Preventive veterinary medicine最新文献

{"title":"A scoping review and taxonomy of epidemiological-macroeconomic models for livestock diseases","authors":"Nathan J. Doyle ,&nbsp;Gabrielle Bonnet ,&nbsp;Matt J. Keeling ,&nbsp;Michael J. Tildesley ,&nbsp;W. John Edmunds","doi":"10.1016/j.prevetmed.2025.106694","DOIUrl":"10.1016/j.prevetmed.2025.106694","url":null,"abstract":"<div><h3>Background and objectives</h3><div>Livestock disease outbreaks can generate significant macroeconomic harms, though few studies have combined epidemiological and macroeconomic analysis in a joint modelling framework. We reviewed existing approaches to epidemiological-macroeconomic modelling for livestock diseases, to identify challenges and opportunities for further model development.</div></div><div><h3>Methods</h3><div>We systematically searched electronic databases for journal articles, preprints, working papers and grey literature. We assessed model methodologies, scope, and application to empirical data. We analysed results descriptively and provided a critical appraisal of the strength, limitations and use cases of existing methods.</div></div><div><h3>Results</h3><div>We found 17 articles modelling the animal health and macroeconomic outcomes of livestock disease outbreaks. Models used outputs from epidemiological model simulations linked to a macroeconomic model, with no feedback from the economy to epidemiology. Macroeconomic modelling approaches consisted of the use of computable general equilibrium models, input output models and social accounting models. The literature was heavily dominated by considerations for hypothetical foot-and-mouth disease incursions in high-income settings.</div></div><div><h3>Conclusion</h3><div>Priority for future research includes efforts to model a wider array of livestock diseases in endemic and low-income settings, further consideration for socioeconomic impacts and the need for improved data.</div></div>","PeriodicalId":20413,"journal":{"name":"Preventive veterinary medicine","volume":"245 ","pages":"Article 106694"},"PeriodicalIF":2.4,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145118069","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Modelling PRRS transmission between pig herds in Denmark and prediction of interventions impact","authors":"You Chang ,&nbsp;Ana Rita Pinheiro Marques ,&nbsp;Mette Fertner ,&nbsp;Nils Toft ,&nbsp;Bjørn Lorenzen ,&nbsp;Mossa Merhi Reimert ,&nbsp;Hans Houe ,&nbsp;Beate Conrady","doi":"10.1016/j.prevetmed.2025.106692","DOIUrl":"10.1016/j.prevetmed.2025.106692","url":null,"abstract":"<div><div>Porcine reproductive and respiratory syndrome (PRRS) is an endemic viral disease in most pig-producing countries, including Denmark. In 2022, Denmark launched a control program to reduce PRRS prevalence, with legislative changes in 2023 making testing and status reporting mandatory. The program also enforces the loss of PRRS-free status for farms that purchase pigs from non-PRRS-free sources and implements region-specific control measures to coordinate PRRS elimination within herds. However, the effectiveness of these interventions remains uncertain and requires thorough evaluation through transmission modelling and analysis of data before and after legislation changes.</div><div>To understand PRRS transmission prior to legislative changes in 2023 and predict the impact of control measures, we developed a between-herd stochastic compartmental model. This model includes compartments for susceptible (<em>S</em>), highly infectious <span><math><msub><mrow><mo>(</mo><mi>I</mi></mrow><mrow><mi>h</mi></mrow></msub></math></span>), lowly infectious <span><math><msub><mrow><mo>(</mo><mi>I</mi></mrow><mrow><mi>l</mi></mrow></msub></math></span>) and detected (<em>D</em>) pig herds, using data from 2020 to 2021. The model (i) quantifies the relative contributions of pig movements and local transmission to the spread of PRRS; (ii) generate herd-level maps of the basic reproduction (<span><math><msub><mrow><mi>R</mi></mrow><mrow><mn>0</mn></mrow></msub></math></span>); and (iii) assess the effectiveness of targeted interventions for eradicating PRRS in Denmark.</div><div>Model results indicated that more than 50 % of herds had an <span><math><msub><mrow><mi>R</mi></mrow><mrow><mn>0</mn></mrow></msub></math></span> greater than 1, suggesting a potential for sustained transmission if no interventions had been implemented after 2022. Both local spread and movement-mediated transmission play important roles, but local transmission drives the spatial heterogeneity in observed PRRS prevalence across Denmark. Although only 17 % of infectious herds remain undetected under current surveillance, they are responsible for 60 % of total transmission. Local control via depopulation and repopulation, is the fastest measure to reduce the observed prevalence of PRRS, but it has a lower effect on true transmission due to the hidden infections. Therefore, achieving eradication may require a combination of more frequent testing, targeted within-herd PRRS elimination and stricter risk-based trading. This study identifies PRRS hotspots and transmission routes, offering evidence-based recommendations for control.</div></div>","PeriodicalId":20413,"journal":{"name":"Preventive veterinary medicine","volume":"245 ","pages":"Article 106692"},"PeriodicalIF":2.4,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145105553","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving understanding of household level biosecurity challenges for smallholder pig producers in Vietnam: A pilot for data collection and capacity building opportunities","authors":"Jennifer Manyweathers ,&nbsp;Ninh Thi Huyen ,&nbsp;Ngo Thi Kim Cuc ,&nbsp;Gang Xie ,&nbsp;Nguyen Cong Dinh ,&nbsp;Tran Trung Thong ,&nbsp;Le Thu Ha ,&nbsp;Nguyen Pham Trung Nguyen ,&nbsp;Ngo Thi Le Quyen ,&nbsp;Lynne Hayes ,&nbsp;Marta Hernandez-Jover ,&nbsp;Yin Li ,&nbsp;Jennifer Kelly ,&nbsp;Jen Bond","doi":"10.1016/j.prevetmed.2025.106691","DOIUrl":"10.1016/j.prevetmed.2025.106691","url":null,"abstract":"<div><div>Understanding smallholder pig producers’ biosecurity practices in Vietnam has been identified as a priority to better support the pork industry’s management of African swine fever. A cross-sectional survey was undertaken in three provinces, Bac Giang, Ha Nam, and Hoa Binh, in 2022/23, with the aim of piloting data collection approaches and gaining a better understanding of the lived experience of smallholders affected by African swine fever. A total of 160 smallholders, with less than 100 pigs, responded to the questionnaire. Bayesian Network modelling was used to analyse the data, using a vulnerability framework (likelihood of exposure and response capacity) to segment the respondent population based on the potential susceptibility to an African swine fever outbreak. Results suggest that the diversity within the population, across provinces, number of pigs kept, and percentage of income from pigs should drive the development of tailored interventions. Recommendations developed from the analysis include focussing on support to reduce swill feeding in smaller herds (under 20 pigs), restricting animal access to production areas in medium sized farms (21–49 pigs)and examining barriers to quarantine of both new and sick animals in larger farms (50–100 pigs). The findings from this pilot could be useful in the development of a vulnerability and resilience assessment tool to identify gaps and innovation opportunities to support smallholder pig producer biosecurity capabilities both in Vietnam and other countries affected by or preparing for African swine fever.</div></div>","PeriodicalId":20413,"journal":{"name":"Preventive veterinary medicine","volume":"245 ","pages":"Article 106691"},"PeriodicalIF":2.4,"publicationDate":"2025-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145081532","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Global antibiotic resistance trends in Enterococcus faecalis from animals, food, and environmental sources: A meta-analysis","authors":"Yongwu Yang ,&nbsp;Reihaneh Yeganeh ,&nbsp;Mahya Abedi Moghadam ,&nbsp;Zahra Teymouri ,&nbsp;Faezeh Motallebi Tabaei ,&nbsp;Samaneh Moradi ,&nbsp;Masoumeh Beig","doi":"10.1016/j.prevetmed.2025.106689","DOIUrl":"10.1016/j.prevetmed.2025.106689","url":null,"abstract":"<div><div>This systematic review investigates the prevalence of antibacterial resistance in <em>Enterococcus faecalis</em> outside clinical settings, focusing on animals, food, and the environment. Using a \"One Health\" approach connecting human, animal, and environmental health, we analyze resistance patterns across diverse geographical regions and timeframes, emphasizing the impact of geographical factors. We thoroughly searched various databases, including Scopus, PubMed, Web of Science, and EMBASE. We assessed study quality using the Newcastle-Ottawa Scale Checklist. A random-effects model estimated resistance proportions, while meta-regression, subgroup analyses, and identification of outliers and influential studies provided additional insights. We employed R software (version 4.2.1) and the metaphor package (version 3.8.1) for comprehensive data management and analysis. Our analysis of 75 studies revealed significant variability in antibiotic resistance across countries, continents, and World Health Organization (WHO) regions in non-clinical <em>E. faecalis</em>. Resistance rates differed based on antibacterial agents, geographic regions, periods, and isolation sources. Resistance was highest to tetracycline, rifampicin, quinupristin-dalfopristin, doxycycline, and erythromycin (≈48–62 %). Moderate resistance was observed for streptomycin, kanamycin, ciprofloxacin, gentamicin, chloramphenicol, and linezolid (≈10–26 %). In contrast, low resistance rates (&lt;10 %) were found for vancomycin, teicoplanin, levofloxacin, nitrofurantoin, and fosfomycin. Considerable heterogeneity was noted across most antibiotics. This review emphasizes the influence of geographic variations and isolated sources on <em>E. faecalis</em> resistance patterns. We underscore the urgent need for a collaborative One Health approach, integrating human, animal, and environmental health sectors, to tackle antibacterial resistance in <em>E. faecalis</em> effectively.</div></div>","PeriodicalId":20413,"journal":{"name":"Preventive veterinary medicine","volume":"245 ","pages":"Article 106689"},"PeriodicalIF":2.4,"publicationDate":"2025-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145102972","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Risk analysis of infectious disease in pigs in Gifu prefecture, Japan, through network analysis","authors":"Naotoshi Kuninaga ,&nbsp;Emi Yoshita ,&nbsp;Miki Koyasu ,&nbsp;Hibiki Morimoto ,&nbsp;Noboru Hayashi ,&nbsp;Takahiro Hiono ,&nbsp;Satoshi Ito ,&nbsp;Yoshihiro Sakoda ,&nbsp;Norikazu Isoda","doi":"10.1016/j.prevetmed.2025.106687","DOIUrl":"10.1016/j.prevetmed.2025.106687","url":null,"abstract":"<div><div>Movements of livestock, humans, vehicles and wildlife are recognized as critical contributors to the spread of infectious diseases in livestock and can be modeled as a network to assess and predict disease transmission. This study developed a comprehensive multilayer network incorporating the movements of pigs, humans, vehicles, and, presumably, wild boars to estimate the risks of disease introduction and transmission for each husbandry stakeholder and to identify key clusters and modes of movement involved. A questionnaire-based study was conducted across 22 pig farms in Gifu Prefecture, Japan, collecting data on pig, human, and vehicle movements to establish networks. The wild boar movement network was estimated using data on pig farm locations and wild boar habitats collected from vegetation cover data. Movement-associated effects in each network based on movement frequency were assigned to combinations of the four networks, resulting in a four-layered network. The network exhibited small-world characteristics and was clustered into four groups. Disease containment schemes in livestock are commonly established along administrative boundaries, however these four epidemiological clusters, comprising 31, 28, 24, and 22 nodes, did not align exactly with administrative districts, suggesting the significance of managing livestock infectious diseases beyond governmental borders. In the Partial Least Squares Regression (PLSR) analysis, pig, vehicle, and wild boar movement made comparable positive contributions to PageRank-based node importance within the multilayer network. This study highlights the significance of epidemiological links among husbandry and nonhusbandry stakeholders, emphasizing the need to develop effective risk management tools considering the probable disease transmission pathways.</div></div>","PeriodicalId":20413,"journal":{"name":"Preventive veterinary medicine","volume":"245 ","pages":"Article 106687"},"PeriodicalIF":2.4,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145081477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Associations between animal health indicators, antimicrobial use, and antimicrobial resistance on Swiss dairy farms","authors":"Adrian Minnig ,&nbsp;Véronique Bernier Gosselin ,&nbsp;Silja Griss ,&nbsp;Gertraud Schüpbach-Regula ,&nbsp;Mireille Meylan ,&nbsp;Guy-Alain Schnidrig ,&nbsp;Beat Thomann","doi":"10.1016/j.prevetmed.2025.106688","DOIUrl":"10.1016/j.prevetmed.2025.106688","url":null,"abstract":"<div><div>Farm animal health and welfare have been ongoing concerns for farmers, consumers, and society for a long time. Antimicrobial resistance (AMR) and its relationship with antimicrobial use (AMU) have been identified as global public health challenges. A set of established health and welfare indicators (HWI) for dairy cows and the increasing availability of detailed farm-level AMU data enabled us to carry out this study with the objectives to (i) investigate the relationship between HWI and farm-level AMU, (ii) capture the current situation of AMR in dairy cows and calves on the study farms, and (iii) identify associations between AMR presence, AMU, and management practices in Swiss dairy farms. High AMU farms (n = 22) and low AMU farms (n = 28) were recruited based on prescription records in the national AMU database. Current farm AMU was assessed quantitatively using an animal treatment index (ATI) based on antimicrobial prescription data. A combination of animal-based and data-based health indicators were assessed on all study farms and used to form a total health and welfare score (HWS). Associations between lower ATI and good health and welfare were found for the total HWS (estimate: −0.144, p = 0.03) and the partial score for HWI related to health (-0.111, p = 0.01). Several individual HWI (e.g. productive lifespan of cows, participation in animal welfare programs) were found to be significantly associated with either AMU group (high vs. low), ATI, or both. Pooled fecal samples were used to assess the presence of AMR in the study farms. Commensal <em>Escherichia coli</em> isolates from calves and cows were most commonly resistant to tetracyclines (36 % and 3 %, respectively), followed by penicillins (34 % and 1 %) and sulfonamides (32 % and 1 %). No significant associations were observed between AMU and the presence of AMR (overall and for specific antimicrobial classes) on the study farms. Several management practices reported by farm managers during interviews (e.g. feeding of waste milk with antimicrobial residues to calves) were significantly associated with AMU group and ATI, but none were associated with the presence of AMR. The results of this study may help to understand how AMU is related to HWI in dairy cows and support monitoring and benchmarking efforts to reduce AMU while maintaining high levels of cow health and welfare. Further research on a larger scale is needed to obtain more generalizable results and to further investigate the complex and multifactorial relationship between AMU and AMR.</div></div>","PeriodicalId":20413,"journal":{"name":"Preventive veterinary medicine","volume":"245 ","pages":"Article 106688"},"PeriodicalIF":2.4,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145092533","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Corrigendum to \"Exploring canine mast cell tumors: An investigation into demographic characteristics, and grading system analysis from a pathology lab data (2019-2021)\" [Prev. Vet. Med. 236 (2025) 106416].","authors":"J Catarino, K Pinello, J Niza-Ribeiro, J Santos, R Payan-Carreira, J Reis, P Faísca","doi":"10.1016/j.prevetmed.2025.106686","DOIUrl":"https://doi.org/10.1016/j.prevetmed.2025.106686","url":null,"abstract":"","PeriodicalId":20413,"journal":{"name":"Preventive veterinary medicine","volume":" ","pages":"106686"},"PeriodicalIF":2.4,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145016014","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Economic viability of different surveillance strategies for the control of paratuberculosis in Swiss dairy cattle","authors":"Silja Griss ,&nbsp;Gertraud Schüpbach-Regula ,&nbsp;Luís P. Carmo ,&nbsp;Mireille Meylan ,&nbsp;Beat Thomann","doi":"10.1016/j.prevetmed.2025.106685","DOIUrl":"10.1016/j.prevetmed.2025.106685","url":null,"abstract":"<div><div>Paratuberculosis (PTB), or Johne's disease, is a globally prevalent disease caused by <em>Mycobacterium avium</em> subsp. <em>paratuberculosis</em>, leading to significant economic losses in the dairy industry. This study assessed the economic viability of potential PTB surveillance strategies in Switzerland using stochastic simulation models.Two non-targeted surveillance strategies (SS1 and SS2) were assessed under three participation scenarios over ten years. Additionally, a more targeted surveillance strategy (SS3) was evaluated, focusing on historically positive herds.In SS1, herds are screened using sock swab PCR followed by individual faecal PCR testing in sock swab-positive herds. SS2 follows a risk-based approach: in high-risk herds (i.e., previous PTB cases or animals with suggestive symptoms), all cows older than 24 months are tested using serum ELISA, with faecal PCR confirmation. In other herds, only high-risk cows (somatic cell count &gt;200,000/ml or &gt;3 lactation) undergo ELISA testing to determine herd status.In SS3, only animals from herds with recent outbreaks or clinical cases are tested.Neither of the two non-targeted strategies (SS1 and SS2) proves economically beneficial over a ten-year period, with Benefit-Cost Ratios (BCR) below 0.1. In contrast, the targeted strategy (SS3) achieves a BCR of 1.42, a net present value of CHF 568,545, and reaches the break-even point after seven years.These results indicate that targeted control strategies are more economically viable in low-prevalence settings.</div></div>","PeriodicalId":20413,"journal":{"name":"Preventive veterinary medicine","volume":"245 ","pages":"Article 106685"},"PeriodicalIF":2.4,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145019831","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessment of seasonality in antimicrobial susceptibility testing and resistance of urinary Escherichia coli from dogs and cats in the United States (2019 – 2022)","authors":"Rasaq A. Ojasanya ,&nbsp;J. Scott Weese ,&nbsp;Kurtis E. Sobkowich ,&nbsp;Anne Deckert ,&nbsp;Donald Szlosek ,&nbsp;Andy Plum ,&nbsp;Theresa M. Bernardo ,&nbsp;Zvonimir Poljak","doi":"10.1016/j.prevetmed.2025.106684","DOIUrl":"10.1016/j.prevetmed.2025.106684","url":null,"abstract":"<div><div>Antimicrobial resistance (AMR) poses a significant global health challenge, impacting humans, animals, and the environment. Dogs and cats are vulnerable to urinary tract infections (UTIs), mostly caused by antimicrobial-resistant <em>Escherichia coli</em>, necessitating antimicrobial susceptibility testing (AST) for optimal treatment. This study investigated and evaluated the seasonality of AST and AMR in urinary <em>E. coli</em> isolates from dogs and cats in the USA and evaluated the potential influence of climatic zones on these patterns. Retrospective data from IDEXX Laboratories, from January 2019 to December 2022, were analyzed. The dataset included 344,862 urinary <em>E. coli</em> isolates (74.2 % from dogs, 25.8 % from cats) tested against seven antimicrobials. Linear regression and negative binomial regression models assessed seasonality and trends, accounting for climatic zone variability. An increasing trend in AST was observed, with a seasonal peak in the summer for both species. Urinary <em>E. coli</em> isolates from dogs and cats had the highest resistance to amoxicillin at 27.9 % (95 % CI: 27.7–28.1) and 28.4 % (95 % CI: 28.1–28.7), respectively, among all antimicrobials tested. Resistance rates significantly declined (<em>p</em> &lt; .01) for all drugs tested in dogs, while in cats, declines were significant only for cefovecin, marbofloxacin, and enrofloxacin. No seasonality in AMR was found at the national level or across climatic zones, though AMR rates varied significantly by climatic zone (<em>p</em> &lt; .01). The hot-humid zone had the highest resistance rates but the lowest AST rates per one million dogs and cats. Seasonality in AST suggests a seasonal pattern for UTIs; however, no seasonal pattern in AMR could be observed nationally or regionally for urinary <em>E. coli</em> isolates.</div></div>","PeriodicalId":20413,"journal":{"name":"Preventive veterinary medicine","volume":"245 ","pages":"Article 106684"},"PeriodicalIF":2.4,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145019830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application of computer vision for automated detection of different lesions in pig lungs: An exploratory study","authors":"Matteo Recchia ,&nbsp;Lucia Scuri ,&nbsp;Camilla Allegri ,&nbsp;Claudia Romeo ,&nbsp;Federico Scali ,&nbsp;Antonio Marco Maisano ,&nbsp;Giovanni Santucci ,&nbsp;Giorgio Bontempi ,&nbsp;Simone Canesi ,&nbsp;Laura Sala ,&nbsp;Camilla Recordati ,&nbsp;Eugenio Scanziani ,&nbsp;Sara Panseri ,&nbsp;Sergio Ghidini ,&nbsp;Emanuela Zanardi ,&nbsp;Adriana Ianieri ,&nbsp;Lis Alban ,&nbsp;Giovanni Loris Alborali","doi":"10.1016/j.prevetmed.2025.106672","DOIUrl":"10.1016/j.prevetmed.2025.106672","url":null,"abstract":"<div><div>Accurate classification of lung lesions at necropsy is crucial for guiding the diagnostic process and ensuring effective management of porcine respiratory diseases. <em>Post-mortem</em> inspection of the lungs during slaughter also provides valuable insights into disease occurrence, offering useful feedback on the efficacy of on-farm prevention and control strategies. However, manual assessment protocols may be impaired by high slaughtering speeds and low inter-rater agreement, which limits continuous data collection and hinders comparability. Artificial intelligence, particularly computer vision (CV), may offer a promising alternative. This study aimed to train and test a commercial CV model for segmenting both anatomical structures and lesions in pig lungs. Overall, 1742 lungs were collected at slaughter, examined macroscopically, and photographed laterally. Two veterinarians with expertise in swine pathology manually annotated the acquired images to outline anatomical (i.e., lung, heart, lung lobes), pathological (i.e., bronchopneumonia, fibrinous pleuropneumonia, chronic pleuritis), and artefactual (i.e., parenchymal laceration, bronchoinhalation of blood) classes, forming the reference dataset for model training and testing. Model performance in segmenting these classes varied by confidence threshold, with sensitivity (36–84 %), positive predictive value (62–93 %) and F1 score (52–78 %) indicating imperfect yet improvable performance. Overall, anatomical structure segmentation outperformed lesion detection, likely due to class imbalance in the training dataset and the complexity of pulmonary pathology. Integrating standardized and real-time detection of lung lesions via digital imaging could improve respiratory health surveillance, thereby enhancing the role of abattoirs as strategic epidemiological observatories.</div></div>","PeriodicalId":20413,"journal":{"name":"Preventive veterinary medicine","volume":"245 ","pages":"Article 106672"},"PeriodicalIF":2.4,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145004730","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}