PLS-ANN判别分析自体荧光光谱识别胃癌  

PLS-ANN Discriminant Analysis on Autofluorescence Spectra to Identify Gastric Cancer

作　　者：马君[1] 史晓凤[1] 郑荣儿[1] 朱玉平[1] 李颖[1] 毛伟征[2] 孟继武[1] 

机构地区：[1]中国海洋大学物理系,中国山东青岛266071 [2]青岛大学医学院附属医院普外科,中国山东青岛266003

出　　处：《激光生物学报》2005年第6期432-436,共5页Acta Laser Biology Sinica

基　　金：国家自然科学基金项目(30070728);青岛市自然科学基金项目(03-2-jz-11)(KGCX2-405)

摘　　要：本文对58例胃癌病人离体标本的癌浆膜和正常浆膜进行以308nm为激发光的自体荧光光谱检测,采用多因素分析法进行光谱信息提取,以识别胃癌。研究表明偏最小二乘法结合神经网络法(简称PLS-ANN)进行判别分析,诊断胃癌的灵敏度为86%,特异度为100%,准确率为93%,有望成为手术中快速识别胃癌在胃壁的浸润范围的有效方法。Measurement of fluorescence spectra was performed at excitation wavelength of 308 nm and emission wavelength in the range of 328 nm -596 nm. The partial Least-squares and artificial neural network （PLS-ANN） method was used to analyze autofluorescenee spectra of gastric cancer. The 58 cancer samples and normal samples were taken from stomach serosa. The normalized and centerized spectra of two kinds of samples showed similar hut divergent patterns. PLS-ANN classification algorithm could differentiate cancer tissues from normal tissues with a sensitivity of 86%, a specificity of 100% and a total success rate of 93%. We concluded therefore that the PLS-ANN method was a fast, more effective choose for identification of gastric cancer.

关 键 词：胃肿瘤 自体荧光光谱 偏最小二乘法 人工神经网络 

分 类 号：R735.2[医药卫生—肿瘤] Q657.319[医药卫生—临床医学]