Using Artificial Intelligence to Predict Survivability Likelihood and Need for Surgery in Horses Presented With Acute Abdomen (Colic).

This research article investigates how artificial intelligence (AI) and machine learning tools can assess the need for surgery and predict survival rates in horses afflicted with severe abdominal pain, commonly known as colic.

Research Purpose and Rationale

• The authors have set out to demonstrate the potential of AI and machine learning in veterinary medicine, an area where the use of such technology is relatively underdeveloped.
• They specifically focus on testing these tools on assessing the need for surgical treatment and predicting survival rates among horses presented with acute abdomen, better known as colic.

Approach and Methodology

• The machine learning algorithms put to use in this study are Decision Trees, Multilayer Perceptron, Bayes Network, and Naïve Bayes.
• The system was trained on available clinical data, such as the patient’s history, clinical examination findings, and diagnostic procedures.

Key Findings

• The machine learning algorithms predicted with 76% accuracy whether surgery would be required for the horses affected with colic.
• In terms of survival likelihood, the algorithms were slightly more accurate, predicting with 85% accuracy whether the horses would survive.

Implications

• The research carries significant implications for the veterinary medicine field, showcasing the potential for AI and machine learning to improve outcomes and decision-making in clinical scenarios.
• The authors advocate for further investigation and testing to expand the application of this technology in various clinical fields of veterinary medicine.