Advanced medical imaging techniques hold great potential to facilitate personalized medicine and health research. Imaging modalities such as MRI, CT, PET, and Ultrasound are increasingly used to diagnose patients and guide treatment decisions. Emerging artificial intelligence (AI) methods (radiomics, machine learning, deep learning) allow for computer-assisted diagnosis based on medical images. Examples are automated cancer detection, early diagnosis of dementia, disease subtyping, prognosis, treatment response prediction, and monitoring of disease progression. Combining medical images with other data types such as genetic data, therapy responses, histopathology findings, and other clinical data provides insight into disease etiology and may suggest new avenues for treatments.
An advanced technical infrastructure is required to collect, anonymize, clean-up & structure, store, share, inspect & annotate, process & analyze imaging data, and to combine these data with other data types. Imaging data sets tend to be large, complex, and heterogeneous in nature. Their analysis requires substantial computing power. These factors pose challenges at multiple levels: individual researchers, institutes, and multicenter studies.
The Health-RI Imaging community aims to give the research community access to the medical imaging infrastructure that is available at Dutch research institutions, including software tools, services, an imaging data archive, computing power, training, and networking opportunities. In addition, the community aims to jointly improve this infrastructure.
Please contact the Imaging community manager Stefan Klein if you want to learn more about the Imaging community or if you want to contribute.
Stefan Klein, Erasmus MC
By jointly tackling the infrastructural challenges in medical imaging, we aim to improve the efficiency, quality and reproducibility of our research.
Joining forces and aligning practices in the field of medical imaging will improve harmonization and standardization. This will lead to more reproducible results, and thus will increase the quality of research in this field. Sharing and combining imaging data will facilitate the analysis of larger data sets (more cases, leading to more reliable conclusions), and will thereby enable the development and validation of novel AI methods for automated image analysis. Combining imaging data sets with other types of data will allow researchers to answer new research questions.
The Imaging community consists of radiologists; IT experts; researchers in imaging physics, image analysis, and machine learning; and other professionals that are already collaborating in various projects.
Mode of operation & governance structure
The work is performed ‘by and for’ the community rather than imposed in a top-down manner. Activities often take place in the context of financed projects, of which current examples are the Horizon 2020 EuCanImage project and the ZonMW-funded Netherlands Consortium of Dementia Cohorts (NCDC). Specifically on the topic of XNAT, a national knowledge exchange network is being set up, and a workshop is being organized. Similar focused interest groups will be formed when there is sufficient interest from the community to cooperate on a particular infrastructural challenge.
Related communities and projects, websites and initiatives
Services by Health-RI
Related projects and initiatives
- EPI2: European Population Imaging Infrastructure
- Grand Challenges
- H2020 EUCanImage: A European Cancer Image Platform Linked to Biological and Health Data for Next-Generation Artificial Intelligence and Precision Medicine in Oncology
- NCDC: ZonMW Netherlands Consortium of Dementia Cohorts
- OpenMR Benelux
Selected open-source software tools