NetZeroAICT Digital Contrast for Computerised Tomography Towards Climate Neutral and Sustainable Health Systems

Project Information

NetZeroAICT

Grant agreement ID: 101136679

DOI

10.3030/101136679

EC signature date 7 December 2023

Start date 1 December 2023

End date 30 November 2027

Funded under

• Health

Total cost

€ 3 791 075,00

EU contribution

€ 3 791 075,00

Coordinated by

COLLECTIVE MINDS RADIOLOGY AB
Sweden

Periodic Reporting for period 1 - NetZeroAICT (NetZeroAICT Digital Contrast for Computerised Tomography Towards Climate Neutral and Sustainable Health Systems)

Reporting period: 2023-12-01 to 2025-05-31

Summary of the context and overall objectives of the project

A computerised tomography (CT) scan is one of the most common medical imaging performed in healthcare, over 300 million CT scans performed globally each year , from which ~ 60% are enhanced by iodinated contrast. A study in Lancet Regional Health estimated the average carbon footprint is 9.2Kg per CT scan.
NetZeroAICT aims to reduce the carbon footprint created by CT imaging, by developing state-of-the-art AI software to synthesise ‘CT Digital Contrast’ This will revolutionise the existing CT imaging pathway and promote climate neutral and sustainable health systems, as well as reduce the risk of adverse events that could otherwise be associated with iodinated contrast usage (e.g. renal damage and hypersensitivity).This is through the following objectives
1. Implement legal, ethical and sustainable frameworks that promotes the trustworthiness of NetZeroAICT
2. Establish a centralised, trusted, CT image repository for medical AI development which champions the FAIR principles.
3. Through a central computation platform, CT images in the repository will be accurately classified by its characteristics
4. Implement a ‘green’, sustainable and integrated computational pipeline for the training, validation and deployment of medical AI.
5. Develop and validate 5 clinical applications (Indication For Use, IFU) of ‘Digital Contrast’
6. Validation of the trustworthy NetZeroAICT products with though engaging stakeholders
7. Demonstrate the environmental impact and economical impact of our innovations through comprehensive social-life cycle analyses and health technology assessment
8. Define the roadmap to wider exploitation and impact for the NetZeroAICT ecosystem – towards sustainable and climate neutral health systems
9. Promote awareness of NetZeroAICT innovations

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

During this reporting period, the NetZeroAICT consortium reached a shared understanding of its core values, explicitly aligning with the ethical principles for trustworthy AI as set out by the European Commission’s High-Level Expert Group on AI. Given the interdisciplinary nature of the consortium—comprising industry partners, healthcare professionals, and academic researchers—collaboration was established as an additional foundational value.
Achievements
• Legal, ethical and sustainable frameworks for NetZeroAICT have been established
• CT image repository established with more than 590 k CT datasets with continuous data transfer from each 8 clinical sites
• A contrast phase classifier algorithm was created to classify with 95 % accuracy of CT images in the repository, through the Rhino Federated Computing Platform
• Green computing provider selected to be utilized during the training and validation steps of the “Digital Contrast” algorithm
• Development of a robust methodology applicable to multiple AI functionalities (IFUs). Protocols defined study design, inclusion criteria, reader selection, ground truth methodology, and scoring metrics.Feedback improvement loops will come after each IFU.
• Stakeholder engagement activities initiated since M1 of the project, through the establishment of the Public Advisory Group. In addition, an External Advisory Board has been set up.
• Progress on advancing both the environmental life cycle analysis (E-LCA), updating key assumptions and datapoints and refining the lifecycle inventory and the social life cycle analysis (S-LCA), which develops a system boundary specific to social impacts and reviews literature to contextualise stakeholder perspectives. In addition, the demonstration of the economic impact through the Health Assessment Technology model developed.
• Preparations for the first Innovation management and Consortium exploitation plans due at year 2, through the set-up of an asset register and establishment of the Innovation Management Group (IMG) who meets monthly to review the register.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The established governance model has the potential to become a transferable approach for other AI-driven projects. The advances on the clinical validation of AI-generated synthetic imaging, moving beyond isolated pilots toward multicenter validation using real-world datasets. The use of blinded comparisons between real and synthetic images, supported by software for pairwise quality scoring, represents an innovative evaluation design. This methodology incorporates both objective anatomical metrics and reader confidence scoring, better reflecting clinical decision-making. The results will support CE/FDA certification and offer a reference model for future validation studies.
The HTA framework and regulatory mapping provide a foundation for future assessment and market readiness of the NetZeroAICT technology. The systematic review is expected to offer insights into both cost-effectiveness and sustainability, informing health policy and technology adoption decisions. The potential impact lies in supporting decision-makers with evidence on economic value and environmental responsibility in medical imaging.

To ensure further uptake and long-term impact, the following areas have been identified as key enablers:
• Further validation and demonstration: Continued use across diverse project contexts will drive iterative improvements, further strengthening the tools’ robustness and adaptability.
• Engagement with standardization efforts: Alignment with emerging EU AI governance standards will enhance the tool’s relevance.
• Demonstrated equivalence or non-inferiority in diagnostic performance
• Further development of the regulatory strategy, regulatory-ready documentation of protocols and results
• A scalable framework reusable for other AI tools in radiology and beyond
• Continued access to clinical data and validation results.
• Completion of remaining HTA chapters, particularly those on clinical effectiveness and organisational aspects.