We’re proud to report that four new Horizon 2020 projects with a strong focus on medical imaging just kicked off in September and October 2020.
The new projects include the EIBIR-coordinated SINFONIA and EURAMED rocc-n-roll projects in the field of radiation protection research. The CHAIMELON and EuCanImage projects focus on AI in health imaging and aim to set up and contribute to populate large interoperable repositories of health images. In these projects, EIBIR is leading the dissemination activities.
The 3-year EURAMED rocc-n-roll project aims to propose an integrated and coordinated European approach to research and innovation in medical applications of ionising radiation and related radiation protection based on stakeholder consensus and existing activities in the field (incl. existing SRAs of radiation protection platforms, EC health and digitisation programmes, EURATOM-funded projects, SAMIRA initiative). To achieve this, research and radiation protection needs in the clinical disciplines using ionising radiation will be analysed with the aim to generate the largest benefit for the European population in an equal, safe, high-quality way throughout Europe, by fostering clinical translation, while at the same time strengthening economic growth and industrial competitiveness, supported by research and innovation in the field.
Representation of relevant radiation protection disciplines like radiation biology, dosimetry for medical applications, ethics as well as clinical expertise in the fields of radiology, nuclear medicine, radiation therapy, oncology, cardiovascular diseases, neurology, paediatrics complemented by regulatory and health policy, AI and industry experts in the consortium will ensure a fully integrative approach and allow embedding the project work into a broader framework of quality and safety in healthcare. Based on wide stakeholder input and consultation on above aspects, EURAMED rock-n-roll will produce a strategic research for medical applications of ionising radiation and related radiation protection and a corresponding roadmap, as well as an interlink document, integrating the views and identifying synergies from the areas of radiation protection, health research and digital, with impactful guidance to the EC and stakeholders on future research in this area. This will be accompanied by proposed and tested education and training schemes for health workforce and scientists to increase Europe’s research capacity in the field.
Reinforced risk appraisal of medical exposure is needed due to an extensive use of ionizing radiation for diagnosis and therapy. The main objective of the 4-year SINFONIA project is to develop novel methodologies and tools that will provide a comprehensive risk appraisal for detrimental effects of radiation exposure on patients, workers, carers and comforters, the public and the environment during the management of patients suspected or diagnosed with lymphoma and brain tumours.
The scientific work will (1) develop novel AI-powered personalised dosimetry and risk appraisal methods and tools to estimate the radiation burden on patients undergoing state-of-the-art radiological, nuclear medicine and radiation therapy procedures, (2) reinforce risk appraisal for exposed staff, comforters, the public and the environment during nuclear medicine and proton therapy procedures, (3) determine the degree of patient variability in radiation sensitivity for the risk of developing secondary malignancies and (4) design and develop data management techniques for managing data from imaging and non-imaging examinations and radiation therapy sessions. A data repository will be developed for storing data as well as for the deployment of AI algorithms on an online platform.
SINFONIA research outcomes for the two clinical examples, lymphoma and brain tumours, will be also applicable to other diseases. AI-powered personalised dosimetry tools will provide advanced knowledge on parameters affecting radiation detriment. This will help balancing risks and benefits of ionising radiation procedures and developing dose optimisation strategies. Additionally, radiation biology studies will identify individuals with increased susceptibility of developing cancer from ionising radiation exposure. SINFONIA also will organise high-level multidisciplinary training in the field of radiation dosimetry, risk appraisal and radiation protection and develop recommendations on radiological protection.
CHAIMELEON aims to set up a structured repository for health imaging data to be openly reused in AI experimentation forcancer management. An EU-wide repository will be built as a distributed infrastructure in full compliance with legal andethics regulations in the involved countries. It will build on partner ́s experience (e.g. PRIMAGE repository for paediatriccancer and the Euro-BioImaging node for Valencia population, by HULAFE; the Radiomics Imaging Archive by MaastrichtUniversity; the national repository DRIM AI France, the Oncology imaging biobank by Pisa University). Clinical partners andexternal collaborators will populate the Repository with multimodality (MR, CT, PET/CT) imaging and related clinical data forhistoric and newly diagnosed lung, prostate and colorectal cancer patients.
A multimodal analytical data engine will facilitate to interpret, extract and exploit the right information stored at theRepository. An ambitious development and implementation of AI-powered pipelines will enable advancement towardsautomating data deidentification, curation, annotation, integrity securing and images harmonisation, the latest being of thehighest importance for enabling reproducibility of Radiomics when using large multiscanner/multicentre image datasets.
The usability and performance of the Repository as a tool fostering AI experimentation will be validated, including avalidation subphase by other world-class European AI developers, articulated via the organisation of Open Challenges to theAI Community. A set of selected AI tools will undergo early on-silico validation in observational (non-interventional) clinicalstudies coordinated by leading experts in Gustave Roussy (lung cancer), San Donato (breast), Sapienza (colorectal) and LaFe (prostate) hospitals. Their performance will be assessed, including external independent validation, on hallmark clinicaldecisions in response to some of the currently most important clinical end points in cancer
The goal of EuCanImage is to build a highly secure, federated and large-scale European cancer imaging platform, with capabilities that will greatly enhance the potential of artificial intelligence (AI) in oncology. Firstly, the EuCanImage platform will be populated with a completely new data resource totaling over 25,000 single subjects, which will allow to investigate unmet clinical needs like never before, such as for the detection of small liver lesions and metastases of colorectal cancer, or for estimating molecular subtypes of breast tumours and pathological complete response.
Secondly, the cancer imaging platform, built by leveraging the well-established Euro-Bioimaging infrastructure, will be cross-linked to biological and health repositories through the European Genome-phenome Archive, allowing to develop multi-scale AI solutions that integrate organ-level, molecular and other clinical predictors into dense patientspecific cancer fingerprints.
To deliver this platform, the consortium will build upon several key European initiatives in data sharing for personalised medicine research, including EUCANCAn (cancer genomics and health data sharing), euCanSHare (cardiac imaging and omics data sharing) and EUCAN-Connect (federated data analytics). Furthermore, to foster international cooperation and leverage existing success stories, the consortium comprises the coordinators of The Cancer Imaging Archive (TCIA), the US cancer imaging repository funded by the National Cancer Institute. This will allow EuCanImage to leverage a unique 10-year long experience in cancer imaging storage, anonymisation, curation and management. Finally, a close collaboration between world renown clinical, radiomics, AI and legal experts within the consortium and beyond will establish well-needed guidelines for AI development and validation named FUTURE, for delivering Fair, Universal, Traceable, Usable, Robust and Explainable decision support systems for future cancer care.