Variational methods for medical ultrasound imaging (Q2844744)

In the past decade, medical ultrasound imaging became an important preferred choice for diagnostics in clinical routing ranging from parential diagnostics, inspection of stomacks, colon, kidney, to echocardiography due to its low costs, no radiation, and real-time capacity. Hence, the research in processing and analysis of ultrasound images has been intensified.NEWLINENEWLINE This thesis is concentrated on variational methods for fully-automatic processing and analysis of medical ultrasound images including the problem of low-level segmentation, the concept of shape modeling and shape analysis to perform high-level segmentation of medical ultrasound images, and the motion estimation in medical ultrasound imaging. The main contribution in this work is the incorporation of a priori knowledge about the image formation process in ultrasound images and the development of novel variational formulations which are based on non-standard data fidelity terms. The work uses both synthetic and real patient data in the analysis. Some real-life application examples are given.