Fast computation of mutual information in the frequency domain with applications to global multimodal image alignment

Correlative geochemical imaging of Desmophyllum dianthus reveals biomineralisation strategy as a key coral vital effect

The chemical and isotopic composition of stony coral skeletons form an important archive of past climate. However, these reconstructions are largely based on empirical relationships often complicated by "vital effects" arising from uncertain physiological processes of the coral holobiont. The skeletons of deep-sea corals, such as Desmophyllum dianthus, are characterised by micron-scale or larger geochemical heterogeneity associated with: (1) centres of calcification (COCs) where nucleation of new skeleton begins, and (2) fibres that thicken the skeleton. These features are difficult to sample cleanly using traditional techniques, resulting in uncertainty surrounding both the causes of geochemical differences and their influence on environmental signals. Here we combine optical, and in-situ chemical and isotopic, imaging tools across a range of spatial resolutions (~ 100 nm to 10 s of μm) in a correlative multimodal imaging (CMI) approach to isolate the microstructural geochemistry of each component. This reveals COCs are characterised by higher organic content, Mg, Li and Sr and lower U, B and δ11B compared to fibres, reflecting the contrasting biomineralisation mechanisms employed to construct each feature. CMI is rarely applied in Environmental/Earth Sciences, but here we illustrate the power of this approach to unpick the "vital effects" in D. dianthus, and by extension, other scleractinian corals.

INSPIRE: Intensity and spatial information-based deformable image registration

We present INSPIRE, a top-performing general-purpose method for deformable image registration. INSPIRE brings distance measures which combine intensity and spatial information into an elastic B-splines-based transformation model and incorporates an inverse inconsistency penalization supporting symmetric registration performance. We introduce several theoretical and algorithmic solutions which provide high computational efficiency and thereby applicability of the proposed framework in a wide range of real scenarios. We show that INSPIRE delivers highly accurate, as well as stable and robust registration results. We evaluate the method on a 2D dataset created from retinal images, characterized by presence of networks of thin structures. Here INSPIRE exhibits excellent performance, substantially outperforming the widely used reference methods. We also evaluate INSPIRE on the Fundus Image Registration Dataset (FIRE), which consists of 134 pairs of separately acquired retinal images. INSPIRE exhibits excellent performance on the FIRE dataset, substantially outperforming several domain-specific methods. We also evaluate the method on four benchmark datasets of 3D magnetic resonance images of brains, for a total of 2088 pairwise registrations. A comparison with 17 other state-of-the-art methods reveals that INSPIRE provides the best overall performance. Code is available at github.com/MIDA-group/inspire.