Dynamic Temperature Mapping - Real-time Strategies and Model-based Reconstructions

by Zhongshuai Zhang
Doctoral thesis
Date of Examination:2016-12-14
Date of issue:2017-02-09
Advisor:Prof. Dr. Jens Frahm
Referee:Prof. Dr. Jens Frahm
Referee:Prof. Dr. Jörg Enderlein
Persistent Address: http://dx.doi.org/10.53846/goediss-6114

 

 

Files in this item

Name:Dissertation_Zhongshuai_Zhang_finial.pdf
Size:106.Mb
Format:PDF

The following license files are associated with this item:

Abstract

English

The purpose of this thesis was to explore the potential of real-time MRI methods for temperature mapping during thermal procedures. This mainly included a comprehensive analysis of two different methods, i.e. proton resonance frequency (PRF) and T1 thermometry, with respect to speed, robustness, and accuracy in different experimental settings. These methods were chosen because of their pronounced temperature sensitivities and relatively rapid data sampling schemes. Basic implementations and comparisons of suitable data acquisition and image reconstruction strategies were accomplished with use of a home-made temperature phantom. Secondly, PRF thermometry as the resulting preferred method was further analyzed using heating studies of ex vivo tissues. And thirdly, in order to even better mimic clinical thermal therapies, focal heating with use of a laser system was studied on swine kidneys. In both kidney cortex and medulla, the temperature changes for different output powers could be monitored by PRF thermometry in real time. A further development in preparation of future in vivo studies referred to an optimized reference selection procedure in order to minimize temperature errors in the PFR method due to inter-scan motion between the reference and actual scan. Lastly, a model-based reconstruction approach for phase-contrast MRI, which yields temperature maps and anatomic images directly from undersampled raw data sets, has been adopted to the present conditions. Efforts were made especially with respect to temporal regularization to obtain temperature maps with considerably increased quality.
Keywords: MR thermometry; real-time MRI; nonlinear reconstruction; model-based reconstruction; laser induced thermal therapy
 
Statistik