Abstract The aim of the study was to evaluate the biochemical characteristics of normal CCD18-Co, CRL-1831 and cancer CaCo-2 human colon cell lines. Statistic assisted analysis of label-free Raman spectra demonstrates that normal and cancer human colon cells can be distinguished with high confidence based on vibrational properties. Moreover, vibrational spectra can be used to differentiate normal cells of epithelial CRL-1831 and fibroblast CCD18-Co origin with high sensitivity and specificity. Principal Component Analysis and Partial Least Squares Discriminant Analysis yielded effective and comparable cell classification based on vibrational features. We have proved that Raman spectroscopy is a powerful tool for cancer diagnostics, and could play an important clinical role for cell phenotyping.

中文翻译：

---

人类结肠细胞系的拉曼光谱和成像的统计辅助分析——结肠直肠癌的无标记光谱诊断

摘要 本研究的目的是评估正常 CCD18-Co、CRL-1831 和癌 CaCo-2 人结肠细胞系的生化特征。无标记拉曼光谱的统计辅助分析表明，可以基于振动特性以高可信度区分正常和癌症人类结肠细胞。此外，振动光谱可用于区分上皮 CRL-1831 和成纤维细胞 CCD18-Co 来源的正常细胞，具有高灵敏度和特异性。主成分分析和偏最小二乘判别分析产生了基于振动特征的有效且可比较的细胞分类。我们已经证明拉曼光谱是一种强大的癌症诊断工具，可以在细胞表型分析中发挥重要的临床作用。