The PHAROS – Adaptation, Fairness, Explainability in AI Medical Imaging Workshop, organised on October 20th, 2025, by PHAROS AI Factory, a project coordinated by GRNET, in the context of the International Conference on Computer Vision (ICCV 2025), brought together leading experts and researchers to explore innovative AI approaches for predictive modeling on large-scale medical image datasets. The workshop emphasized deep learning, Generative AI (GenAI), and Large Language Models (LLMs, VLMs) in healthcare, highlighting transparency, fairness, and human-centered integration in digital health services.
Addressing major challenges at the intersection of computer vision and healthcare AI—including multi-disease diagnosis, model explainability, domain adaptation, fairness, and continual learning—the event provided a vibrant forum for discussion on trustworthy and interpretable AI and its responsible deployment in sensitive medical applications.
Organized by Prof. Stefanos Kollias (GRNET Chairman, Prof. Emeritus at National Technical University of Athens and PHAROS Coordinator), Dr. Dimitrios Kollias (Queen Mary University of London), Prof. Xujiong Ye (University of Exeter, UK), and Dr. Francesco Rundo (University of Catania, Italy), the workshop took place at Ballroom A, on Monday, 20 October, from 4 p.m. to 9 p.m. PDT, as part of the International Conference on Computer Vision (ICCV 2025).
As part of the PHAROS AI Factory, the workshop ensured that its topics were grounded in real-world applications and cutting-edge research. This connection strengthened collaboration between academia, healthcare institutions, and industry, fostering innovation in trustworthy AI for medical imaging.
The event featured two major AI competitions designed to benchmark state-of-the-art solutions:
- Multi-Source Covid-19 Detection Challenge
- Fair Disease Diagnosis Challenge
Participants were given access to large, curated medical datasets, enabling global research teams to develop, test, and compare their AI models in realistic medical scenarios.
About ICCV
The International Conference on Computer Vision (ICCV) is the premier biennial global conference in the field of computer vision, organized by the Computer Vision Foundation (CVF). ICCV gathers top researchers, academics, and industry leaders to present the latest advances in image understanding, machine learning, and AI applications. Through its main conference, workshops, tutorials, and demos, ICCV defines the state of the art in vision-based AI research and promotes the translation of innovation into real-world impact.
