Early detection of swine diseases is crucial to minimizing health risks and economic losses in pig farming. To address this need, the proponents developed Swine Scan, an AI-driven health monitoring system specifically designed for fattener white pigs in medium to large-scale farms. The system utilizes computer vision and neural network technologies to detect and classify pig behaviors associated with early signs of illness. It employs the YOLOv8 algorithm for pig detection, DeepSORT for real-time tracking, and a Self-Organizing Map (SOM) neural network for behavior classification based on patterns of activity and environmental sensor data. Behavior and environmental data such as duration of activity, temperature, humidity, and luminance are logged and batch-evaluated every four hours.