The AI in Oncology Lab (ANN) centers its research on three core themes: improving cancer diagnosis, optimizing therapy and risk assessment, and exploring novel AI concepts in oncology.
To drive these advancements, the lab leverages key technical AI components, focusing primarily on:
With a commitment to explainable and responsible AI systems.
The AI in Oncology Lab is dedicated to bridging the gap between technical AI research and practical oncological applications.
The lab’s mission and vision are to develop innovations in artificial intelligence to build the fundamental knowledge base, technological tools, and human capacities necessary to transition to a new generation of cancer care. This clinical transition is supported and enhanced by smart, tailored technology designed to reduce breast cancer overtreatment, improve risk assessment for breast and ovarian cancer, track tumors in real-time during radiotherapy, and augment overall treatment planning.
The impact of the lab lies in improving cancer diagnosis and therapy while training and retaining the next generation of interdisciplinary talent. This is achieved by validating breakthroughs through close collaboration between The Netherlands Cancer Institute and the University of Amsterdam, involving clinical end-users (such as oncologists and pathologists) in the validation and hands-on training of these solutions, and embedding explainable and responsible AI frameworks to build trust across the medical ecosystem.
Breast cancer screening and overtreatment reduction – Focuses on using AI to discover why certain screen-detected Ductal carcinoma in situ (DCIS) lesions remain harmless while others progress into invasive breast cancer, helping clinicians make more informed treatment decisions.
Risk assessment and rare cancer diagnostics – Focuses on using AI to predict breast cancer treatment outcomes through histopathological and genetic data, and to accurately diagnose rare ovarian cancer subtypes.
Novel AI concepts in oncology – Focuses on exploring boundary-pushing AI concepts, including deep learning on non-Euclidean topologies (such as graphs or manifolds) to predict drug responses and analyzing causal relationships between data and treatment outcomes.
Real-time tumor tracking on MR-linac – Focuses on developing AI models to predict real-time tumor movement and deformation using prior imaging to enable faster, more precise radiotherapy tracking.
Multimodal tumor segmentation – Focuses on developing AI approaches that incorporate shape-based inductive biases and extra clinical data to capture complex tumor geometry and micro-invasions rather than simple voxel-by-voxel classification.
Netherlands Cancer Institute is among the top 10 comprehensive cancer centres, combining world-class fundamental, translational, and clinical research with dedicated patient care.
University of Amsterdam (UvA) is the Netherlands’ largest university, offering the widest range of academic programmes.
