Morphological characterization of the Polatli sheep in terms of live weight using data mining algorithms.

The aim of this research is both to estimate the live weight (LW) of Polatli sheep (Ile de France × Akkaraman (G1)) by considering some body measurements (withers height (WH), rump height (RH), body length (BL), chest depth (CD), chest width (CD), chest girth (CG), cannon bone circumference (CBC)), age, and sex variables as independent variables using C&RT (Classification and Regression Tree), CHAID (Chi-square Automatic Interaction Detector), and MARS (Multivariate Adaptive Regression Splines) algorithms and to determine the significant independent variables in the estimation of live weight. For this purpose, a total of 210 sheep were used, including 180 females and 30 males of different ages, for the estimation of LW. The calculated Pearson correlation coefficients between LW and WH, RH, BL, CD, CW, CG, and CBC characteristics are 0.897, 0.896, 0.853, 0.948, 0.550, 0.914, and 0.798, respectively (p < 0.05). In the application of data mining algorithms as prediction models, a cross-validation of 10 was used, while for tree-based algorithms, the parent node was set to 10 and the child node to 5. While CHAID and C&RT algorithms each used 8 independent variables to explain the variation observed in LW, the MARS algorithm used 9 independent variables. In Polatli sheep, the sheep with the highest live weight was found in the node with age > 3 and CD > 36 cm cutting point in the CHAID algorithm (93.571 kg). In the C&RT algorithm, it was predicted to be (91.316 kg) when age > 0, CD > 36.5 cm, and CBC > 9.5 cm. When evaluated considering commonly used criteria, the prediction performances of CHAID, C&RT, and MARS algorithms were calculated as follows: RMSE (root mean square error) values are "5.788, 5.103, 4.005"; SDR (standard deviation ratio) values are "0.254, 0.224, 0.176"; MAPE (mean absolute percentage error) values are "7.555, 6.675, 5.682"; Adj-Rsq (adjusted R-squared) values are "0.935, 0.950, 0.969"; and AIC (Akaike information criterion) values are "741.436, 688.489, 582.792," respectively. In terms of prediction performance, among the tree algorithms (CHAID and C&RT), C&RT was found to be the best, while considering all performance measures, it was observed that the MARS algorithm exhibited the best performance. Consequently, it has been determined that C&RT and MARS algorithms can be safely employed in morphological characterization studies for the identification of indirect criteria and the formation of elite herds in terms of LW. This decision allows for the reliable use of these algorithms to facilitate the selection of indirect variables and the establishment of elite populations in breeding programs focusing on live weight.

Diagnosis of lameness via data mining algorithm by using thermal camera and image processing method in Brown Swiss cows.

Lameness is one of the culling factors such as mastitis, low milk yield, and infertility that cause economic losses in herd management as they threaten animal health and welfare. The purpose of this study was to evaluate the early detection of lameness in Brown Swiss cattle by using a data mining algorithm by both integrating lameness scores and some image parameters such as Lab (CIE L*, a*, b*), HSB (hue, saturation, brightness), RGB (red, green, blue) by processing thermal images with ImageJ program. In the study, the variables obtained as a result of processing the skin surface temperatures and thermal images taken at the fetlock joint of 33 Brown Swiss cattle were used as independent variables. Also, healthy cows (lameness scores 1 and 2) and unhealthy cows (lameness scores 3, 4, and 5) used in the diagnosis of lameness were used as a binary response variable. Classification and regression tree (CART) was used as a data mining algorithm in the diagnosis of lameness. As a result, the CART algorithm correctly classified 12 of the 13 heads unhealthy cows according to locomotion scores. According to locomotion scores by using CART analysis in this study, independent variables that are used to classify healthy and unhealthy (lame) animals were determined as maximum temperature (Tmax), green (mean), L (max), and age (P<0.05). The cut-off values of these independent variables were predicted as 32.40, 149.14, 97.11, and 5.50 for Tmax, green (mean), L (max), and age, respectively. Also, the sensitivity, specificity, and area under the ROC curve (AUC) of the CART algorithm for locomotion scoring were found as 92.31%, 95%, and 93.7% respectively. The area under ROC curve (AUC) was found to be significant in the diagnosis of lameness (P<0.01). Results showed that the use of CART classification algorithm together with thermal camera and image processing methods is a usefull tool in the detection of lameness in the herds. It is recommended that more comprehensive studies by increasing the number of animals in the future would be more beneficial.

Prediction of the live weight at breeding age from morphological measurements taken at weaning in indigenous Honamli kids using data mining algorithms.

The purpose of this study was to predict live weight at breeding age (LW) based on weaning morphological traits, which birth weight (BW), weaning weight (WW), withers height (WH), back height (BH), rump height (RH), chest depth (CD), body length (BL), tail length (TL), chest girth (CG), leg girth (LG), front shark circumference (FSC), head length (HL), head width (HW), nose length (NL), ear length (EL), and ear width (EW). For this purpose, measurements were taken from 84 Honamli kids born in 2018. The study also included sex, birth type (BT), and ear type as the nominal predictors. For this purpose, two MARS (Multivariate Adaptive Regression Splines), which are interaction (MARS2) and non-interaction (MARS1), and based-tree algorithms, such as CART (Classification and Regression Tree), CHAID (Chi-Square Automatic Interaction Detector), and Exhaustive CHAID, were used by cross-validation 5 and compared with each other considering the predictive performance by taking into account nine predictive performances criteria. LW has a significantly positive and high linear relationship with WH (0.770), BH (0.770), RH (0.750), BL (0.750), and CG (0.770), respectively (p < 0.01). According to these criteria, second-order interaction MARS2 model had the best performance among all data mining algorithms. Also, the CHAID algorithm was the best predictor of LW among regression tree-based algorithms. The CHAID algorithm predicted that the Honamli goat with 14.426 < WW < 15.575 kg and HW > 16.464 cm had the heaviest LW at 56.268 kg. The MARS2 model showed that the heaviest LW could be produced by WW > 16.10 kg, HW > 17 cm, Sex-Male × BL > 60 cm, WW × BL < 50 cm, BT-twin × WW < 15.60 kg, BL > 50 cm × CG > 62.4 cm and male goats. Also, CHAID and MARS2 algorithms explain 92.00% and 94.50% of the variation in LW, respectively. According to the results, it can be concluded that the CHAID and MARS algorithms used in the prediction of LW at breeding age could give an idea to reveal the breed standards examined for breeding purposes. While determining that there are important statistical methods in defining body characteristics at weaning in a complex way, the body characteristics determined by these models can be used as indirect selection criteria.

Relationships between FAMACHA© scores and parasite incidence in sheep and goats.

One of the most important activities in animal raising is parasite control. Since parasites share what animals eat on one hand, they have disease-causing effects on the other hand, they are parasites that must be fought. In fact, they can be responsible for the decrease in yield and can also cause death. Haemonchus contortus parasite from the Trichostrongylidae family is a blood parasite that causes anemia in sheep and goats. In this study, using the FAMACHA© card, which is widely used in the determination of anemia caused by Haemonchus contortus in the world, the color of the conjunctiva was scored in Akkaraman sheep and Hair goats raised in the Sizma district of Konya, and the relations between the score and the parasite type and density were determined. Accordingly, while no animals with 1 and 5 scores were observed in sheep, the numbers of animals with scores of 2, 3, and 4 were determined as 7.84%, 40.2%, and 52.96%, respectively. Parasite eggs belonging to the Trichostrongylidae family were observed in 96.07% of the sheep. On the other hand, no animal with score 1 or 2 was observed in goats. The numbers of animals with 3, 4, and 5 scores were determined as 6.12%, 54.08%, and 39.80%, respectively. Eimeria, the most common parasite species, was found in 68.18% of the goats and parasite eggs from the Trichostrongylidae family were observed in 17.05%. As a result, it was concluded that the card in question was a practical, easily available, and cheap tool in the detection of anemia by using the relationships between FAMACHA© card scoring and parasite loads.

Dentinal tubule penetration of endodontic sealers after nonthermal plasma treatment: A confocal laser scanning microscopy study.

One of the factors affecting the success of endodontic treatment is to fill the root canal system hermetically. The aim of this in vitro study was to evaluate the effect of nonthermal plasma (NP) on dentinal tubule penetration of root canal sealers using confocal laser scanning microscopy. Forty mandibular premolar teeth were selected and the root canals were prepared with large-Waveone-Gold rotary-files. Specimens were divided into four experimental groups according to sealer and NP treatment (n = 10). G1: AH-Plus (AH) G2: nonthermal plasma application + AH-Plus(AH-P) G3: Endosequence-BC(BC) G4: nonthermal plasma application + Endosequence-BC(BC-P). Cold lateral-condensation technique was used for the obturation of root canals. The roots were sectioned horizontally and the sections were examined under confocal laser scanning microscopy. The maximum tubule penetration and percentage of penetration values were obtained from the microscopy images and were statistically analyzed with repeated measurements-ANOVA and the Tukey (HSD) test (p < 0.05). The percentages of dentinal tubule penetration of the groups were not statistically different. The maximum tubule penetration of the AH-P was statistically lower than that of the BC-P (p < 0.05). Plasma application had no affect on the percentage of dentinal tubule penetration. Under the conditions of this in vitro Endosequence-BC sealer showed higher maximum tubule penetration values than AH-Plus after NP treatment. Percentage of dentinal tubule penetration values of experimental groups was similar.