Data registration tool for cross-modality 3D elemental imaging, with a focus on laser ablation ICP-MS and X-ray imaging
by Stijn J. M. Van Malderen (Ghent University, Department of Chemistry)
at 25b ( 109 )
Data registration tool for cross-modality 3D elemental imaging,
with a focus on laser ablation ICP-MS and X-ray imaging
Stijn J. M. Van Malderen1, Laszlo Vincze1 , Frank Vanhaecke1
1 Department of Chemistry, Ghent University, Campus Sterre, Krijgslaan 281 S12, 9000 Ghent, Belgium
This talk will discuss the fundamental principles and features of laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS), and how this technique compares to state-of-the-art lab-based X-ray instrumentation and SR-based XRF. LA-ICP-MS is a highly sensitive elemental and isotopic analysis and imaging technique which is performed in situ on solid samples. Its principles are based on the removal of particles from the surface through ablation, followed by transport, atomization, and ionization of these particles. The excited ions then undergo mass analysis in a mass spectrometer. Major developments took place in the last few years in this field; so-called low-dispersion ablation cells have yielded gains in sensitivity and imaging speed of 2-3 orders of magnitude and the spatial resolution achievable by this technique has come down to < 1 µm. There are synergistic effects when combining data from MS-based and X-ray based techniques, which has given rise to cross-modality approaches. Both techniques are applicable on a wide application field – from bio-imaging to geo-imaging and solid-state research. Combining the data of multiple modalities through (unsupervised) data fusion approaches has been a challenge as each modality can have its own coordinate system, probe volume, sensitivity and detection mode. In mass spectrometry imaging (MSI) methods as well as XRF data, the volume of data has been steadily increasing, and the time required by the user to process this data, as a result of advances in lateral resolution, increases in the dimensionality and the emergence of multiplexed, complex data sets. The image processing tools to efficiently deal with large volumes of cross-modal data are still lacking power and flexibility. The presence and uniformity in the formatting of meta-data, such as positional data has become a critical aspect. Significant efforts to create multi-modal data analysis tools have been ongoing for years. One of the tools developed, a 3D co-registration tool based on the HDF5 format will be discussed, and an example of data fusion of LA-ICP-MS and µCT on the single-cell level will be shown.