@inproceedings{10.1117/12.3005413,
author = {Noah Heldt and Cornelia Holzhausen and Martin Ahrens and Mario Pieper and Peter K{\"o}nig and Gereon H{\"u}ttmann},
title = {{Improved image quality in dynamic OCT imaging by reduced imaging time and machine learning based data evaluation}},
volume = {PC12830},
booktitle = {Optical Coherence Tomography and Coherence Domain Optical Methods in Biomedicine XXVIII},
editor = {Joseph A. Izatt and James G. Fujimoto},
organization = {International Society for Optics and Photonics},
publisher = {SPIE},
pages = {PC128302A},
abstract = {Dynamic Optical Coherence Tomography combines high resolution tomographic imagery with a cell specific contrast by Fourier analysis. However, the conversion from frequency space into RGB images by binning requires a priori knowledge and artifacts due to global movements provide another obstacle for in vivo application.
We could show that an automated binning based on the Neural Gas algorithm can yield the highest spectral contrast without a priori knowledge and that motion artifacts can be reduced with shorter sequence lengths. Imaging murine airways, we observed that even just 6 frames are enough to generate dOCT images without losing important image information.},
keywords = {Dynamic OCT, Optical Coherence Tomography, Airways, Artificial Intelligence},
year = {2024},
doi = {10.1117/12.3005413},
URL = {https://doi.org/10.1117/12.3005413}
}

@InProceedings{pmlr-v143-gruening21a,
  title = 	 {Direct Inference of Cell Positions using Lens-Free Microscopy and Deep Learning},
  author =       {Gruening, Philipp and Nette, Falk and Heldt, Noah and de Souza, Ana Cristina Guerra and Barth, Erhardt},
  booktitle = 	 {Proceedings of the Fourth Conference on Medical Imaging with Deep Learning},
  pages = 	 {219--227},
  year = 	 {2021},
  editor = 	 {Heinrich, Mattias and Dou, Qi and de Bruijne, Marleen and Lellmann, Jan and Schläfer, Alexander and Ernst, Floris},
  volume = 	 {143},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {07--09 Jul},
  publisher =    {PMLR},
  pdf = 	 {https://proceedings.mlr.press/v143/gruening21a/gruening21a.pdf},
  url = 	 {https://proceedings.mlr.press/v143/gruening21a.html},
  abstract = 	 {With in-line holography, it is possible to record biological cells over time in a three-dimensional hydrogel without the need for staining, providing the capability of observing cell behavior in a minimally invasive manner. However, this setup currently requires computationally intensive image-reconstruction algorithms to determine the required cell statistics. In this work, we directly extract cell positions from the holographic data by using deep neural networks and thus avoid several reconstruction steps. We show that our method is capable of substantially decreasing the time needed to extract information from the raw data without loss in quality.}
}