Integration of Image Segmentation Method in Inverse Scattering for Brain Tumour Detection

• Main Authors: Kismet, Hong Ping, Joseph, Eustacius J., Kuryati, Kipli, Shafrida, Sahrani, Dyg Norkhairunnisa, Abang Zaidel, Mohd Iqbal, Saripan, Mohamad Hamiruce, Marhaban, Dayang Azra, Awang Mat
• Format: Article
• Language: English
• Published: EMW Publishing 2017
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://ir.unimas.my/id/eprint/31288/1/INTEGRATION%20OF%20IMAGE%20SEGMENTATION%20METHOD%20IN%20INVERSE%20SCATTERING%20FOR%20BRAIN%20TUMOUR%20DETECTION%20-%20Copy.pdf; http://ir.unimas.my/id/eprint/31288/; http://www.jpier.org/PIERM/pier.php?paper=17070603

This paper presents a microwave imaging for brain tumour detection utilizing ForwardBackward Time-Stepping (FBTS) inverse scattering technique. This technique is applied to solve electromagnetic scattered signals. It is proven that this technique is able to detect the presence of tumour in the breast. The application is now extended to brain imaging. Two types of results are presented in this paper; FBTS and FBTS integrated with image segmentation as a pre-processing step to form a focusing reconstruction. The results show that the latter technique has improved the reconstructions compared to the primary technique. Integration of the image segmentation step helps to reduce the variation of the estimated dielectric properties of the head tissues. It is also found that the optimal frequency used for microwave brain imaging is at 2 GHz and able to detect a tumour as small as 5 mm in diameter. The numerical simulations show that the integration of image segmentation with FBTS has the potential to provide useful quantitative information on the head internal composition.