Computational Pathology Research Group

The Computational Pathology Group of the Department of Pathology of Radboudumc works in close collaboration with DIAG. The history of the group reaches back to the early 1980s, while the current group has been active since 2013. We have a strong translational focus, owing to our strong embedding in a clinical environment. We have a leading position in our research field, witnessed for instance by the highly successful CAMELYON16 and CAMELYON17 grand challenges which we organized and by our active role in committees and conference organisation (e.g. the Computational Pathology Symposium, held in conjunction with the European Congress of Pathology).

Faculty

Technicians

Researchers

Students

Projects

Projects focus on two areas:

1. increasing efficiency of the current pathology diagnostic process by developing and validating computer aided diagnosis (CAD) workflows;

2. supporting personalized medicine by developing image based biomarkers that are predictive for patient outcome or therapy response. The latter may be based either on H&E stained tissue sections or immunohistochemistry.

Below you will find an overview of current research projects. Feel free to approach us with any questions or suggestions!

CAMELYON16 Grand Challenge
CAMELYON16 Grand Challenge The goal of this challenge is to evaluate new and existing algorithms for automated detection of metatases in hematoxylin and eosin (H&E) stained whole-slide images of lymph node sections. This task has a high clinical relevance but reqiures large amounts of reading time from pathologists. More...
MULTIMOT
MULTIMOT MULTIMOT is a H2020 EU funded project that aims to build an open data ecosystem for cell migration research, through standardization, dissemination and meta-analysis efforts. More...
Prim4BC
Prim4BC No biomarkers currently exist for assessing the risk of triple negative breast cancer recurrence. Availability of such a biomarker would enable selection of patients at increased risk, requiring adjuvant chemotherapy. More...
AQUILA
AQUILA The development and prospective, multicentre validation of algorithms for extraction of tissue based biomarkers from WSI resulting in a robust and reproducible method that will yield optimal prognostic data in a cost effective way. More...
AMI
AMI Fraunhofer MEVIS and DIAG have started a large joint project called Automation in Medical Imaging (AMI). Funded as an ICON project, AMI will develop a generic platform for automatic medical image analysis. More...
OralTil
OralTil In this project we focus on evaluating potentially prognostic biomarkers for oral squamous cell carcinomas using deep learning algorithms. More...
Pronn4BC
Pronn4BC This project will apply digital pattern recognition to objectively quantify stromal characteristics in digitally scanned tissue sections of breast cancer tissue. To develop a model that is capable of assessing the most discriminative features, ‘hypothesis free’ deep neural networks will be applied. More...
Sys-Mifta
Sys-Mifta Interstitial fibrosis and tubular atrophy (IF/TA) are major clinical challenges in kidney transplantation. Systems medicine opens a new avenue towards unraveling these complex interactions involving chronic cellular and humoral rejection, interstitial inflammation, and macrophage-mediated tissue remodeling processes.. More...
STITPRO (finished)
STITPRO (finished) This project established a software environment for research and application of digital image analysis algorithms to support the daily work of the pathologist. More...
VPH-PRISM (finished)
VPH-PRISM (finished) In this project we focus on automated recognition and grading of breast tumors. Methods will be developed to classify the identified regions of interest into tumor and normal regions. Subsequently, algorithms will be developed to estimate tumor aggressiveness and predict patient outcome. More...
Rectum
Rectum Visual estimation of tumor and stroma proportions in microscopy images yields a strong, Tumor-(lymph)Node- Metastasis (TNM) classification-independent predictor for patient survival in colorectal cancer. More...
Prostate
Prostate In this project, we will combine deep learning and digitized whole-slide imaging of prostate cancer for reproducible extraction of quantitative biomarkers. Furthermore, due to the ability of deep learning systems to learn relevant features without human intervention, we expect to identify novel biomarkers which allow us to further improve the current risk models. More...