Classification of FAMACHA© Scores with Support Vector Machine Algorithm from Body Condition Score and Hematological...

Area of Science:

• Veterinary Epidemiology
• Computational Biology
• Parasitology

Background:

• The FAMACHA© scoring system is crucial for managing parasitic infections in animals.
• Accurate classification of FAMACHA© scores aids in early diagnosis and treatment.
• Evaluating computational models for FAMACHA© score estimation is essential for improving diagnostic accuracy.

Purpose of the Study:

• To assess the performance of Support Vector Machines (SVMs) in classifying FAMACHA© scores.
• To evaluate the accuracy and reliability of SVMs for estimating parasitic load via FAMACHA© scoring.
• To identify areas for improvement in SVM model performance for FAMACHA© score classification.

Main Methods:

• Utilized Support Vector Machines (SVMs) for the classification of FAMACHA© scores.
• Analyzed model performance using metrics including sensitivity, specificity, and predictive values.
• Investigated SVM accuracy across different FAMACHA© score classes (1, 2, and 3).

Main Results:

• SVM model demonstrated high sensitivity and specificity for FAMACHA© classes 1 and 3.
• Model performance was relatively lower for FAMACHA© class 2.
• Achieved a high overall accuracy rate of 97.26% and a kappa value of 0.9588.

Conclusions:

• Support Vector Machines (SVMs) are a reliable tool for FAMACHA© score estimation in veterinary epidemiology.
• The SVM model shows significant potential for improving the diagnosis and management of parasitic infections.
• Further refinement of SVM models is needed to enhance classification accuracy for all FAMACHA© score classes, particularly class 2.