Esmraldi: efficient methods for the fusion of mass spectrometry and magnetic resonance images

Abstract

Background Mass spectrometry imaging (MSI) is a family of acquisition techniques producing images of the distribution of molecules in a sample, without any prior tagging of the molecules. This makes it a very interesting technique for exploratory research. However, the images are difficult to analyze because the enclosed data has high dimensionality, and their content does not necessarily reflect the shape of the object of interest. Conversely, magnetic resonance imaging (MRI) scans reflect the anatomy of the tissue. MRI also provides complementary information to MSI, such as the content and distribution of water. Results We propose a new workflow to merge the information from 2D MALDI–MSI and MRI images. Our workflow can be applied to large MSI datasets in a limited amount of time. Moreover, the workflow is fully automated and based on deterministic methods which ensures the reproducibility of the results. Our methods were evaluated and compared with state-of-the-art methods. Results show that the images are combined precisely and in a time-efficient manner. Conclusion Our workflow reveals molecules which co-localize with water in biological images. It can be applied on any MSI and MRI datasets which satisfy a few conditions: same regions of the shape enclosed in the images and similar intensity distributions.

MALDI-MSI Towards Multimodal Imaging: Challenges and Perspectives

“…When using non-linear transformations, it is important to select parameters such as to obtain a compromise between shape-matching and intensity fidelity. For instance, Grélard et al (2021) estimated the shape matching by F-measure, and the intensity fidelity by computing the mutual information between the images before and after applying non-linear registration.…”

“…Segmentation methods are frequently based on pixel intensities or use the geometrical properties of the objects. In a multimodal context, MALDI-MSI segmentation can involve simple thresholding ( Anyz et al, 2017 ), region growing ( Grélard et al, 2021 ), methods assisting in selecting spatially coherent ion images ( Alexandrov and Bartels, 2013 ), or deep learning ( Abdelmoula et al, 2022 ).…”

“…After highlighting relevant methodological developments, multimodal imaging better contributes or provides evidence towards the biological understanding. Grélard et al (2021) correlated water distribution given by MR images with the distribution of specific polysaccharides in MS images. This study hints at the role of these molecules in the cell-wall porosity on tissue, and corroborates previous observations made in vitro .…”

Fusing ToF‐SIMS Images for Spatial‐Spectral Resolution Enhancement using a Convolutional Neural Network

“…[ 16 ] Other groups have simply visualized complementary images as overlays and completed correlative analyses, also referring to these workflows as fusion methods. [ 17–19 ]…”

“…[16] Other groups have simply visualized complementary images as overlays and completed correlative analyses, also referring to these workflows as fusion methods. [17][18][19] While many of these approaches have proven useful, and have demonstrated enhanced ion image resolution, they are not suitable for the fusion of two hyperspectral ToF-SIMS imaging data sets. This is because the fusion of individual ion images will be ineffective when two or more high mass resolution peaks correspond to a single low mass resolution peak, the exact problem we are attempting to address in this work.…”