5 papers
Self-supervised Pre-training Helps Retinal Disease Progression Modelling Most When Data Is Scarce
Ifeoma Veronica Nwabufo, Julius Gervelmeyer, Sarah Müller +1
Modelling how a disease progresses over time requires longitudinal imaging cohorts, which are scarce and small, whereas cross-sectional data -- one image per participant -- is abun…
Ordinal Diffusion Models for Color Fundus Images
Gustav Schmidt, Philipp Berens, Sarah Müller
Generative image models such as diffusion models can improve performance on clinically relevant tasks by offering deep learning models supplementary training data. However, most co…
Mitigating Shortcut Learning via Feature Disentanglement in Medical Imaging: A Benchmark Study
Sarah Müller, Philipp Berens
Although deep learning models in medical imaging often achieve excellent classification performance, they can rely on shortcut learning, exploiting spurious correlations or confoun…
Learning Disease State from Noisy Ordinal Disease Progression Labels
Gustav Schmidt, Holger Heidrich, Philipp Berens +1
Learning from noisy ordinal labels is a key challenge in medical imaging. In this work, we ask whether ordinal disease progression labels (better, worse, or stable) can be used to…
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…