Realization of an Iterative Reconstruction Algorithm for Brain Stroke Detection Using Microwave Tomography Technique

The work aims to evaluate the performance of microwave scanning to identify and detect the stroke affected brain cells. The researches on this subject are motivated by the need to make continuous control of the brain stroke affected patient. In this paper, a system has been proposed to accurately detect the complex dielectric perturbations of the water content of the stroke affected brain model and predict the exact location of the affected region inside the head model. With the help of proper field pattern analysis and an iterative Exact reconstruction algorithm, the model has been reconstructed with better accuracy, both for normal and diseased cases. The reconstruction is made based on the complex dielectric values of the affected cells. A comparative study has been performed on different sizes of stroke affected brain regions. It has been shown that, the algorithm is efficient enough to detect the presence of the stroke affected cells in a small region (1% of total head model) with considerable error margins. The required 2D-cross sectional images are shown for proper visualization of the model along with the positioning of the stroke affected region. The changing dielectric properties of brain tissues in diseased condition can be measured using Microwave Tomography Technique (MTT). With this kind of experimental implementation, the concept of MTT has been utilized with proper algorithmic and mathematical modelling to iustify the effectiveness of our algorithm.