Fibers to Cells

Virtual Cresyl violet staining from 3D-PLI.

Quickstart

Clone the repository and install it using pip:

    git clone https://jugit.fz-juelich.de/inm-1/bda/personal/aoberstrass/projects/pli2cells.git
    cd pli2cells
    pip install -e .

Installation requires gxx_linux-64 and openmpi or mpich packages.

Usage

Debugging the pipeline:

sbatch scripts/debug.sbatch

Full-scale training:

sbatch scripts/train.sbatch unetdftstyle_affine

where the unetdftstyle_affine experiment can be replaced with any configuration under configs/experiment.

To apply trained models adjust and run

sbatch scripts/apply_models.sbatch

Project Organization

├── configs                       <- Hydra configuration files
│   ├── callbacks                     <- Callbacks configs
│   ├── datamodule                    <- Datamodule configs
│   ├── debug                         <- Debugging configs
│   ├── experiment                    <- Experiment configs
│   ├── hparams_search                <- Hyperparameter search configs
│   ├── local                         <- Local configs
│   ├── log_dir                       <- Logging directory configs
│   ├── logger                        <- Logger configs
│   ├── model                         <- Model configs
│   ├── trainer                       <- Trainer configs
│   │
│   └── train.yaml                    <- Main config for training
│
├── data                          <- Project data
│   └── vervet1818-stained            <- Repository containing train and test data
│
├── logs
│   ├── experiments                   <- Logs from experiments
│   ├── slurm                         <- Slurm outputs and errors
│   └── tensorboard/mlruns/...        <- Training monitoring logs
│
├── scripts                       <- Scripts used in project
│   ├── apply_model.py                <- Script to apply a trained model
│   ├── apply_models.sbatch           <- Job submission to apply a model
│   ├── debug.sbatch                  <- Debug training job
│   ├── train.py                      <- Run training
│   └── train.sbatch                  <- Training job
│
├── src/pli_cyto                  <- Source code
│   ├── datamodules                   <- Lightning datamodules
│   ├── eval                          <- Code for evaluation scores
│   ├── models                        <- Lightning models
│   ├── utils                         <- Utility scripts
│   │
│   ├── testing_pipeline.py           <- Model evaluation workflow
│   └── training_pipeline.py          <- Model training workflow
│
├── LICENSE.txt                   <- Apache License Version 2.0
├── pyproject.toml                <- Build configuration.
├── setup.cfg                     <- Declarative configuration of the project.
└── README.md                     <- This file

DataLad

To retrieve the training data run

datalad get datasets/vervet1818-stained/

or

datalad get --reckless=ephemeral datasets/vervet1818-stained

if you just want to link to the data on a remote without copying the files. Additional sources of submodules are specified as datalad.get.subdataset-source-candidate in .datalad/config (See the doc).

Please note that access to the data can only be provided on request.

How to Cite

If you use this work in your research, please cite it as follows:

@article{oberstrass2025,
  title={From Fibers to Cells: Fourier-Based Registration Enables Virtual Cresyl Violet Staining From 3D Polarized Light Imaging},
  author={Oberstrass, Alexander and Vaca, Esteban and Upschulte, Eric and Niu, Meiqi and {Palomero-Gallagher}, Nicola and Graessel, David and Schiffer, Christian and Axer, Markus and Amunts, Katrin and Dickscheid, Timo},
  journal={arXiv preprint arXiv:2505.11394},
  year={2025}
}