Multiphoton imaging with computational specificity

Digital staining based on machine learning models can provide cellular specificity to label-free optical imaging.

This concept is particularly interesting for in vivo applications in fundamental research of auto-immune diseases as well as for future clinical translations.

In this project MICS Multiphoton imaging with computational specificity, I will develop and implement computational specificity for label-free multiphoton microscopy (MPM) using artificial intelligence (AI).

The direct outcome of this project will be two AI modules to perform (i) automated classification of mucosal inflammation based on 3D images from colon tissue and (ii) digital staining of un-stained immune cells.

This integration of computational specificity to label-free multiphoton microscopy will allow direct investigation of global tissue alteration as well as specific immune cell localization during inflammatory tissue remodelling.

Digital staining is an emerging concept in the field of computational microscopy but has not yet been implemented for immune cells based on label-free MPM images.

Building on my previous expertise in label-free in vivo imaging via endomicroscopy, future implementations of multiphoton endomicroscopy would profit from tools for computational specificity, developed during this project.