University of Twente Student Theses


Generating high frame rate MR images using surrogate signals

Veenstra, G.W. (2019) Generating high frame rate MR images using surrogate signals.

[img] PDF
Abstract:This paper continues the work on creating a reconstruction algorithm capable of generating synthetic MR images online both while the patient is in- \& outside the scanner. The ability to construct free-breathing real-time MR images will be beneficial in both augmenting medical personal and for the development of robotic tools in the medical field by improving cancer tumor segmentation and tracking. Two retrospective gating approaches capable of constructing such synthetic MR images were implemented for performance assessment: one using full MR images as datapoints, Retrospective Image Gating (RIG); and one using k-space lines, Retrospective K-space Gating (RKG). Input data was acquired by letting five participants perform breathing exercises in the MRI scanner in order to induce specific Respiratory Motion (RM) patterns. At the same time, sagittal MR images of the liver were being acquired at 1.27Hz alongside one-dimensional ultrasound measurements of the upper right abdomen at 50Hz. This information enabled the performance assessment of synthetic MRI construction algorithms for different types of RM patterns. The algorithms performed better on the regular, breath-hold and shallow breathing types and considerably worse on the deep and coughing breathing types. Also, several enhancements were successfully implemented to improve reconstruction quality or execution time.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Embedded Systems MSc (60331)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page