5 papers
Benchmarking Dependence Measures to Prevent Shortcut Learning in Medical Imaging
Sarah Müller, Louisa Fay, Lisa M. Koch +3
Medical imaging cohorts are often confounded by factors such as acquisition devices, hospital sites, patient backgrounds, and many more. As a result, deep learning models tend to l…
This actually looks like that: Proto-BagNets for local and global interpretability-by-design
Kerol Djoumessi, Bubacarr Bah, Laura Kühlewein +2
Interpretability is a key requirement for the use of machine learning models in high-stakes applications, including medical diagnosis. Explaining black-box models mostly relies on…
Benchmarking Retinal Blood Vessel Segmentation Models for Cross-Dataset and Cross-Disease Generalization
Jeremiah Fadugba, Patrick Köhler, Lisa Koch +2
Retinal blood vessel segmentation can extract clinically relevant information from fundus images. As manual tracing is cumbersome, algorithms based on Convolution Neural Networks h…
Disentangling representations of retinal images with generative models
Sarah Müller, Lisa M. Koch, Hendrik P. A. Lensch +1
Retinal fundus images play a crucial role in the early detection of eye diseases. However, the impact of technical factors on these images can pose challenges for reliable AI appli…
Generating Realistic Counterfactuals for Retinal Fundus and OCT Images using Diffusion Models
Indu Ilanchezian, Valentyn Boreiko, Laura Kühlewein +5
Counterfactual reasoning is often used in clinical settings to explain decisions or weigh alternatives. Therefore, for imaging based specialties such as ophthalmology, it would be…