The University of Thessaly’s interest and central role in the SEARCH project stem from the significant potential that synthetic data holds for advancing healthcare. The core motivation is to explore how high-quality synthetic data can serve as a catalyst for developing the next generation of clinical decision support systems (CDSS). By overcoming the data accessibility and privacy challenges that often hinder medical research, synthetic data offers a robust pathway to train more sophisticated AI models, ultimately leading to more accurate, personalized tools for clinicians and improved patient outcomes.

UTH’s Role and Expertise

The University of Thessaly (UTH) serves as the scientific coordinator for the SEARCH project; an initiative focused on the application of synthetic data and artificial intelligence (AI) in healthcare.

The University of Thessaly, through its Department of Computer Science and Biomedical Informatics, provides significant expertise in data science, which is central to the project’s goals. The department’s background in artificial intelligence and machine learning informs its leadership of the project’s scientific objectives.

Key Contributions to the SEARCH Project

UTH’s primary responsibilities within the SEARCH project are concentrated in two main areas: synthetic data generation and the development of clinical decision support systems.

• Synthetic Data Generation:A major focus of UTH’s work is the creation of high-quality synthetic data. This involves developing models to produce realistic, non-identifiable datasets that replicate real-world patient information. UTH has leading roles in generating two complex data types:
  • Synthetic Medical Images: Creating artificial images for training AI models, particularly for applications in the project’s focus areas of cancer, cardiovascular and neurovascular diseases.
  • Synthetic Genomic Data: Generating synthetic genomic information to support research in personalized medicine; thus protecting the sensitive information contained in actual patient DNA.
• Explainable Clinical Decision Support Systems (CDSS): UTH is also involved in developing advanced CDSS with a strong focus on Explainable AI (XAI). A critical challenge with complex AI models is their “black box” nature, which can make it difficult for clinicians to trust their outputs. By building these systems on XAI principles, the goal is to ensure that the reasoning behind every AI-driven recommendation is transparent and understandable to healthcare professionals. These explainable systems will use the synthetic data generated by the project to provide clear, data-driven insights, assisting with diagnostic accuracy, personalized treatment plans, and optimized clinical workflows, thereby fostering greater trust and adoption in clinical practice.

As the scientific lead, the University of Thessaly is responsible for overseeing the technical execution of the SEARCH project. Its contributions are focused on providing the specialized knowledge required to generate useful synthetic health data and apply it to the development of practical clinical support tools.

Principal Investigator: Prof. Dimitris K. Iakovidis