctors and physicians to diagnose a disease. This project aims to enhance the quality of the PET image by using the filtering image methods. Therefore, the MATLAB software for reconstructing the PET image is used in this project. The filtering method is used to enhance the quality of the PET images. The filtering methods used in this project are None filter, Hann filter, Hamming filter, Cosine filter, Ramp filter, and Sheep-logan filter. Based on the qualitative approach, there are 79 respondents that responds to the survey, and the majority of the respondent agree that the Hann filter is the best filter to enhance the PET image. While from a quantitative approach, based on the image quality metric, the Signal-to-Noise Ratio (SNR) shows that the Hann filter give the highest SNR value. Overall, the result from this research shows an improvement in the quality of the PET image by implementing filtering methods to remove the noise from the PET image.

Actions (login required)

View Item