Search Results for author: Eike Petersen

Found 9 papers, 2 papers with code

Slicing Through Bias: Explaining Performance Gaps in Medical Image Analysis using Slice Discovery Methods

no code implementations17 Jun 2024 Vincent Olesen, Nina Weng, Aasa Feragen, Eike Petersen

Sex-based differences in the prevalence of these shortcut features appear to cause the observed classification performance gap, representing a previously underappreciated interaction between shortcut learning and model fairness analyses.

Fairness

Fast Diffusion-Based Counterfactuals for Shortcut Removal and Generation

1 code implementation21 Dec 2023 Nina Weng, Paraskevas Pegios, Eike Petersen, Aasa Feragen, Siavash Bigdeli

Via a novel inpainting-based modification we spatially limit the changes made with no extra inference step, encouraging the removal of spatially constrained shortcut features while ensuring that the shortcut-free counterfactuals preserve their remaining image features to a high degree.

counterfactual Counterfactual Explanation +1

Are Sex-based Physiological Differences the Cause of Gender Bias for Chest X-ray Diagnosis?

no code implementations9 Aug 2023 Nina Weng, Siavash Bigdeli, Eike Petersen, Aasa Feragen

In this work, we investigate the causes of gender bias in machine learning-based chest X-ray diagnosis.

Fairness

That Label's Got Style: Handling Label Style Bias for Uncertain Image Segmentation

no code implementations28 Mar 2023 Kilian Zepf, Eike Petersen, Jes Frellsen, Aasa Feragen

Segmentation uncertainty models predict a distribution over plausible segmentations for a given input, which they learn from the annotator variation in the training set.

Image Segmentation Segmentation +1

On (assessing) the fairness of risk score models

no code implementations17 Feb 2023 Eike Petersen, Melanie Ganz, Sune Hannibal Holm, Aasa Feragen

Further, we address how to assess the fairness of risk score models quantitatively, including a discussion of metric choices and meaningful statistical comparisons between groups.

Fairness

Feature robustness and sex differences in medical imaging: a case study in MRI-based Alzheimer's disease detection

1 code implementation4 Apr 2022 Eike Petersen, Aasa Feragen, Maria Luise da Costa Zemsch, Anders Henriksen, Oskar Eiler Wiese Christensen, Melanie Ganz

Instead, while logistic regression is fully robust to dataset composition, we find that CNN performance is generally improved for both male and female subjects when including more female subjects in the training dataset.

Alzheimer's Disease Detection Binary Classification +4

Responsible and Regulatory Conform Machine Learning for Medicine: A Survey of Challenges and Solutions

no code implementations20 Jul 2021 Eike Petersen, Yannik Potdevin, Esfandiar Mohammadi, Stephan Zidowitz, Sabrina Breyer, Dirk Nowotka, Sandra Henn, Ludwig Pechmann, Martin Leucker, Philipp Rostalski, Christian Herzog

This survey provides an overview of the technical and procedural challenges involved in creating medical machine learning systems responsibly and in conformity with existing regulations, as well as possible solutions to address these challenges.

BIG-bench Machine Learning Federated Learning +1

On Approximate Nonlinear Gaussian Message Passing On Factor Graphs

no code implementations21 Mar 2019 Eike Petersen, Christian Hoffmann, Philipp Rostalski

Factor graphs have recently gained increasing attention as a unified framework for representing and constructing algorithms for signal processing, estimation, and control.

Cannot find the paper you are looking for? You can Submit a new open access paper.