AI Models
Research in bioinformatics such as within the Utrecht Bioinformatics Center, the Big Data Core Máxima and within the recently established UMC Utrecht Artificial intelligence (AI) labs and Animal Welfare AI lab, further enhances the development of predictive algorithms on molecular (genomic, proteomic, metabolomic) and imaging data.
Diagnostic accuracy and prediction of therapeutic response are crucial in preventing administration of ineffective cancer treatments. AI and in silico models play an instrumental role in improving these areas.
Breakthroughs and impact
The development of advanced AI and in silico models, applied to genetic data, digital pathology or radiology images, improves diagnostics and prediction of treatment responses.
AI identification of brain tumor
Rapid nanopore sequencing combined with a new 'deep-learning algorithm’, speeds up the identification of a specific brain tumor type. Yearly, 1,400 adults and 150 children are diagnosed with a tumor in the brain or spinal cord.
Surgery is often the first step taken in treatment. Brain tumor characterization normally takes a week, but now this new technology, developed by UMC Utrecht and Princess Máxima Center, allows neurosurgeons to adjust their surgical strategies in real-time.
AI models for breast cancer
AI has been shown to improve diagnostic accuracy in pathology. In the CONFIDENT-B trial, real-time assistance of an IVD-approved algorithm for detection of breast cancer metastases in sentinel lymph nodes proved to be safe, cost-effective and fast, thereby significantly reducing pathologists’ workload. This algorithm is now being used in daily clinical practice. UMC Utrecht is at the forefront of AI validation and implementation to improve cancer diagnostics and treatment strategies.
Mathematics of Somatic Evolution of Cancer
Tumors are complex evolving ecosystems of cells pursuing different strategies for survival and reproduction within the body. Through both theoretical and experimental work, researchers at the Science faculty of Utrecht University model this complex system as an evolutionary game. The evolutionary game theory was used to establish the significance of tissue edges on cancer cell motility; and the importance of treatment timing due to social dilemmas of tumour acidity and vasculature. An experimental game assay was developed to measure and study the evolutionary games played by lung cancer to better understand evolution and improve cancer treatment.
Computer Simulations
Predictive models and computer simulations have been used to uncover the relationship between the rheological and mechanical behavior of healthy tissues and cancerous tumors, and the properties of single cells. Researchers investigate the effect of various cell properties (stiffness, chemical factors, actin filament network) on the emergent collective behavior of whole tissues. A better understanding on the link between individual cells and the properties of tumors is important for cancer prognosis and metastatic potential, and thus is expected to influence cancer therapies.
Dense Tissue and Early Breast Neoplasm Screening (DENSE)
Dense breast tissue increases breast cancer risk and hampers detection with mammography. Studies suggest MRI alongside mammography improves cancer detection, but its impact lacks randomized trial scrutiny. The DENSE trial of the UMC Utrecht is pioneering, exploring MRI's added value for women with dense breasts in Dutch biennial mammography screening.
Bacterial DNA signature
The Hubrecht Institute and Princess Máxima Center for pediatric oncology worked together on new techniques to better detect a specific ‘signature’ in DNA caused by bacteria. This group of abnormalities in the DNA is not enough by itself to cause colon cancer, but it is one of the steps in the development of the disease. The novel techniques could potentially be used for future research into the origins of other types of cancer.
Next generation experts
Our responsibility includes educating and training the next generation of researchers to create a dynamic environment in which innovative science will continue contributing to an ever-improving outcome for our patients.
PhD programmes
Within the Graduate School of Life Sciences we offer the following PhD programmes tailored to this goal.
Students can also visit our other PhD programmes or Master's programmes.