Related Experiment Video
Updated: Apr 15, 2026
08:51
Intraductal Injection of LPS as a Mouse Model of Mastitis: Signaling Visualized via an NF-κB Reporter Transgenic
Published on: September 4, 2012
20.1K
Thermal Image-Based Artificial Neural Network Approach to Determine Mastitis Detection in Holstein Dairy Cattle
Hasan Alp Şahin1, Edit Mikó2, Hasan Önder3
1Hemp Research Institute, Ondokuz Mayis University, 55139 Samsun, Türkiye.
Animals : an Open Access Journal From MDPI
|April 14, 2026
Summary
This study used artificial neural networks (ANN) and thermal imaging to detect mastitis in dairy cows. The AI model accurately identified mastitis levels, offering a reliable tool for farmers.
Area of Science:
- Veterinary Medicine
- Animal Science
- Biotechnology
Background:
- Mastitis significantly impacts dairy farming economics globally.
- Early and accurate detection is crucial for effective management.
Purpose of the Study:
- To develop and validate an Artificial Neural Network (ANN) model for mastitis detection.
- To utilize thermal imaging of cow udders during milking for disease identification.
Main Methods:
- Collected thermal images from 500 Holstein dairy cows during milking.
- Classified mastitis severity using California Mastitis Test (CMT) scores.
- Trained an ANN model using RGB thermal images and somatic cell count (SCC) data.
Main Results:
- The ANN model demonstrated high accuracy with correlation coefficients (R) of 0.91 (training), 0.97 (validation), and 0.97 (test).
- Strong agreement between validation and test results confirmed the model's generalization capability and lack of overfitting.
Conclusions:
- Artificial neural networks combined with thermal imaging offer a reliable and high-quality method for mastitis detection in dairy cows.
- This approach can aid farmers in managing mastitis, reducing economic losses.
