Blood profile testing is a valuable tool for monitoring the metabolism, health, and nutritional status of lactating dairy animals, including buffaloes. However, conducting extensive analyses on a large scale is not feasible due to high costs, labor, the need for invasive procedures, and consequent animals' stress. The utilization of the Fourier-transform mid-infrared (FT-MIR) spectroscopy to predict hematic traits from milk spectral information may represent an effective opportunity for the dairy sector, given that buffaloes are under official testing in various countries. This study aims to test the ability of milk FT-MIR spectra to predict the most relevant buffaloes' hematic components. Blood and milk samples were collected once in 9 Italian farms from 308 buffaloes (98 primiparous and 210 pluriparous) at different stages of lactation. The hematic concentrations of urea, creatinine, glucose, total bilirubin, cholesterol, triglycerides, total protein, albumin, globulin, alanine and aspartate aminotransferases, alkaline phosphatase, lactate dehydrogenase, gamma-glutamyl transferase, and creatine kinase were measured using reference analysis to evaluate the predictive ability of FT-MIR. Spectral data (n = 308) were divided into a calibration set (70%) and a validation set (30%). In external validation, R-2 of 0.72, 0.60, 0.56, and 0.69 were achieved for urea, triglycerides, creatinine, and total bilirubin concentrations, respectively, whereas the R-2 of other hematic traits was <0.50. The spectral regions important for the prediction of triglycerides, creatinine, and total bilirubin fell in those associated with milk protein structures (1,570-1,550 cm(-1)). Although our models were not accurate enough for precise determination of the blood parameters concentration, the predictions of some traits such as urea can be effectively used for herd-level screening (e.g., for nutritional status evaluation). Our results, although preliminary, provide the basis for future large-scale investigations on buffalo health status and metabolism.
Feasibility of milk mid-infrared spectroscopy for large-scale screening of hematic traits in dairy buffaloes
De Marchi M.
2025
Abstract
Blood profile testing is a valuable tool for monitoring the metabolism, health, and nutritional status of lactating dairy animals, including buffaloes. However, conducting extensive analyses on a large scale is not feasible due to high costs, labor, the need for invasive procedures, and consequent animals' stress. The utilization of the Fourier-transform mid-infrared (FT-MIR) spectroscopy to predict hematic traits from milk spectral information may represent an effective opportunity for the dairy sector, given that buffaloes are under official testing in various countries. This study aims to test the ability of milk FT-MIR spectra to predict the most relevant buffaloes' hematic components. Blood and milk samples were collected once in 9 Italian farms from 308 buffaloes (98 primiparous and 210 pluriparous) at different stages of lactation. The hematic concentrations of urea, creatinine, glucose, total bilirubin, cholesterol, triglycerides, total protein, albumin, globulin, alanine and aspartate aminotransferases, alkaline phosphatase, lactate dehydrogenase, gamma-glutamyl transferase, and creatine kinase were measured using reference analysis to evaluate the predictive ability of FT-MIR. Spectral data (n = 308) were divided into a calibration set (70%) and a validation set (30%). In external validation, R-2 of 0.72, 0.60, 0.56, and 0.69 were achieved for urea, triglycerides, creatinine, and total bilirubin concentrations, respectively, whereas the R-2 of other hematic traits was <0.50. The spectral regions important for the prediction of triglycerides, creatinine, and total bilirubin fell in those associated with milk protein structures (1,570-1,550 cm(-1)). Although our models were not accurate enough for precise determination of the blood parameters concentration, the predictions of some traits such as urea can be effectively used for herd-level screening (e.g., for nutritional status evaluation). Our results, although preliminary, provide the basis for future large-scale investigations on buffalo health status and metabolism.
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
