Abstract Connectomics is a developing field aiming at reconstructing the connection of the neural system at nanometer scale. Computer vision technology, especially deep learning methods used in image processing, has promoted connectomic data analysis to a new era. However, the performance of the state-of-the-art methods still falls behind the demand of scientific research. Inspired by the success of ImageNet, we present the U-RISC, an annotated U ltra-high R esolution I mage S egmentation dataset for C ell membrane, which is the largest cell membrane annotated Electron Microscopy (EM) dataset with a resolution of 2.18nm/pixel. Multiple iterative annotations ensured the quality of the dataset. Through an open competition, we reveal that the performance of current deep learning methods still has a considerable gap with human-level, different from ISBI 2012, on which the performance of deep learning is close to human. To explore the causes of this discrepancy, we analyze the neural networks with a visualization method, attribution analysis. We find that in U-RISC, it requires a larger area around a pixel to predict whether the pixel belongs to the cell membrane or not. Finally, we integrate currently available methods to provide a new benchmark (0.67, 10% higher than the leader of competition, 0.61) for cell membrane segmentation on U-RISC and propose some suggestions in developing deep learning algorithms. The U-RISC dataset and the deep learning codes used in this paper will be publicly available.

U-RISC: an ultra-high-resolution electron microscopy dataset challenging existing deep learning algorithms