Instance segmentation of mitochondria in electron microscopy images with a generalist deep learning model trained on a diverse dataset

作者：Ryan Conrad, Kedar Narayan · 发表于：Cell Systems · 年份：2023 · DOI：10.1016/j.cels.2022.12.006 · 被引用次数：122 · 研究领域：Mitochondrial Function and Pathology、Advanced Electron Microscopy Techniques and Applications、Machine Learning in Materials Science

Mitochondria are extremely pleomorphic organelles. Automatically annotating each one accurately and precisely in any 2D or volume electron microscopy (EM) image is an unsolved computational challenge. Current deep learning-based approaches train models on images that provide limited cellular contexts, precluding generality. To address this, we amassed a highly heterogeneous ∼1.5 × 10 6 image 2D unlabeled cellular EM dataset and segmented ∼135,000 mitochondrial instances therein. MitoNet, a model trained on these resources, performs well on challenging benchmarks and on previously unseen volume EM datasets containing tens of thousands of mitochondria. We release a Python package and napari plugin, empanada, to rapidly run inference, visualize, and proofread instance segmentations. A record of this paper's transparent peer review process is included in the supplemental information.