Spine Detection and Extraction

This code segments, detects and extracts the spines from three dimensional in vivo images of dendrites with spines.

Usage

Clone or downlaod the repository and run AutomatedSpineExtractionModule.m to get the segmented image, centerline of the image, detected endpoints, roughly extracted spines and spine count.

Reference

If you use our code for research, please cite our paper P. K. Singh, P. Hernandez-Herrera, D. Labate, M. Papadakis, Automated 3-D Detection of Dendritic Spines from In Vivo Two-Photon Image Stacks, Neuroinformatics (2017). https://doi.org/10.1007/s12021-017-9332-2

Visualization

Download and install ParaView ParaView. To visualize the 3D image ("Test_Volume" in the directory "Data"):

1. Open .mhd ("Test_Volume.mhd") file.
2. Press "Apply".
3. From the drop-down of "Representation", choose "Volume".

To visualize any binary image generated by the code (for example, the segmentation of the image, or extracted spines):

1. Open the .mhd file(s)
2. Press "Apply".
3. From the "Filters" tab, use "Alphabetical" list and select "Iso Volume".
4. In the section "Input Scalars", change "Minimum" to something greater than zero and less than 1.
5. Press "Apply".

Contributions

The segmentation code was developed by Paul Hernandez Herrera and the centerline generation code was developed by David Jiménez.

Data

Thanks to Sajo M and Morishita H, Department of Psychiatry, Department of Neuroscience, Department of Ophthalmology, Mindich Child Health and Development Institute, and Friedman Brain Institute, Icahn School of Medicine at Mount Sinai, New York.