Skip to content

Commit

Permalink
Merge pull request #1 from murphygroup/v1.2
Browse files Browse the repository at this point in the history
V1.2 create and use PyPI packages
  • Loading branch information
rfmurphy authored Feb 8, 2024
2 parents 6d1d0ae + 15b8794 commit 273ce8c
Show file tree
Hide file tree
Showing 40 changed files with 2,384 additions and 414 deletions.
Binary file added .DS_Store
Binary file not shown.
30 changes: 25 additions & 5 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,14 +1,34 @@
# 3DCellComposer - A Versatile Pipeline Utilizing 2D Cell Segmentation Methods for 3D Cell Segmentation
Haoran Chen and Robert F. Murphy\
Carnegie Mellon University\
V1.1 Dec 14, 2023
V1.2 February 7, 2024

We introduce 3DCellComposer, a versatile, open-source software designed as a general solution for 3D cell segmentation. It allows users to choose any existing 2D segmentation model appropriate for their tissue or cell type(s) without requiring any additional training. Moreover, we have enhanced our CellSegmentationEvaluator quality evaluation tool to support 3D images. It allows users to compare and select the most suitable 2D segmentation models for 3D tasks, without the need for human annotations to assess performance.
3DCellComposer is a versatile, open-source software designed as a general solution for 3D cell segmentation. It allows users to choose any existing 2D segmentation model appropriate for their tissue or cell type(s) without requiring any additional training. Moreover, we have enhanced our CellSegmentationEvaluator quality evaluation tool to support 3D images. It allows users to compare and select the most suitable 2D segmentation models for 3D tasks, without the need for human annotations to assess performance.

Reference: Haoran Chen and Robert F. Murphy (2023) 3DCellComposer - A Versatile Pipeline Utilizing 2D Cell Segmentation Methods for 3D Cell Segmentation
It is available as a full-featured GitHub repository, and as a python package on PyPi that provides a simplified implementation that uses just DeepCell as the 2D segmenter.

Reference: Haoran Chen and Robert F. Murphy (2023) 3DCellComposer - A Versatile Pipeline Utilizing 2D Cell Segmentation Methods for 3D Cell Segmentation. Under review.

## Manual
## Using the PyPI package

To use the package,
```bash
pip install ThreeDCellComposer
```

Then import the function,
```bash
from ThreeDCellComposer.ThreeDCellComposer import ThreeDCellComposer
```

To call the function
```bash
ThreeDCellComposer(image_path,nucleus_channel_marker_list,cytoplasm_channel_marker_list,membrane_channel_marker_list,segmentation_method)
```
Where the channel lists are lists of strings consisting of the names (not numbers) of the channels to be used for segmentation. Only 'deepcell' is supported as the segmentation_method by the PyPI package at this time.


## Using the repository version

### Overview
The `3DCellComposer` script processes multiplexed imaging data for cell segmentation. It requires specific inputs including the path to the image file and lists of markers for different cell components (nucleus, cytoplasm, and cell membrane). It also includes an optional input for specifying the 2D segmentation method(s) to be utilized.
Expand Down Expand Up @@ -71,5 +91,5 @@ In this command, the script processes the image located at `./data/3D_IMC_image.
## Contact

Robert F. Murphy - murphy@cmu.edu\
Haoran Chen - hrchen@cmu.edu
Haoran Chen - haoran.chen@stjude.org

Binary file added evaluation/.DS_Store
Binary file not shown.
Binary file removed evaluation/model/pca_cell_culture.pkl
Binary file not shown.
Binary file removed evaluation/model/pca_tissue.pkl
Binary file not shown.
Loading

0 comments on commit 273ce8c

Please sign in to comment.