Flux-based Medical Image Segmentation: A Survey

PhD Qualifying Examination


Title: "Flux-based Medical Image Segmentation: A Survey"

by

Mr. Jierong WANG


Abstract:

Curvilinear structure segmentation is of importance in computer-aided 
diagnosis, which is a fundamental pre-processing step and the foundation 
of a range of applications in medical image analysis. Such structures can 
be commonly found in human anatomy, e.g., blood vasculature, spinal cord, 
lung airway, etc. Therefore, with an effective curvilinear structure 
detector, it will greatly improve the effectiveness of medical image 
segmentation methods. A vast number of approaches have been proposed in 
the past few decades. The flux-based approaches have raised high interest 
in the community, which is the target in this survey. We firstly introduce 
the background of flux-based methods. Then, several variants of the 
flux-based method will be discussed, including optimally oriented flux 
(OOF), cylindrical flux and irregular flux. The above mentioned flux-based 
models are intrinsically order-2 tensors. To deal with more practical and 
challenging problems in tabular modeling, higher-order tensor model is 
presented. A higher-order flux model is proposed recently. Finally, we 
discuss the potential direction for future research respective to the 
flux-based model.


Date:			Monday, 30 September 2019

Time:                  	10:00am - 12:00noon

Venue:                  Room 2408
                         Lifts 17/18

Committee Members:	Prof. Albert Chung (Supervisor)
 			Dr. Xiaojuan Ma (Chairperson)
 			Prof. Long Quan
 			Prof. Chiew-Lan Tai


**** ALL are Welcome ****