Tissue abnormality localization based on the internal structure analysis and inhomogeneous modeling

Abstract

The mechanical property of biological tissue is one of important indicators for their pathological states. The pathological findings exhibit that cancerous tissues have stiffer elastic modulus compared to normal tissues. The mechanical property characterization of these tissues could be used to diagnosis and localization by using the mechanical property difference between the normal and cancer tissues. Combined with anatomical and diagnostic knowledge, the palpation has been applied by using the surgeon’s sense of touch to detect and localize stiff nodules in normal tissue. However, palpation has not been able to provide objective and precise information due to its skill dependence on surgeons. By combining the mechanical characterization based on a well-defined biomechanical model, palpation could provide significant information to surgeons and achieve clinical acceptance. The mechanical localization of the cancer using relative mechanical property between the normal tissue and cancer could give more valuable information to surgeons during the diagnosis and robotic surgery. In cell level, the mechanical indicator could be another approach to detect the cancerous cells since conventional method are not enough to diagnose the cancerous cells due to chemical diversity and morphological variation of cells. In this dissertation, we develop local inhomogeneous models that consider the location and volume of abnormality and geometric information of normal tissues for local normal property criteria. The mechanical properties of abnormality and normal tissues were estimated by using the developed model simulation and the previous experiment results. For mechanical localization, we propose and perform the sweeping palpation. The estimated properties of normal tissues are used to localization of prostate cancer by comparing with newly performed ex-vivo experiment results for clinical validation of our methodologies.