Steffen Czolbe

I am a PhD Fellow of Machine Learning with the Department of Computer science at the University of Copenhagen, co-advised by Aasa Feragen and Oswin Krause.

My PhD research focuses on developing deep learning methods for medical imaging, with a focus on image registration. In my work I dive into unsupervised & semi-supervised learning, with an emphasis on the registration of tricky, real-world anatomies and an ambition to develop general algorithms that apply to a multitude of problem domains and modalities. I also published on uncertainty quantification and topology detection.

Previously, I obtained a M.Sc. in computer Science from the University of Copenhagen. During my Masters I worked tgether with the LEGO group on teaching reinforcement learning with LEGO Mindstorms Robots, and designed a lightweight perceptual similarity metric for image generation and distance measuring, advised by Christian Igel.

News

Nov 11, 2021	My paper on topological change detection in images has been accepted to NeurIPS 2021! check out the summary here
Nov 8, 2021	My revamped website is online! 🎉
Jul 9, 2021	I am honored to be awarded Best Paper Runner Up at MIDL2021 for my paper on semantic similarity metrics

Selected publications

Find the full list here

NeurIPS

A Loss Function for Generative Neural Networks Based on Watson’s Perceptual Model

Czolbe, Steffen, Krause, Oswin, Cox, Ingemar, and Igel, Christian

In Advances in Neural Information Processing Systems Dec 2020

Summary Talk Code PDF BibTeX

@inproceedings{czolbe2020loss,
  title = {A Loss Function for Generative Neural Networks Based on Watson’s Perceptual Model},
  author = {Czolbe, Steffen and Krause, Oswin and Cox, Ingemar and Igel, Christian},
  booktitle = {Advances in Neural Information Processing Systems},
  volume = {33},
  month = dec,
  year = {2020},
  abbr = {NeurIPS},
  pdf = {neurips2020.pdf},
  github = {https://github.com/SteffenCzolbe/PerceptualSimilarity},
  talk = {https://www.youtube.com/watch?v=qPmHQbR4DeI&ab_channel=Steffen},
  project = {_projects/2020-12-neurips2020_watsons_loss.md},
  selected = {true},
  bibtex_show = {true}
}

NeurIPS
Spot the Difference: Detection of Topological Changes via Geometric Alignment

Czolbe, Steffen, Feragen, Aasa, and Krause, Oswin

In Advances in Neural Information Processing Systems Dec 2021

Summary Talk Code Abstract PDF BibTeX

Geometric alignment appears in a variety of applications, ranging from domain adaptation, optimal transport, and normalizing flows in machine learning; optical flow and learned augmentation in computer vision and deformable registration within biomedical imaging. A recurring challenge is the alignment of domains whose topology is not the same; a problem that is routinely ignored, potentially introducing bias in downstream analysis. As a first step towards solving such alignment problems, we propose an unsupervised algorithm for the detection of changes in image topology. The model is based on a conditional variational auto-encoder and detects topological changes between two images during the registration step. We account for both topological changes in the image under spatial variation and unexpected transformations. Our approach is validated on two tasks and datasets: detection of topological changes in microscopy images of cells, and unsupervised anomaly detection brain imaging.
@inproceedings{czolbe2021topology, author = {Czolbe, Steffen and Feragen, Aasa and Krause, Oswin}, file = {:home/steffen/.local/share/data/Mendeley Ltd./Mendeley Desktop/Downloaded/Anonymous - 2021 - Spot the Difference Detection of Topological Changes via Geometric Alignment.pdf:pdf}, booktitle = {Advances in Neural Information Processing Systems}, title = {{Spot the Difference: Detection of Topological Changes via Geometric Alignment}}, volume = {34}, year = {2021}, month = dec, abbr = {NeurIPS}, github = {https://github.com/SteffenCzolbe/TopologicalChangeDetection}, talk = {https://www.youtube.com/watch?v=MQ4GvM0up5c&ab_channel=Steffen}, project = {_projects/2021-11-neurips2021_topology.md}, pdf = {spot_the_difference_detection_.pdf}, selected = {true}, bibtex_show = {true} }
MIDL Award
Semantic similarity metrics for learned image registration

Czolbe, Steffen, Krause, Oswin, and Feragen, Aasa

Best Paper Award Runner-Up

In Proceedings of Machine Learning Research Jul 2021

Summary Talk Code Abstract PDF BibTeX

We propose a semantic similarity metric for image registration. Existing metrics like Euclidean Distance or Normalized Cross-Correlation focus on aligning intensity values, giving difficulties with low intensity contrast or noise. Our approach learns dataset-specific features that drive the optimization of a learning-based registration model. We train both an unsupervised approach using an auto-encoder, and a semi-supervised approach using supplemental segmentation data to extract semantic features for image registration. Comparing to existing methods across multiple image modalities and applications, we achieve consistently high registration accuracy. A learned invariance to noise gives smoother transformations on low-quality images.
@inproceedings{czolbe2021semantic, archiveprefix = {arXiv}, arxivid = {2104.10051}, author = {Czolbe, Steffen and Krause, Oswin and Feragen, Aasa}, eprint = {2104.10051}, booktitle = {Proceedings of Machine Learning Research}, month = jul, keywords = {Deep Learning,Image Registration,Representation Learning}, title = {{Semantic similarity metrics for learned image registration}}, url = {https://arxiv.org/abs/2104.10051v1}, year = {2021}, abbr = {MIDL}, tag = {Award}, github = {https://github.com/SteffenCzolbe/DeepSimRegistration}, talk = {https://www.youtube.com/watch?v=Hs9X3wSO774&ab_channel=Steffen}, pdf = {semantic_similarity_metrics_fo.pdf}, additional = {<b>Best Paper Award Runner-Up</b>}, selected = {true}, project = {_projects/2021-07-midl2021_deepsim.md}, bibtex_show = {true} }