A narrative review of AI in hospital-acquired infections

Abstract

Hospital-Acquired Infections (HAI) are a longstanding challenge in patient safety efforts and an important public health matter. The development of healthcare information technology has been improving the surveillance of hospital-acquired infections, making artificial intelligence technologies a promising approach to mitigate the burden of HAIs. This thesis constitutes a thorough narrative review of the existing literature on the use of Artificial Intelligence (AI) to combat hospital-acquired infections, as well as qualitative insights from professionals in the area. It aims to critically evaluate the challenges and goals of resorting to artificial intelligence for the early detection of HAIs, allowing the improvement of patient safety and healthcare outcomes. Findings from this review reveal significant results in HAI detection accuracy when AI is used. Demonstrating that, when compared to conventional methods, AI-based models are capable of greatly increasing the accuracy of hospital-acquired infection detection, which can result in earlier and more accurate identification of HAIs. The discussion reflects how AI can be a powerful support tool rather than a replacement for clinical judgment. It also addresses the importance of tailoring AI applications to specific hospital contexts to maximize their impact. The conclusion reinforces the promising potential of AI in reducing the burden of HAIs, while acknowledging that its success largely depends on overcoming technical, organizational and ethical challenges.

Date of Award	23 Oct 2025
Original language	English
Awarding Institution	Universidade Católica Portuguesa
Supervisor	Henrique Martins (Supervisor)