Developing the "Lesion-Detect" system that can identify and score gross lung lesions automatically during the slaughtering process

The importance of monitoring respiratory health in pigs is well known as respiratory problems are causing significant economic damage to the European and international pig industry every year. Currently, the Pig Health Monitoring Scheme (PHMS) is used internationally to evaluate the health status of pig sty after slaughter. Unfortunately, the PHMS system is expensive and labour intensive and the need for highly trained staff to undertake manual data collection. Automation using machine inspections via automated tools could be one solution. The PHMS more efficient and reliable tool for monitoring respiratory health at slaughterhouses. The aim of this study was to demonstrate the initial results of our “Lesion-Detect” system, which was developed with the aim of assisting the respiratory health of pigs at slaughter by providing real time scoring of human respiratory health.

This new system was designed based on the use of two Intel day cameras and a modern computing device (an MS Surface i300-case containing Intel i7 1165G7) and an eight core processor that has been installed using a Windows 10 2004 operating system. The recording system was custom made for the i-4. The lung experimental procedure was simple and involved having a researcher “spreading” the lung lobes, so the lung samples can be better viewed by the cameras. Images were taken, selected automatically and the lungs images were extracted from the background clutter. Various image analysis techniques have been used to ensure that only one lung can be seen in an image and only images with the full lung outline are accepted by the system. Finally manual scoring and the automated scoring were compared. It is hoped that by the end of 2025 a prototype Lesion-Detect system will be built and used to evaluate the respiratory health of pigs in real time in slaughterhouses.