MR functional cardiac imaging: segmentation, measurement and WWW based visualisation of 4D data

摘要: Abstract This paper considers the problem of ventricular segmentation and visualisation from dynamic (4D) MR cardiac data covering an entire patient cycle, in a format that is compatible with web. Four different methods are evaluated for process objects interest: The K-means clustering algorithm, fuzzy (FKM) self-organizing maps (SOMs) seeded region growing algorithm. technique active surface then subsequently applied to refine results, employing deformable generalised cylinder as geometric primitive. final models presented VRML 2.0 format. same repeated all 3D volumes cycle. radial displacement between end systole diastole calculated each point encoded colour on vertex, using RGB model. Using specifications, morphing performed showing phases real time. expert has ability view interact them simple internet browser. Preliminary results normal abnormal cases indicate very important pathological situations (such infarction) can be visualised thus easily diagnosed localised assistance proposed technique.