Related Experiment Video
Updated: Dec 25, 2025
09:21
Author Spotlight: Generating Neuronal Phenotypic Profiles - A Protocol to Culture and Image Human Midbrain Dopaminergic Neurons
Published on: July 7, 2023
1.9K
PigLeg: prediction of swine phenotype using machine learning
Siroj Bakoev1, Lyubov Getmantseva1, Maria Kolosova2
1L.K. Ernst Federal Science Center for Animal Husbandry, Moscow, Russia.
Peerj
|April 2, 2020
Summary
Machine learning accurately predicts pig leg weakness using early growth and meat data. Random Forest and K-Nearest Neighbors algorithms identified key predictors like muscle thickness for improved animal welfare and reduced economic loss.
Area of Science:
- Animal Science
- Agricultural Technology
- Machine Learning Applications
Background:
- Industrial pig farming faces challenges with leg weakness and lameness, causing significant economic losses.
- Identifying predictors of limb condition is crucial for improving pig welfare and farm profitability.
Purpose of the Study:
- To assess pig limb condition using growth and meat characteristic indicators.
- To evaluate the predictive accuracy of nine machine learning (ML) classification algorithms for pig leg weakness.
Main Methods:
- Compared nine ML algorithms: Random Forest, K-Nearest Neighbors, Artificial Neural Networks, C50Tree, Support Vector Machines, Naive Bayes, Generalized Linear Models, Boost, and Linear Discriminant Analysis.
- Utilized measurements of Muscle Thickness, Back Fat, and Average Daily Gain as predictors.
- Identified best-performing algorithms for early-stage prediction.
Main Results:
- Random Forest and K-Nearest Neighbors demonstrated the highest accuracy in predicting pig leg weakness.
- Muscle Thickness, Back Fat amount, and Average Daily Gain were identified as significant predictors of limb conformation.
- The study highlights the effectiveness of simple, early-stage measurements for prediction.
Conclusions:
- Machine learning offers a practical and effective approach to assessing pig limb health.
- Early identification of leg weakness predictors can mitigate economic losses and enhance animal welfare in pig farming.
