Application of Ultrasonic Nondestructive Testing for Breeding of Meat Pigeons

Featured Application

The first application of ultrasonic nondestructive testing technology for pectoral musculature yield in the breeding of pigeons intended for meat production.

Abstract

The aims of this study were to assess the correlation of breast muscle thickness (BMT) with the breast muscle weight (BMW), body weight (BW), breast width, and breast depth (BD) of meat pigeons, and to explore the feasibility of ultrasonic nondestructive testing (UNDT) for non-invasive measurements of BMT. It has been proven that using ultrasonic nondestructive testing specifically for evaluating the pectoral musculature yield in other poultry species such as Peking ducks and Yangzhou geese is feasible. The BMT of Carneau and Silver King pigeons (n = 103 each; age, 28 days) was measured at two points by UNDT. The correlation between the BMT at two points and BMW was analyzed. The results demonstrated that the correlation coefficients between BMT at point A and BMW were 0.907 and 0.897 in Carneau and Silver King pigeons, respectively, with significant regression relationships (p < 0.01). Variety and sex had little effect on BMT measurements. Six optimal univariate linear regression equations were established. The regression relationship was very significant, as determined by the F-test (p < 0.01), indicating an actual regression relationship between the variables of each equation and that UNDT can be applied for the breeding of meat pigeons. This represents the first application of ultrasonic nondestructive testing specifically for evaluating pectoral musculature yield in pigeons intended for meat production.

1. Introduction

China is the largest producer of meat pigeons, accounting for about 80% of global production [1]. The meat pigeon is currently the fourth most commonly bred poultry species in China after chickens, ducks, and geese [2]. In recent years, the global demand for meat pigeons has continued to grow. The meat pigeon industry in China not only meets domestic market needs but also holds an important position in international exports. Its industry chain involves various aspects such as feed processing, disease prevention and control, breeding technology promotion, product processing, and brand establishment, resulting in significant economic stimulation [3]. Pigeon meat is rich in nutrients, various amino acids, high-quality proteins, and trace elements [4,5]. An old Chinese saying states that “one pigeon is better than nine chickens”. The breast muscle (BM) of meat pigeons is the main edible part and the most valuable. Breast muscle weight (BMW) is the most important economic indicator of the production performance of meat pigeons [6]. BMW and carcass weight are established indicators for breeding high-quality BM traits of meat pigeons. However, the measurement of BMW requires slaughtering, which is time-consuming and increases production costs [7]. Moreover, high-quality birds selected after slaughter can no longer be used for breeding [8].

Ultrasonic nondestructive testing (UNDT) was recently developed for non-invasive measurements of poultry and other livestock [9]. The main advantages of UNDT include safety, reliability, and non-invasive measurement [10], which can be repeated and observed in real time [11], making it particularly advantageous for preserving breeding stock. By contrast, traditional methods, such as manual palpation or destructive sampling, are limited by precision, cost, and the inability to retain high-value breeding stock. UNDT was introduced in the 1980s in China for use in animal husbandry and veterinary medicine [12], and is currently widely applied in clinical and biological research [13]. In livestock breeding, UNDT has been used to evaluate the composition of rabbit carcasses [14], the backfat thickness of live lambs [15], the correlation between the live and carcass weights of beef cattle [16], and the gestational age of Korean black sheep [17]. UNDT has also been widely applied in poultry breeding. Zhu et al. used UNDT to identify correlations between the fat content and carcass traits of Peking ducks, which provided the basis for the application of UNDT in the breeding of meat ducks [7]. Yang et al. reported a significant positive correlation between breast muscle thickness (BMT) and body weight (BW), keel length, and BMW for the selection of Yangzhou geese [8]. However, with the rapid development of meat pigeon breeding, new requirements for testing production efficiency and accuracy have emerged. Additionally, a comprehensive review of the limitations of these applications and the rationale for extending UNDT to meat pigeons has yet to be detailed. Therefore, it is necessary to accelerate the utilization of measurable traits to evaluate the breeding performance of meat pigeons [6].

The aim of the present study was to assess the feasibility of UNDT to measure the traits of meat pigeons and identify potential correlations of BMT with BMW, dressed weight, live BW, breast width, and breast depth (BD) to improve breeding production, and to attempt to address the high time and resource consumption associated with traditional slaughter measurement methods, and promote efficient development in the pigeon meat industry.

2. Materials and Methods

2.1. Animals and Sample Collection

A total of 206 pigeons were selected randomly, including 103 Carneau pigeons and 103 Silver King pigeons, aged 28 days. All samples were reared under identical environmental and dietary conditions in facilities operated by Shanghai Jinhuang Industry Co., Ltd. (Shanghai, China).

2.2. Live and Carcass Measurements

All measurements were conducted after fasting for 12 h. The live BW, breast width, and BD of each pigeon were measured in accordance with the Chinese standard NY/T 823-2020 [18]. The feathers were removed from the breast to establish the measurement points A and B (Figure 1). The left breast of each pigeon was smeared with a coupling agent and a probe was placed perpendicular to point A or B and slowly moved back and forth over a small range until a bright “V”-shaped image appeared (Figure 2), where the upper white band is the superficial muscle membrane of the BM and the lower white band is the sternum. The vertical distance between the upper white band and the lowest point of the lower white band is the thickness of the BM, which was measured in millimeters to two decimal places. After the measurement of live BMT, the pigeon was sacrificed by the neck cutting method and the carcass was weighed after bleeding. Each measurement was conducted in triplicate to ensure repeatability. The ultrasound equipment was calibrated before each session using manufacturer-recommended standards, and operator variability was minimized through standardized training.

2.3. Statistical Analysis

All data were collected in an Excel file (Microsoft Corporation, Redmond, WA, USA) and analyzed using IBM SPSS Statistics for Windows, version 26.0 (IBM Corporation, Armonk, NY, USA). The Pearson correlation coefficients between BMT and BMW at points A and B of two breeds of meat pigeons were calculated. A probability (p) value of <0.05 was considered statistically significant. In addition, residual analysis and regression diagnostics were performed to assess the adequacy and reliability of the linear regression models.

3. Results

3.1. Live and Carcass Traits of 28-Day-Old Carneau and Silver King Pigeons

As shown in

Table 1. Body size and slaughter traits of two species of meat pigeons.

Variety	BMT at Point A (mm)	BMT at Point B (mm)	Breast Width (mm)	BD (mm)	Live BW (g)	Dressed Weight (g)	BMW (g)
Carneau pigeon	15.30 ± 1.75	11.87 ± 1.69	66.73 ± 5.09	71.70 ± 2.95	524.51 ± 52.82	456.49 ± 48.18	95.19 ± 15.26
Silver King pigeon	14.71 ± 1.29	11.10 ± 1.40	62.64 ± 3.47	70.87 ± 3.11	478.26 ± 40.16	418.73 ± 35.63	85.23 ± 11.51
p	0.007	0.225	0.001	0.988	0.017	0.016	0.094

Note: p > 0.05 indicates no significant difference, p < 0.05 indicates a significant difference, and p < 0.01 indicates an extremely significant difference. The text continues here (Figure 2 and Table 2).

, there were significant differences at point A in the BMT, breast width, live BW, and carcass weight between Carneau and Silver King pigeons. The average BMT, breast width, BD, BW, and BMW were greater in Carneau pigeons than Silver King pigeons. Overall, Carneau pigeons are larger than Silver King pigeons [18,19].

3.2. Correlation Analyses Between BMT, BMW, BW, Breast Width, and BD

As shown in

Table 2. Correlation analysis of slaughter traits with BW and size traits of Carneau pigeons.

Item	BMT at Point A	BMT at Point B	BMW	Breast Width	BD	Live BW	Dressed Weight
BMT at Point A	1
BMT at Point B	0.904 **	1
BMW	0.907 **	0.874 **	1
Breast width	0.395 **	0.494 **	0.510 **	1
BD	0.187	0.213 *	0.273 **	0.518 **	1
Live BW	0.548 **	0.527 **	0.672 **	0.416 **	0.322 **	1
Dressed weight	0.665 **	0.638 **	0.711 **	0.476 **	0.359 **	0.967 **	1

Note: * p < 0.05; ** p < 0.01.

and

Table 3. Correlation analysis of slaughter traits with BW and size traits of Silver King pigeons .

Item	BMT at Point A	BMT at Point B	BMW	Breast Width	BD	Live BW	Dressed Weight
BMT at Point A	1
BMT at Point B	0.844 **	1
BMW	0.897 **	0.760 **	1
Breast width	0.512 **	0.486 **	0.590 **	1
BD	0.388 **	0.497 **	0.471 **	0.291 *	1
Live BW	0.366 **	0.420 **	0.533 **	0.545 **	0.275 **	1
Dressed weight	0.388 **	0.567 **	0.634 **	0.579 **	0.469 **	0.906 **	1

Note: * p < 0.05; ** p < 0.01.

, BMW was significantly positively correlated with live BW, carcass weight, breast width, BD, and BMT at points A and B (p < 0.05), with higher correlations between BMW and BMT at point A. Although breast width, BD, and BMW were also significantly positively correlated, the correlation coefficient was far lower than that between the BMT and BMW at point A.

The results of correlation analysis showed that BMT at point A had the highest correlation with the BMW of the two breeds of meat pigeons (r = 0.907 and 0.897, respectively).

3.3. Regression Analysis of BMT and BMW

Univariate linear regression was conducted to investigate the relationship between the BMT and BMW. The BMW of Carneau and Silver King pigeons (Y1 and Y2, respectively) was the dependent variable (Y) and the BMT at point A (X) was the independent variable. The correlations between the BMT and BMW of the two breeds of meat pigeons are shown in Figure 3, Figure 4, Figure 5, Figure 6, Figure 7 and Figure 8 and Table 2 and Table 3. The correlation coefficient between the BMW and BMT at point A of Carneau pigeons was 0.907 (Table 2 and Table 3). The formula describing the relationship between the BMT and BMW at point A of Carneau pigeons was Y1 = 7.898X − 25.663. In Figure 3, X and Y1 are the BMT at point A and BMW of Carneau pigeons, respectively. The correlation coefficient between the BMW and BMT at point A of Silver King pigeons was 0.897. The formula describing the relationship between the BMT and BMW at point A of Silver King pigeons was Y2 = 7.991X − 32.357. In Figure 4, X and Y2 are the BMT at point A and the BMW of Silver King pigeons.

To further investigate the influence of sex on this method for BM measurement, the BMWs of Carneau and Silver King pigeons (Y3 and Y4, respectively) were used as dependent variables (Y), while the BMTs at point A of male and female Carneau and Silver King pigeons (X1, X2, X3, and X4, respectively) were used as independent variables for univariate linear regression. As shown in

Table 4. Correlation coefficients between BMT and BMW at point A of two breeds of meat pigeons.

Sample Group	Sample Size	Correlation Coefficient
Male Carneau pigeons	66	0.912
Female Carneau pigeons	37	0.896
Male Silver King pigeons	49	0.882
Female Silver King pigeons	54	0.911

, the correlation coefficients between the BMW and BMT at point A of male Carneau pigeons was 0.912. The formula describing the relationship between the BMT and BMW at point A of male Carneau pigeons was Y3 = 7.704X1 − 23.006. In Figure 5, X1 and Y3 are the BMT and BMW of male Carneau pigeons, respectively. The formula describing the relationship between the BMW and BMT at point A of male Carneau pigeons was Y3 = 8.058X2 − 27.700. In Figure 6, X2 and Y3 are the BMT and BMW at point A of female Carneau pigeons, respectively. The correlation coefficient between the BMW and BMT at point A of female Silver King pigeons was 0.911. The formula describing the relationship between the BMW and BMT at point A of female Silver King pigeons was Y4 = 7.766X3 − 28.864. In Figure 7, X3 and Y4 are the BMT and BMW at point A of male Silver King pigeons, respectively. The formula describing the relationship between the BMW and BMT at point A of male Silver King pigeons was Y4 = 8.235X4 − 36.059. In Figure 8, X4 and Y4 are the BMT and BMW at point A of female Silver King pigeons, respectively.

The results of the regression analysis showed that the BMT at point A (1 cm to the left of the top of the keel of the meat pigeon) had the highest correlation with BMW (r = 0.907 and 0.897; R² = 0.822 and 0.804, respectively). The equation indicated that the selected characteristics had significant effects on the variables and were not affected by variety with high reference values and linear fitting of the six characteristics of the two pigeon varieties. Also, the results showed that the method was not affected by pigeon sex. Therefore, this site could be selected as the point for UNDT of BMT to improve the BM of meat pigeons.

4. Discussion

Meat pigeons are mainly bred to provide BM. The percentage of BM of meat pigeons is about 18–30% [20]. Therefore, it is necessary to continuously select to improve BM in order to meet the requirements of the meat pigeon industry. Traditionally, the BMW of meat pigeons is measured after slaughter, and thus individual birds with excellent BM cannot be further used for breeding, resulting in a waste of resources and prolonging the breeding cycle. UNDT has the advantages of being non-invasive, real-time, and operationally simple, which makes it very applicable in examining fattening results or, in our research, determining the yield of the most valuable musculature in commercial breeding. However, while this study confirms the reliability and efficiency of UNDT, a critical analysis of its limitations is essential. For instance, operator dependency can significantly influence the consistency of measurements, emphasizing the need for rigorous training protocols. Additionally, environmental factors such as lighting and temperature may impact imaging quality, highlighting the importance of standardized measurement conditions. BMW is an important economic indicator that is affected by various traits, especially BMT at point A, BMT at point B, live BW, slaughter weight, breast width, and BD. Our results indicated that the correlation coefficients between BMT at Point A and BMW in Carneau pigeons and Silver King pigeons were 0.907 and 0.897, respectively. These values were significantly higher than those at Point B or for other body measurement traits. These higher correlation coefficients compared to earlier studies may be attributed to the larger sample size, refined measurement protocols, and precise selection of anatomical sites, such as point A. Further comparative studies with other poultry species and equipment types would provide deeper insights into these findings. This suggests that Point A, located 1 cm to the left side of the top of the keel of the meat pigeons, is a stable and representative measurement site. This finding aligns with previous studies, such as Yang et al., who found a significant correlation between breast muscle thickness and weight in Yangzhou geese, underscoring the importance of precise anatomical site selection [8]. Zhang et al. were the first to use UNDT to measure the BMT of meat pigeons in China and created a univariate linear model of weight, BMT, BD and BMW of meat pigeons with a correlation coefficient between the BMT and BMW of meat pigeons of 0.658 [6], which was lower than in the present study due to the small sample number. More than 100 samples were selected for this study, thus demonstrating stronger credibility.

This study further validated that this method is unaffected by sex and breed, enhancing its versatility. For instance, Zhu et al. used ultrasound to measure breast muscle thickness in Peking ducks and found a strong positive correlation with slaughter traits [7]. Scheuermann et al. used ultrasound to measure the muscle thickness of broilers and found that the correlation between BMW and carcass value was relatively high at 0.64–0.69 [21]. Oviedo-Rondón et al. developed a multiple linear regression model to estimate the carcass weight of broilers. The results showed that BMW and other carcass indicators could be estimated with high accuracy by ultrasound and that measurements of the right BM were consistently more accurate than the left [22]. Prior studies of UNDT for the measurement of BMT and correlation analysis with BMW reported notable differences in correlation coefficients [6,8,13,21,23]. Potential reasons for these disparities may be differences in poultry species, the number of samples used for correlation analysis, the accuracy of the B-ultrasound detector, and especially the measurement of BM at different sites. Therefore, it is particularly important to determine an accurate and easily identifiable site for UNDT. In the present study, BMT and BMW were significantly correlated at points A and B (p < 0.01). The correlation coefficients between the BMT and BMW at points A and B were 0.907 and 0.897, respectively. Therefore, points A and B should be selected as the most appropriate sites for UNDT of the BMT of meat pigeons.

UNDT is widely applied in the livestock and poultry industries in many countries [13,14,15,16] and commonly used in China for the breeding of Peking ducks [7,13,24] and, to a lesser extent, for chickens [9], geese [8], and meat pigeons [6]. A study conducted by Bochno et al. to assess the correlation between carcass yield and BW, BMT, and the sternal crest of Beijing ducks found that the correlation coefficient between BW and BMT measured by ultrasound was 0.82 [23]. A study by Hou et al. found that the correlation coefficients of the slaughter weight, live BW, and BMT of three species of Peking ducks was 0.50–0.62 [13]. Moreover, Xu et al. reported that the correlation coefficient between the BMT and BW of Peking ducks was 0.46 [24]. Kleczek et al. found a high correlation coefficient between the live BW and weight of the measured tissue components of Muscovy ducks of 0.70–0.86 [25]. Meanwhile, the results of the present study showed a significant positive correlation between the live BW, carcass weight, and BMT of Carneau and Silver King pigeons (p < 0.01); the correlation coefficients between breast muscle thickness and weight in meat pigeons (0.907 and 0.897) are significantly higher. Hence, UNDT is a useful method for the breeding of high-quality meat pigeons and to further develop the meat pigeon industry and market.

Despite its demonstrated value, the widespread adoption of ultrasonic measurement technology faces challenges. Equipment costs may restrict its use in small-scale farms, and measurement stability can be influenced by operator skill and environmental conditions [22]. Therefore, it is possible to develop automatic ultrasound image analysis system using artificial intelligence technology, reducing human-induced variability [9]; develop cost-effective and portable ultrasound devices to improve accessibility [15]; and combine UNDT with molecular genetics to further enhance breeding efficiency [21].

5. Conclusions

There was a significant positive correlation of BMT with the BMW, live BW, slaughter weight, and breast width of Carneau and Silver King pigeons. BMT was measured by UNDT at 1 cm on the left side of the top of the keel of the meat pigeon, which had a significant positive correlation with the BMW and was not affected by variety or sex. This site is suitable for UNDT measurement of the BMT of meat pigeons and to indirectly measure the BMW to reduce the cost of breeding meat pigeons. To enhance the application of UNDT in breeding programs, it is recommended to develop portable and cost-effective ultrasound devices, suitable for use on small-scale farms. Operator training programs should be standardized to reduce variability in measurements. Additionally, combining UNDT data with genetic selection methods could accelerate breeding efficiency by identifying superior breeding stock with greater precision. Further research should explore the integration of artificial intelligence for automated image analysis to enhance accuracy and consistency in measurement processes.