Images and annotations for training and testing of Glycogen-segmenter

These uploads contain raw images and annotations for training and testing the Glycogen-segmenter model presented in this paper:

A semantic segmentation model to predict subcellular glycogen localization using transmission electron microscopy images. Anders Aagaard Hansen, Jacob Marcus Egebjerg, Kristian Solem, Kristoffer Jensen Kolnes, Daniel Wüstner, Jørgen F. P. Wojtaszewski, Jørgen Jensen, and Joachim Nielsen bioRxiv 2026.02.04.703703; doi: https://doi.org/10.64898/2026.02.04.703703

The folders are:
region: contain 859 pairs of images and annotations for training the region-model
glycogen: contain 100 pairs of images and annotations for training the glycogen-model
test: contain 92 pairs of images and annotations for testing the model
test_ss: contain 84 images for testing the model on subsarcolemmal glycogen

The model weights are also uploaded (region_model.pth and glycogen_model.pth)

The code for traning, testing, and also a GUI can be found here: https://github.com/JoachimNielsenSDU/Glycogen-segmenter