DigiPatICS consolidates the use of AI to improve cancer diagnosis and share results
Pathologist viewing histopathological samples using DigiPatICS tools. Credit: Vall d'Hebron
The eight hospitals of the Catalan Institute of Health have optimised breast cancer diagnosis thanks to artificial intelligence algorithms developed by the UPC as part of the DigiPatICS project. These algorithms automate the analysis of tissue samples, enabling earlier and more accurate detection of the disease. Results and diagnoses are shared in real time across the eight centres, creating Europe’s largest digital pathology network. The project received the 15th UPC Research Valorisation Award on 19 November.
Jan 07, 2026
The Image Processing Group (GPI) of the Universitat Politècnica de Catalunya - BarcelonaTech (UPC) is participating in DigiPatICS, a project coordinated by the Catalan Institute of Health (ICS) to improve cancer diagnosis in Catalonia’s public hospitals. It involves digitising histopathological samples (microscopic samples of biological tissue) and incorporating AI tools that automate analysis to make diagnoses faster and more reliable. This system allows a greater number of samples to be analysed more quickly, so patients receive earlier and more accurate diagnoses.
Previously, tissue samples were examined under microscopes, requiring pathologists to manually count a small number of cells reactive to immunohistochemical stains (techniques used to detect cancer). Since 2022, when the first phase of the project began, focusing on digitisation, ICS hospitals have had 24 high-precision scanners connected to 13 stations to digitise samples. This has significantly increased the number of cells analysed per sample and enabled images and diagnoses to be shared securely and in real time across the network, creating a common database of more than four million digitised images.
Ferran Marqués, GPI researcher, a professor at the UPC’s Barcelona School of Telecommunications Engineering (ETSETB) and project lead, explains: “From the digitised samples, we have developed deep learning-based AI algorithms that automatically detect, quantify and classify cells reactive to immunohistochemical stains.” Marqués emphasises that “the process is not only faster but also statistically more accurate and ensures diagnoses are reproducible, which is one of DigiPatICS’s goals.” Just one year after implementing the algorithms, in 2024 13,885 images had been analysed, supporting treatment for more than 4,400 patients.
The analysis results generated by the algorithms are accessible to over 170 pathologists at the seven ICS hospitals offering this service: Hospital Universitari Arnau de Vilanova (Lleida), Hospital Universitari de Bellvitge (L’Hospitalet de Llobregat), Hospital Universitari Germans Trias i Pujol (Badalona), Hospital Universitari de Girona Doctor Josep Trueta, Hospital Universitari Joan XXIII (Tarragona), Hospital Universitari Vall d’Hebron (Barcelona) and Hospital de Tortosa Verge de la Cinta. Diagnoses and images are also shared with Hospital de Viladecans, completing the ICS network. This connectivity enables joint supervision of results and remote diagnosis, creating one of the largest digital pathology networks in the world. At the same time, these tools have advanced ICS’s digital transformation in medical services.
Initially, the project focused on breast cancer due to its high prevalence—one in eight women is estimated to have experienced or will experience this type of cancer. It is now expanding to other oncological pathologies such as lung, colon and endometrial cancer.
The research team also includes Josep Ramon Casas, Montse Pardàs, Philippe Salembier and Verónica Vilaplana, all members of the GPI, a research group affiliated with the Intelligent Data Science and Artificial Intelligence (IDEAI-UPC) research centre.
Winner of the 15th UPC Research Valorisation Award
On 19 November, DigiPatICS received the 15th UPC Research Valorisation Award in the category of best technology transfer project, for the success of its implementation, the social impact of the transfer and its contribution to the UN’s Sustainable Development Goals.
Organised by the UPC’s Board of Trustees with the support of Fractus and the Fractus-UPC Deep Tech Hub, the award also recognised the project Permanently Polarized Hydroxyapatite, a ceramic and biocompatible catalyst developed by the UPC’s Innovation in Materials and Molecular Engineering - Biomaterials for Regenerative Therapies (IMEM-BRT) research group, as the best market-ready invention. This catalyst captures greenhouse gases and converts them into useful chemicals for industry, such as methanol and ammonia.
Virmedex Virtual Experiences was also recognised as the best technology-based company or spin-off. The company trains healthcare professionals with virtual simulators based on video games, virtual reality and AI.
