Robustness of near-infrared calibration models for the prediction of milk constituents during the milking process

Combining genetic markers, on-farm information and infrared data for the in-line prediction of blood biomarkers of metabolic disorders in Holstein cattle

BACKGROUND

Various blood metabolites are known to be useful indicators of health status in dairy cattle, but their routine assessment is time-consuming, expensive, and stressful for the cows at the herd level. Thus, we evaluated the effectiveness of combining in-line near infrared (NIR) milk spectra with on-farm (days in milk [DIM] and parity) and genetic markers for predicting blood metabolites in Holstein cattle. Data were obtained from 388 Holstein cows from a farm with an AfiLab system. NIR spectra, on-farm information, and single nucleotide polymorphisms (SNP) markers were blended to develop calibration equations for blood metabolites using the elastic net (ENet) approach, considering 3 models: (1) Model 1 (M1) including only NIR information, (2) Model 2 (M2) with both NIR and on-farm information, and (3) Model 3 (M3) combining NIR, on-farm and genomic information. Dimension reduction was considered for M3 by preselecting SNP markers from genome-wide association study (GWAS) results.

RESULTS

Results indicate that M2 improved the predictive ability by an average of 19% for energy-related metabolites (glucose, cholesterol, NEFA, BHB, urea, and creatinine), 20% for liver function/hepatic damage, 7% for inflammation/innate immunity, 24% for oxidative stress metabolites, and 23% for minerals compared to M1. Meanwhile, M3 further enhanced the predictive ability by 34% for energy-related metabolites, 32% for liver function/hepatic damage, 22% for inflammation/innate immunity, 42.1% for oxidative stress metabolites, and 41% for minerals, compared to M1. We found improved predictive ability of M3 using selected SNP markers from GWAS results using a threshold of > 2.0 by 5% for energy-related metabolites, 9% for liver function/hepatic damage, 8% for inflammation/innate immunity, 22% for oxidative stress metabolites, and 9% for minerals. Slight reductions were observed for phosphorus (2%), ferric-reducing antioxidant power (1%), and glucose (3%). Furthermore, it was found that prediction accuracies are influenced by using more restrictive thresholds (-log10(P-value) > 2.5 and 3.0), with a lower increase in the predictive ability.

CONCLUSION

Our results highlighted the potential of combining several sources of information, such as genetic markers, on-farm information, and in-line NIR infrared data improves the predictive ability of blood metabolites in dairy cattle, representing an effective strategy for large-scale in-line health monitoring in commercial herds.

LED-Based Desktop Analyzer for Fat Content Determination in Milk

In dairy, there is a growing request for laboratory analysis of the main nutrients in milk. High throughput of analysis, low cost, and portability are becoming critical factors to provide the necessary level of control in milk collection, processing, and sale. A portable desktop analyzer, including three light-emitting diodes (LEDs) in the visible light region, has been constructed and tested for the determination of fat content in homogenized and raw cow's milk. The method is based on the concentration dependencies of light scattering by milk fat globules at three different wavelengths. Univariate and multivariate models were built and compared. The red channel has shown the best performance in prediction. However, the joint use of all three LED signals led to an improvement in the calibration model. The obtained preliminary results have shown that the developed LED-based technique can be sufficiently accurate for the analysis of milk fat content. The ways of its further development and improvement have been discussed.

Invited review: integration of technologies and systems for precision animal agriculture-a case study on precision dairy farming

Precision livestock farming (PLF) offers a strategic solution to enhance the management capacity of large animal groups, while simultaneously improving profitability, efficiency, and minimizing environmental impacts associated with livestock production systems. Additionally, PLF contributes to optimizing the ability to manage and monitor animal welfare while providing solutions to global grand challenges posed by the growing demand for animal products and ensuring global food security. By enabling a return to the "per animal" approach by harnessing technological advancements, PLF enables cost-effective, individualized care for animals through enhanced monitoring and control capabilities within complex farming systems. Meeting the nutritional requirements of a global population exponentially approaching ten billion people will likely require the density of animal proteins for decades to come. The development and application of digital technologies are critical to facilitate the responsible and sustainable intensification of livestock production over the next several decades to maximize the potential benefits of PLF. Real-time continuous monitoring of each animal is expected to enable more precise and accurate tracking and management of health and well-being. Importantly, the digitalization of agriculture is expected to provide collateral benefits of ensuring auditability in value chains while assuaging concerns associated with labor shortages. Despite notable advances in PLF technology adoption, a number of critical concerns currently limit the viability of these state-of-the-art technologies. The potential benefits of PLF for livestock management systems which are enabled by autonomous continuous monitoring and environmental control can be rapidly enhanced through an Internet of Things approach to monitoring and (where appropriate) closed-loop management. In this paper, we analyze the multilayered network of sensors, actuators, communication, networking, and analytics currently used in PLF, focusing on dairy farming as an illustrative example. We explore the current state-of-the-art, identify key shortcomings, and propose potential solutions to bridge the gap between technology and animal agriculture. Additionally, we examine the potential implications of advancements in communication, robotics, and artificial intelligence on the health, security, and welfare of animals.

Milk Components by In-Line Fiber Optic Probe-Based FT-NIR: Commercial Scale Evaluation of a Potential Alternative Measurement Approach for Milk Payment

BACKGROUND

Mid-infrared (MIR) spectroscopy has traditionally been used to determine the macronutrients in bovine milk, as the basis of milk payment. Recent studies have demonstrated that NIR/FT-NIR spectroscopic systems can not only achieve MIR measurement performance, but are also generally simpler, more robust, and thus much more amenable to actual industrial process applications.

OBJECTIVE

The goal of this unique study was to investigate the feasibility of in-line FT-NIR spectroscopy for milk fat, protein, and total solids (TS) determination in a large industrial dairy processing facility, as an alternative basis for milk payment.

METHOD

Multivariant chemometric models using partial least squares (PLS) regression were built to predict the milk components. Over 1000 composite FT-NIR results gathered from the milk unloading process were compared directly to independent third-party FT-IR results.

RESULTS

Accuracy, precision, and linearity of the method were shown by Standard Error of Prediction (SEP) and Range/SEP of individual components. The SEP for fat, protein, and TS models were 0.09, 0.11, and 0.52, respectively. Range/SEP were 25.10, 12.60, and 6.40 for fat, protein, and TS, respectively. Accuracy and precision for the three components were further evaluated by the mean differences (0.01, 0.05, and 0.51) from dairy FT-IR results and the standard deviations of the mean difference (0.09, 0.09, and 0.13). Robustness was demonstrated by evaluating milk with natural variation over 6 months and using multiple instrumentation setups. The repeatability was also evaluated.

CONCLUSIONS

Overall, the in-line FT-NIR technology was found to have accurate, reliable, consistent performance similar to dairy FT-IR technology.

Appropriate Data Quality Checks Improve the Reliability of Values Predicted from Milk Mid-Infrared Spectra

Simple Summary

There is a growing interest in using milk mid-infrared (MIR) spectrometry to obtain new phenotypes to assist in the complex management of dairy farms. These predictive values can be erroneous for many reasons, even if the prediction equations used are accurate. Unfortunately, there is no quality protocol routinely implemented to detect those abnormal predictive values in the database recorded by dairy herd improvement (DHI) organizations, except for fat and protein contents. However, for financial and practical reasons, it is unfeasible to adapt the quality protocol commonly used in milk laboratories to improve the accuracy of those traits. So, this study proposes three different statistical methods that would be easy to implement by DHI organizations to detect abnormal values and limit the spectral extrapolation in order to improve the accuracy of MIR-based predictive values.

Abstract

The use of abnormal milk mid-infrared (MIR) spectrum strongly affects prediction quality, even if the prediction equations used are accurate. So, this record must be detected after or before the prediction process to avoid erroneous spectral extrapolation or the use of poor-quality spectral data by dairy herd improvement (DHI) organizations. For financial or practical reasons, adapting the quality protocol currently used to improve the accuracy of fat and protein contents is unfeasible. This study proposed three different statistical methods that would be easy to implement by DHI organizations to solve this issue: the deletion of 1% of the extreme high and low predictive values (M1), the deletion of records based on the Global-H (GH) distance (M2), and the deletion of records based on the absolute fat residual value (M3). Additionally, the combinations of these three methods were investigated. A total of 346,818 milk samples were analyzed by MIR spectrometry to predict the contents of fat, protein, and fatty acids. Then, the same traits were also predicted externally using their corresponded standardized MIR spectra. The interest in cleaning procedures was assessed by estimating the root mean square differences (RMSDs) between those internal and external predicted phenotypes. All methods allowed for a decrease in the RMSD, with a gain ranging from 0.32% to 41.39%. Based on the obtained results, the “M1 and M2” combination should be preferred to be more parsimonious in the data loss, as it had the higher ratio of RMSD gain to data loss. This method deleted the records based on the 2% extreme predictions and a GH threshold set at 5. However, to ensure the lowest RMSD, the “M2 or M3” combination, considering a GH threshold of 5 and an absolute fat residual difference set at 0.30 g/dL of milk, was the most relevant. Both combinations involved M2 confirming the high interest of calculating the GH distance for all samples to predict. However, if it is impossible to estimate the GH distance due to a lack of relevant information to compute this statistical parameter, the obtained results recommended the use of M1 combined with M3. The limitation used in M3 must be adapted by the DHI, as this will depend on the spectral data and the equation used. The methodology proposed in this study can be generalized for other MIR-based phenotypes.

Accuracy and application of milk fatty acid estimation with diffuse reflectance near-infrared spectroscopy

Near infrared spectroscopy (NIRS) has the potential to estimate contents of fatty acids (FA) in milk frequently at-farm or during daily milking routine. In this study, a total of 738 raw milk spectra collected from 33 Holstein cows over a period of 30 weeks were recorded. Reference data on FA composition in milk and in milk fat were analysed in laboratory. Calibration models were calculated for single FA and groups of FA in milk and in milk fat. Validation resulted in sufficient Ratio of Prediction to Deviation (RPD) values for some single FA and in higher RPD values for groups of FA when concentrations of FA in milk were predicted. Since the concentrations of most FA in milk are highly correlated with milk fat content, the prediction of FA contents in milk fat is more meaningful when independent predictions are intended. The accuracy of predicting single FA concentrations in milk fat is rather poor for most FA but still comparable to alternative analysing methods such as MIR analysis. The estimation of different groups of FA in milk fat resulted in an improved accuracy based on higher RPD values, which was sufficient to mirror the development in the different lactation phases. The course of cow individual long chain fatty acid (LCFA) concentration in the early lactation stage can be an indicator for body fat mobilisation. The accurate estimation of the extent and duration of body fat mobilisation in cow individuals was rather difficult with NIR predicted LCFA concentrations and would require a higher measuring frequency than applied in this study.

Computational optimization of the configuration of a spatially resolved spectroscopy sensor for milk analysis

A global optimizer has been developed, capable of computing the optimal configuration in a probe for spatially resolved reflectance spectroscopy (SRS). The main objective is to minimize the number of detection fibers, while maintaining an accurate estimation of both absorption and scattering profiles. Multiple fibers are necessary to robustify the estimation of optical properties against noise, which is typically present in the measured signals and influences the accuracy of the inverse estimation. The optimizer is based on a robust metamodel-based inverse estimation of the absorption coefficient and a reduced scattering coefficient from the acquired SRS signals. A genetic algorithm is used to evaluate the effect of the fiber placement on the performance of the inverse estimator to find the bulk optical properties of raw milk. The algorithm to find the optimal fiber placement was repeatedly executed for cases with a different number of detection fibers, ranging from 3 to 30. Afterwards, the optimal designs for each considered number of fibers were compared based on their performance in separating the absorption and scattering properties, and the significance of the differences was tested. A sensor configuration with 13 detection fibers was found to be the combination with the lowest number of fibers which provided an estimation performance which was not significantly worse than the one obtained with the best design (30 detection fibers). This design resulted in the root mean squared error of prediction (RMSEP) of 1.411 cm(-1) (R(2) = 0.965) for the estimation of the bulk absorption coefficient values, and 0.382 cm(-1) (R(2) = 0.996) for the reduced scattering coefficient values.