Review on Smart Surface Plasmon Resonance: Taxonomy, Applications, and Trends

Surface Plasmon Resonance (SPR) has become an emerging powerful label-free sensing technology for the real-time detection in biomedical application, environmental monitoring and industrial application. In recent years, combining SPR with optical fiber platforms has gained more and more research interest because of its benefits such as the compact sensor design, remote sensing, high sensitivity, and in-situ measurement capability. In parallel, new possibilities of signal interpretation and sensing performance have been opened up by recent advances in artificial intelligence (AI), in particular machine learning (ML) and deep learning (DL). Despite the recently increasing number of studies focusing on fiber-integrated SPR sensors, as well as AI-assisted sensing approaches, there is still a lack of systematic and critically comparative analysis that can link these studies. This review is dedicated to this lack of information and aims to give a structured overview of the optical fiber-integrated SPR sensing technologies and the emerging integration of these technologies with intelligent data-driven methods. A taxonomy-based framework is proposed to classify the existing studies based on different methodological approaches, application domains, material platforms, system integration strategies and performance optimization methods. Representative sensor architectures are compared with one another for sensitivity, detection limit, robustness and structural complexity to find out the important design trends and performance trade-offs. In addition, the review discusses some of the major technical challenges in fiber-integrated SPR systems, such as noise sensitivity, complexity in signal interpretation, multiplexing limitations, and fabrication limitations. The role of ML and DL techniques in enhancing signal processing, reliability of sensing together with intelligent optimization of sensing platforms is also discussed. Overall, this review opens the possibility of the combination of complex plasmonic sensing architectures with artificial intelligence for the next-generation smart SPR sensing systems.