Determination of prediction equations for estimating body weight of Zulu (Nguni) sheep

Morphological Description and Live Weight Prediction from Body Measurements of Socorro Island Merino Lambs

The aim of this study was to describe the morphology and estimate live weight from body measurements of Socorro Island Merino lambs. A group of Socorro Island Merino lambs was recorded from birth to year for live weight, rump width, rump length, withers height, body length, cannon bone perimeter, and chest girth, width, and depth. The effect of the lamb type on body measurements and live weight was analyzed using ANOVA, Pearson's correlation analysis was performed to estimate the relationship between body measurements and live weight, multiple linear regressions were fitted to obtain prediction equations of live weight from the body measurements and finally, chest girth was used to generate prediction equations using linear and exponential models. At birth and at year, differences were observed in body measurements, especially those related to the thoracic region, with crossbred males showing the highest values. Live weight was correlated with almost all the body measurements, with the highest coefficients observed with chest girth, chest width, and chest depth. Live weight can be accurately predicted from multiple regression equations using several body measurements, but using only chest girth (CG) as a predictor, the exponential equation W_0-365 = 0.9142 exp^{(0.0462 CG)} showed the best accuracy.

Towards the Estimation of Body Weight in Sheep Using Metaheuristic Algorithms from Biometric Parameters in Microsystems

The Body Weight (BW) of sheep is an important indicator for producers. Genetic management, nutrition, and health activities can benefit from weight monitoring. This article presents a polynomial model with an adjustable degree for estimating the weight of sheep from the biometric parameters of the animal. Computer vision tools were used to measure these parameters, obtaining a margin of error of less than 5%. A polynomial model is proposed after the parameters were obtained, where a coefficient and an unknown exponent go with each biometric variable. Two metaheuristic algorithms determine the values of these constants. The first is the most extended algorithm, the Genetic Algorithm (GA). Subsequently, the Cuckoo Search Algorithm (CSA) has a similar performance to the GA, which indicates that the value obtained by the GA is not a local optimum due to the poor parameter selection in the GA. The results show a Root-Mean-Squared Error (RMSE) of 7.68% for the GA and an RMSE of 7.55% for the CSA, proving the feasibility of the mathematical model for estimating the weight from biometric parameters. The proposed mathematical model, as well as the estimation of the biometric parameters can be easily adapted to an embedded microsystem.

Prediction of live body weight based on body measurements in Thalli sheep under tropical conditions of Pakistan using cart and mars

Thalli sheep is a significant breed reared under tropical region of Punjab province of Pakistan. The present study was conducted to predict live body weight (LBW) by means of from some body measurements, i.e., chest girth (CG), belly girth (BG), rump height (RH), withers height (WH), neck girth (NG), and body length (BL) taken from 155 Thalli indigenous sheep of Pakistan. Age factor is determined to be a significant source of variation for BL, BG, CG, BG, WH, and NG (p < 0.05). LBW is correlated significantly with BL (0.850), CG (0.825), BG (0.849), RH (0.579), WH (0.547), and NG (0.7760), respectively (p < 0.01). For LBW prediction, CART and MARS data mining algorithms were comparatively used based on ten cross-validation method. Among 185 candidate MARS models with 1-5 degrees of interaction and 2-38 terms, the MARS model with 7 terms and no interaction effect in R software was the best model for LBW prediction on the basis of the smallest cross-validated RMSE value. Also, the optimal CART tree structure was obtained with 9 terminal nodes for the smallest cross-validated RMSE value. MARS algorithm outperformed CART in LBW prediction and explained 90.3 (%) of variability in LBW of Thalli sheep. Results of the optimal CART structure reflected that Thalli sheep with BL > 75 cm, RH > 83 cm, and NG > 55 cm has the heaviest LBW of 72 kg. The optimal MARS model displays that the heaviest LBW can be produced by Thalli sheep with BL > 71.12 cm, BG > 106.68 cm, WH > 76.2 cm, NG > 50.8 cm in 5^th age group. In conclusion, it coud be recommended that MARS predictive modeling may enable animal breeders to obtain elite Thalli sheep population and to detect body measurement positively influencing LBW as indirect selection criteria for not only describing breed characterization and developing flock management standards, but also ensuring sustainable meat production and rural development in Pakistan.

ASAS-NANP SYMPOSIUM: Applications of machine learning for livestock body weight prediction from digital images

Monitoring, recording, and predicting livestock body weight (BW) allows for timely intervention in diets and health, greater efficiency in genetic selection, and identification of optimal times to market animals because animals that have already reached the point of slaughter represent a burden for the feedlot. There are currently two main approaches (direct and indirect) to measure the BW in livestock. Direct approaches include partial-weight or full-weight industrial scales placed in designated locations on large farms that measure passively or dynamically the weight of livestock. While these devices are very accurate, their acquisition, intended purpose and operation size, repeated calibration and maintenance costs associated with their placement in high-temperature variability, and corrosive environments are significant and beyond the affordability and sustainability limits of small and medium size farms and even of commercial operators. As a more affordable alternative to direct weighing approaches, indirect approaches have been developed based on observed or inferred relationships between biometric and morphometric measurements of livestock and their BW. Initial indirect approaches involved manual measurements of animals using measuring tapes and tubes and the use of regression equations able to correlate such measurements with BW. While such approaches have good BW prediction accuracies, they are time consuming, require trained and skilled farm laborers, and can be stressful for both animals and handlers especially when repeated daily. With the concomitant advancement of contactless electro-optical sensors (e.g., 2D, 3D, infrared cameras), computer vision (CV) technologies, and artificial intelligence fields such as machine learning (ML) and deep learning (DL), 2D and 3D images have started to be used as biometric and morphometric proxies for BW estimations. This manuscript provides a review of CV-based and ML/DL-based BW prediction methods and discusses their strengths, weaknesses, and industry applicability potential.

Comparison of linear and nonlinear models to estimate body weight of Pelibuey ewes from body measurements

The aim of this study was to develop prediction equations for the body weight of Pelibuey ewes from body measurements comparing linear vs nonlinear models. A subsample of 197 ewes was scored for body weight (BW), rump length (RL), rump width (RW), height at withers (HW), chest girth (CG), chest width (CW), chest depth (CD), cannon bone perimeter (CP), and body length (BL). Pearson's correlation analysis was performed on a sub-data set from 197 ewes to estimate the relationship between body weight and body measurements. Multiple linear regressions were fitted to obtain prediction equations of body weight from the eight body measurements, and prediction equations were obtained from the body measurement that showed the highest correlation with body weight using five nonlinear models allometric, saturation growth, exponential, and incomplete gamma. Data from an independent subsample of 196 ewes was used to validate the equation with the best goodness of fit using linear regression analysis. CG was the body measurement that showed the highest correlation with BW, and based on multiple stepwise regression, in the equation BW = - 60.622 + 1.233CG explained 79% of the body weight variation. Moreover, BW prediction was more accurate when other measurements such CW, BL, and RW were added to the model generating to the equation BW = - 68.875 + 0.845CG + 0.866CW + 0.195BL + 0.601RW (R² = 0.85, MSE = 15.51). In the case of nonlinear models, incomplete gamma and exponential models generated the equations with the best goodness of fit and precision: BW = 0.077CG^1.108exp^(0.016CG) (R² = 0.82, MSE = 18.64) and BW = 3.5759exp^(0.0292CG) (R² = 0.82, MSE = 18.65) respectively.

Estimation of body weight in hair ewes using an indirect measurement method

The aim of the present study was to develop and evaluate an equation to predict body weight (BW) using hip width (HW) in Pelibuey ewe lambs and ewes. Five hundred seventy-seven 2-month-old to 3-year-old, non-pregnant, non-lactating, clinically healthy ewe lambs and adult ewes with a mean BW of 34.7 ± 12.4 kg and HW of 15.6 ± 3.4 cm were considered. Three equations were evaluated: BW (kg): - 19.17 + 3.46 × HW (Eq. 1), BW (kg): - 17.79 + 3.25 × HW + 0.007 × HW² (Eq. 2) and BW (kg): 0.39 × HW^1.63 (Eq. 3). Independent data from 80 animals with similar characteristics (BW of 23.4 ± 10.9 kg and HW of 12 ± 3.1 cm) were also considered to evaluate the developed equations. The evaluation was based on the relationship between the observed and predicted values of BW analysed using a linear regression, the mean squared error of prediction (MSEP), the root MSEP (RMSEP) and the concordance correlation coefficients (CCCs). Additionally, cross-validation analyses were performed using the k-folds validation (k = 10) procedure. The correlation coefficient (r) between BW and HW was 0.94 (P < 0.001). The parameters for precision and accuracy showed that the proposed equations had high precision (R² > 0.95%), accuracy (Cb > 0.98) and reproducibility (CCC > 0.96) in predicting the BW of ewe lambs and adult ewes. Equation (1) accurately predicted observed BW, with a bias (observed - predicted) of 4.3 kg and RMSEP of 9.68% with respect to the observed BW (random error of 84.23%); it also generated the best prediction according to the residual mean squared prediction error, coefficient of determination and mean absolute error. In conclusion, the highly correlated relationship between BW and HW in Pelibuey ewe lambs and adult ewes under humid tropic conditions enabled the development of mathematical models herein to estimate BW with an adequate goodness of fit. The linear model showed the best performance according to the goodness-of-fit evaluation and internal and external validation; hence, this model is proposed for use in both the experimental and commercial farms.

Prediction of carcass weight from body measurement traits of Chinese indigenous Dagu male chickens using path coefficient analysis

In the current study, correlation and path coefficient analysis were applied to investigate the relationship among body measurement traits and carcass weight, and to determine the direct and indirect effects of the body measurement traits including body slope length (BSL), breast width (BW), breast depth (BD), pelvis width (PW), shank length (SL) and shank circumference (SC) on carcass weight. Chinese indigenous Dagu male chickens (80) were used at eighteen weeks of age. Pair-wise correlation results showed high significance between carcass weight and body measurement traits. High correlation was between carcass weight and body slope length (0.596) while the lower correlation was between breast depth and shank length (0.112), respectively. Path coefficient analysis results indicated that shank circumference and shank length had the highest direct effect (0.225, 0.223) on carcass weight than other body measurement traits and breast depth had the highest indirect effect (0.125), respectively.The current study might be used by chicken farmers for prediction of carcass weight while the chicken is still alive.

Estimation of egg weight from some external and internal quality characteristics in quail by using various data mining algorithms

The objective of this study was to compare the predictive performance of some data mining algorithmsimplemented in the estimation of egg weight (EW), some egg quality traits measurements in quail. These qualitytraits are albumen weight (AW), yolk weight (YW), specific weight (SW), albumen index (AI), albumen height(AH), yolk index (YI), shell weight (SHW), shell thickness (SHT), Haugh unit (HU) and shape index (SI). Forcomparing the predictive performance of these algorithms in Model, goodness of fit criteria such as coefficient ofdetermination (R2%), adjusted coefficient of determination (Adj. R2%), coefficient of variation (CV%), SD ratio,root mean square error (RMSE), relative approximation error (RAE), and Pearson correlation coefficient, betweenobserved and predicted values were calculated. The heaviest average EW of 13.516 g was obtained from thesubgroup of those having AW > 8 cm. The results showed that the analysis based on exhaustive CHAID might beuseful for further researches linked with characterization of quail egg better than those provided by CART andCHAID algorithms.

Principal component analysis of biometric traits to explain body conformation in Kajali sheep of Punjab, India

Different body biometric traits were analysed in Kajali ewes (395) of Punjab (India) using varimax rotated principal components (PCA) with Kaiser Normalization to explain their body conformation and were subsequently used to predict adult body weight. These traits revealed that Kajali sheep were large in size and suitable for mutton production. The positive and highly significant phenotypic correlations among most of the traits indicates high predictability among these traits. PCA extracted major three components which explained 68.66% of total variation of body biometry. First component described the body size and explained about 36% of total variation. It was represented by high component loadings for chest girth, paunch girth and body weight. The second component described the tail length, height and ear length and explained about 21% variation. The communalities ranges between 0.33 (face length) to 0.87 (chest girth). The lower communality of face length and ear length indicates that these traits are less effective to explain the body conformation. The study suggested that PCA could be used in breeding programme for phenotypic selection of Kajali ewes and PCA is more appropriate compared to multiple regression analysis in predicting adult body weight. In predicting body weight using multiple regression analysis, chest girth alone accounted for 55% of total variation in body weight whereas, use of first principal component as a single predictor explained 79% of total variance and the best prediction equation (R2 = 0.83) was obtained after inclusion of second and third component in the model.

Genetic and phenotypic variation among four Nguni sheep breeds using random amplified polymorphic DNA (RAPD) and morphological features

This study was conducted to investigate phenotypic and genetic differentiation among the four Nguni sheep breeds. Sheep with two permanent incisors and above were sampled from areas, namely KwaZulu-Natal (Zulu sheep), Limpopo (Pedi sheep), Mozambique (Landim sheep) and Swaziland (Swazi sheep). The Dorper was used as an out-group. Eight morphometric variables were measured from each animal, and blood samples were collected (n = 50 per population) for genetic characterization. The mean body weights for sheep were 30.41 ± 0.41, 35.34 ± 0.43, 35.23 ± 0.43, 37.63 ± 0.42 and 52.84 ± 0.30 for Swazi, Zulu, Landim, Pedi and Dorper, respectively. Morphometric cluster analysis showed the Landim, Swazi and Zulu breeds in one cluster. The Pedi sheep were closer to the Dorper than to the other Nguni sheep. Random amplified polymorphic DNA (RAPD) technique was used to assess genetic variation. Eight primers were selected for analysis based on band pattern quality, reproducibility and the presence of distinctive bands. The Swazi sheep formed a cluster with Zulu sheep, and the Pedi formed a cluster with the Dorper. These results confirm indications by other researchers that Pedi sheep are genetically distant from Zulu and Swazi sheep breeds. This could indicate the possibility of cross breeding Zulu and Swazi sheep as a possible conservation strategy to control inbreeding. The mtDNA should be analyzed to trace the relationships between Pedi and the three Nguni sheep breeds through maternal lines.

The estimation of live weight based on linear traits in indigenous Tswana goats at various ages in Botswana

The objective of this study was to determine the usefulness of some linear body measurements in predicting live weight in indigenous Tswana goats at various ages. Data for this study were obtained from 2,783 goats sampled from six agricultural regions of Botswana except for Tsabong and Ghanzi districts. Fifteen farmers keeping goats were randomly selected from each district and records taken on a random sample of 4-12 animals per farm depending on the 1999 Botswana Government average district flock size. Body measurements recorded were heart girth, height at withers, body length, shoulder width, and live weight. Information on age of each animal was estimated from dentition; flock size and sex of the animal were also recorded for each farm. Regression analysis using stepwise selection method in Statistical Analysis System was used to determine prediction equations for live weight with heart girth, height at withers, body length, and shoulder width as independent variables for male and female goats of different ages. In all models, heart girth contributed most in explaining variation in body weight as shown by high partial R(2) which ranged from 0.48 for female mature to 0.80 for mature male goats. Using mean square error, R(2), and Mallows' C(p), the best prediction equations were for female and male kids (R(2) = 0.83 and 0.82, respectively) and mature male goats (R(2) = 0.82). The poorest model was for mature female goats with R(2) = 0.57. Live body weight could be estimated with accuracy from linear body measurements in Tswana goats which are unique for each sex and age. More research is required to explore using these models to develop measuring tapes for use by resource poor farmers who keep indigenous Tswana goats.

A review of some characteristics, socio-economic aspects and utilization of Zulu sheep: implications for conservation

Zulu sheep are Nguni sheep of Zululand and are adapted to the harsh conditions of KwaZulu-Natal. They are used by rural farmers for economic purposes. Their numbers are declining, indicating a potential extinction threat. Knowledge of their phenotypic and genotypic characteristics is essential for conservation planning. In this review, there is a focus on the utilization, socio-economic aspects, phenotypic and genotypic characteristics as well as a proposed breeding programme. A survey has shown that rural farmers in the areas of northern KwaZulu-Natal prefer to keep this breed for its adaptability, resistance to diseases and meat quality. Zulu sheep are small-framed multi-coloured animals. Mature males weigh up to 38 kg and females up to 32 kg. Based on four morphological traits and live weight, phenotypic diversity between three populations was estimated at 48%. A genetic diversity between these three populations was estimated at 22%. Live weight of Zulu sheep can be estimated using the heart girth and wither height measurements. Scrotum circumference of young rams (up to 22 months old) is reliable for estimating the live weight. Animals that were characterized in the studies were grazed extensively and no supplements were provided. There is therefore a potential of weight increase if these animals are reared in a semi-extensive environment. An open nucleus breeding scheme is thus recommended for a sustainable use and conservation of this breed. For more conclusive results, larger numbers of phenotypic and genetic characteristics, in larger numbers of Zulu sheep populations, should be investigated.

The Damara in the context of Southern Africa fat-tailed sheep breeds

Sheep production is one of the most important activities in animal production in tropical regions. Southern Africa (SA) fat-tailed sheep such as the Afrikaner, Namaqua Afrikaner, Nguni, Persian Black Head, Tswana, Pedi, Sabi, and particularly the Damara are important animal genetic resources very well adapted to their harsh home environments. In this review, autochthonous sheep genetic resources of SA are reviewed regarding history and productive performance. A special focus is made to the most internationalized of such breeds: the Damara. The following aspects are reviewed: history, standards, productive performance in SA, and importation to Australia. Finally, a reference is made to the influence of fat-tailed sheep formation of composite breeds such as the Dorper or the Meatmaster.

Morphological traits in Portuguese Bordaleira de Entre Douro e Minho sheep: divergence of the breed

Production of the autochthonous Portuguese sheep breed Bordaleira de Entre Douro e Minho (BEDM) is directly associated with the concept of sustainable agriculture. This concept focuses on the adaptation of animal populations to adverse environmental conditions, high biological efficiency and traditional management. Native BEDM sheep can be divided into the following two types on the basis of their geographic location in Portugal and how they are managed: the Várzea, which is found mainly in the lower Entre Douro e Minho region, and the Mountain type, which traditionally prefers common areas. In the present work, we analysed records of liveweight, 15 body measurements and eight indexes of the BEDM sheep, to explore the ability to distinguish different characteristics between BEDM sheep populations. The records that were obtained and analysed by principal component analysis for 412 BEDM sheep (68 males and 344 females) included the following: heights at withers (HW), at back (HB), at rump (HR) and at chest (HC); lengths of trunk (LT), of rump (LR), of head (LH) and of perineum (LP); widths of head (WH), of chest (WC), of biiliac (WBIIL) and of biischiatic (WBIIS); perimeters of the shin (PS), thorax (PT), and abdomen (PA); and finally indexes such as cephalic index (CI), thoracic index (TI), pelvic index (PI), body index (BI), relative chest depth index (RDCI), lateral corporal index (LCI), relative shin thickness index (RTSI) and dactyl–thoracic index (DTI). The Várzea BEDM was morphologically superior to the Mountain variety, with this superiority more evident in females. Height measurements, especially HW in adult females, provided an excellent measurement reference for biometrical breed studies. Functional index values confirmed the BEDM Mountain variety’s appropriateness for mountainous regions. Relationships among body measurements suggested that the breed standard needed to be updated. The research presented here was used by the Portuguese Ministery to split BEDM into two new breeds; the traditional variety called Várzea now represents the new BEDM breed, whereas the Mountain variety represents the new Churra do Minho breed.