胃粘膜组织共焦显微拉曼光谱分类研究

Classification of normal and malignant human gastric mucosa tissue with confocal Raman microspectroscopy

采用基于小波变换的光谱去噪和背景扣除预处理技术,对32例(其中:13例正常,19例癌变)胃粘膜组织拉曼光谱进行分析,克服了手动背景扣除的一些缺点,并观察到一些新的光谱特征。基于此,给出能对所有正常组织和癌变组织进行有效分类的特征量和判据。通过研究这些特征量与胃粘膜组织癌变发展阶段之间的定量关系,可望实现胃癌的早期诊断。

32 samples from the human gastric mucosa tissue, including 13 normal and 19 malignant tissue samples were measured by confocal Raman microspectroscopy. The low signal - to -background ratio spectra from human gastric mucosa tissues were obtained by this technique without any sample preparation. A preprocessed algorithm based on wavelet analysis was used to reduce noise and eliminate background/baseline of Raman spectra. Comparing preprocessed spectra of malignant gastric mucosa tissues with those of counterpart normal ones, there were obvious spectral changes, including intensity increase at -1156 cm^-1 and intensity decrease at 1587 cm^-1 The quantitative criterion based upon the intensity ratio of the -1156 and 1587 cm^-1 was extracted for classification of the normal and malignant gastric mucosa tissue samples, This could result in a new diagnostic method, which would assist the early diagnosis of gastric cancer.