close
close

Computers can very well carry out the evaluation of pork loungers

Cada evaluation in the slaughter line can inform the industry about pathogens and problems in the pig house. This image was not taken in the context of the research described in the article. Photo: Irma Regien

The gate of pigs in a slaughter line is an interesting field of research. In this way it is easy to determine whether there were lung problems during the animal's life or not. The evaluation process can be credibly facilitated by the use of computer technology.

The development was described in a recently published publication in the publication examined by experts Veterinary research. A team of researchers from institutes in the United States and Spain worked together on the project, supported by a Hipra team. The team tested a computer vision system called AI Diagnos and wanted to know whether the results were comparable to the human eye.

Lung lesion connected to M. Hyo Infections

In order to determine this, the team focused on a lung lesion called “Craniovovral Lung consolidation” (CVPC). The lesion is often associated with infection of the bacterium Mycoplasm Hyopneumonia ((M. Hyo) Infection. The team also evaluated differences in the interpretations between the different evaluations – and these were compared with variability that were found within the results of the computer vision system.

The team and the system analyzed 1,050 dorsal views of pigs. The total value of the lung lesion, the lesion value per lung lobe and the percentage of the affected lung surface were used as results for the evaluation.

The computer system showed a “moderate accuracy” (62-71%) in the distinction between non-clinated and lesioned lung lobes in all except for the diaphragm. The team observed a “low classification accuracy with a low multi-class” (24-36%) at the level of the lung lobes.

A good alternative to recognizing lesions

As a conclusion of the study, the researchers wrote that the computer vision system used between Pneumonian and non-pneumonic lung lobes was discriminated against with “moderate accuracy”, which indicates that the system could be a good alternative to recognizing CVPC during the slaughter inspections compared to the human reviews.

The prototype is further refined in order to correctly predict the degree of pneumonia, the researchers wrote. “Overall, the computer vision system evaluated in this study has the potential to automatically evaluate M. Hyo-Feded lung lesions and facilitating the processing of data to solve problems that are connected to the conventional CVPC evaluation when slaughtering. ”

Publisher of the pig progress / topic: pigs all over the world
Zipline already delivers all types of medical services with drones, including blood. - Photo: Zipline
June up up up: What is new in the world of pigs?

Leave a Comment