FORECAST OF BLOOD FAT CONCENTRATION BASED ON SERUM UV ABSORPTION SPECTRA AND NEURAL NETWORK
The work was supported by the Innovation Foundation of Nanjing University of Aeronautics and Astronautics.
Blood plays an important role in the clinical diagnosis and treatment, the analysis of blood will be of very important practical significance. Study shows that the UV absorption spectrum is of complex shape, there is more absorption peak in 200 to 300nm, it shows that there is a complex absorption phenomenon in blood group macromolecules; there is a degree of displacement of absorption peak to different samples; there is no significant correlation between absorbance at some a wavelength and blood fat concent, but random. Based on the evident correlation between serum UV absorption spectrum and blood fat concentration in the wave band of 265 to 282nm, a neural network model was built to forecast the blood fat concentration, and a relatively good prediction was obtained. It provides a new spectral test method of blood fat concentration.
