Sakudo Akikazu, Kato Yukiko Hakariya, Kuratsune Hirohiko, Ikuta Kazuyoshi
Department of Virology, Center for Infectious Disease Control, Research Institute for Microbial Diseases, Osaka University, Yamadaoka, Suita, Osaka 565-0871, Japan.
Clin Chim Acta. 2009 Oct;408(1-2):123-7. doi: 10.1016/j.cca.2009.08.005. Epub 2009 Aug 16.
After blood donation, in some individuals having polycythemia, dehydration causes anemia. Although the hematocrit (Ht) level is closely related to anemia, the current method of measuring Ht is performed after blood drawing. Furthermore, the monitoring of Ht levels contributes to a healthy life. Therefore, a non-invasive test for Ht is warranted for the safe donation of blood and good quality of life.
A non-invasive procedure for the prediction of hematocrit levels was developed on the basis of a chemometric analysis of visible and near-infrared (Vis-NIR) spectra of the thumbs using portable spectrophotometer. Transmittance spectra in the 600- to 1100-nm region from thumbs of Japanese volunteers were subjected to a partial least squares regression (PLSR) analysis and leave-out cross-validation to develop chemometric models for predicting Ht levels.
Ht levels of masked samples predicted by this model from Vis-NIR spectra provided a coefficient of determination in prediction of 0.6349 with a standard error of prediction of 3.704% and a detection limit in prediction of 17.14%, indicating that the model is applicable for normal and abnormal value in Ht level.
These results suggest portable Vis-NIR spectrophotometer to have potential for the non-invasive measurement of Ht levels with a combination of PLSR analysis.
献血后，在一些患有红细胞增多症的个体中，脱水会导致贫血。尽管血细胞比容（Ht）水平与贫血密切相关，但目前测量Ht的方法是在采血后进行的。此外，监测Ht水平有助于健康生活。因此，为了安全献血和提高生活质量，有必要进行Ht的非侵入性检测。
基于使用便携式分光光度计对拇指的可见和近红外（Vis-NIR）光谱进行化学计量分析，开发了一种预测血细胞比容水平的非侵入性程序。对日本志愿者拇指在600至1100纳米区域的透射光谱进行偏最小二乘回归（PLSR）分析和留一法交叉验证，以建立预测Ht水平的化学计量模型。
该模型根据Vis-NIR光谱预测的掩蔽样本的Ht水平，预测决定系数为0.6349，预测标准误差为3.704%，预测检测限为17.14%，表明该模型适用于Ht水平的正常和异常值。
这些结果表明，便携式Vis-NIR分光光度计结合PLSR分析具有非侵入性测量Ht水平的潜力。
