Comparison of the decision tree, artificial neural network and multiple regression methods for prediction of carcass tissues composition of goat kids

A supervised machine learning approach for the prediction of antioxidant activities of Amaranthus viridis seed

This research aimed at modelling and predicting the antioxidant activities of Amaranthus viridis seed extract using four (4) data-driven models. Artificial Neural Network (ANN), Support Vector Machine (SVM), k-nearest Neighbour (k-NN), and Decision Tree (DT) were used as modelling algorithms for the construction of a non-linear empirical model to predict the antioxidant properties of Amaranthus viridis seed extract. Datasets for the modelling operation were obtained from a Box Behnken design while the hyperparameters of the ANN, SVM, k-NN and DT were determined using a 10-fold cross-validation technique. Among the Machine Learning algorithms, DT was observed to exhibit excellent performance and outperformed other Machine Learning algorithms in predicting the antioxidant activities of the seed extract, with a sensitivity of 0.867, precision of 0.928, area under the curve of 0.979, root mean square error of 0.184 and correlation coefficient of 0.9878. It was closely followed by ANN which was used to analyze and explain in detail the effect of the independent variables on the antioxidant activities of the seed extracts. This result affirmed the suitability of DT in predicting the antioxidant activities of Amaranthus viridis.

Using fat thickness and longissimus thoracis traits real-time ultrasound measurements in Black Belly ewe lambs to predict carcass tissue composition through multiresponse multivariate adaptive regression splines algorithm

The main idea of the current study was to estimate carcass tissue composition using fat thickness and longissimus thoracis (LT) traits real-time ultrasound measurements (USM) in Black Belly ewe lambs through multiresponse multivariate adaptive regression splines (MARS) algorithms. Twenty-four hours before slaughter, subcutaneous (SFT) and kidney-fat thickness (KFT), LT depth (LTD), width (LTA, cm) and area (LTMA) were measured in 60 lambs (BW of 26.40 ± 7.01 kg). Information on carcass and non-carcass components was recorded after slaughter. The total carcass muscle (TCM), total carcass bone (TCB), and total carcass fat (TCF) had a low to high correlation (P < 0.01) with BW, cold carcass weight (CCW), and LTD, SFT, KFT, and LDMA. The CCW (%65.58) and SFT (%16.70) were the most effective variables, whilst LTD (%9.57) and LTMA (%8.15) were the lowest variables for determining TCB, TCM, and TCF. The multiresponse MARS algorithm provides an accurate and efficient means of estimating TCF, TCB, and TCM.

In vivo prediction of abdominal fat and breast muscle in broiler chicken using live body measurements based on machine learning

The purpose of this study was to predict the carcass characteristics of broilers using support vector regression (SVR) and artificial neural network (ANN) model methods. Data were obtained from 176 yellow feather broilers aged 100-day-old (90 males and 86 females). The input variables were live body measurements, including external measurements and B-ultrasound measurements. The predictors of the model were the weight of abdominal fat and breast muscle in male and female broilers, respectively. After descriptive statistics and correlation analysis, the datasets were randomly divided into train set and test set according to the ratio of 7:3 to establish the model. The results of this study demonstrated that it is feasible to use machine learning methods to predict carcass characteristics of broilers based on live body measurements. Compared with the ANN method, the SVR method achieved better prediction results, for predicting breast muscle (male: R² = 0.950; female: R² = 0.955) and abdominal fat (male: R² = 0.802; female: R² = 0.944) in the test set. Consequently, the SVR method can be considered to predict breast muscle and abdominal fat of broiler chickens, except for abdominal fat in male broilers. However, further revaluation of the SVR method is suggested.

Estimation of milk yield based on udder measures of Pelibuey sheep using artificial neural networks

Udder measures have been used to assess milk yield of sheep through classical methods of estimation. Artificial neural networks (ANN) can deal with complex non-linear relationships between input and output variables. In the current study, ANN were applied to udder measures from Pelibuey ewes to estimate their milk yield and this was compared with linear regression. A total of 357 milk yield records with its corresponding udder measures were used. A supervised learning was used to train and teach the network using a two-layer ANN with seven hidden structures. The globally convergent algorithm based on the resilient backpropagation was used to calculate ANN. Goodness of fit was evaluated using the mean square prediction error (MSPE), root MSPE (RMSPE), correlation coefficient (r), Bayesian’s Information Criterion (BIC), Akaike’s Information Criterion (AIC) and accuracy. The 15–15 ANN architecture showed that the best predictive milk yield performance achieved an accuracy of 97.9% and the highest values of r² (0.93), and the lowest values of MSPE (0.0023), RMSPE (0.04), AIC (− 2088.81) and BIC (− 2069.56). The study revealed that ANN is a powerful tool to estimate milk yield when udder measures are used as input variables and showed better goodness of fit in comparison with classical regression methods.

Estimation of Carcass Tissue Composition from the Neck and Shoulder Composition in Growing Blackbelly Male Lambs

This study was designed to develop predictive equations estimating carcass tissue composition in growing Blackbelly male lambs using as predictor variables for tissue composition of wholesale cuts of low economic value (i.e., neck and shoulder). For that, 40 lambs with 29.9 ± 3.18 kg of body weight were slaughtered and then the left half carcasses were weighed and divided in wholesale cuts, which were dissected to record weights of fat, muscle, and bone from leg, loin, neck, rib, and shoulder. Total weights of muscle (CM), bone (CB) and fat (CF) in carcass were recorded by adding the weights of each tissue from cuts. The CM, CF and CB positively correlated (p < 0.05; 0.36 ≤ r ≤ 0.86), from moderate to high, with most of the shoulder tissue components, but it was less evident (p ≤ 0.05; 0.32≤ r ≤0.63) with the neck tissue composition. In fact, CM did not correlate with neck fat and bone weights. Final models explained (p < 0.01) 94, 92 and 88% of the variation observed for CM, CF and CB, respectively. Overall, results showed that prediction of carcass composition from shoulder (shoulder) tissue composition is a viable option over the more accurate method of analyzing the whole carcass.

Non-Destructive Imaging and Spectroscopic Techniques for Assessment of Carcass and Meat Quality in Sheep and Goats: A Review

In the last decade, there has been a significant development in rapid, non-destructive and non-invasive techniques to evaluate carcass composition and meat quality of meat species. This article aims to review the recent technological advances of non-destructive and non-invasive techniques to provide objective data to evaluate carcass composition and quality traits of sheep and goat meat. We highlight imaging and spectroscopy techniques and practical aspects, such as accuracy, reliability, cost, portability, speed and ease of use. For the imaging techniques, recent improvements in the use of dual-energy X-ray absorptiometry, computed tomography and magnetic resonance imaging to assess sheep and goat carcass and meat quality will be addressed. Optical technologies are gaining importance for monitoring and evaluating the quality and safety of carcasses and meat and, among them, those that deserve more attention are visible and infrared reflectance spectroscopy, hyperspectral imagery and Raman spectroscopy. In this work, advances in research involving these techniques in their application to sheep and goats are presented and discussed. In recent years, there has been substantial investment and research in fast, non-destructive and easy-to-use technology to raise the standards of quality and food safety in all stages of sheep and goat meat production.