Research Of Quantitative Microwave Imaging Algorithm For Intracranial Tumor Detection

As a malignant disease,intracranial tumor threatens the health of residents in our country.Early medical imaging diagnosis can save the patient's life to the greatest extent.As a new medical imaging method,microwave imaging system has the advantages of non-damage detection,low cost and good portability.Microwave imaging technology that is base on the principle of inverse scattering transmits microwave signal into the human body and obtains the distribution of complex relative dielectric constant in the detection area by receiving and processing the scattered signal.This paper studies the application of microwave imaging algorithm which is base on inverse scattering for head imaging.Studied a variety of quantitative microwave imaging algorithm based on inverse scattering,respectively is the Born iterative method(BIM),the iterative shrinkage thresholding enhanced Born iterative method(IST-BIM),the distorted Born iterative method(DBIM),the contrast source method(CSI),a quantitative algorithm based on linear sampling method(LSM)and the back-propagation method(BP).On the basis of these algorithm,a algorithm combines the BP and the the shrinkage thresholding enhanced BIM(BP-ST-BIM)and a formula enhanced CSI are proposed.And the computer program corresponding to all the algorithms mentioned above is written.Then the influence of relevant parameters of each algorithm on the imaging results is determined and optimized through preliminary simulation experiments.The imaging effect of the proposed algorithm is verified by simulation experiments.The two models used in the simulation experiment are the simple two-objective model and the complex human brain profile model.For the simple model,the imaging results of the BP-IST-BIM show that,compared with the BIM,the overall calculation time is shorter and the background ripple is less.Compared with the IST-BIM,it can avoid divergent results.The imaging effect of the improved method is also better than that of the original CSI in that the relative error of the dielectric constant is smaller and the error of the image is less.For the human brain model,the relative error of the permittivity of the BP-IST-BIM is larger than that of the formula enhanced CSI and the quantitative algorithm based on the LSM.In addition,in order to further verify the imaging effect of the algorithm under actual conditions,the experimental data measured by our research group was used as the input of the algorithm program for calculation,and the final imaging result could show the shape and position of the target object.