Li Pengpeng, Chen Changshui, Deng Xiaoyuan, Mao Hua, Jin Shaoqin
South China Normal University, Photon and Nano Research Centre for Biosciences, MOE Key Laboratory of Laser Life Science, Guangzhou, Guangdong 510631, China.
Southern Medical University, Zhujiang Hospital, Department of Gastroenterology, Guangzhou, Guangdong 510282, China.
J Biomed Opt. 2015 Mar;20(3):037004. doi: 10.1117/1.JBO.20.3.037004.
We have recently applied the technique of drop coating deposition Raman (DCDR) spectroscopy for colorectal cancer (CRC) detection using blood plasma. The aim of this study was to develop a more convenient and stable method based on blood plasma for noninvasive CRC detection. Significant differences are observed in DCDR spectra between healthy (n = 105) and cancer (n = 75) plasma from 15 CRC patients and 21 volunteers, particularly in the spectra that are related to proteins, nucleic acids, and β-carotene. The multivariate analysis principal components analysis and the linear discriminate analysis, together with leave-one-out, cross validation were used on DCDR spectra and yielded a sensitivity of 100% (75/75) and specificity of 98.1% (103/105) for detection of CRC. This study demonstrates that DCDR spectroscopy of blood plasma associated with multivariate statistical algorithms has the potential for the noninvasive detection of CRC.
我们最近应用滴涂沉积拉曼（DCDR）光谱技术，通过血浆检测结直肠癌（CRC）。本研究的目的是开发一种基于血浆的更便捷、稳定的非侵入性CRC检测方法。在15名CRC患者和21名志愿者的健康（n = 105）和癌症（n = 75）血浆的DCDR光谱中观察到显著差异，特别是在与蛋白质、核酸和β-胡萝卜素相关的光谱中。对DCDR光谱进行多变量分析主成分分析和线性判别分析，并采用留一法交叉验证，检测CRC的灵敏度为100%（75/75），特异性为98.1%（103/105）。本研究表明，与多变量统计算法相关的血浆DCDR光谱具有非侵入性检测CRC的潜力。
