Exploring Dry-Film FTIR Spectroscopy to Characterize Milk Composition and Subclinical Ketosis throughout a Cow's Lactation

Spectral Profiling (Fourier Transform Infrared Spectroscopy) and Machine Learning for the Recognition of Milk from Different Bovine Breeds

The Podolica cattle breed is widespread in southern Italy, and its productivity is characterized by low yields and an extraordinary quality of milk and meats. Most of the milk produced is transformed into "Caciocavallo Podolico" cheese, which is made with 100% Podolica milk. Fourier Transform Infrared Spectroscopy (FTIR) is the technique that, in this research work, was applied together with machine learning to discriminate 100% Podolica milk from contamination of other Calabrian cattle breeds. The analysis on the test set produced a misclassification percentage of 6.7%. Among the 15 non-Podolica samples in the test set, 2 were misclassified and recognized as Podolica milk even though the milk was from other species. The correct classification rate improved to 100% when the same method was applied to the recognition of Podolica and Pezzata Rossa milk produced by the same farm. Furthermore, this technique was tested for the recognition of Podolica milk mixed with milk from other bovine species. The multivariate model and the respective confusion matrices obtained showed that all the 14 Podolica samples (test set) mixed with 40% non-Podolica milk were correctly classified. In addition, Pezzata Rossa milk produced by the same farm was detected as a contaminant in Podolica milk from the same farm down to concentrations as little as 5% with a 100% correct classification rate in the test set. The method described yielded higher accuracy values when applied to the discrimination of milks from different breeds belonging to the same farm. One of the reasons for this phenomenon could be linked to the elimination of the environmental variable. However, the results obtained in this work demonstrate the possibility of using FTIR to discriminate between milks from different breeds.

Two-Channel Detecting Sensor with Signal Cross-Correlation for FTIR Instruments

This paper's purpose was to demonstrate a performance of a novel approach in a low-noise optical sensor for an FTIR spectrometer.

METHODS

Compared to the standard FTIR detection setup, our sensor ensures a higher signal-to-noise ratio (SNR) and lower signal standard deviation by reducing the uncorrelated noise components (e.g., thermal and 1/f noises of the detection module). Its construction is based on two-channel detection modules and a processing unit with implemented cross-correlation signal analyses. Each module was built of LWIR HgCdTe photodiodes and low-noise transimpedance amplifiers.

RESULTS

the experiments demonstrated a decrease in a signal standard deviation of about 1.7 times with a 10 dB-improvement in the SNR.

CONCLUSION

this result indicates our sensor's main benefit, especially in registered "weak" and noisy interferograms.

The Use of Multilayer Perceptron Artificial Neural Networks to Detect Dairy Cows at Risk of Ketosis

Simple Summary

Ketosis is a serious metabolic disease in high-yield dairy cows, that affects productive herds throughout the world. Subclinical ketosis is one of the most dominant metabolic disorders in dairy herds during early lactation, so early detection and prevention are important for both economic and animal welfare reasons. Neural networks, which offer a high degree of accuracy in predicting various phenomena and processes where there is no clear causal correlation or there are no rules that allow the establishment of a logical cause-and-effect relationship, can be used to address problems related to prediction, classification, or control. A Multi-Layer perceptron (MLP) is a feedforward artificial neural network model that takes input data for a set of proper output. This study investigated the performance of four algorithms used to train MLP networks. The experimental results demonstrate that the MLP network model improved the accuracy of process recognition of subclinical ketosis in dairy cows. The received artificial model’s results were saved in the predictive model markup language (PMML) and can be used to describe the learning set, the algorithm used in the data mining application and related information.

Abstract

Subclinical ketosis is one of the most dominant metabolic disorders in dairy herds during lactation. Cows suffering from ketosis experience elevated ketone body levels in blood and milk, including β-hydroxybutyric acid (BHB), acetone (ACE) and acetoacetic acid. Ketosis causes serious financial losses to dairy cattle breeders and milk producers due to the costs of diagnosis and management as well as animal welfare reasons. Recent years have seen a growing interest in the use of artificial neural networks (ANNs) in various fields of science. ANNs offer a modeling method that enables the mapping of highly complex functional relationships. The purpose of this study was to determine the relationship between milk composition and blood BHB levels associated with subclinical ketosis in dairy cows, using feedforward multilayer perceptron (MLP) artificial neural networks. The results were verified based on the estimated sensitivity and specificity of selected network models, an optimum cut-off point was identified for the receiver operating characteristic (ROC) curve and the area under the ROC curve (AUC). The study demonstrated that BHB, ACE and lactose (LAC) levels, as well as the fat-to-protein ratio in milk, were important input variables in the network training process. For the identification of cows at risk of subclinical ketosis, variables such as BHB and ACE levels in milk were of particular relevance, with a sensitivity and specificity of 0.84 and 0.61, respectively. It was found that the back propagation algorithm offers opportunities to integrate artificial intelligence and dairy cattle welfare within a computerized decision support tool.