Meyke Hermsen et al present the first multi-class segmentation network for the histopathological analysis of renal tissue. Their work was published online yesterday by the Journal of the American Society of Nephrology.
David Tellez et al present a new method to train neural networks on gigapixel whole-slide images directly, avoiding the need for fine-grained annotations. Their work appeared online yesterday in IEEE TPAMI.
Mehmet Dalmis will defend his PhD thesis with the title' Automated Analysis of Breast MRI: from Traditional Methods into Deep Learning' the 12th of September at 12.30.
Congratulations to Thomas de Bel for winning the Best Poster Award at the second edition of the International Conference on Medical Imaging with Deep Learning held in London this week.
The final meeting of the AMI-project took place last week. The AMI-project was a close collaborative project between the Diagnostic Image Analysis Group and the Fraunhofer Institute for Digital Medicine MEVIS. With the development of a generic platform for automatic medical image analysis, the project was a succes.
Jeroen van der Laak was honored as Nathan Kaufman timely topics lecturer at the 108th annual meeting of the United States and Canadian Academy of Pathology (USCAP). This lecture is regarded as a great honor within the USCAP sphere.
Computational pathology group member Oscar Geessink and colleagues investigated the potential of computer-aided quantification of intratumoral stroma in rectal cancer whole-slide images. This month, their work has been accepted for publication by Celullar Oncology.
20 European researchers gathered last week at the Techno-pôle in Sierre, Switserland to kick-off the European H2020 project ExaMode. The objective of the project is to develop new prototypes for processing large volumes of medical data on exascale computing facilities.
Last month, Computational pathology group-members Francesco Ciompi and Jeroen van der Laak have been awarded 2 grants for the EXAMODE and PROACTING project. Both projects have a focus on cancer research based on deep learning techniques.
Counting of mitotic tumor cells in tissue sections constitutes one of the strongest prognostic markers for breast cancer. David Tellez et al developed a method to automatically detect mitotic figures in H&E stained breast cancer tissue sections based on convolutional neural networks.