R. W. Conrad

Instance segmentation of mitochondria in electron microscopy images with a generalist deep learning model.

The paper develops a deep learning model for mitochondria instance segmentation in EM images and releases a tool for inference and visualization.

CEM500K - A large-scale heterogeneous unlabeled cellular electron microscopy image dataset for deep learning.

The paper introduces CEM500K, a large-scale unlabeled EM dataset for deep learning, showing pre-trained models achieve state-of-the-art results on segmentation tasks.