论文部分内容阅读
人体肿瘤的中红外光谱法早期检测是生物医学领域的前沿课题之一。研究了中红外FTIR光谱技术与化学计量方法 (模式识别技术 )相结合对人体恶性肿瘤进行检测的方法 ,并讨论了如何通过采用光谱预处理方法以及波长范围的选择来提高模式识别技术 (SIMCA法 )分类判别效果。研究了 6 3个胃癌病人的癌变组织和癌旁正常组织光谱 ,选择有代表性的 4 0个样品作为训练集 ,剩余的 2 3个样品作为预测集 ,该方法的准确率为 91%。这一研究结果表明傅里叶变换红外光谱法与化学计量学方法相结合可以对良、恶性组织进行鉴别诊断 ,有希望发展成为一种肿瘤快速诊断的新方法。
The early detection of human tumors by mid-infrared spectroscopy is one of the frontier topics in the field of biomedicine. The method to detect human malignant tumor by combining the mid-infrared FTIR spectroscopy with the stoichiometric method (pattern recognition) was studied and how to improve the pattern recognition technology by using the method of spectral pretreatment and the selection of the wavelength range (SIMCA method Classification discrimination effect. The spectrum of cancerous tissues and adjacent normal tissues of 63 gastric cancer patients was studied. 40 representative samples were selected as training set and the remaining 23 samples were used as prediction sets. The accuracy of this method was 91%. The results of this study show that the combination of Fourier transform infrared spectroscopy and chemometric methods can differentiate benign and malignant tissues and hopefully develop into a new method for rapid tumor diagnosis.