Abstract
Invasive fungal infections (IFI) primarily occur in immunocompromised patients, particularly those in intensive care units (ICU). Due to the use of immunosuppressive agents, invasive therapeutic procedures, and advancements in diagnostic technologies, the detection rate of IFI has shown a significant upward trend. This review aims to explore epidemiological changes in the field of IFI, early detection techniques, and the application of artificial intelligence (AI) technologies. We conducted a literature review using PubMed data up to April 2025, focusing on studies related to IFI. Specifically, we focus on three aspects of IFI research: first, the epidemiology of IFI is undergoing significant changes, with Candida auris rapidly spreading across more than 40 countries worldwide, and rare fungal infections such as Mucor spp. and Fusarium spp. becoming increasingly prevalent; simultaneously, resistance to antifungal drugs among various pathogens continues to rise. Second, breakthroughs have been achieved in early detection technologies, including molecular detection techniques, biomarker testing, imaging technologies, and other emerging diagnostic methods, significantly enhancing the sensitivity and specificity of diagnosis. Thirdly, with the widespread application of AI technology, the development of clinical predictive models, the establishment of scoring rules, and the formulation of AI-based treatment decision-making tools are advancing the exploration of early diagnosis for IFI. In summary, as early diagnostic technologies for IFI continue to advance and AI algorithms are integrated into clinical practice, there is potential to improve the early diagnosis and treatment outcomes for critically ill patients with IFI.