Insight into the potential of algorithms using AI technology as in vitro diagnostics utilizing microbial extracellular vesicles.

Abstract

Recently, the microbiome has been gaining significant attention in the healthcare sector as a next-generation factor. However, there remains a substantial gap in our understanding of the fundamental mechanisms of microbes, particularly regarding the effector microbial products exchanged between the microbiota and the host. Consequently, research on microbial extracellular vesicles (MEVs) has increased. MEVs, which are nano-sized, can circulate throughout the body and penetrate the bloodstream, carrying diverse information. Consequently, they are increasingly being utilized in medical applications. Additionally, AI technologies are being utilized in medicine. The combination of MEVs and AI technology is being explored for the development of algorithm-based in vitro diagnostics (IVD). Therefore, this study aims to review the integration of MEVs and AI technology as diagnostic tools for personalized medicine. This paper reviewed the MEV-based algorithms developed by a variety of human samples and AI technology. Additionally, most of MEV-based diagnostic models showed higher clinical performance. Several important factors are crucial for accurate diagnosis. First, optimizing sample types according to specific diseases is essential. Second, AI technology with higher diagnostic power yields more accurate results. Finally, incorporating additional markers can enhance diagnostic power. However, applying this tool in situ faces several limitations, including method standardization, sample size, and analysis techniques. In the future, we anticipate that research on MEVs will advance our understanding of their role in disease and establish the foundation for precision medicine strategies.