Several models combined with ultrasound techniques to predict breast muscle weight in broilers

Li, Zhengda; Zheng, Jumei; An, Bingxing; Ma, Xiaochun; Fan, Ying; Kong, Fuli; Wen, Jie; Zhao, Guiping

doi:10.1016/j.psj.2023.102911

Cited by 2 publications

(7 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Section: Overall Processmentioning

confidence: 99%

“…In addition, live weight is included as one of the features in the breast muscle weight prediction model. The reasons for this are listed as follows: (1) it is frequently employed in research on prediction methods for broiler breast muscle weight [7,8,15]; (2) it is highly correlated with breast muscle weight in poultry [37]; and (3) it is relatively easy to obtain.…”

Section: Breast Muscle Weight Prediction Modelmentioning

confidence: 99%

“…Therefore, breeding broiler chickens with a high yield of breast muscle holds significant economic value, which requires accurate measurement of the breast muscle weight [4][5][6]. The conventional approach to broiler breeding infers the carcass performance of an individual to be cultivated by using the slaughter value from individuals within the same family [7,8]. However, such methods may consume a large amount of time, compromise animal welfare, and incur high economic costs [9].…”

Section: Introductionmentioning

confidence: 99%

“…However, this manual method for determining the area is time-consuming and highly depends on the operator. Two recent studies on the breast muscle weight of broilers employed external body traits and breast muscle thickness determined by ultrasound images to develop machine learning models for breast muscle weight [7,8]. Nevertheless, the process of ultrasonic measurement is tedious, including feather separation, gel application, and probe positioning, which is not suitable for the automation of data collection.…”

Section: Introductionmentioning

confidence: 99%

“…Next, 15 related shape features were extracted from the pectoral muscle mask images obtained with MAEFNet. Meanwhile, inspired by the previous studies on breast muscle weight [7,8], live weight was also included as one of the features for breast muscle weight prediction. To further enhance the model's accuracy and reliability, the recursive feature elimination (RFE) method was employed to screen these features.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

In Vivo Prediction of Breast Muscle Weight in Broiler Chickens Using X-ray Images Based on Deep Learning and Machine Learning

Zhu,

Li,

Yang

et al. 2024

Animals

View full text Add to dashboard Cite

Accurately estimating the breast muscle weight of broilers is important for poultry production. However, existing related methods are plagued by cumbersome processes and limited automation. To address these issues, this study proposed an efficient method for predicting the breast muscle weight of broilers. First, because existing deep learning models struggle to strike a balance between accuracy and memory consumption, this study designed a multistage attention enhancement fusion segmentation network (MAEFNet) to automatically acquire pectoral muscle mask images from X-ray images. MAEFNet employs the pruned MobileNetV3 as the encoder to efficiently capture features and adopts a novel decoder to enhance and fuse the effective features at various stages. Next, the selected shape features were automatically extracted from the mask images. Finally, these features, including live weight, were input to the SVR (Support Vector Regression) model to predict breast muscle weight. MAEFNet achieved the highest intersection over union (96.35%) with the lowest parameter count (1.51 M) compared to the other segmentation models. The SVR model performed best (R2 = 0.8810) compared to the other prediction models in the five-fold cross-validation. The research findings can be applied to broiler production and breeding, reducing measurement costs, and enhancing breeding efficiency.

show abstract

Section: Overall Processmentioning

confidence: 99%