Pattern Recognition group & AIMI @ BVM 2025 in Regensburg

At this year’s BVM workshop in Regensburg, researchers and students from the Pattern Recognition Group and the AIMI lab showcased their work on diverse medical imaging topics.

Jingna Qiu showcased her work on prototype sampling via text-to-image retrieval to reduce the annotation effort in histopathological tasks, originally presented at MICCAI 2024 [link & blog post].

From Marc Aubrevilles group, Sweta Banerjee (FH Flensburg, PhD student at CAIDAS) presented a Comprehensive Dataset of Coarse Tumor Annotations for The Cancer Genome Atlas Breast Invasive Carcinoma [link]. Jonathan Ganz and Jonas Ammeling (TH Ingolstadt, PhD student with the AIMI group) presented their work on foundation model evaluation in the context of mitotic figure detection [link & blog post] and abstracts on re-identification from histopathology images [link] and considerations when using IHC staining to generate reference annotations for mitosis detections [link].

Luis Carlos Rivera Monroy introduced his work on multi-level cancer profiling through joint cell-graph representations, which was presented in Smart Health last year [link].

It was great to also meet our collaboration partners from FH Flensburg, Prof. Marc Aubreville, and from VetMetUni Wien, including Prof. Christof Bertram and his group.

For further details on our work, feel free to explore the related papers or reach out to us directly.

Many thanks to this year’s BVM organizers for facilitating great presentations, a wonderful social event and many engaging discussions; we look forward to BVM 2026 in Lübeck.