The PathAI organization aims to foster collaboration and streamline code-sharing across all digital pathology projects conducted at the Leiden University Medical Center (LUMC). Our overarching goal is to standardize digital pathology workflows, from quality control (QC) and slide processing to downstream tasks like multiple instance learning (MIL), segmentation, and graph neural networks (GNNs).
Standardization enhances efficiency, reduces errors, and accelerates research by ensuring reproducibility and resource optimization. Access to the organization is restricted to LUMC researchers, who can be invited by one of the moderators. However, select repositories, such as PathBench-MIL, will be publicly available.
- Deep Learning–Based Classification of Early-Stage Mycosis Fungoides and Benign Inflammatory Dermatoses
- PathBench-MIL: A Comprehensive, Flexible Benchmarking / AutoML Framework for Multiple Instance Learning in Histopathology
- HECTOR: Multimodal Deep Learning to Predict Distant Recurrence-Free Probability from Digitized H&E Tumor Slides and Tumor Stage
- Im4Mec: Interpretable Deep Learning Model to Predict the Molecular Classification of Endometrial Cancer
🚧 To be announced!
🚧 To be announced!
🚧 To be announced!
Access is available to researchers working in or collaborating with LUMC’s digital pathology research projects. Please contact a moderator for more information.
