Horse surgery and survival prediction with artificial intelligence models: performance comparison of original, imputed, balanced, and feature- selected datasets
Artificial intelligence (AI) technology, while less advanced than in human medicine, holds significant potential in the field of veterinary medicine. This technology offers a range of essential benefits, such as disease diagnosis, treatment planning, disease control, and overall animal health improv...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Kafkas University, Faculty of Veterinary Medicine
2024-01-01
|
| Series: | Kafkas Universitesi Veteriner Fakültesi Dergisi |
| Subjects: | |
| Online Access: | https://vetdergikafkas.org/pdf.php?id=3076 |
| _version_ | 1797319689095872512 |
|---|---|
| author | Pınar CİHAN |
| author_facet | Pınar CİHAN |
| author_sort | Pınar CİHAN |
| collection | DOAJ |
| description | Artificial intelligence (AI) technology, while less advanced than in human medicine,
holds significant potential in the field of veterinary medicine. This technology offers
a range of essential benefits, such as disease diagnosis, treatment planning, disease
control, and overall animal health improvement. Based on clinical data, this study
uses 15 AI models to predict the necessity of surgery and the likelihood of survival in
horses displaying symptoms of acute abdominal pain (colic). By comparing surgical and
survival predictions across the original, imputed missing values, and balanced datasets,
we determine the most effective dataset based on the average accuracy of the 15 AI
models. Furthermore, we explore the potential for improved accuracy with a reduced
feature set by calculating feature importance scores for surgery and survival predictions.
Our results indicate that the balanced dataset achieved the highest average accuracy for
predicting surgery and survival, with 80.76% and 77.96%, respectively. The Random
Forest (RF) model outperformed others as the most accurate model for both surgery
(accuracy = 85.83, Area Under the Curve [AUC] = 0.906) and survival prediction
(accuracy = 80.75, AUC = 0.888). It was observed that reducing the number of features
in the dataset by 56% led to an increase in surgery prediction accuracy to 86.38%.
Similarly, when the number of features was reduced by 24% for survival prediction, the
prediction performance increased to 83.75%. This study emphasizes the importance of
the precise implementation of artificial intelligence techniques in veterinary medicine,
which can significantly enhance model performance. |
| first_indexed | 2024-03-08T04:10:34Z |
| format | Article |
| id | doaj.art-7a51ea8435484302907572ff3377c6bf |
| institution | Directory Open Access Journal |
| issn | 1309-2251 |
| language | English |
| last_indexed | 2024-03-08T04:10:34Z |
| publishDate | 2024-01-01 |
| publisher | Kafkas University, Faculty of Veterinary Medicine |
| record_format | Article |
| series | Kafkas Universitesi Veteriner Fakültesi Dergisi |
| spelling | doaj.art-7a51ea8435484302907572ff3377c6bf2024-02-08T20:14:21ZengKafkas University, Faculty of Veterinary MedicineKafkas Universitesi Veteriner Fakültesi Dergisi1309-22512024-01-0130223324110.9775/kvfd.2023.309083076Horse surgery and survival prediction with artificial intelligence models: performance comparison of original, imputed, balanced, and feature- selected datasetsPınar CİHAN0Tekirdağ Namık Kemal University, Faculty of Corlu Engineering, Department of Computer Engineering, TR-59860 Tekirdağ - TÜRKİYEArtificial intelligence (AI) technology, while less advanced than in human medicine, holds significant potential in the field of veterinary medicine. This technology offers a range of essential benefits, such as disease diagnosis, treatment planning, disease control, and overall animal health improvement. Based on clinical data, this study uses 15 AI models to predict the necessity of surgery and the likelihood of survival in horses displaying symptoms of acute abdominal pain (colic). By comparing surgical and survival predictions across the original, imputed missing values, and balanced datasets, we determine the most effective dataset based on the average accuracy of the 15 AI models. Furthermore, we explore the potential for improved accuracy with a reduced feature set by calculating feature importance scores for surgery and survival predictions. Our results indicate that the balanced dataset achieved the highest average accuracy for predicting surgery and survival, with 80.76% and 77.96%, respectively. The Random Forest (RF) model outperformed others as the most accurate model for both surgery (accuracy = 85.83, Area Under the Curve [AUC] = 0.906) and survival prediction (accuracy = 80.75, AUC = 0.888). It was observed that reducing the number of features in the dataset by 56% led to an increase in surgery prediction accuracy to 86.38%. Similarly, when the number of features was reduced by 24% for survival prediction, the prediction performance increased to 83.75%. This study emphasizes the importance of the precise implementation of artificial intelligence techniques in veterinary medicine, which can significantly enhance model performance.https://vetdergikafkas.org/pdf.php?id=3076artificial intelligencedata balancingfeature selectionhorse colicpredictionsmote |
| spellingShingle | Pınar CİHAN Horse surgery and survival prediction with artificial intelligence models: performance comparison of original, imputed, balanced, and feature- selected datasets Kafkas Universitesi Veteriner Fakültesi Dergisi artificial intelligence data balancing feature selection horse colic prediction smote |
| title | Horse surgery and survival prediction with artificial intelligence models: performance comparison of original, imputed, balanced, and feature- selected datasets |
| title_full | Horse surgery and survival prediction with artificial intelligence models: performance comparison of original, imputed, balanced, and feature- selected datasets |
| title_fullStr | Horse surgery and survival prediction with artificial intelligence models: performance comparison of original, imputed, balanced, and feature- selected datasets |
| title_full_unstemmed | Horse surgery and survival prediction with artificial intelligence models: performance comparison of original, imputed, balanced, and feature- selected datasets |
| title_short | Horse surgery and survival prediction with artificial intelligence models: performance comparison of original, imputed, balanced, and feature- selected datasets |
| title_sort | horse surgery and survival prediction with artificial intelligence models performance comparison of original imputed balanced and feature selected datasets |
| topic | artificial intelligence data balancing feature selection horse colic prediction smote |
| url | https://vetdergikafkas.org/pdf.php?id=3076 |
| work_keys_str_mv | AT pınarcihan horsesurgeryandsurvivalpredictionwithartificialintelligencemodelsperformancecomparisonoforiginalimputedbalancedandfeatureselecteddatasets |