1. UKM Learning and Research Repository
2. PHD Thesis / Tesis PHD
3. Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina

Title:	Microwave imaging system using heuristically optimized synthetic focusing technique
Authors:	Salehin Kibria (P78646)
Supervisor:	Mohammad Tariqul Islam, Prof. Dr.
Keywords:	Universiti Kebangsaan Malaysia -- Dissertations Dissertations, Academic -- Malaysia Microwave imaging technique Breast cancer
Issue Date:	17-Jul-2019
Description:	Breast cancer is the second most common form of cancer with over two million new cases every year worldwide. The survival rate can even reach up to 97% with early detection, which emphasizes the urgent requirement of a reliable and highly efficient method for early breast cancer detection. X-ray mammography, Computed Tomography, Ultra-Sound and Magnetic Resonance Imaging are frequently used diagnostic systems for detecting breast cancer. However, these techniques have some limitations like false positives, low-resolution scan, higher cost, uncomfortable compression and ionization, performance degradation for a deep-lying or a solid tumour and time-consuming diagnostic process. Thus, microwave imaging techniques have attracted significant attention as a complement to existing diagnostic techniques. In this research, a new, complete and comprehensive near-field microwave imaging system (MIS) is presented to non-invasively detect the tumours inside the breast tissue. In the first stage of this research, several antenna designs are developed and fabricated according to the requirements for MIS application using an optimized synthetic focusing technique. The existing data independent MIS techniques, which rely on coherent summation of delayed signals, treat the antennas as point sources. It is a valid assumption for far-field applications, but this approach becomes ill posed in near-field setups. Thus, the position of the point source with respect to the imaging domain was determined using heuristic optimization techniques to minimize background noise. Furthermore, since such techniques are designed to detect one target, the presence of multiple tumours may cause misdetection of one or more tumours. Thus, an iterative technique was utilized to enhance further the capability to detect multiple tumours. Afterwards, experimental validation of the focusing technique for different types of antennas was performed by designing and developing a microwave imaging system to detect tumours of approximately 1.5 cm radii inside multiple breast phantoms of different configurations. The data collected from the proposed system were processed using the frequency domain-imaging algorithm, which reconstructs the image of breast interior to detect and localize the tumour using Matlab. The change of backscattered signals with the change of dielectric contents inside the structure of breast phantom is analysed to produce a 3D reconstruction with 2.5 mm resolution. Overall, the optimizing algorithm accurately predicted the position of the point source for each antenna and reduced the ambient noise in the final image by up to 9.09 dB for a metamaterial based antenna with peak gain of 9.30 dBi, operating frequency of 2.70 GHz to 8.0 GHz. The optimized system detected up to 4 distinct tumours in homogenous breast phantom, which is potentially viable for a low cost, portable and robust medical screening system.,Ph.D.
Pages:	101
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.