Method and apparatus for probabilistically classifying tissue in vitro and in vivo using fluorescence spectroscopy

摘要: Fluorescence spectral data acquired from tissues in vivo or vitro is processed accordance with a multivariate statistical method to achieve the ability probabilistically classify tissue diagnostically useful manner, such as by histopathological classification. The apparatus includes controllable illumination device for emitting electromagnetic radiation selected cause produce fluorescence intensity spectrum. Also included are an optical system applying plurality of wavelengths sample, and spectrum detecting spectra emitted sample result device. also include processor, connected device, analyzing detected calculate probability that belongs particular processor analyzes using method. five primary steps involved (i) preprocessing each patient account inter-patient variation, (ii) partitioning preprocessed all patients into calibration prediction sets, (iii) dimension reduction set principal component analysis, (iv) selection most components two-sided unpaired student's t-test (v) development optimal classification scheme based on logistic discrimination scores inputs.