Dong Ding | Search PubMed

PubMed2005

[Improving partial least square regression precision in NIR multi-component analysis using artificial neural network].

The present paper presents a new NIR multi-component analysis method with Artificial Neural Network(ANN) and Partial Least Square Regression(PLS). First, this method divides the concentration range of training samples into some sub-ranges, and respectively computes a PLS correlation model in each sub-range with the sub-range's training samples. Then, the authors classify prediction samples according to its concentration sub-range with ANN and judge which sub-range theprediction sample belongs to. Finally, the authors compute the concentration of prediction component with the PLS correlation model of the sub-range according to ANN. The experiment and the result of data processing show that this method improves the model's applicability, and evidently enhances prediction precision compared to traditional PLS.

Least-Squares Analysis↗

PubMed2005

[Achievement of the noninvasive measurement for human blood glucose with NIR diffusion reflectance spectrum method].

The noninvasive measurement of human blood glucose was achieved with NIR diffusion reflectance spectrum method. The thumb fingertip NIR diffusion reflectance spectra of six different age healthy volunteers were collected using Nexus-870 and its NIR fiber port smart accessory. The test was implemented with changing the blood glucose concentration for the limosis and satiation of every volunteer. The calibration model was set up using PLS method with the smoothing, baseline correction and first derivatives pretreatment spectrum in the 7500-8500 cm(-1) region for single volunteer, the same age combination and that of different age. When the spectrum was obtained, the actual blood glucose value of every spectrun sample was demarcated using ultraviolet spectrophotometer. The correlation between the calibration value and true value for single volunteer is better than that for the combination of volunteers, the correlative coefficients are all over 0.90471, RMSECs are all less than 0.171.

Adult↗