NewSpiralTOF™ High-Speed MS Imaging × High Mass-Resolution
In MALDI mass spectrometry imaging (MALDI-MSI), molecules are visualized directly within a specimen by coating the specimen surface with matrix, scanning it with a laser, and acquiring a mass spectrum at each position. This makes it possible to see where specific molecules are located and how much of them is present across a tissue section or other specimen surface.
High Mass-Resolution for Accurate MS Imaging
Thanks to its 17 m flight path, the NewSpiralTOF™ delivers high mass-resolution even when analyzing biological tissue sections with nonuniform surface conditions. In measurements of approximately half of a mouse brain over a region of about 5 × 7 mm, the system achieved a mass resolution of approximately 40,000 in the average mass spectrum. This high resolving power enabled the separation of isobaric lipid species such as phosphatidylcholine (PC), phosphatidylethanolamine (PE), and galactosylceramide (GalCer), making it possible to obtain the correct spatial distribution for each molecule.
PE: Phosphatidyl ethanolamine, PC: Phosphatidyl Choline, GalCer: Galactosylceramide
This data was acquired in a joint research project with the Mass Spectrometry Group, Project Research Center for Fundamental Sciences, Graduate School of Science, Osaka University. The tissue section specimen was provided by Awazu Laboratory, Division of Sustainable Energy and Environmental Engineering, Graduate School of Engineering, Osaka University.
AI-Enhanced Image Quality for MS Imaging
JEOL has a long history of innovation in image processing, driven by its leadership in electron microscopy. Building on this expertise, JEOL adapted its AI-based image enhancement technology LIVE-AI (Live Image Visual Enhancer-AI), originally developed for SEM, to MS imaging data processing and implemented it as the FINE-AI Filter. The result is a substantial improvement in the clarity and interpretability of mass images.
Automatic Extraction of Important Features
With the high mass-resolution of the NewSpiralTOF™, more than 100 lipid species were detected in mouse brain sections. However, interpreting the distribution of each lipid individually is not practical. Meaningful insight requires statistical tools that can identify groups of components that vary together and summarize complex datasets efficiently.
To address this, vertex component analysis (VCA), a method that models all spectra in a mass imaging dataset as mixtures of a small number of "vertex components" is implemented. Compared with methods such as principal component analysis, VCA provides results that are easier to interpret in a shorter time. Because VCA is sensitive to noise, applying it after denoising with the FINE-AI Filter makes the analysis far more effective. In the example shown, this approach revealed three key vertex components characterizing lipids in mouse brain tissue.