Measuring spatial resolution in Mass Spectrometry Imaging
File(s)
Author(s)
Metodiev, Martin
Type
Thesis
Abstract
Mass spectrometry imaging (MSI) is a term for a collection of imaging techniques that allow for the spatial distribution of different molecules found natively within thin tissue sections to be visualised. The hyperspectral nature of those imaging techniques makes these methods extremely powerful for the detailed analysis of complex systems, for example characterising living systems. Unfortunately, the concept of image formation, which can be thought of as a set of mathematical rules that describe how information is mapped from an object to an image, has remained unexplored. As a consequence, the definition of spatial resolution has been ill-defined. This thesis introduces a new image formation model which mends the currently ill-defined notion of spatial resolution and at the same time exposes some common misconceptions about it. With the aid of this model the processes that lead to the loss of spatial resolution are identified as blur and noise and the origin of those parameters is derived and demonstrated through computational simulations. Ultimately, a new method for measuring spatial resolution is proposed. The method allows for the study of spatial resolution across different pixel sizes and modalities. Apart from the practical use of measuring spatial resolution, the image formation model also allows for theoretical studies of MSI tissue imaging experiments. As MSI detects hundreds or thousands of ions per pixel and resolution is shown to be intricately linked to the signal intensity as well as blur, resolution in MSI is best described by a range representing resolution calculated for different ion images. The model is used to asses resolution from standard samples and tissue imaging experiments acquired using NanoSIMS, OrbiSIMS, MALDI and DESI.
Version
Open Access
Date Issued
2023-03-30
Date Awarded
2024-11-01
Copyright Statement
Attribution-Non Commercial-No Derivatives 4.0 International Licence (CC BY-NC-ND)
Advisor
Bunch, Josephine
Takats, Zoltan
Sponsor
Cancer Research UK
Publisher Department
Department of Metabolism, Digestion and Reproduction
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)