Research And Implementation Of The Imaging Platform For Early-stage Breast Cancer Under Near-Field Microwave Sounding

Breast tumor is a significant health issue for women nowadays, and its early diagnosis of breast tumor will greatly enhance the success of therapy and the forward liability of sufferer percentage. Therefore, early breast tumor detection technology has been an international highlight of medical imaging community. Microwave Near-Field Breast Tumor Detection relies on differences in electrical properties between malignant and fatty tissues. Based on the technology of microwave near-field breast tumor detection technology has such following excellent features: low radiation energy of electromagnetic signal, a large amount of objective carrier information, and high resolution and Millimeter level accurate location. Using this new detection technology, it is possible now to detect Early Breast Tumor safely, quickly and accurately.This dissertation is a part of Shanghai Municipal High Technology Tackle KeyProject--"Research on the imaging method of breast tumor in UWB near-fieldenvironment and diagnosis of early breast tumor (Project Number: 03JC14026)." Based on the research and implementation of the imaging platform for early-stage breast cancer under near-field microwave sounding, the dissertation discusses the imaging algorithm of early-stage breast tumor under near-field microwave sounding environment, and also discusses the design and implementation of medical picture archiving used in this special field.Based on the concept of medical picture archiving system, the design and layout of the imaging platform for early-stage breast cancer under near-field microwave sounding are detailed introduced, including system structure, module structure and module functions of the software, database planning, etc. Moreover, the importance in the whole detection system of the signal analysis and the imaging algorithm is pointed out.This dissertation gives out the implementation of signal analysis tools used in the imaging software system, which enable end users perform mathematical transforms to the input discrete digital signals, such as Fourier transform, Wavelet transform, and filters, and time delay parameters used in tumor imaging also can be retreated.On the basis of time delay parameters, an imaging algorithm called Finite Element Confocal Imaging Algorithm is presented according to the special application of early-stage breast tumor's detection. Results of imaging show that this new method based on the hemisphere model of breast, can effectively detects those tumors which are no more than 10mm diameters. And the following advanced algorithms are proposed: Meshing Step by Step Algorithm used to improve efficiency and accuracy, and Tracking and Searching Algorithm used to reduce the false-negative-rate.In order to realize Patients achieved information and data sharing, and to make a further step in electronic medical diagnose, the imaging platform proposed in this dissertation can be integrated with those large PACS system used in hospitals. And also, the imaging algorithm mentioned in this dissertation can be applied to other diagnose applications of hypodermic tumors like skin cancer and lymphoma detections. Furthermore, this detection technology under near-field microwave sounding also can be a new imaging method for target imaging on tissues with different electrical properties.