Luo Ying, Xia Xuliang, Yu Wenrou, Chen Zhaoxian, Xiong Hongyuan, Li Shunbo, Xu Yi, Wang Li, Huang Yingzhou
Key Laboratory of Optoelectronic Technology & Systems (Ministry of Education), Chongqing University, Chongqing 400044, China; Chongqing Key Laboratory of Interface Physics in Energy Conversion, College of Physics, Chongqing University, Chongqing 401331, China.
The Second Affiliated Hospital of Chengdu Medical College, Nuclear Industry 416 Hospital, Chengdu, Sichuan 610051, China.
Spectrochim Acta A Mol Biomol Spectrosc. 2026 Jan 5;344(Pt 1):126648. doi: 10.1016/j.saa.2025.126648. Epub 2025 Jul 14.
Over recent years, surface-enhanced Raman spectroscopy (SERS) has shown its unparalleled sensitivity and molecular specificity in biomedical applications. However, noninvasive and sensitive detection of biomarkers with conventional SERS for oral cancer diagnosis remains challenges. For early diagnosis of oral cancer, there are two difficulties in spectral recognition: one involving low-concentration and dynamic biomarkers capturing, and the other is the complex interference in breath and saliva samples. To address these challenges, we developed a plasmonic covalent organic framework (COF) synthetic SERS biosensor to enhance the SERS signal and improve the biomarker adsorption efficiency. This sensing platform integrates the silver nanowire network and the COF-TpPa film with micropores. Both the liquid and gaseous molecules are demonstrated to have excellent SERS performance based on this plasmonic COF substrate without any pretreatment. Taking advantage of the deep learning analysis with the Light Gradient Boosting (LGB) algorithm based on the spliced SERS spectra of gaseous and liquid analytes, the oral cancer identification accuracy reaches to 98 % in the experiment using artificial exhaled methyl mercaptan and salivary uric acid. This work explores SERS application in detecting gas-liquid fluid analytes and noninvasive clinical cancer diagnosis, and may have an important future role in food safety and environmental monitoring.
近年来，表面增强拉曼光谱（SERS）在生物医学应用中展现出了无与伦比的灵敏度和分子特异性。然而，利用传统SERS对生物标志物进行无创且灵敏的检测以用于口腔癌诊断仍然存在挑战。对于口腔癌的早期诊断，在光谱识别方面存在两个难题：一是涉及低浓度和动态生物标志物的捕获，另一个是呼吸和唾液样本中的复杂干扰。为应对这些挑战，我们开发了一种等离子体共价有机框架（COF）合成SERS生物传感器，以增强SERS信号并提高生物标志物的吸附效率。该传感平台将银纳米线网络与具有微孔的COF-TpPa薄膜集成在一起。基于这种等离子体COF基底，液体和气态分子在无需任何预处理的情况下均表现出优异的SERS性能。利用基于气态和液态分析物拼接SERS光谱的轻梯度提升（LGB）算法进行深度学习分析，在使用人工呼出的甲硫醇和唾液尿酸的实验中，口腔癌识别准确率达到了98%。这项工作探索了SERS在检测气液流体分析物和无创临床癌症诊断中的应用，并且可能在食品安全和环境监测方面具有重要的未来作用。
