Application Of Various Mathematical Models In DWI Examination Of Liver Cancer

Liver tumors do great harm to people, the early diagnosis of differentiating begin or malignant lesions was important for the appropriate treatment options and surgical methods. And the imageological examination has a great significance in the identification of benign or malignant bone tumors, observation of curative effect and postoperative follow-up.Magnetic resonance diffusion weighted magnetic resonance imaging (MRI) is a kind of widely used imaging method. It uses the water molecules in the biological tissues of Brownian motion for noninvasive imaging. The mono-exponential model and apparent diffusion coefficient (ADC) were wildly used in analyzing. But the ADC value of integrates the information of the two aspects of the true diffusion and pseudo perfusion in the biological tissue, which reflects the diffusion characteristics of the capillary microcirculation. Bi-exponential model solved this problem, the model separated the D and D*. D for the simple diffusion coefficient of water molecules and D* as the pseudo diffusion coefficient. Stretched-exponential model reflects more biological features of organization of information through the distribution of internal water molecules diffusion index(DDC) and anisotropic diffusion index.In this paper, the above three kinds of mathematical models are analyzed and studied. The results (ADC,D,D*,f,DDC,α) are analyzed by SPSS software. The parameters of normal liver tissue and tumor region were calculated, and the relationship between the parameters with statistical significance and better effect was analyzed by Pearson correlation analysis. The parameter data of different patients were compared with two two t test, and the diagnostic efficacy of each model was analyzed.Experimental results show that the three models can aid for the diagnosis of liver tumors to provide convenient, provide numerical basis for qualitative check, and can easily see the therapeutic effect on tumor surgery to convenient the late diagnosis. Among them, the parameter D of the double exponential decay model and the parameter DDC of the tensile index model are the best.