Biomedical Imaging Group Rotterdam
The Biomedical Imaging Group Rotterdam (BIGR) is at the forefront of research in medical image analysis & artificial intelligence (AI). We aim to improve efficiency and quality of healthcare by developing innovative AI methods in medical imaging.
We focus on both fundamental and applied research, covering the topics of image analysis, machine learning, image reconstruction, quantitative imaging biomarkers, image-guided interventions, making use of both research data and routine clinical data.
We have a strong outward look: towards other imaging sources, other diagnostic modalities, integrated diagnostics, collaboration with clinical departments. We have strong collaborations with many researchers, clinicians and industry partners.
EOSC4Cancer has been featured by the Research Executive Agency as a success story in the health domain.
October 13, 2025
Héctor Cadavid et al.’s paper, “MyDigiTwin: A privacy-preserving framework for personalized cardiovascular risk prediction and scenario exploration,” has now been officially published open access in Computers in Biology and Medicine!
September 24, 2025
Jiaqi Tang et al.’s paper, “Manual Registration in AR-Assisted Surgical Navigation: A Comparative Evaluation,” has now been officially published in the International Journal of Computer Assisted Radiology and Surgery!
Imaging infrastructure for the COllaboration for New TReatments of Acute Stroke Consortium
2023 - 2028
The Rotterdam Study is a large population-based cohort designed to investigate causes and consequences of age-related diseases. Since 2005, brain imaging has been performed in over 6,000 participants using a 1.5T MRI scanner. A hardware upgrade in 2020 resulted in changes in signal reception across brain regions, introducing variability in subjects scanned directly before and after the scanner ugrade. This complicates longitudinal analysis, which is essential to study long-term disorders such as dementia. The aim of this project is to evaluate and apply feature-based harmonization methods (e.g., ComBat) and to explore deep learning-based approaches to correct for scanner-related variability in imaging biomarkers. Once harmonized, the data can be used to address longitudinal research questions, such as the association between cortical thickness and dementia risk. There is flexibility to explore specific research questions in the context of dementia. We are looking for a motivated student with an interest in neuroimaging analysis and epidemiology.
The aim is to develop an AI-based algorithm that uses qMRI of a patient to synthesize diagnostic-quality contrast images and/or recommend personalized acquisition settings for subsequent scans. This approach seeks to enhance diagnostic precision, reduce scan time, and enable efficient, patientspecific imaging without redundant acquisitions. This project designed for a motivated master student interested in MR physics, medical image analysis, and AI seeking a 6-9 month master’s thesis starting Aug/Sept 2025. Experience with Python and some familiarity with deep learning (preferably PyTorch) is expected.
Desmoid tumors, also known as aggressive fibromatosis, are rare soft tissue neoplasms that exhibit locally invasive growth but lack metastatic potential. While not typically life-threatening, their infiltrative nature can lead to significant morbidity depending on their location. A large proportion of sporadic desmoid tumors harbor activating mutations in the CTNNB1 gene, which encodes the β-catenin protein. Among these, the S45F mutation is of particular clinical interest as it is associated with increased tumor aggressiveness, higher recurrence rates, and reduced response to conservative therapies.Routine detection of the S45F mutation currently relies on molecular assays such as PCR or sequencing, which, although accurate, are time-consuming, costly, and not always available in low-resource settings. In contrast, hematoxylin and eosin (H&E) stained histopathology slides are inexpensive and routinely available in all diagnostic workflows. Recent advances in computational pathology have shown that deep learning models can infer underlying molecular alterations from subtle morphologic features in H&E slides. This MSc project aims to develop and evaluate a deep learning model that predicts the presence of the S45F mutation in desmoid tumors directly from digitized H&E-stained slides. Such a model could serve as a cost-effective screening tool to complement molecular testing, reduce unnecessary assays, and enhance decision-making in clinical practice. For more information contact: m.starmans@erasmusmc.nl or k.prathaban@erasmusmc.nl.
Biomedical Imaging Group Rotterdam is a part of Erasmus MC
