Artificial Intelligence @ The Netherlands Cancer Institute

Central repository of WSI for AI

We develop deep learning techniques to enhance the capabilities of breast imaging including digital breast tomosynthesis, mammography and dynamic contrast-enhanced breast MRI

Cell annotation game

Immunotherapy response prediction

HRD assessment on WSI

Medical imaging is the cornerstone of modern medicine. We build deep learning algorithms to reconstruct the measured machine data to an image of the patient anatomy.

Our AI research in immunology investigates interactions between T-cell receptors and peptides, and fundamental steps in the development of personalized cancer vaccines.

Immunotherapy is a systemic cancer treatment that exploits the power of the body’s immune system to fight cancer cells by boosting the immune response. We develop algorithms to predict immunotherapy outcomes.

AI enables new diagnostic and treatment paradigms. However, its application to oncology brings many questions and challenges. Inspired by the oncological application, we research and develop new AI methodologies.

Multiplex image analysis

Patient monitoring in cancer care is crucial to track the efficacy of anti-cancer treatments and overall patient condition, utilizing methods as follow-ups and response assessments. Our goal is to leverage AI methods to propose optimized treatment strategies, objective progression definition and diagnostic precision.

Automated artifact detection on WSI

Radiogenomics brings together multiple disciplines. Our goal is to study the impact of somatic mutations on the morphology of the tumour, to identify radiomic signatures predictive of clinically relevant molecular profiles, and to explore MR techniques to predict tumour microenvironment.

An accurate risk assessment of breast cancers or their precursors is paramount to deciding on the appropriate treatment regime.

Pre-treatment imaging has the power to give insights into response to therapy before it is given. Our goal is to leverage advanced machine learning combining imaging modalities to craft more personalized treatment plans. Currently these algorithms focus on prostate, rectal and oesophageal cancer, by using self-supervised and semi-supervised learning, image-to-image translation and multimodal AI.

It is extremely common that Deep Learning algorithm performs well in one medical dataset but is unable to generalize to a different dataset. Our goal is the development of robust algorithms, techniques to share data and understanding of AI decisions, by focusing on generalizability, privacy and explainable AI.

Image encoders are often trained for project-specific tasks, failing to capture an optimal representation of the medical scan. Our goal is to develop powerful encoders capable of generating high-quality embeddings, and to subsequently use them for clinical-use cases, such as automatic radiology report generation or promptable segmentation via textual queries.

Genomic landscape of Breast cancer

Virtual (immuno) staining