BioSpecNorway / IRmHiRegistration

Code for the paper: Grayscale representation of infrared microscopy images by extended multiplicative signal correction for registration with histological images

Installation

pip install git+https://github.com/BioSpecNorway/IRmHiRegistration.git

Remarks

1. We have encountered issues when writing a huge histological image to a file using PIL. For such a case, pyvips worked like a charm. However, for simplicity, we excluded pyvips from the dependencies of the current project.
2. In the original paper, we tested different grayscale transformations of H&E images. The luminance conversion based on the linear formula performed equally well compared to more sophisticated approaches based on deconvolution. Therefore, we didn't include deconvolution-based grayscale transformations.
3. Registration API (ABCRegistrator, LandmarkRegistrator, etc.) was created to solve the author's needs. If you have suggestions on how to improve it please create PR.

Citation

If you find this repository useful in your research, please consider citing the following paper:

@article{trukhan2020grayscale,
  title={Grayscale representation of infrared microscopy images by extended multiplicative signal correction for registration with histological images},
  author={Trukhan, Stanislau and Tafintseva, Valeria and T{\o}ndel, Kristin and Gro{\ss}erueschkamp, Frederik and Mosig, Axel and Kovalev, Vassili and Gerwert, Klaus and Kohler, Achim},
  journal={Journal of biophotonics},
  volume={13},
  number={8},
  pages={e201960223},
  year={2020},
  publisher={Wiley Online Library}
}