Non-rigid local registraton of three-dimensional medical images
Funding: 2000: $21,080
Project Member(s): Braun, M.
Funding or Partner Organisation: Australian Research Council (ARC SPIRT (Strategic Partnerships with Industry Re)
Westmead Hospital (Westmead Hospital Partner Funds)
Start year: 2000
Summary: In image registration, one image is deformed to match another. Current techniques cannot account for local deformations in organ shape due to factors such as patient positioning, pathology, and individual anatomical differences between subjects. We propose to develop a registration method that allows for non-rigid local deformations, based on local optimization of information-theoretic measures and incorporating knowledge of the anatomy to constrain the registration process. A implementation of the method will be subject to a preliminary investigation of clinical utility. The method will help improve the accuracy of image analysis in many areas of clinical diagnosis, treatment and medical research.
Keywords: image registratio; image matching; medical image analysis
FOR Codes: Image Processing, Medical Physics, Diagnostic methods, Other, Provision of health and support services not elsewhere classified, Diagnosis of human diseases and conditions