The evolution of analytical techniques has opened the possibilities of accurate analyte detection through a straightforward method and short acquisition time, leading towards their applicability to identify medical conditions. Surface-enhanced Raman spectroscopy (SERS) has long been proven effective for rapid detection and relies on SERS spectra that are unique to each specific analyte. However, the complexity of viruses poses challenges to SERS and hinders further progress in its practical applications. The principle of SERS revolves around the interaction among substrate, analyte, and Raman laser, but most studies only emphasize the substrate, especially label-free methods, and the synergy among these factors is often ignored. Therefore, issues related to reproducibility and consistency of results, which are crucial for medical diagnosis and are the main highlights of this review, can be understood and largely addressed when considering these interactions. Viruses are composed of multiple surface components and can be detected by label-free SERS, but the presence of non-target molecules in clinical samples interferes with the detection process. Appropriate spectral data processing workflow also plays an important role in the interpretation of results. Furthermore, integrating machine learning into data processing can account for changes brought about by the presence of non-target molecules when analyzing spectral features to accurately group the data, for example, whether the sample corresponds to a positive or negative patient, and whether a virus variant or multiple viruses are present in the sample. Subsequently, advances in interdisciplinary fields can bring SERS closer to practical applications.

中文翻译：

---

创新且多功能的表面增强拉曼光谱启发的病毒检测方法走向临床应用：综述


分析技术的发展开辟了通过简单的方法和短的采集时间准确检测分析物的可能性，从而使其适用于识别医疗状况。表面增强拉曼光谱 (SERS) 早已被证明可有效进行快速检测，并且依赖于每种特定分析物所独有的 SERS 光谱。然而，病毒的复杂性给SERS带来了挑战，阻碍了其实际应用的进一步进展。 SERS的原理围绕底物、分析物和拉曼激光之间的相互作用，但大多数研究只强调底物，特别是无标记方法，而这些因素之间的协同作用往往被忽视。因此，与结果的可重复性和一致性相关的问题对于医学诊断至关重要，也是本次综述的主要亮点，在考虑这些相互作用时可以理解并在很大程度上得到解决。病毒由多种表面成分组成，可以通过无标记的SERS进行检测，但临床样本中非目标分子的存在会干扰检测过程。适当的光谱数据处理工作流程在结果解释中也发挥着重要作用。此外，将机器学习集成到数据处理中可以在分析光谱特征时考虑非目标分子的存在所带来的变化，以准确地对数据进行分组，例如，样本是否对应于阳性或阴性患者，以及是否是病毒样本中存在变种或多种病毒。随后，跨学科领域的进步可以使SERS更接近实际应用。