Imaging of the heart anatomy and function using magnetic resonance imaging (MRI) is an important diagnosis tool for heart diseases. Several techniques have been developed to increase the contrast-to-noise ratio (CNR) between myocardium and background. Recently, a technique that acquires cine cardiac images with black-blood contrast has been proposed. Although the technique produces cine sequence of high contrast, it suffers from elevated noise which limits the CNR. In this paper, we study the performance and efficiency of applying a Bayes classifier to remove background noise. Real MRI data is used to test and validate the proposed method; In addition, a quantitative comparison is done between the proposed method and other thresholding-based classifications techniques.
|Title of host publication||Proceedings of the Biomedical Engineering Conference (CIBEC 2008, 18-20 December 2008, Cairo, Egypt|
|Place of Publication||Piscataway|
|Publisher||Institute of Electrical and Electronics Engineers|
|Publication status||Published - 2008|