Aims: This special session aims to explore the paradigm of Augmented Intelligence in healthcare, focusing on AI systems designed to enhance, rather than replace, clinicians’ decision-making processes. The session will highlight advances in multimodal deep learning for diagnostic and prognostic modelling, and the emerging role of generative AI as a cognitive support tool for physicians. Emphasis will be placed on clinically grounded, interpretable, and trustworthy AI solutions.

Scope: The session will focus on computational methodologies that integrate heterogeneous biomedical data, such as medical imaging, omics profiles, electronic health records, and wearable sensor data, into unified multimodal deep learning frameworks for diagnosis and prognosis. Particular attention will be devoted to models that support clinical reasoning, including foundation and generative AI systems capable of synthesising complex information, assisting in report generation, and enabling scenario-based decision support. Contributions addressing model interpretability, robustness, uncertainty quantification, and human-AI interaction will be especially encouraged, as these elements are critical to ensuring safe and effective deployment in clinical environments. The session aims to bridge methodological innovation with real-world validation, promoting translational approaches aligned with regulatory and ethical requirements.

Impact: By fostering interdisciplinary dialogue between computational scientists and clinicians, this session aims to advance the development of reliable AI systems that meaningfully extend physicians’ analytical capabilities. The expected outcome is a clearer roadmap for safe, effective, and human-centred augmented intelligence solutions that improve diagnostic accuracy, prognostic stratification, and personalised treatment planning in real-world healthcare settings.