ReSCU-Nets: recurrent U-Nets for segmentation of multidimensional microscopy data

Segmenting multi-dimensional microscopy data requires high accuracy across many images (e.g. timepoints or Z slices) and is thus a labour-intensive part of biological image processing pipelines. We present ReSCU-Nets, recurrent convolutional neural networks that use the segmentation results from the previous frame as a prompt to segment the current frame. We demonstrate that ReSCU-Nets outperform state-of-the-art image segmentation models in different segmentation tasks on time-lapse microscopy sequences.