My scientific collaborations span national and international networks, connecting research partners across borders.
My collaborations combine methodological research in optimal transport and domain adaptation with applications of machine learning to the biomedical and life sciences. Those presented below offer an overview.
Data-driven approaches to optimize and scale up the microfluidic production of lipid-based nanoparticles.
Structured optimal transport in hyperbolic geometry to compare the phenotypic signatures of monogenic diseases, complex diseases and model organisms, for the interpretation of genome-wide association studies.
Optimal transport methods for domain adaptation, to transfer knowledge from labelled source data to target data that follow a different distribution.
Machine learning to predict the size and encapsulation efficiency of mRNA-loaded lipid nanoparticles, extended within the VECT-PI project to relate lipid chemical structure to endosomal release, efficacy and safety profile.
Deep learning for the segmentation of electron microscopy images and for 3D reconstruction from scanning electron microscopy (SEM) and serial block-face (SBF) image stacks.