Capstone Projects
MSU Affiliation
Data Science Academic Institute
Advisor
Dr. Will Davis
Editors
Dr. Will Davis
Abstract
This research focuses on identifying the key predictors of mastitis and ketosis in dairy cattle, two common and costly diseases affecting milk production and animal health. The primary objective was to identify significant risk factors and evaluate the predictive performance of modeling approaches.
To analyze these complex relationships, a longitudinal dataset following a large sample of dairy cattle is studying using several advanced methods. Survival analysis techniques, like Kaplan-Meier estimations, help measure how long cows remain free from health issues. The Cox Proportional Hazards (CPH) model complements this approach by identifying risk factors associated with health-related events over time. These statistical tools reveal patterns and differences in health outcomes based on various predictors. Additionally, machine learning techniques like gradient boosting are used to improve prediction accuracy and uncover subtle interactions among variables. The results from the CPH model are then validated using k-fold cross-validation and evaluated with the concordance index (C-index).
The findings from this research show influential variables associated with mastitis and ketosis risk. While all approaches performed with solid predictive ability, the gradient boosting models performed with marginal improvements in predicting time-to-occurrence in health events.
Overall, this research pinpoints critical risk factors that affect mastitis and ketosis. The goal is to help farmers improve prediction for early intervention. The approach combines statistical rigor with advanced technology, providing a comprehensive understanding of mastitis and ketosis risk factors.
DOI
https://doi.org/10.54718/BRCJ8049
Publication Date
2025
Temporal Coverage
2021-2024
Requires
Adobe Acrobat Reader
Keywords
dairy cattle, survival analysis, mastitis, ketosis
Disciplines
Dairy Science
Recommended Citation
Blocker, Samantha, "Investigating Health Events in Dairy Cattle" (2025). Capstone Projects. 4.
https://scholarsjunction.msstate.edu/data-capstone/4
This is a User Guide for Mastitis code.
Mastitis_Survival_Analysis_Code_User_Guide.ipynb (31 kB)
This is code for producers and researchers to run the Mastitis Survival Analysis.
Ketosis Survival Analysis User Guide.pdf (141 kB)
This is a User Guide for Ketosis code.
Ketosis_Survival_Analysis_Code_User_Guide.ipynb (33 kB)
This is code for producers and researchers to run the Ketosis Survival Analysis.
Code Used in Project.ipynb (1486 kB)
This is the code file for the code used specifically in this project. It runs with Python version 3.10.
