胃癌的自体荧光特征谱

利用激光诱导荧光法（laser-induced auto fluorescence,LIAF)选择多种波长测试和比较了胃癌和胃正常组织标本，得到区分癌变组织与正常组织的最佳激发波长；选取光谱特征差异最显著的波段425－475nm，采用形状识别技术分析31例胃癌及胃正常组织的自体荧光光谱曲线，经去噪声，曲线拟合，求区间最大曲率，最终以区间最大曲率的理想平均值作为区分阈值标准，获得高达93．5％的敏感性和100％的特异性，为临床实用奠定了基础。

Laser-Induced Autofluorescence Characteristic Spectrum of Cancerous

To determine bio tissue characters by its autofluorescence has developed very fast since the1980s and it has proved to be an efficient tumor diagnostic method. According to laser induced autofluorescence (LIAF) method,some different wavelengths are used to test gastric cancerous and normal samples, then the optimal excitation wavelength to differentiate the cancer from the normal tissue is obtained. Meanwhile, the bands at the wavelength range from 425 nm to 475 nm are chosen as the characteristic bands for cancerous tissues. Our approach based on curvefitting of LIAF spectra includes three parts: noise reduction,curvefitting,and computation of the maximum local curvature. At last, the maximum local curvatures of fitting curves are utilized as the criteria of differentiating gastric tumors from normal tissues.A sensitivity as high as 93.5 % and a specificity up to 100 % are obtained.