Study On Localization Approach Of Breast Tumor Basede On Electrical Impedance Tomography

Breast cancer has become the primary cause of cancer incidence and mortality in women,seriously threatening women’s physical and mental health.In order to enable patients to maximize breast-conserving treatment,early detection and precise treatment of breast lesions are necessary.Electrical Impedance Tomography(EIT)is a safe and non-destructive medical imaging detection method that has been developed in the past forty years.It can repeatedly detect and monitor the human body in a short time.This paper focuses on how to use the principle of electrical impedance tomography to realize the problem of localization of breast foreign bodies.The main research contents include:(1)A three-dimensional truncated mammary gland model,which closer to the shape of the female breast,was constructed,and six new types of electrode configurations with different levels of two layers were designed.The results show that the image reconstruction quality was the best and the positioning accuracy was high in the 8-12 electrode configuration mode.(2)The original average conductivity data solved by the EIT inverse problem was processed by the maximum average method,which can effectively reduce the artifacts of the reconstructed image and improve the image resolution.(3)A method for height localization and center point localization of breast foreign body based on EIT technology was proposed.The spatial distribution of the imaging target was obtained by the multi-layer image,and the K-means clustering algorithm was used to gain the target central point position.The results show that the positioning error was small when the stratification interval was 5 mm.(4)The conjugate gradient algorithm,the improved Tikhonov regularization algorithm and the Tikhonov regularization algorithm based on singular value decomposition were compared,and the model was imaged and located under different parameters.The research results show that the improved regularization algorithm could obtain good image quality and accurate positioning within a certain parameter range.(5)The mammary gland model which simulates the skin layer,fat layer and gland layer was established,and the foreign matter position was set in different layers.The EIT image was reconstructed by the conjugate gradient algorithm,and the position of the foreign object could be accurately obtained.The innovations of this article are:(1)A truncated mammary gland model with two layers of unequal number of electrodes was established to obtain the optimal electrode configuration mode.(2)A method based on electrical impedance image for breast foreign body localization was proposed and verified.