Automatic 3D Nonlinear Registration of Mass Spectrometry Imaging and Magnetic Resonance Imaging Data

Abstract

Multimodal integration between mass spectrometry imaging (MSI) and radiology-established modalities such as magnetic resonance imaging (MRI) would allow the investigations of key questions in complex biological systems such as the central nervous system. Such integration would provide complementary multiscale data to bridge the gap between molecular and anatomical phenotypes, potentially revealing new insights into molecular mechanisms underlying anatomical pathologies presented on MRI. Automatic co-registration between 3D MSI/MRI is a computationally challenging process due to dimensional complexity, MSI data sparsity, lack of direct spatial-correspondences, and non-linear tissue deformation. Here, we present a new computational approach based on stochastic neighbor embedding to non-linearly align 3D MSI to MRI data, identify and reconstruct biologically-relevant molecular patterns in 3D, and fuse the MSI datacube to the MRI space. We demonstrate our method using multimodal high-spectral resolution MALDI 9.4 Tesla MSI and 7 Tesla in vivo MRI data, acquired from a patient-derived, xenograft mouse brain model of glioblastoma following administration of the EGFR inhibitor drug of Erlotinib. Results show the distribution of some identified molecular ions of the EGFR inhibitor

Experimental and Data Analysis Considerations for Three-Dimensional Mass Spectrometry Imaging in Biomedical Research

“…In this context, Abdelmoula et al developed an automatic co-registration between 3D-MSI and MRI that tackles the differences in spatial scales and coordinate systems, the lack of direct spatial-correspondences, or nonlinear tissue deformations (Fig. 5b) [71].…”

“…The visualization on the left shows a molecule that highlights the tumor area and on the right, the same ion is overlaid with the MRI image. Adapted with permission from [71] (copyright American Chemical Society).…”

Discovering new lipidomic features using cell type specific fluorophore expression to provide spatial and biological specificity in a multimodal workflow with MALDI IMS

“…Traditionally, multimodal imaging has relied on manual interpretation of co-registered ion images 13 , which is prone to human bias. Other supervised and unsupervised approaches have been used to improve data analysis [14][15][16][17][18][19][20] . Each of these approaches still requires an independent benchmark to define cells or structures.…”

Light sheet fluorescence microscopy guided MALDI-imaging mass spectrometry of cleared tissue samples

“…Furthermore, MALDI-MSI analyses can also be combined with a variety of other imaging modalities, providing additional morphological and/or biochemical information from the identical tissue section, or sample 14 , 17 . Previously reported multimodal (MALDI)-MSI analysis approaches e.g., include the combination with autofluorescence-microscopy 21 , 22 , Raman spectroscopy 23 , magnetic resonance imaging (MRI) 24 , 25 , histology, immunohistochemistry, and laser scanning confocal microscopy 26 . Using appropriate alignment methods 22 , 24 , 27 , the images acquired by MALDI-MSI and additional imaging modalities can be exactly overlaid and thus directly be related to e.g., distinct morphological features of the examined tissue, such as different structural compartments, cell types, pathological alterations, etc.…”