SANKT AUGUSTIN – How can we detect breast and lung cancer or neurodegenerative diseases at an early stage? How do we develop effective medications and predict who will benefit from them? Often, the answers lie in analyzing large amounts of patient data using artificial intelligence (AI). However, accessing such data can be protracted and expensive, complicated by strict data protection laws. Negotiating access agreements may take years, only to discover that the available data is unsuitable for the intended research.

As AI technologies like ChatGPT have shown how generative AI can create seemingly realistic text, images, and music, a key question arises: can similar techniques be employed to generate patient data? This is where the SYNTHIA project comes into play.

SYNTHIA is committed to responsibly generating high-quality synthetic data using customized generative AI processes. Such synthetic data can help overcome data protection obstacles for sharing real patient data while advancing AI methods for early diagnosis, such as for Alzheimer's disease. Additionally, synthetic patient data can even emulate control groups for clinical studies, accelerating the development and approval of new medications.

The SYNTHIA team is creating a modern IT platform to generate, validate, and apply synthetic patient data across various scenarios. To achieve this, the Fraunhofer Institute for Algorithms and Scientific Computing SCAI, the Institute for Translational Medicine and Pharmacology ITMP, and the Institute for Digital Medicine MEVIS collaborate closely with industry and academic partners. SYNTHIA adopts a comprehensive approach, addressing various data types while adhering to strict data protection requirements. The research incorporates innovative concepts such as digital twins, clinical trial simulations, and federated machine learning.

Prof. Dr. Holger Fröhlich, group leader AI & Data Science at Fraunhofer SCAI and a member of SYNTHIA’s management team, highlights the potential of generative AI methods for creating synthetic patient data for medical purposes: “We have been researching this area for years, and now we have the opportunity to further develop and apply our methods with an outstanding group of partners.” Fraunhofer SCAI also contributes expertise in applying generative AI to clinical trials for Alzheimer's disease. Fröhlich leads this particular initiative in collaboration with Gates Ventures, the private investment firm of Microsoft co-founder Bill Gates. Gates Ventures focuses on tackling global health challenges, including Alzheimer's.

Focusing on six diseases

SYNTHIA will also investigate how synthetic data can be validated and applied effectively. The team validates methods for different data types, including laboratory results, genomic data, and medical imaging. Their research focuses on six diseases: lung cancer, breast cancer, multiple myeloma, diffuse large B-cell lymphoma, Alzheimer's disease, and type 2 diabetes.

The SYNTHIA project brings together 32 partners from medicine, pharmaceutical research, medical technology, and academia. Among them are data scientists, clinical researchers, lawyers, and privacy experts. Together, they strive towards a common goal: accelerating medical innovation through the use of synthetic data.

The Innovative Health Initiative Joint Undertaking (IHI JU) funds SYNTHIA with € 24.1 million from September 2024 to August 2029 (grant agreement no. 101172872). The JU receives support from the European Union's Horizon Europe research and innovation program as well as private stakeholders, including COCIR, EFPIA, EuropaBio, MedTech Europe, Vaccines Europe, and DNV.

Find out more at https://www.ihi-synthia.eu/

Contact
Prof. Dr. Holger Fröhlich
Group leader AI & Data Science
Deputy Head of Business Area Bioinformatics
Fraunhofer Institute for Algorithms and Scientific Computing SCAI
Schloss Birlinghoven 1, 53757 Sankt Augustin
E-Mail: holger.froehlich@scai.fraunhofer.de
www.scai.fraunhofer.de/en/bio