My contributions build upon the ability to ask & translate biomedical questions into math, and the skills to develop new image processing workflows using deep learning, among others. These are the different works that brought me the chance to get such skills and prove myself as a scientist:
Cell motility. During my doctorate, I studied cell motility in 3D mesenchymal cell migration. I’m particularly interested in deciphering the role of the dendritic protrusions formed by the cells to produce contractile movements. Can we relate their mechanics with the cell motility patterns? What is their role in the 3D mesenchymal migration? To answer these questions, I developed automatic image analysis methods to process phase-contrast time-lapse movies of cells embedded in 3D Collagen Type I matrices (segmentation, tracking and detection).
Small extracellular vesicles. Small extracellular vesicles play a critical role in the cell communication of many physiological processes, such as tumor growth. Due to their nanosize, studying their molecular cargo and structure is limited to challenging isolation clearing protocols. Moreover, imaging extracellular vesicles is only possible with super-resolution imaging that then, requires an advance image analysis. Together with collaborators, we contributed an automatic deep-learning based method to analyse the structure of extracellular vesicles in transmission electron microscopy images.
Biostatistics. Automatic image processing allows us generating tones of data that need to be analysed afterwards. Namely, dealing with big numbers requires new analytical approaches. For this, we developed a novel approach to assess statistical null hypotheses (
H0) in the big-data paradigm (yes, p-values are data size-dependent).
FAIRy deep learning. Yet more popular, I’m one of the creators of deepImageJ -an environment to bridge deep learning to ImageJ-, a core team member of the BioImage Model Zoo, a developer of ZeroCostDL4Mic and DeepBacs, and also a collaborator of the Cell Tracking Challenge. All these projects contribute to making the awesome image analysis technology accessible, reusable and reproducible.
Advance bioimaging. Working on the interface between analytical imaging, algorithm development and biology is incredibly enriching to identify the critical limitations of the field and to build multidisciplinary pipelines that enable new approaches in research. At the moment I’m combining microscopy imaging and live cell experiments to understand how the analytical pipelines can be optimised while enabling new scientific discoveries.