Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
Institute for Biomedical Informatics, Faculty of Medicine, University Hospital Cologne, University of Cologne, Center for Molecular Medicine Cologne (CMMC)
3
0
3
You don't need predefined features: a Transformer trained on raw microscopy videos can predict cancer cell fate with high accuracy, revealing the temporal dynamics of cellular decision-making.
DINO, a self-supervised ImageNet model, surprisingly outperforms current state-of-the-art methods for corneal nerve fiber tortuosity grading, even without segmentation maps.
Graph Transformers can beat state-of-the-art image-based models at classifying cancer cells in whole-slide images by explicitly modeling tissue-level context.