Web-Based Application Predicting Cow Hoof Infections with 3D Modeling Scrapping Simulator and Treatment Recommendation

Veterinary students often face challenges in assessing cow's hoof health due to limited access to live animals and insufficient practical materials, which hinder hands-on experience and impact treatment outcomes. To address these gaps, this study developed WounderCow, a web-based application that integrates image processing, 3D modeling, and AI-driven treatment recommendations. The system allows users to predict cow hoof infections, classify lameness types, assess severity, and receive appropriate treatment recommendations. The development process followed the Agile Scrum framework, with iterative stages focusing on refining key components like image processing, 3D modeling, and AI-driven predictions. After system development, a survey was administered to 53 participants-veterinary students, farmers, and IT professionals-using a four-point Likert scale questionnaire. The results were analyzed to evaluate how well the system met the needs of users. To assess the system's accuracy, a confusion matrix was used to measure precision, recall, and overall accuracy in predicting lameness type, severity, and treatment recommendations. The system was also evaluated against ISO/IEC 25010 standards, ensuring that it met criteria for usability, performance, and reliability. The survey results showed significant satisfaction, with weighted mean scores ranging from 3.29 to 3.45, and 61.9% of respondents strongly agreed that the system improved their ability to assess and treat hoof health issues. The application enables early lameness detection and includes a 3D simulator with step-by-step guidance for hoof scraping and treatment. This study highlights the application's potential to bridge educational gaps while advancing diagnostic and treatment practices for cow hoof conditions.