Artificial Intelligence (AI) offers substantial opportunities for healthcare, supporting better diagnosis, treatment, prevention and personalised care. Analysis of health images is one of the most promising fields for applying AI in healthcare, contributing to better prediction, diagnosis and treatment of diseases. In order to develop and test reliable AI applications in the field, access to large-volume of high- quality data is needed.
This action should contribute to testing and developing AI tools and analytics focused on the prevention, prediction and treatment of the most common forms of cancer while providing solutions to securely share health images across Europe.
Proposals should set up and contribute to populate a large interoperable repository of health images, enabling the development, testing and validation of AI–based health imaging solutions to improve diagnosis, disease prediction and follow-up of the most common forms of cancer.
The repository should include high quality, interoperable, anonymised or pseudo-anonymised data sets of annotated cases, based on data donorship, and should comply with relevant ethics, security requirements and data protection legislation. Gender aspects should be considered appropriately. It should ensure data quality and interoperability based on common standards and open Application Programming Interfaces (APIs).
Proposers should specify measures for validating AI-based solutions for health images, such as the effectiveness of clinical decision making. There should be rigorous, peer-reviewed scientific evidence establishing their safety, validity, reproducibility, usability, reliability and usefulness for better health outcomes. It is critical to show how AI-based solutions will deal with and inform about possible failures, inaccuracies and errors. Adequate performance metrics, monitoring and evaluation criteria and procedures should be put in place. The reasoning behind AI-based conclusions and recommendations should be explained so that users can understand their situation and be able to consent or challenge any proposed course of action.
The consortium should build on relevant national and EU activities and bring together:
- expertise to set up the infrastructure, ensuring the appropriate sharing of data quality and interoperability,
- AI developers/expertise to experiment its content while ensuring compliance with relevant legislations
The Commission considers that proposals requesting from the EUR 8 -10 million would allow this specific challenge to be addressed appropriately. Nonetheless, this does not preclude submission and selection of proposals requesting other amounts.
The proposal should provide appropriate indicators to measure its progress and specific impact in the following areas:
- Contributing towards the creation of a EU-wide repository of health images dedicated to the most common forms of cancer, enabling experimentation of AI-based solutions to improve diagnosis, treatment and follow-up and contribute to a more precise and personalised management of cancer.
- Contributing to developing technical, organisational and ethical standards for AI for health imaging
- Promoting access to anonymised health image data sets to be made more openly reusable across the EU for training AI applications.
- Increasing trust in AI solutions among users (healthcare professionals and patients), investors and stakeholders at industry and academia.
 As reported by the World Health Organisation, see for example https://www.who.int/news-room/fact-sheets/detail/cancer
Illustration Photo: Medical Imaging (free photo from pexels.com)