Synthetic and scalable data platform for medical empowered AI

Información del proyecto

AISym4MED

Identificador del acuerdo de subvención: 101095387

DOI

10.3030/101095387

Fecha de la firma de la CE 22 Noviembre 2022

Fecha de inicio 1 Diciembre 2022

Fecha de finalización 30 Noviembre 2026

Financiado con arreglo a

• Health

Coste total

€ 6 341 765,00

Aportación de la UE

€ 6 341 765,00

Coordinado por

ASSOCIACAO FRAUNHOFER PORTUGAL RESEARCH
Portugal

CORDIS proporciona enlaces a los documentos públicos y las publicaciones de los proyectos de los programas marco HORIZONTE.

Los enlaces a los documentos y las publicaciones de los proyectos del Séptimo Programa Marco, así como los enlaces a algunos tipos de resultados específicos, como conjuntos de datos y «software», se obtienen dinámicamente de OpenAIRE .

Resultado final

Websites, patent fillings, videos etc. (3)

Dissemination and Communication Plan – Version 2.0

Report of D&C results, outlining the approach, activities, channels and tools, used until M24, any relevant statistics or evidence concerning the selected KPIs (e.g. conference attendance, material), and timing for D&C activities until M48.

Dissemination and Communication Plan – Version 1.0

Core document outlining the impact-enabling approach, activities, channels (website and social media profiles), tools and timing at the basis of the project’s D&C outreaching strategy, the internal communication processes, templates, responsibilities, KPIs and operative plan for action.

Project website

Project public website for dissemination of project news events and results and connection with project and partners social medias

Data Management Plan (2)

Data Management Plan – Version 1.0

Description of the data management operative approach defined by the partners for project implementation, including references standards to be applied, type of data/research outputs, findability approach, accessibility, interoperability and reusability objectives and approach, curation/storage/preservation approach (M6 release).

Data Management Plan – Version 2.0

Report of data management related activities, results and risk management performed until M24. In addition, this document will contain the mapping of the DB generated by the project, the assessment of the state of implementation of the DMP approach and actions to be implemented until M48.

Demonstrators, pilots, prototypes (1)

Piloting planning and monitoring approach

Document gathering all the requirements needed to guarantee appropriate implementation of the use cases and management of users’ data. This document will include the definition of the legally and ethically compliant engagement process for each end-user, modules/ formats/ checklist/ guidelines for the correct deployment and usage of the developed modules in the platform experimentation. In addition, specific objectives and KPIs for each testing iteration will be defined, as well as the specific timing, minimal configurations (e.g., technical configuration, number, approach for feedback collections, etc.) and tools (e.g. questionnaires, comparison approaches) for KPIs measurement.

Documents, reports (4)

Procedure and solutions for feedbacks implementation – Version 1.0

Study on possible options/tools (technical and non-technical) for active involvement of end-users in the reporting of issues related with the platform, and definition of a project specific solution for this purpose.

State of the art report (Data model auditing)

Analysis of the state of the art for data model auditing, including complete literature review, scouting of relevant R&I projects’ results as well as market solution, and selection of most relevant contributions to be used for AISym4Med (e.g. model/algorithm and how to be applied within the platform)

Legal and ethical requirements report and updates

Analysis of legal references relevant for the platform, ethical assessment and list of mandatory requirements and corresponding strategies to implement them (including in annex any template/format needed, e.g. informed consent model) and matchmaking of these requirements with the corresponding SW components

Procedure and solutions for feedbacks implementation – Final version

Description of the final selected solution for a feedback system that will allow end-users to report impacts, incidents, and malfunctioning experienced with the platform. This will include the means to automatically redirect these inputs into the ethical governance of the platform, becoming a key aspect for incident handling.

Publicaciones

Conference proceedings (7)

synple: A Platform for Privacy Preserving Synthetic Patient Data Generation

Autores: Silveira, I., Silva. L., Veladas. F, Braga, R. & Gamboa, H.
Publicado en: 2024, ISBN 978-3-031-63851-0
Editor: Cham: Springer Nature Switzerland
DOI: 10.1007/978-3-031-63851-0_9

Systematic analysis of the impact of label noise correction on ML Fairness

Autores: Silva, I. Oliveira e; Soares, C.; Sousa, I.; Ghani, R.
Publicado en: 2023, ISBN 978-981-99-8391-9
Editor: Springer, Singapore
DOI: 10.48550/arxiv.2306.15994

Adapting Stable Diffusion Models for Domain-Specific Medical Imaging: A Case Study in Synthetic Retinal Fundus Image Generation

Autores: Façoco, Ivo; Mesquita, Gonçalo; Lúcio, Francisca; Rosado, Luís
Publicado en: Proceedings of the ECML-PKDD workshop SynDAiTE: Synthetic Data for AI Trustworthiness and Evolution, 2025
Editor: Springer
DOI: 10.5281/ZENODO.17207258

GASTeN: Generative Adversarial Stress Test Networks

Autores: Cunha, L., Soares, C., Restivo, A., Teixeira, L.F.
Publicado en: 2023, ISBN 978-3-031-30047-9
Editor: Cham: Springer Nature Switzerland
DOI: 10.1007/978-3-031-30047-9_8

Kernel Corrector LSTM

Autores: Tuna, R., Baghoussi, Y., Soares, C., Mendes-Moreira, J.
Publicado en: 2024
Editor: Cham: Springer Nature Switzerland.
DOI: 10.1007/978-3-031-58553-1_1

Designing for Qualitative Evaluation of Synthetic Medical Data

Autores: Isabella Barbosa Silva; Elsa Oliveira; Ricardo Melo; Luís Rosado; César Gálvez-Barrón; Irene Bernadet Heijink; Sem Hoogteijling; Iñigo Gabilondo
Publicado en: CHI EA '25: Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems, 2025
Editor: ACM
DOI: 10.1145/3706599.3720274

Towards High-Fidelity ECG Generation: Evaluation via Quality Metrics and Human Feedback

Autores: Russo, Maria; Rebelo, Joana; Bento, Nuno; Gamboa, Hugo
Publicado en: Proceedings of the 18th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2025), ISSN 2184-4305
Editor: SCITEPRESS – Science and Technology Publications
DOI: 10.5220/0013400500003911

Peer reviewed articles (1)

Benchmarking deep neural representations for synthetic data evaluation

Autores: Nuno, Bento; Joana, Rebelo; Marilia, Barandas
Publicado en: Intelligent Systems with Applications, 2025, ISSN 2667-3053
Editor: ScienceDirect
DOI: 10.1016/J.ISWA.2025.200580