Research On The Inverse Problem Based On EIT For Breast Imaging

Breast cancer has become one of the most common malignant tumor disease, ranking the top female cancer mortality.Currently, the equipments used for early breast canner diagnosis, such as X-ray, Ultrasound, MRI, et.al, have not easily to distinguish benign and malignant tumors.The accuracy rate of the check is very low and the body has been at risk of electromagnetic radiation in the detection process. These shortcomings limit the applications of the existing equipment in the breast sxamination process.Electrical Impedance Tomography is a new and non-invasive functional imaging technology, which has an important applying prospect. In EIT, a sensing electrode array is attached on the boundary of an object, safe currents are injected through these electrodes, and the resulting voltages are measured which reflecting the electrical characteristics of the biological tissue and organs.After data collection and processing,the conductivity or resistivity distribution is imaged by reconstructed algorithms. The results of impedeance spectroscopy measurements for breast tissue in vitro indicate that the impedeance characteristics of the cancer tissue are significantly different from the normal tissue's. The impedeance information of organisms can not only reflect the structure of biological tissues, and more importantly, give the physiological and pathological information of the organs.EIT has the important clinical value and practical significance in terms of disease prevention and diagnosis. It has become a pop area of the research fields at home and abroad and great progress has been made in EIT technique for breast imaging recently.First, EIT imaging principle and study situation are summarized. And based on three-dimensional hemispheric model for the human breast, the numerical computation methods of EIT forward problem are analyzed. Therefore, the EIT inverse problems, which are based on EIT forward problems, are studied. Because of the ill-conditioned of the inverse problem, satisfactory images in terms of good quality and distinguishability haven't been achieved. In order to solve ill-posed problem and poor resolution in EIT, a new image reconstruction algorithm based on total variation regularization is adopted. Compared with the conventional Tikhonov regularization, this new algorithm not only stabilizes the reconstruction, but also enhances the distinguishability of the reconstruction image in areas with discontinous medium distribution, which possesses a good edge-preserving property. The simulation results show that this algorithm has obvious advantages over other methods in resolution of the reconstruction image, and the real-time performance meets the requirements of application, too.