HistomicsTK
HistomicsTK copied to clipboard
A Python toolkit for pathology image analysis algorithms.
# The problem There is a subtle edge case with with [compute_nuclei_features](https://github.com/DigitalSlideArchive/HistomicsTK/blob/1200e540c0184a2fde3a2e7e85644ce93f1e1a11/histomicstk/features/compute_nuclei_features.py#L10) for the cytoplasm features. In particular, [dilate_xor](https://github.com/DigitalSlideArchive/HistomicsTK/blob/1200e540c0184a2fde3a2e7e85644ce93f1e1a11/histomicstk/features/compute_nuclei_features.py#L184) is not guaranteed to for cytoplasm for every single nuclei i.e....
This addresses Issue #946. .features.graycomatrixext determines appropriate ranges for calculations based on image type, and converts the image to float after. Conversion to uint8 is not necessary and prevents these...
My understanding of [compute_haralick_features](https://github.com/DigitalSlideArchive/HistomicsTK/blob/f737e1dd0010ade64069140c42c8c572a65a31a6/histomicstk/features/compute_haralick_features.py#L9) is that you can provide im_intensity as a 2d array of float intensity values between 0 and 1. However, as written, the function will **always** cast...
Python Histomicstk package should be adjusted to include `xml` files and to expose the clis on the command line.
Demonstrate adjusting an algorithm with a widget (like a slider for a float parameter) in a Jupyter notebook. Postitive pixel count is a good example to start with. This involves...
We should highlight that Jupyter notebook examples that are stand-alone. For instance, the color deconvolution notebook can be run in google colab by just adding a pip install command as...
I installed histomicstk using `pip install histomicstk` then get the following error message when I try to import the package ```python from histomicstk.features import compute_morphometry_features ~/anaconda3/envs/comp_onc/lib/python3.8/site-packages/histomicstk/__init__.py in 1 # import...
When I am trying the nuclei segmentation on Windows 10, basically same code as your demo https://digitalslidearchive.github.io/HistomicsTK/examples/nuclei-segmentation.html I got the ValueError ValueError Traceback (most recent call last) in 22 23...
Create documentation for an example algorithm being wrapped for HistomicsUI using slicer_cli_web, and integrate this into existing documentation.
The new scikit-image version now uses the attribute name "centroid_weighted" instead of "weighted_centroid". In addition to this, there were bugs with accesses the attribute as prop.centroid_weighted, so now changed it...