MALIGNANT TUMOR DETECTION USING FINITE VOLUME TIME DOMAIN

Abstract

For a long time microwave engineers have dreamed of using non-ionizing electromagnetic waves to image the human body in order to detect cancer. Over the past several years, significant progress has been made towards making this dream a reality for breast cancer detection. In the next decade, microwave systems are likely to become viable diagnostic option for many women and men alike. More so than for any other cancers, breast tumors have electrical properties at microwave frequencies that are significantly different from those of healthy breast tissues. The breast can easily be accessed from outside, while internal organs are much less accessible. Normal breast tissue is also more translucent to microwaves than many other tissues, such as muscle or brain. Phenomenal progress in computers and numerical techniques during the past decade allows us to effectively process data acquired through measurements. This work proposes a new methodology for analyzing malignant tumors. The methodology will be based on finite volume time domain (FVTD) modeling approach. In the past finite difference time domain modeling schemes have been used to detect tumors. However, though it is simple and has wide frequency coverage, its main drawback is that it is computationally intensive. To overcome this drawback finite volume time domain (FVTD) is proposed as a suitable modeling technique for the problem.