
Our postdoc Adrià Casamitjana has won the best paper award at the SASHIMI 2021 workshop (Simulation and Synthesis in Medical Imaging), organized as part of MICCAI. The paper, "Synth-by-Reg (SbR): Contrastive Learning for Synthesis-Based Registration of Paired Images", a registration loss for weakly supervised image translation between domains that does not require perfectly aligned training data. This loss uses a registration CNN with frozen weights to drive a synthesis CNN towards the desired translation. Preservation of structures is further encouraged with a complementary loss based on contrastive learning. In the paper, we show results on 3D histology reconstruction of two datasets. You can find a preprint of the article [here], and the code [here].
Big congrats, Adrià!