vesselNN_dataset

An open-source volumetric brain vasculature dataset obtained with two-photon microscopy at Focused Ultrasound Lab, at Sunnybrook Research Institute (affiliated with University of Toronto by Dr. Alison Burgess, Charissa Poon and Marc Santos.

The dataset contains a total of 12 volumetric stacks consisting images of mouse brain vasculature and tumor vasculature.

The structure of the dataset is following:

• denoised contain the BM4D denoised input stacks as multi-layer 8-bit TIFFs
• experiments contain the resulting images from the 1st stage of VD2D-VD2D3D recursive framework that can be fed to the VD2D3D part. Additionally both folders (VD2D_tanh and VD2D3D_tanh) that can be used as they are for inference or can be finetuned further with more data.
• labelscontain the voxel-level dense segmentation labels for the vasculature

Future modifications

• At the moment there is only the denoised stacks (with BM4D which will be accompanied by the raw Poisson noise corrupted stacks straight from the Olympus FV1000MPE multiphoton microscopy system.
• At the moment in the main repository (vesselNN), there are no implementations of the dataset for commonly used frameworks such as TensorFlow, Theano and Caffe which we plan to add making further development easier.

Updated - January 2018

• Added metadata

missing from "santos2015_mixedSizelowContrastVessels.xml" and "santos2015_lowContrastVessels_metadata.xml" which are from the same original .oib acquisition. Settings are similar to "santos2015_tumor_metadata.xml"

Reference

• Teikari, P., Santos, M., Poon, C. and Hynynen, K. (2016) Deep Learning Convolutional Networks for Multiphoton Microscopy Vasculature Segmentation. arXiv:1606.02382.

@article{teikari2016deep,
  title={Deep Learning Convolutional Networks for Multiphoton Microscopy Vasculature Segmentation},
  author={Teikari, Petteri and Santos, Marc and Poon, Charissa and Hynynen, Kullervo},
  journal={arXiv preprint arXiv:1606.02382},
  year={2016}
}