Abstract
About this talk
The world is evolving at an unprecedented pace, with Artificial Intelligence (AI) emerging as a transformative technology. Its applications are becoming increasingly sophisticated, working alongside humans across diverse fields, including healthcare—from rapid disease prediction and diagnostic reporting to AI-assisted surgical procedures.Digital Health (DH) and Artificial Intelligence(AI) in healthcare are rapidly evolving but have often been addressed synonymously by healthcare authorities and practitioners. A clear distinction between these concepts is fundamental for developing robust frameworks and practical guidelines that ensure the safety, efficacy, and effectiveness of DH solutions and AI-enabled technologies.The categorization of DH into technologies (DHTs) and services (DHSs) enables regulatory, HTA, and reimbursement bodies to develop category-specific frameworks for evaluating these solutions. DH is also instrumental in generating real-world data, which is increasingly important for evidence-based decision-making. AI, as a potential tool for DH, can be embedded in technologies such as medical devices and applications to enhance functionality and performance. AI excels in processing numerical and perceptual data. For numerical data, machine-learning algorithms enable prediction, classification, and clustering, while perceptual data include images, videos, voice, and text. In recent years, generative AI, which produces new content using various learning approaches, has gained prominence in research and healthcare. A thorough understanding of DH and AI, together with accurate terminology, would facilitate the timely generation of regulatory- and HTA-grade evidence, supporting improved health outcomes and greater certainty in decision-making.The healthcare sector generates vast amounts of data, providing a foundation for AI applications across public health, preventive care, and clinical domains. However, AI deployment in healthcare requires robust legal and ethical frameworks. At national and European levels, efforts to regulate AI in research and healthcare delivery are intensifying, as demonstrated by the European Artificial Intelligence Act.A key challenge lies in aligning clinical needs, research opportunities, and regulatory requirements. Although technological advances enable the acquisition of large data volumes for impact research, obtaining consent for each specific data-processing purpose can create significant procedural challenges. This is particularly evident in prospective studies, where research data are collected during routine care, potentially slowing clinical visits and procedures. Retrospective studies face additional organizational and legal challenges, including difficulties in contacting patients who are unreachable or deceased. Furthermore, informed consent documents should follow principles of transparency and accessibility, enabling patients without technical expertise to understand research objectives. Who envisions a future where AI serves as a powerful force for innovation, equity, and ethical integrity in healthcare, with the goal of helping member states take AI to the people and enable enhanced, sustainable, and smarter healthcare. In conclusion, DH and AI solutions are rapidly emerging and have the potential to influence healthcare outcomes and efficiency. However, while developers may possess strong technical expertise, they may not always be familiar with established healthcare principles, regulations, and clinical practices. This gap can result in solutions whose impacts on patients and health systems have not been adequately evaluated.
Speaker
About the speaker
