Nearfield Electromagnetic Imaging Technique For Brain Tumour Detection

• Main Author: Eustacius Jude, Anak Joseph Linggong
• Format: Thesis
• Language: English
• Published: Universiti Malaysia Sarawak (UNIMAS) 2019
• Subjects: QA75 Electronic computers. Computer science; TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://ir.unimas.my/id/eprint/26684/1/Eustacius%20Jude%20Anak%20Joseph%20Linggong%20ft.pdf; http://ir.unimas.my/id/eprint/26684/

Brain tumour is among the deadliest disease that is fatal if left untreated. An early detection is vital to ensure a successful control of the tumour from metastatic. The availabilities of head imaging machines such as Magnetic Resonance Imaging (MRI) Machine, Computed Tomography (CT) Scan and Positron Emission Tomography (PET) Scan are widely accessible to patients in hospitals. These machines are able to provide a good diagnostic result. However, imaging machines have limitations, in terms of usage restriction and possible long-term effects. Therefore, a method of utilising electromagnetic wave is proposed. Microwave is safer information's carrier to obtained information of the brain tumour due to the distinction of the electrical properties between tumour and normal head tissue. A 2D head model of a transverse plane is created. The outcomes of the simulation expected are to produce image results of good quality and spatial resolution to determine the shape, size and location of the tumour. A program has been developed in C++ to execute the simulation of a head imaging. The simulation consists of two main techniques; a FDTD technique for computational electromagnetics, and Forward-Backward Time-Stepping (FBTS) technique, an algorithm to solve inverse problems. The simulation setting is setup in an active microwave imaging configuration and the main challenge of utilising microwave imaging is the penetration of signals into the head model weaken with high frequency. Therefore, an image segmentation is employed to enhanced tumour detection on the targeted area. Two type of results exhibits in this thesis; FTBS technique and FBTS technique with image segmentation. The results obtained without image segmentations able to detect a smaller tumour. By applying image segmentation to localised the FBTS technique, the detection of the tumour is enhanced. The estimation of the dielectric properties of the tumour is more accurate with image segmentation. Thus, the tumour is located accurately.