Our tooling has been developed for building powerful and custom AI models to reconstruct neurons, organelles and other features from biological microscopy data.
We offer a full service including training data generation and proof-reading in order to achieve high-quality results for your use cases.
● Neuron Segmentation● Axons & Dendrites Reconstruction● Synapse and Dendritic Spine Detection● Heuristical split and merge error handling● Annotation and proof-reading services available
Using our Voxelytics Connect tooling, we offer a repeatable, automated approach to circuit reconstruction and generating Connectome matrices from raw electron microscopy data.
We collaborate closely with your group in order to make the most out of your data. As part of a full service, we optionally generate the required training data and proof-read the results.
● Neuron reconstruction and synapse detection● Heuristics for split and merge error fixes● Annotation and proof-reading services available● Ready for 3D volume electron microscopy● (SBEM, Multi-SEM. FIB-SEM, TEM)● Fully integrated, collaborative workflow in WEBKNOSSOS
Blood Vessel and Nuclei Segmentation
Segment large biological features, such as blood vessels, nuclei or somata, for analysis or masking purposes.
● Blood Vessel, Somata, and Nuclei Segmentation● No training data required● Excellent for operations requiring masks
1. Book an intro call
Discuss your research goals and data characteristics with us. Define the analysis tasks and arrange data access on WEBKNOSSOS.
2. Receive a free segmentation sample for your data
Once we have access to your data, we will perform a segmentation on a subsample of your data (typically 1 GVx). Based on this, we can discuss the next steps and evaluate the need for re-training.
3. Optional retraining for your data
We have a large selection of pre-trained models for various types of EM images. However, sometimes it is required to retrain models for particular image characteristics. In that case, our annotators can generate the required ground truth and we will train custom models for optimal results.
4. Automated processing
We roll out our machine learning pipeline Voxelytics on your data. The processing pipeline includes stack alignment, neuron segmentation, neurite type detection, nuclei/somata/blood vessel classification, synapse detection, and connectome assembly.
5. Polish your results in WEBKNOSSOS
Visualize and evaluate the results in WEBKNOSSOS. Use the advanced proofreading tools in WEBKNOSSOS to correct any remaining errors on the objects you care about. Benefit from the collaboration features to speed up this process.
6. Work on your scientific analysis
Explore the results in WEBKNOSSOS and use the available Python libraries for scientific analysis. Of course, you can download the data at any time!
Loomba et al. submitted a comparative study of neural structures in 8 Mouse, Macaque, and Human datasets (SBEM, each 0.5-2TB).
For that, a reliable, repeatable, highly-scalable solution for Connectome reconstruction across three different species with significant differences in cellular morphology was required.
The automated image alignment, neuron segmentation, and connectome reconstruction were carried out using Voxelytics.We applied the default alignment workflows of Voxelytics on the respective dataset without manual tweaking. The out-of-the-box neuron segmentation and synapse detection workflows worked well for the mouse datasets. Any parameter tweaks could be made in a config file and consecutive Voxelyics runs would only execute the changed workflow tasks.
Dense neuron segmentation of mouse layer 4 somatosensory cortex
Full dense neuron instance segmentation using modified U-Nets and hierarchical agglomeration. Read blog post.
Synapse, vesicle, and mitochondria detection in cortex
CNN-based segmentation of all synapses, vesicles, and mitochondria in preparation for synaptic connectivity mapping.
Axon and dendrite classification
Integrate semantic segmentation of neuron subtypes (axon, dendrite, glia, etc) into the agglomeration to prevent merger error based on prior biological knowledge.