EuCanImage will build and demonstrate a GDPR-compliant and scalable platform for leveraging large-scale, high-quality and interoperable cancer imaging datasets adequately linked to biological and health cancer data. The platform will integrate advanced capabilities and new standards to develop and validate integrative decision support systems for precision oncology with increased clinical trust and adoption. The project consortium is an experienced and ambitious academic-industrial-clinical partnership, with a proven track record in data management, responsible data sharing, cancer imaging research, and AI for personalised medicine.

During 2024, the EC extended the project by 12 months. As a result of this change, the consortium will continue to work on the platform and AI tools until the end of September 2025. In addition, the project has started to recruit radiologists for collective segmentation efforts, which will contribute to the study on direct analysis of the impact of an annotator on differences in AI training. In terms of dissemination, the project has been presented at conferences such as ECR and CIRSE. A new video featuring and interviewing WP2 leader Dr Maciej Bobowicz has been published on YouTube and other social media channels. In addition, the consortium organised a dedicated hands-on workshop in Rotterdam and a webinar on data repositories for AI development together with the CHAIMELEON and EUCAIM projects. This year was also groundbreaking for communication efforts, as the project launched its own showcase page on LinkedIn.

The CHAIMELEON project aims to create a structured repository of health images and related clinical and molecular data for the most prevalent cancers in Europe: lung, breast, prostate, and colorectal. This EU-wide interoperable repository is designed to facilitate the development and validation of artificial intelligence tools for improved cancer management. Its key objectives include providing access to large-scale databases in line with legal and ethical requirements, establishing a quality-checked repository as a resource for AI tools, building distributed infrastructure based on existing initiatives, and exploring innovative harmonisation approaches. The project also focuses on implementing pipelines to enhance AI integrity and interpretability, validating the repository internally and externally, performing early clinical validation of AI-based solutions, and ensuring the repository’s sustainability beyond the project’s duration while fostering a large and engaged user base.

In 2024, the CHAIMELEON project reached significant milestones and participated in key events, including its presence at ECR and its promotion through webinars with EIBIR and the European Society of Radiology. Several project videos were published on the CHAIMELEON YouTube channel, providing valuable insights into the project’s progress and activities, and major activities such as the re-identification challenges and in-silico validation were completed. The project was extended by six months and is now set to conclude in February 2025, allowing additional time to achieve its goals.

“Transforming brain surgery by advancing functional-guided neuronavigational imaging” will develop a novel cost-effective, transportable, compact optical imaging system (“HyperProbe”) to measure brain function during surgery to achieve better guidance during brain tumour removal by 2027. It will thus critically improve surgical outcomes for patients.

The HyperProbe project will develop and validate a cost- effective, transportable, and easy-to-use device to be integrated with existing clinical instrumentation that is fully suitable for the operating room.

The project will significantly impact clinical practice as the consortium proposes, for the first time, a functional-imaging and machine-based decision-making approach in neurosurgery. The HyperProbe device will be the first multi-biomarker, quantitative optical imaging device and functional imaging device used in image-guided surgery to preserve brain functions and minimise the risk of postoperative neurological impairment and considerably improve the life expectancy of patients.

The project concluded its second year in September 2024. the HyperProbe consortium has made remarkable advances in the design and development of the HyperProbe1.0 system, which was later on upgraded to HyperProbe1.1. This high-performance, versatile, laboratory HSI setup for investigating novel biomarkers for glioma detection and optimisation of hyperspectral parameters has shown promising preliminary results for the clinical translation of its future prototypes. The characterisation of the systems concluded, and major results will be translated to the final device ready for clinical stuy.

Furthermore, in 2024, the project successfully completed the first reporting period and received approval from the Commission with regards to the results submitted to them in this period.

These significant advancements in the project’s first year will help the Consortium advance further and the Consortium will build on this accumulated knowledge to construct together the final device, HyperProbe2, ready for clinical study.

i-Violin was a 24-month project (September 2022 – August 2024) co-funded by the EU, focused on enhancing the quality and safety of radiation technology for cancer diagnosis and treatment. The project, involving 10 partners from 9 countries, concluded in August 2024. It delivered two innovative web-based software tools for image quality assessment and optimisation, available at CTIQA2 and KULeuven. These tools enable the medical community to optimise oncological CT imaging for the chest, pelvis, and abdomen, fostering harmonisation across Europe. The project also highlighted the critical role of multidisciplinary collaboration among radiologists, radiographers, and medical physicists to ensure effective optimisation and patient safety.

RadioVal is breaking new ground as the first international clinical validation study focused on radiomics-based prediction of treatment response to neoadjuvant chemotherapy from breast MRI. With an unwavering commitment to innovation and impact, RadioVal is building a comprehensive and standardised methodological framework for radiomics evaluation, guided by the FUTURE-AI principles of Fairness, Universality, Traceability, Usability, Robustness, and Explainability. At its core, RadioVal aims to establish a new benchmark for radiomics tools by introducing transparent and continuous evaluation and monitoring processes. Through a multi-stakeholder approach, the project has prioritised clinical and healthcare needs alongside socio-ethical and regulatory considerations from its inception. This ensures a well-rounded, patient-centric research trajectory.

The first half of the project successfully laid the foundation, fostering collaboration and dialogue with clinicians, patients, AI developers, and ethicolegal experts. This phase focused on designing the evaluation study and identifying the nuanced needs of all stakeholders, ensuring a robust and practical approach. Now in its second half, RadioVal is transitioning from design to execution. The evaluation study is being conducted using real-world data sourced from eight hospitals across the globe. This pivotal phase not only tests the tools developed but also exemplifies RadioVal’s commitment to validating AI solutions in clinical practice. The active involvement of a multidisciplinary group of stakeholders remains a cornerstone of the project. Engaging clinicians, patients, and developers is critical to ensuring a shared understanding of the project’s goals and its potential impact on healthcare.

The project has also gained significant visibility on international platforms, having been presented at conferences such as ACR. Furthermore, a new video featuring Project Coordinator Oliver Diaz has been published on the project’s YouTube channel, offering insights into the team’s work and the vision driving RadioVal forward.

Investigating the implementation of the Euratom and the EU legal bases with respect to the therapeutic uses of radiopharmaceuticals, the SIMPLERAD project closed its activities by drafting guidelines and recommendations for coherent implementation of these legal requirements and promoting best practices and safer use, including an in-depth analysis of both the interrelations between EU pharmaceutical legislation and Council Directive 2013/59/Euratom and the SIMPLERAD survey of the implementation of European legal requirements for therapeutic nuclear medicine.

A final meeting with the EC and WP MED was held in Luxembourg in March. The changes made according to feedback during the project workshop in December 2023 were presented, and final comments prior to publication of the document were received. The final report was then evaluated by partnering organisations EANM and EFOMP before submission to the EC. Publication of the SIMPLERAD guidelines and recommendations is expected in early 2025.

In 2024, the ODELIA project, a European project focused on developing and implementing a pan-European swarm learning network that enables privacy-preserving and democratic training of medical AI algorithms successfully continued its activities. Moreover, the first periodic reporting to the European Commission was prepared.

During the first reporting period the project embarked on establishing a robust decentralised platform, laying the foundation for collaborative AI model training across multiple institutions without centralised data sharing. A comprehensive manual for implementing SL on-site was compiled, encompassing hardware and software requirements, preprocessing workflows, and seamless integration into existing workflows.

The partners demonstrated the practical application of SL in breast cancer detection using MRI data, marking a significant milestone. The team released a minimal viable product, showcasing decentralised model training capabilities and its efficacy in utilising publicly available data.

A technical documentation website for ODELIA’s open-source SL implementation was set up. This site serves as a valuable resource for those looking to leverage ODELIA’s implementation for diverse use cases, promoting collaborative innovation. The documentation is openly accessible at odelia-ai.github.io.

Furthermore, strides have been made in enhancing the user experience with the development of a user-friendly front-end. This facilitates seamless data gathering for new AI model development while bridging the crucial gap to clinical usability.

First experiments of training a joint model on distributed data from three partners, and publicly available datasets have been performed.

Med-IPUT is a groundbreaking project revolutionizing medical ultrasound imaging. Today, medical ultrasound faces several limitations, such as low resolution, limited penetration depth, and high acoustic peak pressures. Therefore, the MED-IPUT project will develop a high-resolution, high-quality, recyclable medical imaging system based on a disruptive integrated photonic ultrasound transducer concept. Med-IPUT transducers are expected to be up to 100 times more sensitive than current ultrasound sensors. This will be a major leap in ultrasound component performance. The project is aiming to create a new generation of ultrasound imaging sensors that are more accurate, cost-effective, and safer for patients.

During the second year of the project, Med-IPUT:

• Successfully completed the 1st reporting period and two review meetings,
• Prepared a detailed system level design of IPUT demonstrators,
• Provided description of developed SOI technical building blocks and reposted on the design of SOI devices for IPUT demonstrators, and
• Provided description of developed SiN technical building blocks and reposted on the design of SiN devices for IPUT demonstrators.

Additionally, Med-IPUT has been presented during major international events, such as ECR 2024 in Vienna, Austria and IEEE UFFC-JS 2024 in Taipei, Taiwan. Additionally, Med-IPUT has prepared a second explainer video presenting achievements from the first year of the project, which was launched during the first half of 2024.

The project is coordinated by Netherlands Organisation for Applied Scientific Research (TNO), and EIBIR is in charge of the project management and dissemination tasks.

You will find more project information here: https://med-iput.eu/

The SOLACE project (Strengthening the Screening of Lung Cancer in Europe) project, running from April 2023 to March 2026, aims to enhance participation, improve outcomes, and address health inequalities in lung cancer screening (LCS). The project unites 37 partners from 15 EU Member States in a pan-European interdisciplinary consortium. In the first 18 months, SOLACE has achieved significant milestones, including: 1) a systematic evaluation of the LCS landscape across Europe, 2) development of a comprehensive guideline and implementation package for LCS, 3) establishment of the SOLACE Knowledge Hub and Database for screening data, 4) initiation of an assessment on the benefits, harms, and cost-effectiveness of LCS, 5) targeted recruitment efforts for underrepresented groups in LCS, and 6) distribution of a survey on LCS training status across Europe.

EUropean Federation for CAncer IMages (EUCAIM) is the cornerstone of the European Commission-initiated European Cancer Imaging Initiative, a flagship of Europe’s Beating Cancer Plan (EBCP), which aims to foster innovation and deployment of digital technologies in cancer treatment and care to achieve more precise and faster clinical decision making, diagnostics, treatment and predictive medicine for cancer patients.

The 4-year project started in January 2023 and joins 76 partners to deploy a pan-European digital federated infrastructure of FAIR cancer-related, de-identified, real-world images. The infrastructure is designed to preserve the data sovereignty of providers and provide a platform, including an Atlas of Cancer Images, for the development and benchmarking of AI tools towards precision medicine. EUCAIM will address the fragmentation of existing cancer image repositories by building on repositories from the AI4HI initiative, European research infrastructures and national/regional repositories and including clinical images, pathology, molecular and laboratory data.

For more information, visit the project’s website at cancerimage.eu.

This project receives co-funding from the European Union under grant agreement No 945196 and is active until December 31, 2026.

Another Euratom tender project, MARLIN, progressed toward the end of its investigation of the implementation of Council Directive 2013/59/Euratom in the aim of improving reporting and learning from patient-related incidents and near misses in radiotherapy, interventional cardiology, nuclear medicine and interventional and diagnostic radiology. The consortium of experts from EIBIR, EFOMP and ESTRO spent the year finalising their report of guidelines and recommendations to improve implementation of the Directive, current practices of reporting and analysing incidents, and dissemination of lessons learned in each of the fields. The draft document was shared in consultation with important stakeholders across Europe before presentation at the project workshop in Brussels in September, where more than 100 participants learned of the consortium’s work and contributed valuable advice, experience and challenges in discussions. The workshop proceedings and presentations are available on this project webpage.

The final report was then updated based on workshop feedback and submitted to the EC for publication in the Radiation Protection Series. A final meeting with the EC and WP MED is expected in early 2025, followed thereafter by publication.

AI-POD’s mission is to significantly enhance the risk assessment and management of obesity-related vascular diseases. This is achieved through the development of a unique AI-based risk prediction score and Clinical Decision Support System, both grounded in comprehensive individual patient data. By combining these advanced tools with the launch of a Citizen App, AI-POD creates a fully digitalised feedback loop, integrating real-time monitoring of diet and lifestyle into standard risk assessments. This empowers both patients and clinicians, enabling more personalised and timely interventions. Through these innovations, AI-POD aims to redefine the standard of care for obesity-related diseases, with the potential to shape future treatment strategies and clinical guidelines.

In 2024, the project has made significant progress, being featured at international events such as ECR, and welcoming two new partners to the consortium. The establishment of a dedicated Stakeholder Board further strengthens the project’s collaborative framework. Key milestones also include the successful launch of the first end-user webinar, the initiation of patient enrolment, the publication of two major research papers, and the transfer of project coordination to the Medical University of Vienna. Additionally, the release of the AI-POD Citizen App marks an exciting step forward, providing individuals with the tools to actively manage and monitor their cardiovascular health. The project has also released several informative videos on YouTube, further expanding its reach and engagement with both the medical community and the general public.