Lü Guodong, Zheng Xiangxiang, Lü Xiaoyi, Chen Peng, Wu Guohua, Wen Hao
State Key Laboratory of Pathogenesis, Prevention, and Treatment of Central Asian High Incidence Diseases, Clinical Medical Research Institute, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, China.
School of Electronic Engineering, Beijing University of Posts and Telecommunications, Beijing, 100876, China.
Photodiagnosis Photodyn Ther. 2021 Mar;33:102164. doi: 10.1016/j.pdpdt.2020.102164. Epub 2020 Dec 26.
In this paper, we investigated the feasibility of using serum Raman spectroscopy and multivariate analysis method to discriminate echinococcosis and liver cirrhosis from healthy volunteers. Raman spectra of serum samples from echinococcosis, liver cirrhosis, and healthy volunteers were recorded under 532 nm excitation. The normalized mean Raman spectra revealed specific biomolecular differences associated with the disease, mainly manifested as the contents of β carotene in the serum of patients with echinococcosis and liver cirrhosis were lower than those of healthy people. Furthermore, principal components analysis (PCA), combined with linear discriminant analysis (LDA), was adopted to distinguish patients with echinococcosis, liver cirrhosis, and healthy volunteers. The overall diagnostic accuracy based on the PCA-LDA algorithm was 87.7 %. The diagnostic sensitivities to healthy volunteers, patients with echinococcosis, and liver cirrhosis were 92.5 %, 81.5 %, and 89.1 %, and the specificities were 93.2 %, 96.1 %, and 92.4 %, respectively. This exploratory work demonstrated that serum Raman spectroscopy technology combined with PCA-LDA diagnostic algorithm has great potential for the non-invasive identification of echinococcosis and liver cirrhosis.
在本文中，我们研究了使用血清拉曼光谱和多变量分析方法从健康志愿者中鉴别包虫病和肝硬化的可行性。在532 nm激发下记录了包虫病、肝硬化患者及健康志愿者血清样本的拉曼光谱。归一化平均拉曼光谱揭示了与疾病相关的特定生物分子差异，主要表现为包虫病和肝硬化患者血清中β-胡萝卜素的含量低于健康人。此外，采用主成分分析（PCA）结合线性判别分析（LDA）来区分包虫病患者、肝硬化患者和健康志愿者。基于PCA-LDA算法的总体诊断准确率为87.7%。对健康志愿者、包虫病患者和肝硬化患者的诊断敏感性分别为92.5%、81.5%和89.1%，特异性分别为93.2%、96.1%和92.4%。这项探索性工作表明，血清拉曼光谱技术结合PCA-LDA诊断算法在包虫病和肝硬化的无创识别方面具有巨大潜力。
