How can clinicians and patients choose the most suitable treatment? Read more about what our researchers are working on: s.fhg.de/r4g #ClinicalDecisionSupport #AI #DigitalMedicine #Fraunhofer #WeKnowHow

I am happy to talk about image registration in pathology and how it is useful to train AI at the seminar series @TIAwarwick (and a bit sad that it will be remote). Join us on Monday at 3pm CET via zoom, there will be math, images, and (maybe) a surprise!

Mit dezentraler #KI die Behandlung von #Prostatakrebs verbessern: #Forschungsprojekt PROSurvival gestartet: s.fhg.de/5Gc #FederatedLearning

Improved diagnostic performance by multitask pretraining from heterogeneous medical data—find out more in our e-poster at #ECR2023 in #Vienna. s.fhg.de/58e #AI #TransferLearning #ComputationalPathology #DigitalRadiograophy #MRI

Our Tissue Concepts model scored second at the #SemiCol challenge on detection and segmentation of colorectal cancer! 🥳 It is a multi-task learning pipeline aimed at small datasets and generalizability to unseen data. It detected almost all cancer (AUC 0.997) on external data.

In Deutschland werden leider oft die Daten besser geschützt als die Patienten. Ohne Forschungsdaten können wir uns zB an der Spitze der Krebsforschung nicht halten. Auch nutzen wir oft und gerne Daten aus Ländern wo es kaum Datenschutz gibt, auch unethisch tagesschau.de/wissen/gesundh…

Topic of my talk today: Robustness of AI in pathology can be achieved by pre-training medical „foundational“ models with diverse data from pathology. This can best be done together by the community.

New #preprint “Overcoming Data Scarcity in Biomedical Imaging with a Foundational Multi-Task Model”: doi.org/k5fj #AI #DeepLearning #MultitaskLearning #FoundationalModel

New paper “Overcoming data scarcity in biomedical imaging with a foundational multi-task model in Nature Computational Science: nature.com/articles/s4358… #FoundationModel #MedicalImaging #Radiology #ComputationalPathology #MultiTaskLearning

#PressRelease: MEVIS develops a multitasking approach to train foundation models efficiently with minimal data. Data scarcity in medical imaging can thus be overcome. 👉s.fhg.de/JdYY #MedicalImaging #Biomarkers #Pathology #AI #Fraunhofer