Date
Thursday March 14, 2024 from 3:30 PM to 4:30 PMLocation
Neuron 0.262Co-organizer
Biomedical EngineeringPrice
freeDomain Generalization and Representation Learning in Microscopy and Beyond
Katharina Breininger, Head of the Artificial Intelligence in Medical Imaging (AIMI) group at Friedrich-Alexander-University Erlangen-N眉rnberg, Germany.
Title | Domain Generalization and Representation Learning in Microscopy and Beyond
A central goal when applying machine learning in the biomedical research and medical imaging is to answer relevant interdisciplinary questions robustly and reliably across various setups and ideally modalities; however, this is challenged by different imaging systems, differences between a laboratory setting and the real world as well as a multitude of other factors. This talk will explore the challenge of domain generalization in digital pathology, microscopy, and beyond. Specifically, we will discuss the Mitosis Domain Generalization (MIDOG) Challenge that took place at the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in 2021 and 2022 as one example how we can work toward more robust machine learning approaches. Additionally, the talk will highlight the potential of representation learning and synthetic data for bridging domain gaps and understanding the mechanisms of machine learning in data- and annotation-limited settings.
Registation is required and free of charge 鈫�
Professor Katharina Breininger
Katharina Breininger heads the Artificial Intelligence in Medical Imaging (AIMI) group at Friedrich-Alexander-University Erlangen-N眉rnberg, Germany. Previously, she studied computer science in Marburg and Erlangen. During her PhD in a collaboration between Siemens Healthineers and FAU, she worked on algorithms for image fusion during minimally invasive interventions.
Her group focuses on interdisciplinary issues in medical imaging across various applications and modalities with the goal of developing robust and reliable machine learning approaches. She is particularly interested in the advancement of algorithms in intraoperative and multimodal imaging as well as machine learning in relation to microscopy.
Additionally, her group deals with fundamental aspects of machine learning: developing tools for efficient labeling and the generation of datasets, investigating the influence of human behavior in this process, and exploring methods for label-efficient learning.
Eindhoven Artificial Intelligence Systems Institute
The Eindhoven Artificial Intelligence Systems Institute (EAISI) is the central hub for artificial intelligence research at Eindhoven University of Technology (黑料福利网). EAISI brings together researchers across engineering, computer science, and applied domains to develop AI methods, systems, and applications for industry and society.