Predicting Equine Health Outcomes Using Machine Learning Models Trained on Clinical Indicators and Limited Behavioral Data
Bibliographic record
Abstract
The capacity to perceive and anticipate the health status of horses is a critical aspect of equine veterinary care. Recent studies have shown that machine learning algorithms can accurately diagnose and classify animal diseases based on physiological signs. Using properties like heart rate, temperature, and other clinical parameters, the study offers a classification model developed from a publicly available dataset of more than 2,000 equine health records to predict health outcomes. Among the algorithms tested, including k-Nearest Neighbors (KNN), Decision Tree, and Light Gradient Boosting Machine (LightGBM), LightGBM achieved the highest validation accuracy at approximately 76%. Exploratory data analysis was conducted to visualize feature distributions and identify correlations, followed by preprocessing steps such as handling missing values and encoding categorical variables. The model was trained using five-fold cross-validation and fine-tuned for optimal performance. Among the factors contributing to the success of LightGBM were its ability to handle categorical features and its leaf-wise tree growth strategy, which improved learning efficiency on a moderately sized dataset. In addition to structured data, limited behavioral descriptors were incorporated using a language model to provide additional context regarding stress and discomfort. While these features had a smaller role, they created new opportunities for interpreting subtle health cues not included in clinical data alone. This study demonstrates the potential for predictive modeling to assist veterinarians in early diagnosis and treatment planning. Future work may focus on expanding the dataset and implementing more detailed behavioral and physiological data to improve model generalization.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.010 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.000 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.002 | 0.008 |
| Research integrity | 0.001 | 0.005 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".