Generate training data for segmentations

Create dense segmentations for bounding boxes within your dataset. This data will be used to train Machine Learning systems for automatic segmentation. Features of interest can be any biological object from dense neurites, nuclei, blood vessels, organelles to larger features such as brain regions, lesions or entire organs.

Example: Dense segmentation of neurites in EM images of a mouse cortex(EM data by Max Planck Institute for Brain Research, Motta et al. 2019)

Skeletonize neurons

Generate skeleton traces from neurons in your data. Either this can be used directly for scientific discovery or in conjunction with automated analysis, e.g. for evaluation purposes or as training data.

Example: Skeletonization of apical dendrites in the olfactory bulb of a mouse(EM data by Sensory Circuits and Neurotechnology Lab, The Francis Crick Institute)

Proof-read automatic segmentations

State-of-the-art automatic segmentation methods generate good segmentations, but still produce split and merge errors. These can be efficiently corrected through webKnossos' proof-reading tools. 

Example: Fixing over-segmentations by joining falsely split segments(EM data by Max Planck Institute for Brain Research, Motta et al. 2019)

Custom annotations

We are happy to set up custom annotation projects based on your specific needs. Our annotators are flexible and are eager to learn about new kinds of data and biological features. Please get in touch to discuss the specifics of your annotation needs.

Example: Manually segmenting nuclei in mouse cortex(EM data by Max Planck Institute for Brain Research, Motta et al. 2019)