On- and offline-Networks for Event Aware Topological detection

Imaging methods enabling the acquisition of millions of cell contours have opened the path for improved understanding of development, repair, homeostasis and pathology.

The full potential of such imaging methods can only be reached when robust, cost-effective and user-friendly methods are democratized to extract important information from the huge amounts of data generated.

Our project aims to implement a disruptive deep-learning based technology, O-NEAT, that uses these masses of data for training neural networks to automatically and reproducibly explore tissue dynamics.

This would enable researchers in academia and the private sector to quickly and reliably extract information regarding cell dynamics in normal or pathological conditions thus having significant potential market applications.