BCM3D 2.0
About The Project
BCM3D 2.0 is entirely complementary to the approach utilized in BCM3D 1.0. Instead of training CNNs to perform voxel classification, we trained CNNs to translate 3D fluorescence images into intermediate 3D image representations that are, when combined appropriately, more amenable to conventional mathematical image processing than a single experimental image. Using this approach, improved segmentation results are obtained even for very low SBRs and/or high cell density biofilm images.The improved cell segmentation accuracies in turn enable improved accuracies of tracking individual cells through 3D space and time. This capability opens the door to investigating time38 dependent phenomena in bacterial biofilms at the cellular level.
Getting Started
This package was tested on a high-performace computing cluster at University of Virginia with CUDA-enabled GPU and Tensorflow 2.x
Prerequisites
This package uses CSBdeep for training deep leanirng modules. Tensorflow 2.x with its dependencies (CUDA, cuDNN) is required. Please refer to CSBdeep for further instructions (http://csbdeep.bioimagecomputing.com/doc/install.html).
Installation
- create and activate a conda enviroment
conda create -n BCM3D2.0 python=3.8 conda activate BCM3D2.0
- pip install
pip install BCM3D2.0
Usage
Model training
-
Use src/IntermediateImageGenerate.py to generate image representations from ground truth cell arrangement.
-
Use dategen.ipynb to generate training pairs (raw data and the corresponding image representations).
-
Use train.ipynb to train CNNs that will train a model.
Prediction
-
Use src/preprocess.py to apply background substraction to raw data. (optional)
-
Use predict.ipynb to generate segmentations
License
Distributed under the MIT License. See LICENSE.txt for more information.
Contact
yw9et@virginia.com; jz3nc@virginia.edu; ag5vu@virginia.edu
Project Link: https://github.com/GahlmannLab/BCM3D-2.0
Data Link: https://osf.io/m4637
1 maintainer
Metadata
1 maintainer
Release files for BCM3D 2.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| BCM3D-2.0.1.tar.gz | 6.9 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| BCM3D-2.0.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 15.1 kB
Release files / BCM3D-2.0.1.tar.gz
| Download URL | BCM3D-2.0.1.tar.gz |
|---|---|
| Size | 6.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
7df72116b6741102c65e24c99252bf11d3053de6df932e9dbbb428f8053772ad
|
|
BLAKE2b-256 checksum How to use checksums |
e342c15f7c8567c800f6848bacfae34b0870ab9ec9d8b560d1348130b9b9c6bf
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.1 CPython/3.8.5
|
Release files / BCM3D-2.0.1-py3-none-any.whl
| Download URL | BCM3D-2.0.1-py3-none-any.whl |
|---|---|
| Size | 8.2 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
af16b3d3c51db8c87241c24572edf1915189c8e32974611c3ca5ba51285436f3
|
|
BLAKE2b-256 checksum How to use checksums |
9e2fae3735e8f596768e7e223de0071d3965bf197eae82c793ee0b88091f5b22
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.1 CPython/3.8.5
|
