基于FTIR-SVM的正常甲状腺及甲状腺癌组织的分类研究  

Classification of normal thyroid tissues and thyroid carcinoma based on FTIR analysis with SVM

作　　者：成则丰[1] 程路遥[2] 金文英[3] 程存归[1] 

机构地区：[1]浙江师范大学浙江省固体表面反应化学重点实验室,浙江金华321004 [2]长春理工大学计算机科学技术学院,吉林长春130022 [3]义乌工商职业技术学院计算机工程系,浙江义乌322000

出　　处：《哈尔滨工程大学学报》2006年第B07期366-369,共4页Journal of Harbin Engineering University

摘　　要：支持向量机（SVM）是根据统计理论提出的一种新的学习算法．为了进行临床中经常出现的正常甲状腺组织与甲状腺癌组织分类，文章以82对正常甲状腺组织与甲状腺癌组织为实验材料，通过FTIR—SVM建立了正常甲状腺组织与甲状腺癌组织识别的模型．试验结果显示，对学习训练集中的70个样品模型识别率为100％，对94个预测样品的识别准确率为98．9％．研究结果表明，FTIR—SVM可以用于正常甲状腺组织与甲状腺癌组织的区别．The support vector machine （SVM） is a new learning technique based on the statistical learning theory. In order to classify normal thyroid tissues and thyroid carcinoma in clinics, 82 pairs of normal thyroid tissues and thyroid carcinoma samples are used as experimental materials in this paper. The classification models are established with Fourier transform infrared spectra （FTIR）-SVM training method for the intention of identifying normal thyroid tissues and thyroid carcinoma samples. The seventy samples in training set are identified by the classifying models with accurate rate of 100%, while ninety-four estimate samples has an accurate rate of 98.9%. The research result shows the feasibility of establishing the models with FTIR-SVM method to identify normal thyroid tissues and thyroid carcinoma.

关 键 词：傅里叶变换红外光谱法 支持向量机 正常甲状腺组织 甲状腺癌组织 

分 类 号：O675.3[理学—化学] TP181[自动化与计算机技术—控制理论与控制工程]