College des Enseignants de Radiologie de France (CERF)
Collège des Enseignants de Radiologie de France, CERF, is a non-benefit association that is in charge of education and research for students and residents in Radiology in France. This organisation with around 350 physicians (active members, retired, invited, external personalities, etc.) is present in all university hospitals (32 main hospitals). Education is mainly delivered to students in courses during the first 6 years of medicine (2 books). The program for the residency during years 5 to 6 is carried out by e-learning, in-class conferences and hands-on workshops. Research is organised to drive multicentre studies, to publish national publication indicators to evaluate studies with an ethical comity in order to obtain IRB.
Role in the project
CERF is running the project DRIM France IA: a national initiative address to building and operate the biggest qualified health imaging database in France, aimed at enabling the development of programs and applications in medical imaging for patients and public health. The DRIM database will be one of the largest imaging databases in the world. Its potential is rich in images produced each year in public and private radiology offices and services, that is to say about 100 million. DRIM France IA will make its database available to industrial companies, majors or startups, under the control of a scientific committee.
CERF represents the main academic hospitals as well as the cancer centres. Its expertise lies in multicentre studies, but also ethical aspects underlying the sharing of imaging data. Several groups are working to define the uses of informatics and mathematical solutions (deep learning for example) in real-world clinical practice. The CERF emits recommendations, such as clinical guides to guarantee the best practice.
The DRIM and CERF team involved in CHAIMELEON will lead, mainly, the lung cancer research together with CHUP as experts in the field and, as such, leads the pupulation of the respository eith quality-checked imaging data (WP5).
Medexprim, created in 2015, provides IT services in research and big data projects in biomedical imaging, support healthcare institutions and radiology practices in the efficient and secure exploitation of their imaging exams and associated data, for research, and in the valorisation of these resources in the context of external collaborations. Medexprim Radiomics Enabler® solution automates the identification and extraction of large numbers of image sequences from a PACS for secondary use in research (e.g. radiomics, AI algorithms feeding). Radiomics Enabler® is combined with the Clinical Trials Processor, an open-source solution edited by the RSNA, for the de-identification, curation of DICOM fields, filtering, indexing, and routing. Medexprim also deploys, configure and provide professional support on this solution. We configure it for your specific needs related to a particular research project.
Role in project
Medexprim brings in-depth knowledge of the radiology workflow and IT environment in radiology (PACS, RIS, EHR), and DICOM experts to the project. Medexprim is well-known for its experience in the management, deployment and support of solutions that will integrate with the radiologists workflow to:
- Efficiently identify cohorts of patients that are of interest.
- Extract relevant information from PACS, RIS, EHR.
- Annotate/segment images.
- De-identify data.
- Pre-process data towards the desired format.
- Securely route these data to the centralized database.