Dr. ir. Martijn Starmans is Assistant Professor at Erasmus MC (Rotterdam, the Netherlands), where he leads the AI for Integrated Diagnostics (AIID) research line. He holds a dual appointment in the Department of Radiology & Nuclear Medicine (BIGR) and the Department of Pathology (PHANTOM).

His research focuses on developing clinically valuable, trustworthy, and scalable medical AI. A central theme throughout his work is that AI systems should learn not only from individual datasets, but also from prior studies, clinical applications, and existing knowledge. To this end, he develops automated machine learning, foundation-model-based, and multimodal AI methods that generalize across healthcare domains. His primary research interest is integrated diagnostics, combining information from radiology (“radiomics”) and pathology (“pathomics”) into unified AI systems (“radiopathomics”) to support precision oncology.

Martijn’s work spans multiple cancer types, including sarcoma, liver cancer, colorectal cancer, bladder cancer, melanoma, and neuroendocrine tumors. To establish his independent research line, he received a prestigious €2 million NGF AINed Personal Fellowship in 2024. Together with collaborators from the BCN-AIM Lab led by Prof. Karim Lekadir, he contributed to the development of the FUTURE-AI initiative for trustworthy medical AI. For his contributions to equitable and globally accessible medical AI, he received the MICCAI Health Equity Award.

Collaborations & Consortia

Beyond methodological research, Martijn is actively involved in building the international medical AI community through leadership roles in open science, research infrastructure, and global collaboration. Together with Prof. Josien Pluim, he serves as General Chair of MICCAI 2030 in Rotterdam. The conference is organized by an international committee spanning 32 countries across all six continents and aims to advance clinically valuable, trustworthy, and globally accessible medical AI.

Martijn is co-founder and co-lead of the MICCAI SIG Open Data, which promotes the sharing and reuse of medical imaging datasets and benchmarks. Since its launch at MICCAI 2024, the initiative has become an annual event at MICCAI, bringing together researchers, clinicians, industry partners, and dataset creators through scientific presentations, keynote lectures, and community discussions on open data standards. To support underrepresented diseases and populations, he established the AFRICAI repository for sharing accepted Open Data datasets.

Martijn is one of the initiators and principal investigators of the Sarcoma Artificial Intelligence (SAI) consortium (grant awarded), the Liver AI (LAI) consortium (grant awarded), and project lead of the Colorectal Liver Metastases AI (COLIMA) consortium. Across these initiatives, 51 clinical centers, companies, professional societies, and patient organizations from 18 countries collaborate on developing clinically impactful AI solutions. He is also affiliated with the Trustworthy AI for MRI ICAI Lab.

Martijn has a strong affinity with research infrastructure and data sharing. Besides the MICCAI Open Data initiative, he contributes to several large international projects focused on enabling collaborative and trustworthy AI research. He is work package leader of AFRICAI-RI, a Pan-African research infrastructure for collaborative biomedical imaging and artificial intelligence; was work package leader of EuCanImage, which developed a European cancer imaging platform for AI in oncology; and EOSC4Cancer, which established foundations for large-scale European data-driven cancer research. Additionally, he serves as external advisor to RadioVal and contributes to EUCAIM, Europe’s federated cancer imaging infrastructure.

For a more complete overview of ongoing projects and collaborations, see his BIGR profile page.

PhD Degree

Martijn obtained his PhD degree cum laude on February 1, 2022 with the thesis Streamlined Quantitative Imaging Biomarker Development: Generalization of Radiomics through Automated Machine Learning. Motivated by a fascination for automation and generalization, he developed the automated machine learning framework WORC, enabling the efficient development of radiomics models across a wide variety of clinical applications. This work is described in this publication and forms the basis of many subsequent methodological developments in automated machine learning for medical imaging.

His thesis was nominated for the Fredrik Philipsprijs 2023 for Best Dutch Radiology Thesis and was selected among the top five finalists.

Download my PhD Thesis.

Educational Activities

Martijn co-founded two courses and has taught multiple courses within the BSc and MSc Technical Medicine programmes of Erasmus MC and TU Delft. He currently teaches within the MSc Applied Physics programme at TU Delft. He also contributed to the first AFRICAI Summer School as mentor and programme committee member, and to the inaugural Rotterdam Radiology AI Course organized by the Dutch Radiological Society.

Martijn enjoys mentoring the next generation of researchers and has supervised a large number of BSc, MSc, internship, and thesis projects across medical imaging, AI, and integrated diagnostics.

Download my (slightly outdated) resumé / CV .