Recursive Deep Prior Video: A super resolution algorithm for time-lapse microscopy of organ-on-chip experiments

Biological experiments based on organ-on-chips (OOCs) exploit light Time-Lapse Microscopy (TLM) for a direct observation of cell movement that is an observable signature underlying biological processes. A high spatial resolution essential to capture dynamics and interactions from recorded by TLM. Unfortunately, due physical cost limitations, acquiring videos not always possible. To overcome the problem, we present here new deep learning-based algorithm extends well-known Deep Image Prior (DIP) TLM Video Super Resolution without requiring any training. The proposed Recursive method introduces some novelties. weights DIP network architecture are initialized each frames according recursive updating rule combined with efficient early stopping criterion. Moreover, loss function penalized two different Total Variation-based terms. has been validated synthetic, i.e., artificially generated, as well real OOC related tumor-immune interaction. achieved results compared several state-of-the-art trained learning algorithms showing outstanding performances.