Patrick Köhler

Patrick Köhler
Patrick is s PhD student in the Department of Data Science at the Hertie Institute for AI in Brain Health at the University of Tübingen and works with medical images. He develops validation mechanisms for diagnostic AI tools to ensure that they are only employed in safe environments. To this end, he studies the progress in various branches of Machine Learning and reflects on its potential and relevance in clinical contexts. His major fields of interest include Distribution Shifts, Disentangled Representations, Confounded Predictors and Performance Estimators.
Trustworthy AI requires carefully designed validation procedures that are tailored to the exact use case at hand.
01
Köhler, P., Fadugba, J., Berens, P., Koch, L.M.

Efficiently correcting patch-based segmentation errors to control image-level performance in retinal images

Jun 06, 2024 | Medical Imaging with Deep Learning 2024