PhD defence by Marie Garnæs

On Friday 2 May, Marie Garnæs will defend her PhD thesis "Endpoint-driven measurement designs for hyperpolarized Magnetic Resonance".

Time: 13:30-16:30
Place: Bldg. 341, Auditorium 23 & zoom: https://dtudk.zoom.us/meeting/register/rn5KKRdpS3WErSLaNcP9JQ
Please be aware that the PhD defense may be recorded - This will also be informed at the beginning of the PhD defense.

Principal supervisor: Associate Professor Lars G. Hanson
Co-supervisor: Associate Professor Krístoffer H. Madsen
Co-supervisor: 

Assessment committee:
Associate Professor Henrik Lundell, DTU Health Tech
Dr. Mor-Miri Mishkovsky, École Polytechnique de Lausanne 
Associate Professor Mads Sloth Vinding, Aarhus University

Chairperson:
Associate Professor Pernille Rose Jensen, DTU Health Tech.

Abstract:
Hyperpolarized carbon-13 magnetic resonance is a cutting-edge technique that enables the non-invasive assessment of rapid metabolism in vivo. Altered metabolism is a hallmark of numerous diseases and both preclinical and clinical studies have demonstrated significant potential of this technique in e.g. tumor diagnostics, staging and monitoring of early treatment response. 

To gain meaningful insights into metabolic flux, pharmacokinetic models are fitted to experimental data. The models include conversion rate parameters. These parameters are crucial for characterizing and quantifying the metabolism making their accurate estimation the main focus of this thesis.
 
In vivo experiments involve a single injection of a hyperpolarized substrate, which constitutes a finite pool of magnetization that can be converted into signals in a series of measurements. The magnetization decays over time and is depleted with each measurement.  The measurement scheme determines how the spending of magnetization is distributed over time, which in turn affects the uncertainty of the conversion rate estimates obtained through the subsequent pharmacokinetic model fitting. Therefore, employing a scheme that most efficiently exploit the available magnetization is essential for the accuracy of metabolism quantification, and ultimately, patient outcomes. This thesis describes how to  achieve such schemes through mathematical optimization. The approach was validated using simulations and in vitro experiments.