Comparing Visual-Only and Visual-Palpation Post-Mortem Lung Scoring Systems in Slaughtering Pigs

An Integrated Analysis of Abattoir Lung Lesion Scores and Antimicrobial Use in Italian Heavy Pig Finishing Farms

Respiratory diseases significantly affect intensive pig finishing farms, causing production losses and increased antimicrobial use (AMU). Lesion scoring at slaughter has been recognized as a beneficial practice to evaluate herd management. The integrated analysis of abattoir lesion scores and AMU data could improve decision-making by providing feedback to veterinarians and farmers on the effectiveness of antimicrobial treatments, thus rationalizing their use. This study compared lung and pleural lesion scores collected at Italian pig slaughterhouses with on-farm AMU, estimated through a treatment index per 100 days (TI100). Overall, 24,752 pig carcasses, belonging to 236 batches from 113 finishing farms, were inspected. Bronchopneumonia and chronic pleuritis were detected in 55% and 48% of the examined pigs, respectively. Antimicrobials were administered in 97% of the farms during the six months prior to slaughter (median TI100 = 5.2), notwithstanding compliance with the mandatory withdrawal period. EMA category B (critical) antimicrobials were administered in 15.2% of cases (median TI100 = 0.06). The lung score was not associated with the total AMU, but significant, positive associations were found with the past use of critical antimicrobials (p = 0.041) and macrolides (p = 0.044). This result highlights the potential of abattoir lung lesion monitoring to rationalize antimicrobial stewardship efforts, contributing to AMU reduction.

Feasibility of Near-Infrared Spectroscopy in the Classification of Pig Lung Lesions

Respiratory diseases significantly affect intensive pig farming, causing production losses and increased antimicrobial use. Accurate classification of lung lesions is crucial for effective diagnostics and disease management. The integration of non-destructive and rapid techniques would be beneficial to enhance overall efficiency in addressing these challenges. This study investigates the potential of near-infrared (NIR) spectroscopy in classifying pig lung tissues. The NIR spectra (908-1676 nm) of 101 lungs from weaned pigs were analyzed using a portable instrument and subjected to multivariate analysis. Two distinct discriminant models were developed to differentiate normal (N), congested (C), and pathological (P) lung tissues, as well as catarrhal bronchopneumonia (CBP), fibrinous pleuropneumonia (FPP), and interstitial pneumonia (IP) patterns. Overall, the model tailored for discriminating among pathological lesions demonstrated superior classification performances. Major challenges arose in categorizing C lungs, which exhibited a misclassification rate of 30% with N and P tissues, and FPP samples, with 30% incorrectly recognized as CBP samples. Conversely, IP and CBP lungs were all identified with accuracy, precision, and sensitivity higher than 90%. In conclusion, this study provides a promising proof of concept for using NIR spectroscopy to recognize and categorize pig lungs with different pathological lesions, offering prospects for efficient diagnostic strategies.

Scoring Enzootic Pneumonia-like Lesions in Slaughtered Pigs: Traditional vs. Artificial-Intelligence-Based Methods

Artificial-intelligence-based methods are regularly used in the biomedical sciences, mainly in the field of diagnostic imaging. Recently, convolutional neural networks have been trained to score pleurisy and pneumonia in slaughtered pigs. The aim of this study is to further evaluate the performance of a convolutional neural network when compared with the gold standard (i.e., scores provided by a skilled operator along the slaughter chain through visual inspection and palpation). In total, 441 lungs (180 healthy and 261 diseased) are included in this study. Each lung was scored according to traditional methods, which represent the gold standard (Madec's and Christensen's grids). Moreover, the same lungs were photographed and thereafter scored by a trained convolutional neural network. Overall, the results reveal that the convolutional neural network is very specific (95.55%) and quite sensitive (85.05%), showing a rather high correlation when compared with the scores provided by a skilled veterinarian (Spearman's coefficient = 0.831, p < 0.01). In summary, this study suggests that convolutional neural networks could be effectively used at slaughterhouses and stimulates further investigation in this field of research.