PhD defense of Mariëlle Jansen

On Tuesday 10 December @ 14:30, Mariëlle Jansen will defend her thesis entitled: “Analysis of liver lesions in dynamic contrast enhanced MR images.”

Abstract

Lesion detection and characterization is an important step in the diagnosis and treatment planning of focal liver lesions such as hemangiomas, adenomas, and liver tumors. Image processing and image analysis can aid radiologists performing these tasks.

The aim of this thesis is to develop and evaluate methods for image registration, liver lesion detection, and lesion classification using MRI, and in particular dynamic contrast enhanced (DCE) MR images. DCE-MR images are very useful for image analysis since they contain both structural and functional information.

Image registration is the first step in the analysis of DCE-MR images. A groupwise registration method was optimized and evaluated for abdominal DCE-MR images. Next, a liver lesion detection method using convolutional neural networks (CNN) was developed. First, the liver is segmented to define the region of interest. Then the liver lesion detection CNN method using both DCE-MR images as well as diffusion weighted MR images was applied.

The lesion detection method is a general method and could fail if a patient presents features not known or seen before by the CNN. So, we also proposed a patient-specific fine-tuning approach to obtain a detection method dedicated towards the features of one patient.

Once the lesions are detected, the type of lesion needs to be determined. We proposed lesion classification method, which was able to successfully distinguish benign from malignant lesions as well as differentiating five focal liver lesions: adenoma, cyst, hemangioma, HCC, and liver metastasis.