Artificial intelligence for literature reviews: opportunities and challenges

Artificial intelligence (AI) methods have the potential to support evidence synthesis.  In particular, the widespread availability of large language models (LLMs) has accelerated the development and adoption of AI-based tools for literature reviews. In this seminar, we will examine how these tools can support evidence synthesis while also introducing new methodological challenges. We will (i) explore some of the existing tools, and highlight their strenghts and limitations and (ii) discuss best practices for integrating AI into evidence synthesis workflows without compromising transparency, reproducibility, or methodological rigour.

Ricardo Correia is an Associate Professor at the Faculty of Medicine of the University of Porto, in the field of Health Informatics. With a background in Computer Science and a PhD focused on the integration of hospital information systems, his work lies at the intersection of interoperability, data quality, and artificial intelligence in healthcare. His recent research and innovation activities focus on the practical use of AI and large language models in clinical information systems, education, pharmacovigilance, and health data reuse, with a strong emphasis on real-world implementation, governance, and impact. He is also actively involved in academic entrepreneurship, contributing to the creation and scaling of digital health solutions through multiple university spin-offs and international research projects.

Manuel Cruz is an Assistant Professor at the Faculty of Medicine of University of Porto. He is a Public Health MD. He has completed is PhD in Health Data Science. He has been working on AI for Evidence Synthesis, namely on the use of LLM to support multiple stages of the systematic review process and of quality assessment of published and unpublished systematic reviews.