Histopathological Image
Histopathological image analysis focuses on extracting meaningful information from microscopic images of tissue samples, primarily to aid in disease diagnosis and prognosis. Current research heavily utilizes deep learning, employing convolutional neural networks (CNNs), transformers, and graph neural networks (GNNs) for tasks like image segmentation, classification, and multimodal data integration (e.g., combining H&E and immunofluorescence images, or integrating genomic data). These advancements are significantly impacting healthcare by improving diagnostic accuracy, accelerating workflows, and potentially enabling more personalized medicine through improved prediction of treatment response and disease progression.
Papers
Contrastive Deep Learning Reveals Age Biomarkers in Histopathological Skin Biopsies
Kaustubh Chakradeo (1), Pernille Nielsen (2), Lise Mette Rahbek Gjerdrum (3 and 6), Gry Sahl Hansen (3), David A Duchêne (1), Laust H Mortensen (1 and 4), Majken K Jensen (1), Samir Bhatt (1 and 5) ((1) University of Copenhagen, Section of Epidemiology, Department of Public Health, Copenhagen, Denmark, (2) Technical University of Denmark, Department of Applied Mathematics and Computer Science, Denmark, (3) Department of Pathology, Copenhagen University Hospital- Zealand University Hospital, Roskilde, Denmark, (4) Danmarks Statistik, Denmark, (5) Imperial College London, United Kingdom, (6) Department of Clinical Medicine, University of Copenhagen, Copenhagen, Denmark)
PriorPath: Coarse-To-Fine Approach for Controlled De-Novo Pathology Semantic Masks Generation
Nati Daniel, May Nathan, Eden Azeroual, Yael Fisher, Yonatan Savir