Cancer Research: FBK is studying AI that analyzes blood for early diagnosis and personalized treatment
treatment The results of 3DSecret were presented in Trento: the European project is developing an artificial intelligence model integrated with biosensors and genetics. For FBK, the project resulted in a publication in PLOS and recognition from the European Union.
Trento, September 28, 2026 – Understanding how cancer cells travel through the bloodstream and developing smart tools to detect them before they form metastases. This is the goal of 3DSecret, a European research project that held its final meeting in Trento, bringing to a close a collaborative effort invo ving international centers of excellence.
Within the project, FBK coordinated the artificial intelligence research through the Data Science for Health Unit, led by Giuseppe Jurman at the Center for Digital Health and Wellbeing. Researchers at the Center developed a generative artificial intelligence system capable of integrating highly diverse biomedical data, including microscopic images, genetic information, metabolic parameters, and medical records. The approach, validated on more than 10,000 samples across 20 types of cancer, makes it possible to reconstruct a coherent picture of the patient even when data is incomplete and identify which additional tests would be most useful.
“No single modality can describe a patient,” says Nicolò Lazzaro, a researcher at FBK’s Data Science for Health Unit. “An X-ray, a blood test, the tumor’s DNA and gene activity, as well as the patient’s treatment history, describe different aspects of the same system. Our goal is to bring these signals together into a unified model.”
The scientific results and system validation were published in the journal PLOS Computational Biology.
The technology has also received significant recognition from the European Commission’s Innovation Radar.
In 2025, the project was selected for the “PoC by FBK: Test Your Idea!” initiative, created to support the transfer of scientific results into practical applications. With support from PoC funding, the Foundation launched the UK Biobank project “Systemic GenAI Models from Peripheral Samples,” extending the analysis to a cohort of 500,000 participants to test the possibility of detecting diseases at an early stage through simple blood tests.
3DSecret is a frontier research project funded by Horizon Europe through the EIC Pathfinder Open program, which supports high-risk scientific ideas with the potential to generate radically new technologies.
The Data Science for Health unit within FBK’s Center for Digital Health and Wellbeing develops data science and artificial intelligence methods for analyzing biomedical and clinical data, with the goal of contributing to the research and development of precision medicine solutions. Its activities include the study of clinical data, bioimaging, and omics data through multimodal approaches, with applications in diagnosis, prognosis, and prediction. Particular attention is given to developing robust, interpretable models that can be validated in clinical settings, in collaboration with healthcare facilities and research partners.
