近赤外分光法を用いた乳牛における発情及び妊娠の診断方法の開発

寻找更合适、更可靠的发情诊断方法，对于奶牛养殖业来说，达到更好的繁殖是至关重要的。日本的这项联合研究的目的是开发一种通过近红外光谱分析和水分离技术检测发情的奶牛体液(尿液、血液和乳汁)和血管的方法。该方法计划适应农场条件，便携式设备计划通过人工操作进行测试，并作为挤奶过程中的在线设备使用。采集到的奶牛血管光谱没有提供实际适用的结果。在实验奶牛场，尿液采集被证明是不利的，因此，该样本类型的采集被取消。从被调查的奶牛身上取血，并对血清进行分析。结果与牛奶相似。由于基于血液数据的模型精度低于基于牛奶数据的模型，因此研究重点放在牛奶数据上，牛奶的近红外光谱提供了奶牛发情状态的信息，所应用的模型能够高精度地识别处于发情状态的奶牛。所取得的结果反映了该技术作为一种在线牛奶扫描系统应用于实践的可能性。一份已经发表，另一份正在同行评议的科学期刊上发表。此外，研究结果还在国际会议和三次日本会议上发表。

Finding more appropriate and reliable methods for diagnosis of estrus is crucial in dairy farming to reach better reproduction. The purpose of the joint research in Japan was to develop a method to test body fluids (urine, blood, and milk) and blood vessels of dairy cows by means of NIR spectroscopy and aquaphotomics for estrus detection. The method was planned to be adapted to farm conditions and portable devices were planned to be tested by manual operation and also as on-line equipment used in the process of milking.Spectra collected on blood vessels of cows did not provide practically applicable results. Collection of urine was proven to be unfavorable in the experimental dairy farm, thus, the collection of this sample type has been cancelled. Blood was taken from the investigated cows and blood serum was analyzed. Results were similar to those of milk. Since accuracy of models performed on blood data was lower than the models on milk data, the research was focused on milk data.It has been shown that NIR spectra of milk provide information on the estrus conditions of cows and the applied model was able to recognize the cows in heat with high accuracy. The achieved results reflect the possibility of the application of the technique in practice as an on-line milk scanning system.Two manuscripts related to the research topic have been written. One has been published and the other is in the publication process at peer-reviewed scientific journal. Furthermore, the results of the research were presented at international conferences and three Japanese conferences.