AISym4MED, the new platform that will expect to improve the healthcare data system in Europe

Recently, the kick-off meeting of the AISym4MED project took place at the UPTEC University in Oporto, Portugal. The event, hosted by Fraunhofer Portugal AICOS, saw as protagonists all the consortium in the planification of AISym4MED, aiming to make quality data available for medical research and help create more responsible AI through a platform that combines Machine Learning techniques and Synthetic Data Generation. Zabala Innovation leads the dissemination work package with stakeholders of the entire health value chain and participate in co-creation processes.

During the meeting, the consortium made up of partners from eight different countries across Europe, under the aegis of the project manager Fraunhofer Portugal AICOS and under the coordination of HaDEA- European Health and Digital Executive Agency, presented the next operational phases in the implementation process and the agenda of the next steps.

The partners are: Fraunhofer Portugal Research, Imperial College of Science technology and Medicine, Inycom Innovation Technologies, Consorci Sanitari de l’Alt Penedès i Garraf, TIGA Bilgi Teknolojileri Anonim Sirketi, Zabala Innovation, Asociacion Instituto de Investigacion Sanitaria Biocruces Bizkaia, Servicio Vasco de Salud Osakidetza, Time.Lex, Universidade do Porto, Nova ID FCT – Associacao Para a Inovacao e Desenvolvimento da Fct, Ibermatica, Saidot, Utrecht Hospital, and University of Zurich.

More about AISym4MED

“The AISym4Med project aims to advance the research and development of healthcare, which is a demanding, high-risk, and essential area”, says David Belo, AISym4MED project coordinator. “Our goal is to remove barriers that prevent the deployment of high-quality solutions. To achieve this, we will develop data generation, model auditing, and visualization tools that respect the core rights and diversity of humanity, giving us a new perspective and vision for how we can help people achieve a better state of well-being. We envision that in the future, we will transcend the way we interact with medical data – enabling us to make more informed and responsible decisions that will have a significant impact on the lives of individuals regardless of their origin.”, adds Belo.

The idea is trying to solve some challenges that are found especially when it comes to data processing. Especially in the past few years, Europe’s healthcare systems must deal with an aging population, an increase in the number of people who need support for chronic illnesses such as type II diabetes, cardiovascular disorders, and hypertension, as well as relevant pandemic risks. In this sense, Technology is becoming more and more important; since the COVID-19 pandemic, the Healthcare industry is integrating more and more technologies in their market so much to predict revenue of 39€ million by 2026 especially when it comes to Artificial Intelligence.

Recent medical and IT/sensor technology developments are supporting a broader understanding of P4 medicine, also known as predictive, preventative, personalized, and participatory health. In this context, it is becoming increasingly important to profile the risk of diseases, simulate the transmission of diseases, and conduct controlled experiments using data that can be easily, reliably, affordably, and repeatedly collected.

Not to mention the challenge faced when dealing with multiple data sources and the hardships of developing and perfecting consumer-ready Machine Learning solutions. The data science community’s traditional emphasis on accessibility and reuse, however, needs to be tempered with the need to meet ethical and legal constraints, particularly when working with personal health data (PHD).

Another important issue is the processing, exchange, and management of data, especially sensitive ones. In this sense, the project is aligning with the European Commission (EC) which is actively supporting the creation of a European Health Data Space (EHDS) to promote health data exchange and support research on new preventive measures, treatments, medications, and medical devices while ensuring that individuals have control over their own personal data.

Another challenge faced, is that the healthcare data often suffer from incompleteness and lack of quality and does not always follow a standard format/representation. Furthermore, the fact it is dispersed across many hospitals, clinics, and governmental databases inhibits its full potential for research purposes in the healthcare industry.

In this context, AISym4Med aims at developing a platform that will provide healthcare data engineers, practitioners, and researchers access to a trustworthy dataset system augmented with controlled data synthesis for experimentation and modeling purposes. This platform will address data privacy and security by combining new anonymization techniques, attribute-based privacy measures, and trustworthy tracking systems. Furthermore, this platform will exploit federated technologies for reproducing unidentifiable data from closed borders, promoting the indirect assessment of a broader number of databases, while respecting privacy, security, and GDPR-compliant guidelines.

AISym4Med will help in the creation of more robust machine learning (ML) algorithms for real-world readiness while considering the most effective computation configuration. Furthermore, a machine-learning meta-engine will provide information on the quality of the generalized model by analyzing its limits and breaking points, contributing to the creation of a more robust system by supplying on-demand real and/or synthetic data. This platform will be validated against local, national, and cross-border use cases for data engineers, ML developers, and aid for clinicians’ operations.