AISYM4MED – Synthetic and Scalable Data Platform for Medical Empowered AI



One of the hindrances of using Artificial Intelligence applied to the healthcare context is associated with the existing data that is often untrustworthy, disperse, with privacy issues, and risks perpetuating bias if generalized. Being this a setback well known to the research community, Fraunhofer Portugal AICOS (FhP-AICOS) took on the challenge of making quality data available for medical research through a platform that combines Machine Learning techniques and Synthetic Data Generation: the AISYM4MED project.

The AISYM4MED project will address data privacy and security issues by combining new anonymization techniques and privacy measures. This will allow the use of diversified datasets from several countries safeguarding the security of the data. Synthetic data generation will be put through rigorous controlling measures to guarantee representativeness, thus contributing to more robust digital solutions.

The AISYM4MED project counts on the expertise of 14 partners from across Europe and beyond to address the many doubts and questions that arise when combining healthcare with Artificial Intelligence (AI). To add to the consortium’s technical expertise in privacy, security, blockchain, ethical AI, data inspection, healthcare management systems, and Information Technology (IT) law, the project also includes partners from diversified sociodemographic backgrounds to enrich the datasets provided.

In addition to the coordination of the AISYM4MED project, FhP-AICOS will contribute with user-journey, auditing Models and Synthetic Data generation, two areas of expertise that FhP-AICOS’ researchers have worked on and will be further developed within this project.

David Belo, senior researcher at FhP-AICOS and project coordinator, hopes that in 4 or 5 years, the research and medical community will see this as a trustworthy platform. In the words of David, the platform “will support the development of digital solutions and services for medical applications by enabling innovators to evaluate and improve AI systems in a secure and trustable environment.”

The consortium includes Fraunhofer Center for Assistive Information and Communication (AICOS) as coordinator, the Imperial College of Science Technology and Medicine, INYC - Instrumentacion y Componentes SA, Consorci Sanitari de l'Alt Penedès i Garraf, TIGA - Bilgi Teknolojileri Anonim Sirketi, Zabala Brussels, Asociacion Instituto de Investigacion Sanitaria Biocruces Bizkaia, Servicio Vasco de Salud Osakidetza, Time.Lex, Universidade do Porto, Universidade Nova de Lisboa, Ibermatica, Saidot, Utrecht Hospital, and the University of Zurich.