Development of microwave brain stroke imaging system using multiple antipodal vivaldi antennas based on Raspberry Pi technology

• Main Authors: Azahari Salleh,, Yang, Ching Chiou, Touhidul Alam,, Mandeep Singh Jit Singh,, Md. Samsuzzaman,, Mohammad Tariqul Islam,
• Format: Article
• Language: English
• Published: Penerbit Universiti Kebangsaan Malaysia 2020
• Online Access: http://journalarticle.ukm.my/14841/1/06.pdf; http://journalarticle.ukm.my/14841/; http://www.ukm.my/jkukm/volume-321-2020/

This paper proposes a Microwave Imaging System (MIS) for brain stroke detection. In the MIS, the primary challenge is to improve in terms of cost, size, and stroke image quality. Thus, the main contribution of this work is the economy and the compact rotation platform integrated with an array of nine antipodal Vivaldi antenna in circular arrangement and single computer board, Raspberry Pi Module (RPM) as microcontroller developed. The design and fabrication of wideband antenna based on Computer Simulation Technology (CST) software and Rogers RO4350B substrate, which operated from 2.06 GHz to 2.61 GHz. In the RPM, the Python programming language used for regulating the angle of rotation and antenna switching process. The process of receiving reflection signals from the head phantom for each antenna supervised by Single-Pole 8-Throw (SP8T) Radio Frequency (RF) switch. The fabricated head phantom based on the primary tissues of the brain, white matter using inexpensive materials, and located in the middle of the platform. Platform rotation is a combination of wood-based platform with the size 0.36m2 and material Perspex. Then, through an interfacing process between Python script and Vector Network Analyzer (VNA), the raw data in S-Parameters transferred to the MATLAB software for analysis. The fabricated antenna able to realize high directivity, 86.92% efficiency, and 2.45 dBi gain. Overall, the proposed system offers the cost-effective, compact, and able to collect the data effectively around the head phantom that consist of a target clot and without a target clot at 50 different positions. It successfully tracked the presence of stroke clots through color differences in color plots.