在进出口检测检疫部门, 血液制品的检验与分类是件重要且复杂的事情。 对于全血样品, 开放式的采集可能带来污染, 且血样中的致病因子可能会对检测人员造成危害。 因此急需非接触式的全血分类鉴别方法。 常用流式细胞术中的光谱方法由于需要对血细胞进行采样, 所以无法在非接触全血分类鉴定中采用。 红外吸收光谱学是一种可用来分析样品分子结构和化学键的技术, 可以在不直接接触样品的情况下对样品进行探测。 为寻找一种可实现非接触式血液样品种属差异性状探测的可行光谱方法, 采用近红外谱段(4 497.669～7 506.4 cm-1)对犬猫鸡三类常见动物全血样品进行了透射光谱测量。 结果发现所测样品均在5 184～5 215 cm-1之间有个明显的吸收峰, 在7 000 cm-1附近有个较平缓的吸收峰, 且同种动物个体之间的透射光谱分布相似, 只在整体透射率上有些差别。 采用相关系数比较三类动物全血样品近红外透射光谱的区别, 计算得出同种动物不同个体光谱曲线的相关系数均大于0.99, 而不同种动物光谱曲线的相关系数在0.509 48～0.916 13之间。 其中鸡与猫光谱曲线的相关系数在0.857 23～0.912 44之间; 鸡与犬光谱曲线的相关系数在0.509 48～0.664 82之间; 猫与犬光谱曲线的相关系数在0.872 75～0.916 13之间。 犬猫同属哺乳纲, 两者全血的近红外透射光谱相关系数比犬鸡或猫鸡非同纲动物的大, 即光谱曲线的相似度更高。 研究结果表明近红外透射光谱是一种非接触式动物全血鉴别的可行方法。

The inspection and classification for blood products are important but complicated in import-export ports or inspection and quarantine departments. For the inspection of whole blood products, open sampling can cause pollution and virulence factors in bloods samples may even endanger inspectors. Thus non-contact classification and identification methods for whole bloods of animals are needed. Spectroscopic techniques adopted in the flowcytometry need sampling blood cells during the detection; therefore they can not meet the demand of non-contact identification and classification for whole bloods of animals. Infrared absorption spectroscopy is a technique that can be used to analyze the molecular structure and chemical bonds of detected samples under the condition of non-contact. To find a feasible spectroscopic approach of non-contact detection for the species variation in whole blood samples, a near infrared transmitted spectra (NITS, 4 497.669～7 506.4 cm-1) experiment of whole blood samples of three common animals including chickens, dogs and cats has been conducted. During the experiment, the spectroscopic resolution is 5 cm-1, and each spectrogram is an average of 5 measured spectral data. Experimental results show that all samples have a sharp absorption peak between 5 184 and 5 215 cm-1, and a gentle absorption peak near 7 000 cm-1. Besides, the NITS curves of different samples of same animals are similar, and only have slight differences in the whole transmittance. A correlation coefficient (CC) is induced to distinguish the differences of the three animals’ whole bloods in NITS curves, and the computed CCs between NITS curves of different samples of the same animals, are greater than 0.99, whereas CCs between NITS curves of the whole bloods of different animals are from 0.509 48 to 0.916 13. Among which CCs between NITS curves of the whole bloods of chickens and cats are from 0.857 23 to 0.912 44, CCs between NITS curves of the whole bloods of chickens and dogs are from 0.509 48 to 0.664 82, and CCs between NITS curves of the whole bloods of cats and dogs are from 0.872 75 to 0.916 13. The cat and the dog belong to the class of mammal, and the CCs between their whole bloods NITS curves are greater than those between chickens and cats, or chickens and dogs, which are hetero-class animals. Namely, the whole bloods NITS curves of the cat and the dog have higher similarity. These results of NITS provide a feasible method of non-contact identification of animal whole bloods.