Pu Yang, Sordillo Laura A, Yang Yuanlong, Alfano R R
Opt Lett. 2014 Dec 15;39(24):6787-90. doi: 10.1364/OL.39.006787.
The native fluorescence (NFL) spectra of human cancerous and normal breast tissues were excited by a selected wavelength of 300 nm to investigate the efficacy of two key fluorophores: tryptophan and reduced nicotinamide adenine dinucleotide (NADH), as cancer biomarkers. The basis spectra of these key fluorophores' subspaces spanned by the corresponding emission spectra are obtained by the Gram-Schmidt method. A support vector machine (SVM) classifier is trained in the subspace to evaluate the sensitivity, specificity, and accuracy. This research demonstrates that the NFL spectroscopy measurements are effective to detect changes of fluorophores compositions in tissues due to the development of cancer.
通过选择300nm的波长激发人癌性和正常乳腺组织的固有荧光（NFL）光谱，以研究两种关键荧光团——色氨酸和还原型烟酰胺腺嘌呤二核苷酸（NADH）作为癌症生物标志物的功效。通过Gram-Schmidt方法获得由相应发射光谱所跨越的这些关键荧光团子空间的基础光谱。在该子空间中训练支持向量机（SVM）分类器，以评估其敏感性、特异性和准确性。本研究表明，NFL光谱测量对于检测由于癌症发展导致的组织中荧光团组成变化是有效的。
