Whole slide digital pathology image tissue localization.
Clone or download

README.md

tissueloc: Whole slide digital pathology image tissue localization

Build Status Documentation Status PyPI version DOI DOI

Banner

Installation

  1. Install OpenSlide.
$ apt-get install openslide-tools
  1. Installing Python dependencies.
$ pip install opencv-python
$ pip install scikit-image
$ pip install openslide-python
  1. Install tissueloc.
$ pip install tissueloc

Usage example

Interface

def locate_tissue_cnts(slide_path,
                       max_img_size=2048,
                       smooth_sigma=13,
                       thresh_val = 0.80,
                       min_tissue_size=10000):
    """ Locate tissue contours of whole slide image
    Parameters
    ----------
    slide_path : valid slide path
        The slide to locate the tissue.
    max_img_size: int
        Max height and width for the size of slide with selected level.
    smooth_sigma: int
        Gaussian smoothing sigma.
    thresh_val: float
        Thresholding value.
    min_tissue_size: int
        Minimum tissue area.
    Returns
    -------
    cnts: list
        List of all contours coordinates of tissues.
    d_factor: int
        Downsampling factor of selected level compared to level 0
    """

Demo

Testing slide can be downloaded from Figshare.

import tissueloc as tl
slide_path = "../data/SoftTissue/TCGA-B9EB312E82F6.svs"
# locate tissue contours with default parameters
cnts, d_factor = locate_tissue_cnts(slide_path,
                                    max_img_size=2048,
                                    smooth_sigma=13,
                                    thresh_val=0.80,
                                    min_tissue_size=10000)

Documentation

Hosted in https://tissueloc.readthedocs.io, powered by readthedocs and Sphinx.

Contributing

tissueloc is an open source project and anyone is welcome to contribute. An easy way to get started is by suggesting a new enhancement on the Issues. If you have found a bug, then either report this through Issues, or even better, make a fork of the repository, fix the bug and then create a Pull Requests to get the fix into the master branch.

We would like to test this package on more diversified digital slides. Slides (low level images would be better) and their corresponding results are also very welcome as Pull Requests.

License

tissueloc is free software made available under the MIT License. For details see the LICENSE file.

Contributors

See the AUTHORS.md file for a complete list of contributors to the project.

Citing

tissueloc is published in the Journal of Open Source Software - if it's ever useful for your research please consider citing it:

@article{chen2019tissueloc,
  author    = {Pingjun Chen and Lin Yang},
  title     = {tissueloc: Whole slide digital pathology image tissue localization},
  journal   = {J. Open Source Software},
  volume    = {4},
  number    = {33},
  pages     = {1148},
  year      = {2019},
  url       = {https://doi.org/10.21105/joss.01148},
  doi       = {10.21105/joss.01148}
}