A Survey on Medical Image Registration

PhD Qualifying Examination


Title: "A Survey on Medical Image Registration"

Miss Yishan Luo


Abstract:

Image registration is an important technology in medical image analysis. It is 
aimed to align two or more images by finding a suitable transformation that 
relates the involved images. As a pre-step to compare and combine information 
taken from different images, image registration plays an important role in many 
medical applications, such as information fusion, 3D volume construction, atlas 
building and so on.

For a mathematical treatment of this problem, image registration can be 
considered as an optimization problem which tries to maximize an objective 
energy function with respect to the transformation that measures the similarity 
between the two involved images under some regularization constraints. In this 
survey, several core components of the image registration system, i.e., the 
similarity measure, transformation model and optimization, are discussed 
respectively. The state-of-art techniques utilized in each component are 
introduced and compared. After the literature review, a new registration model 
will be proposed. Finally, the paper will be concluded with some possible 
future directions of this research topic.


Date:     		Thursday, 28 January 2010

Time:                   1:30pm - 3:30pm

Venue:                  Room 4483
 			lifts 25/26

Committee Members:      Dr. Albert Chung (Supervisor)
 			Prof. Chi-Keung Tang (Chairperson)
 			Dr. Huamin Qu
 			Dr. Weichuan Yu (ECE)


**** ALL are Welcome ****