Privacy compliant health data as a service for AI development

Informazioni relative al progetto

PHASE IV AI

ID dell’accordo di sovvenzione: 101095384

DOI

10.3030/101095384

Data della firma CE 9 Ottobre 2023

Data di avvio 1 Ottobre 2023

Data di completamento 30 Settembre 2026

Finanziato da

• Health

Costo totale

€ 6 640 205,75

Contributo UE

€ 6 640 205,00

Coordinato da

TURUN YLIOPISTO
Finland

CORDIS fornisce collegamenti ai risultati finali pubblici e alle pubblicazioni dei progetti ORIZZONTE.

I link ai risultati e alle pubblicazioni dei progetti del 7° PQ, così come i link ad alcuni tipi di risultati specifici come dataset e software, sono recuperati dinamicamente da .OpenAIRE .

Risultati finali

Documents, reports (14)

PHASE IV AI Initial Architecture

A description of the initial technical architecture.

Study initiation package – UC1-UC3

A compiled version of the mandatory deliverable for the retrospective, registry-based studies in PHASE IV AI.

User stories, usage scenarios and use case validation v2

User perspectives on PHASE IV AI use cases.

Dissemination, Communication and Exploitation Activities v1

Report on the project's dissemination, communication and exploitation activities.

User stories, usage scenarios and use case validation v1

User perspectives on PHASE IV AI use cases.

Legal and ethical framework and requirements v1

An overview of legal and ethical requirements for the PHASE IV AI technologies.

Dissemination, Communication and Exploitation Plan

PHASE IV AI Dissemination, Communication and Exploitation Plan. Some sections of the exploitation plan could be considered sensitive (SEN).

PHASE IV AI Initial Technology Assessment and List

A document reflecting the developed PHASE IV AI technologies to the original requirements and specifications.

Security and Privacy measures

Description of the PHASE IV AI approaches to security anf privacy (technical measures).

Data Specifications

Data Specifications based on the selected use cases.

Guideline on EHDS Alignment v1

Guidelines for aligning the PHASE IV AI technologies with the EHDS developments.

Data Harmonization for DaaS, MaaS and Specifications plan

The deliverable will identify the heterogeneities of the data arising from the different medical imaging and labelling expert that exists in the data of the same data owner and the heterogeneities among different data owners arising from different EMRs, language and logging process etc. Following this, it will setup the roadmap for the documentation of both processes (OMOP CDM, semantic mismatches, metadata of demographic characteristics) indicating the right harmonisation tools and methodologies to prepare the data in a common and standardized data model within the health data hub.

Project Handbook, Quality Plan & Risk Management

The project handbook prepared by the coordinator will guide the project's implementation.

Technologies for de-identification and synthetic data generation v1

Algorithms for advanced anonymization and synthetic data generation. Lead beneficiaries UTU (de-identification) and EUT (synthetic data generation).

Other (5)

DaaS Toolbox v1

ML algorithms for Data as a Service concept.Type: R+OTHER

Secure Multiparty Computation v1

Improved methods for secure multi-party computation.Type: R+OTHER

Data Hub Design and Data Market v1

Description of the PHASE IV AI Data Hub and Data Market.Type: R+OTHER

MaaS Toolbox v1

ML algorithms for Model as a Service concept.Type: R + OTHER

Integrated Data Services v1

Integration of PHASE IV AI Data ServicesType: R+OTHER

Websites, patent fillings, videos etc. (1)

PHASE IV AI web site & social media

PHASE IV AI web site & social media presence

Pubblicazioni

Conference proceedings (3)

VisTAD: A Vision Transformer Pipeline for the Classification of Alzheimer’s Disease

Autori: Noushath Shaffi, Vimbi Viswan, Mufti Mahmud
Pubblicato in: 2024 International Joint Conference on Neural Networks (IJCNN), 2024
Editore: IEEE
DOI: 10.1109/IJCNN60899.2024.10650975

Understanding Feature Importance of Prediction Models Based on Lung Cancer Primary Care Data

Autori: Teena Rai, Yuan Shen, Jun He, Mufti Mahmud, David J Brown, Jaspreet Kaur, Emma O’Dowd, David R Baldwin, Richard Hubbard
Pubblicato in: 2024 International Joint Conference on Neural Networks (IJCNN), 2024
Editore: IEEE
DOI: 10.1109/IJCNN60899.2024.10650819

Characterization of Synthetic Lung Nodules in Conditional Latent Diffusion of Chest CT Scans

Autori: Roger Marí Molas, Paula Subías-Beltrán, Carla Pitarch Abaigar, Mar Galofré Cardo, Rafael Redondo Tejedor
Pubblicato in: Frontiers in Artificial Intelligence and Applications, Artificial Intelligence Research and Development, 2024
Editore: IOS Press
DOI: 10.3233/FAIA240408

Peer reviewed articles (6)

Does Differentially Private Synthetic Data Lead to Synthetic Discoveries?

Autori: Ileana Montoya Perez, Parisa Movahedi, Valtteri Nieminen, Antti Airola, Tapio Pahikkala
Pubblicato in: Methods of Information in Medicine, Numero 63, 2024, ISSN 0026-1270
Editore: Georg Thieme Verlag KG
DOI: 10.1055/a-2385-1355

Cognitive Computation

Autori: Md. Easin Arafat; Md. Wakil Ahmad; S. M. Shovan; Towhid Ul Haq; Nazrul Islam; Mufti Mahmud; M. Shamim Kaiser
Pubblicato in: Accurate Prediction of Lysine Methylation Sites Using Evolutionary and Structural-Based Information, 2024, ISSN 1866-9956
Editore: Springer Nature
DOI: 10.1007/S12559-024-10268-2

Ensemble of vision transformer architectures for efficient Alzheimer’s Disease classification

Autori: Noushath Shaffi; Vimbi Viswan; Mufti Mahmud
Pubblicato in: Brain Informatics, 2024, ISSN 2198-4018
Editore: Springer Open
DOI: 10.1186/S40708-024-00238-7

Towards practical federated learning and evaluation for medical prediction models

Autori: Andrei Kazlouski, Ileana Montoya Perez, Faiza Noor, Mikael Högerman, Otto Ettala, Tapio Pahikkala, Antti Airola
Pubblicato in: International Journal of Medical Informatics, Numero 204, 2025, ISSN 1386-5056
Editore: Elsevier BV
DOI: 10.1016/J.IJMEDINF.2025.106046

Interpreting artificial intelligence models: a systematic review on the application of LIME and SHAP in Alzheimer’s disease detection

Autori: Viswan Vimbi; Noushath Shaffi; Mufti Mahmud
Pubblicato in: Brain Informatics, 2024, ISSN 2198-4026
Editore: Springer Nature
DOI: 10.1186/s40708-024-00222-1

Benchmarking Evaluation Protocols for Classifiers Trained on Differentially Private Synthetic Data

Autori: Parisa Movahedi, Valtteri Nieminen, Ileana Montoya Perez, Hiba Daafane, Dishant Sukhwal, Tapio Pahikkala, Antti Airola
Pubblicato in: IEEE Access, Numero 12, 2024, ISSN 2169-3536
Editore: Institute of Electrical and Electronics Engineers (IEEE)
DOI: 10.1109/ACCESS.2024.3446913

Book chapters (2)

Medical AI in the EU: Regulatory Considerations and Future Outlook

Autori: Ranttila, Pertti; Sahebi, Golnaz; Kontio, Elina; Salmi, Jussi
Pubblicato in: AI - Ethical and Legal Challenges [Working Title], 2024, ISBN 978-0-85466-497-9
Editore: IntechOpen
DOI: 10.5772/INTECHOPEN.1007443

Medical AI in the EU: Regulatory Considerations and Future Outlook

Autori: Pertti Ranttila, Golnaz Sahebi, Elina Kontio, Jussi Salmi
Pubblicato in: AI - Ethical and Legal Challenges [Working Title], 2024
Editore: IntechOpen
DOI: 10.5772/intechopen.1007443

Other (1)

Response to Letter by Dehaene et al. on Synthetic Discovery is not only a Problem of Differentially Private Synthetic Data

Autori: Ileana Montoya Perez, Parisa Movahedi, Valtteri Nieminen, Antti Airola, Tapio Pahikkala
Pubblicato in: Methods of Information in Medicine, 2025
Editore: Georg Thieme Verlag
DOI: 10.1055/A-2540-8346