Radial Basis Functions Method for Solving Medical Imaging Problems


Student thesis: Doctoral Thesis

View graph of relations


Related Research Unit(s)


Awarding Institution
Award date4 Jan 2018


The success of radiotherapy treatment relies in accurately localizing the target region close to the tumor as well as avoiding damage to the nearby healthy organ. In order to control the tumor growth and minimize side effects to the patient, it is crucial to make some substantial modifications on the patient's anatomy due to the reduction of the patient's weight, shrinkage of the tumor, and the edema of the tissue before the treatment. For surgery consideration, an efficient registration method to delineate the clinically critical objects in computed tomography images obtained from the radiation treatment process is needed. Over the last decades, the Deformable Image Registration Model has undergone intensive investigation from researchers in the fields of computer vision, remote sensing, etc. Despite the significant progress that has been made, deformable registration remains a challenging problem in the field of radiotherapy.

Hepatocellular carcinoma (HCC) is the third most common cancer and the second most deadly cancer in Asia-Pacific region. Liver transplantation is one of the treatments that offers a chance of cure for the tumor and the underlying cirrhosis. Living donor liver transplantation is performed as an alternative way to cadaveric transplantation. Prior screening of suitable donor candidates is essential in which the efficiency and accuracy in detecting images of the liver by using computed tomography is vital to the correct diagnosis of surgery operation for the success of liver transplantation.

In this thesis, we solve the Deformable Image Registration and Image Segmentation for Medical Imaging Problems using Radial Basis Functions method.

    Research areas

  • Radial basis functions, Deformable Image Registration, Image segmentation