Stefan Widgren
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Stefan Widgren has a total of 129 co-authors. Here are the top co-authors by number of shared publications:
- Stefan Engblom
Division of Scientific Computing, Department of Information Technology, Uppsala University, SE-751 05 Uppsala, Sweden
8 shared publications Ann Lindberg
National Veterinary Institute (SVA), Uppsala, Sweden
5 shared publications- Ann Lindberg
Department of Disease Control and Epidemiology, National Veterinary Institute, Uppsala, Sweden
5 shared publications - Pavol Bauer
Uppsala University, Uppsala, Sweden
5 shared publications - Maria Nöremark
National Veterinary Institute (SVA), Uppsala, Sweden
4 shared publications - Ulf Emanuelson
Department of Clinical Sciences, Swedish University of Agricultural Sciences, Uppsala, Sweden
4 shared publications
Latest Publications
- Open Access
- 11 May 2026
- Published by Elsevier in Preventive Veterinary Medicine
- Vol. 254, 106912
Swedish cattle farmers’ preferences for control measures against Mycoplasma bovis: A discrete choice experiment
- 11 May 2026
- Published by Elsevier in Preventive Veterinary Medicine
- Vol. 254, 106912
Open AccessSwedish cattle farmers’ preferences for control measures against Mycoplasma bovis: A discrete choice experiment
- Open Access
- 10 July 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-18
Automated tick classification using deep learning and its associated challenges in citizen science
Lyme borreliosis and tick-borne encephalitis significantly impact public health in Europe, transmitted primarily by endemic tick species. The recent introduction of exotic tick species into northern Europe via migratory birds, imported animals, and travelers highlights the urgent need for rapid detection and accurate species identification. To address this, the Swedish Veterinary Agency launched a citizen science initiative, resulting in the submission of over 15,000 tick images spanning seven species. We developed, trained, and evaluated deep learning models incorporating image analysis, object detection, and transfer learning to support automated tick classification. The EfficientNetV2M model achieved a macro recall of 0.60 and a Matthews Correlation Coefficient (MCC) of 0.55 on out-of-distribution, citizen-submitted data. These results demonstrate the feasibility of integrating AI with citizen science for large-scale tick monitoring while also highlighting challenges related to class imbalance, species similarity, and morphological variability. Rather than robust species-level classification, our framework serves as a proof of concept for infrastructure that supports scalable and adaptive tick surveillance. This work lays the groundwork for future AI-driven systems in One Health contexts, extendable to other arthropod vectors and emerging public health threats.
- 10 July 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-18
Open AccessAutomated tick classification using deep learning and its associated challenges in citizen science
Lyme borreliosis and tick-borne encephalitis significantly impact public health in Europe, transmitted primarily by endemic tick species. The recent introduction of exotic tick species into northern Europe via migratory birds, imported animals, and travelers highlights the urgent need for rapid detection and accurate species identification. To address this, the Swedish Veterinary Agency launched a citizen science initiative, resulting in the submission of over 15,000 tick images spanning seven species. We developed, trained, and evaluated deep learning models incorporating image analysis, object detection, and transfer learning to support automated tick classification. The EfficientNetV2M model achieved a macro recall of 0.60 and a Matthews Correlation Coefficient (MCC) of 0.55 on out-of-distribution, citizen-submitted data. These results demonstrate the feasibility of integrating AI with citizen science for large-scale tick monitoring while also highlighting challenges related to class imbalance, species similarity, and morphological variability. Rather than robust species-level classification, our framework serves as a proof of concept for infrastructure that supports scalable and adaptive tick surveillance. This work lays the groundwork for future AI-driven systems in One Health contexts, extendable to other arthropod vectors and emerging public health threats.
- Open Access
- 1 January 2025
- Published by American Dairy Science Association in Journal of Dairy Science
- Vol. 108 (1) , 845-855
A longitudinal study of the dynamics of Mycoplasma bovis antibody status in primiparous cows and bulk tank milk in Swedish dairy herds
Mycoplasma bovis is an important pathogen causing pneumonia, mastitis, and arthritis in cattle all over the world entailing reduced animal welfare and economic losses. In this longitudinal study, we investigated the presence of M. bovis antibodies in bulk tank milk (BTM) and in milk from primiparous (PP) cows at 4 sampling occasions over 2 yr. Herd characteristics associated with a positive antibody test result in PP cows were investigated. The participating dairy herds (n = 149) were situated in southern Sweden, samples were collected and analyzed with ID Screen antibody ELISA. Information on herd characteristics was retrieved from the national DHI database. To identify herd characteristics associated with the presence of antibodies in PP cows, mixed linear regression with herd and sample as random factors were used. The apparent herd-level prevalence of M. bovis infection based on antibodies in BTM was 17%, but with the addition of PP cows, the prevalence increased to 28%. The results showed that larger herds and introduction of cattle was associated with higher antibody levels in PP cows. In conclusion, this study showed a clear difference in the apparent prevalence of M. bovis infection based on antibodies in BTM or in PP cows: The number of positive herds was almost doubled when including PP cows. This motivates repeated sampling of a few PP cows to find newly infected herds in an early stage. Finally, the results showed that introduction of cattle influences the level of M. bovis antibodies. This is important in the control and prevention of further spread of the infection. It is essential for free herds to know their M. bovis status, and antibody testing is highly recommended if introducing cattle.
- 1 January 2025
- Published by American Dairy Science Association in Journal of Dairy Science
- Vol. 108 (1) , 845-855
Open AccessA longitudinal study of the dynamics of Mycoplasma bovis antibody status in primiparous cows and bulk tank milk in Swedish dairy herds
Mycoplasma bovis is an important pathogen causing pneumonia, mastitis, and arthritis in cattle all over the world entailing reduced animal welfare and economic losses. In this longitudinal study, we investigated the presence of M. bovis antibodies in bulk tank milk (BTM) and in milk from primiparous (PP) cows at 4 sampling occasions over 2 yr. Herd characteristics associated with a positive antibody test result in PP cows were investigated. The participating dairy herds (n = 149) were situated in southern Sweden, samples were collected and analyzed with ID Screen antibody ELISA. Information on herd characteristics was retrieved from the national DHI database. To identify herd characteristics associated with the presence of antibodies in PP cows, mixed linear regression with herd and sample as random factors were used. The apparent herd-level prevalence of M. bovis infection based on antibodies in BTM was 17%, but with the addition of PP cows, the prevalence increased to 28%. The results showed that larger herds and introduction of cattle was associated with higher antibody levels in PP cows. In conclusion, this study showed a clear difference in the apparent prevalence of M. bovis infection based on antibodies in BTM or in PP cows: The number of positive herds was almost doubled when including PP cows. This motivates repeated sampling of a few PP cows to find newly infected herds in an early stage. Finally, the results showed that introduction of cattle influences the level of M. bovis antibodies. This is important in the control and prevention of further spread of the infection. It is essential for free herds to know their M. bovis status, and antibody testing is highly recommended if introducing cattle.
- Open Access
- 1 January 2024
- Published by Wiley in Transboundary and Emerging Diseases
- Vol. 2024 (1) , 4734277
Herd‐Level Modeling of Bovine Viral Diarrhea Virus (BVDV) Transmission in Cattle Herds in Southern Chile: Linking Within and Between‐Herd Dynamics
Bovine viral diarrhea (BVD) represents a serious threat to the cattle sector in Chile, indicating the need for a regionally defined control program. Ex-ante evaluations of program options using simulation modeling have proven to be a successful approach in providing decision‐makers with relevant supporting insights in that respect. Given the complexity of bovine viral diarrhea virus (BVDV) infection dynamics, simulation of BVD spread in a metapopulation requires detailed consideration of both within and between herd transmission dynamics. The aims of the study are (i) to investigate the dynamics of BVDV transmission in cattle herds in southern Chile by linking a within‐herd transmission model (WHM) that accounts for the BVDV’s unique characteristics with a between‐herd model (BHM) that meets the demands for further regional control strategy evaluation; (ii) to suggest and discuss criteria for evaluation of the model approach and plausibility for later research and for support decision‐making. This resulted in bringing forth a modeling rationale for complex disease spread simulation in metapopulations. BHM simulations under this approach show outcomes that agree with BVDV’s known situation in Chile; dairy herds prevalence at endemic equilibrium reaches and maintains 75%, which agrees with estimations of BVDV active infection in dairy herds in southern Chile (77%). For the entire herd population, the infection always reaches endemic levels with a large proportion of infected herds (median = 60%), where herd prevalence was higher in the dairy herd class than in the remaining categories. Transmission probability variation affects the new infections picked, prevalence at endemic levels, and the velocity in which the infection spreads between herds. The fact that the presented approach was able to model a complex infection dynamic such BVDV, with sufficient confidence, provides evidence that this approach can be used to explore mitigation strategies to control BVDV in southern Chilean herds.
- 1 January 2024
- Published by Wiley in Transboundary and Emerging Diseases
- Vol. 2024 (1) , 4734277
Open AccessHerd‐Level Modeling of Bovine Viral Diarrhea Virus (BVDV) Transmission in Cattle Herds in Southern Chile: Linking Within and Between‐Herd Dynamics
Bovine viral diarrhea (BVD) represents a serious threat to the cattle sector in Chile, indicating the need for a regionally defined control program. Ex-ante evaluations of program options using simulation modeling have proven to be a successful approach in providing decision‐makers with relevant supporting insights in that respect. Given the complexity of bovine viral diarrhea virus (BVDV) infection dynamics, simulation of BVD spread in a metapopulation requires detailed consideration of both within and between herd transmission dynamics. The aims of the study are (i) to investigate the dynamics of BVDV transmission in cattle herds in southern Chile by linking a within‐herd transmission model (WHM) that accounts for the BVDV’s unique characteristics with a between‐herd model (BHM) that meets the demands for further regional control strategy evaluation; (ii) to suggest and discuss criteria for evaluation of the model approach and plausibility for later research and for support decision‐making. This resulted in bringing forth a modeling rationale for complex disease spread simulation in metapopulations. BHM simulations under this approach show outcomes that agree with BVDV’s known situation in Chile; dairy herds prevalence at endemic equilibrium reaches and maintains 75%, which agrees with estimations of BVDV active infection in dairy herds in southern Chile (77%). For the entire herd population, the infection always reaches endemic levels with a large proportion of infected herds (median = 60%), where herd prevalence was higher in the dairy herd class than in the remaining categories. Transmission probability variation affects the new infections picked, prevalence at endemic levels, and the velocity in which the infection spreads between herds. The fact that the presented approach was able to model a complex infection dynamic such BVDV, with sufficient confidence, provides evidence that this approach can be used to explore mitigation strategies to control BVDV in southern Chilean herds.
- Open Access
- 1 January 2024
- Published by Wiley in Transboundary and Emerging Diseases
- Vol. 2024 (1) , 6071781
First Outbreak of African Swine Fever in Sweden: Local Epidemiology, Surveillance, and Eradication Strategies
The first case of African swine fever (ASF) was confirmed in Sweden in September 2023. This article describes the local epidemiology, including the spatiotemporal dynamics of the outbreak and some of the factors that may have contributed to its apparently successful eradication. Upon detection of the outbreak, strict control measures were put in place in a preliminarily defined infected zone. A carcass search, including geo‐localisation, removal, sampling, and destruction of found carcasses, was initiated and a preliminary core area was defined based on the results. Six months after confirmation of the first case, 93 wild boar carcasses had been found in the infected zone, of which 62 tested positive for ASF virus (ASFV). All ASFV‐positive carcasses were found inside the core area. Based on two taphonomy methods, it was assumed that the infection was introduced between early May and late June 2023. The data also indicated that the epidemic curve peaked between mid‐August and mid‐September, with the last death occurring in late September 2023. Based on the average estimated time of death, geo‐localisation of carcasses and two‐dimensional kernel density estimation, clustering in space and time was identified. An online questionnaire with questions about hunting and the wild boar population was sent to all leaders of hunting groups in the infected zone. The results showed that the wild boar population had increased in the last 10 years but with large variations and geographical heterogeneity in space use. Disease introduction through natural wild boar movements was excluded and it was assumed that the long‐distance translocation of the virus had occurred through human activities. A municipal waste collection centre without wild boar‐proof fencing is located close to the epicentre of the outbreak, attracting many wild boar and contributing to the spread of the virus once it had been introduced to the population.
- 1 January 2024
- Published by Wiley in Transboundary and Emerging Diseases
- Vol. 2024 (1) , 6071781
Open AccessFirst Outbreak of African Swine Fever in Sweden: Local Epidemiology, Surveillance, and Eradication Strategies
The first case of African swine fever (ASF) was confirmed in Sweden in September 2023. This article describes the local epidemiology, including the spatiotemporal dynamics of the outbreak and some of the factors that may have contributed to its apparently successful eradication. Upon detection of the outbreak, strict control measures were put in place in a preliminarily defined infected zone. A carcass search, including geo‐localisation, removal, sampling, and destruction of found carcasses, was initiated and a preliminary core area was defined based on the results. Six months after confirmation of the first case, 93 wild boar carcasses had been found in the infected zone, of which 62 tested positive for ASF virus (ASFV). All ASFV‐positive carcasses were found inside the core area. Based on two taphonomy methods, it was assumed that the infection was introduced between early May and late June 2023. The data also indicated that the epidemic curve peaked between mid‐August and mid‐September, with the last death occurring in late September 2023. Based on the average estimated time of death, geo‐localisation of carcasses and two‐dimensional kernel density estimation, clustering in space and time was identified. An online questionnaire with questions about hunting and the wild boar population was sent to all leaders of hunting groups in the infected zone. The results showed that the wild boar population had increased in the last 10 years but with large variations and geographical heterogeneity in space use. Disease introduction through natural wild boar movements was excluded and it was assumed that the long‐distance translocation of the virus had occurred through human activities. A municipal waste collection centre without wild boar‐proof fencing is located close to the epicentre of the outbreak, attracting many wild boar and contributing to the spread of the virus once it had been introduced to the population.
- Open Access
- 22 September 2023
- Published by Springer Nature in Acta Veterinaria Scandinavica
- Vol. 65 (1) , 1-9
Understanding the interface between European wild boar (Sus scrofa) and domestic pigs (Sus scrofa domesticus) in Sweden through a questionnaire study
Background: In recent years, the wildlife/livestock interface has attracted increased attention due to disease transmission between wild and domestic animal populations. The ongoing spread of African swine fever (ASF) in European wild boar (Sus scrofa) emphasize the need for further understanding of the wildlife/livestock interface to prevent disease spill-over between the wild and domestic populations. Although wild boar may also act as a potential source for other infectious disease agents, ASF is currently the most severe threat from wild boar to domestic pigs. To gather information on the wild boar situation at commercial pig producing farms in Sweden, a digital questionnaire survey was distributed through the animal health services. Results: Most pigs produced for commercial purposes in Sweden are raised without outdoor access. Of the 211 responding pig producers, 80% saw wild boar or signs of wild boar activity in the vicinity of their farm at least once during the year. Observations were significantly correlated with geographical region, but there was no correlation between farm characteristics (farm size, main type of production, outdoor access) and observed wild boar presence or proximity. However, a reported higher frequency of wild boar observations was positively correlated with the observations being made in closer proximity to the farm. Hunting and strategic baiting were the most common mitigation strategies used to keep wild boar at bay. Of the 14 farms raising pigs with outdoor access, 12 responded that these pigs could be raised solely indoors if needed. Pigs with outdoor access are required to be fenced in, but double fencing in these outdoor pig enclosures was not practiced by all. A perimeter fence surrounding any type of pig farm was very rare. More than half of the producers that grew crops with intended use for pigs reported crop damage by wild boar. Conclusion: This study shows that although pigs raised for commercial purposes in Sweden are, to a large extent, kept indoors the potential for indirect contact with wild boar exists and must be considered. Variable local situations regarding wild boar abundance may require an adaptive approach regarding biosecurity efforts.
- 22 September 2023
- Published by Springer Nature in Acta Veterinaria Scandinavica
- Vol. 65 (1) , 1-9
Open AccessUnderstanding the interface between European wild boar (Sus scrofa) and domestic pigs (Sus scrofa domesticus) in Sweden through a questionnaire study
Background: In recent years, the wildlife/livestock interface has attracted increased attention due to disease transmission between wild and domestic animal populations. The ongoing spread of African swine fever (ASF) in European wild boar (Sus scrofa) emphasize the need for further understanding of the wildlife/livestock interface to prevent disease spill-over between the wild and domestic populations. Although wild boar may also act as a potential source for other infectious disease agents, ASF is currently the most severe threat from wild boar to domestic pigs. To gather information on the wild boar situation at commercial pig producing farms in Sweden, a digital questionnaire survey was distributed through the animal health services. Results: Most pigs produced for commercial purposes in Sweden are raised without outdoor access. Of the 211 responding pig producers, 80% saw wild boar or signs of wild boar activity in the vicinity of their farm at least once during the year. Observations were significantly correlated with geographical region, but there was no correlation between farm characteristics (farm size, main type of production, outdoor access) and observed wild boar presence or proximity. However, a reported higher frequency of wild boar observations was positively correlated with the observations being made in closer proximity to the farm. Hunting and strategic baiting were the most common mitigation strategies used to keep wild boar at bay. Of the 14 farms raising pigs with outdoor access, 12 responded that these pigs could be raised solely indoors if needed. Pigs with outdoor access are required to be fenced in, but double fencing in these outdoor pig enclosures was not practiced by all. A perimeter fence surrounding any type of pig farm was very rare. More than half of the producers that grew crops with intended use for pigs reported crop damage by wild boar. Conclusion: This study shows that although pigs raised for commercial purposes in Sweden are, to a large extent, kept indoors the potential for indirect contact with wild boar exists and must be considered. Variable local situations regarding wild boar abundance may require an adaptive approach regarding biosecurity efforts.
- Open Access
- 7 April 2022
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 17 (4) , e0266457
Complex network analysis to understand trading partnership in French swine production
The circulation of livestock pathogens in the pig industry is strongly related to animal movements. Epidemiological models developed to understand the circulation of pathogens within the industry should include the probability of transmission via between-farm contacts. The pig industry presents a structured network in time and space, whose composition changes over time. Therefore, to improve the predictive capabilities of epidemiological models, it is important to identify the drivers of farmers’ choices in terms of trade partnerships. Combining complex network analysis approaches and exponential random graph models, this study aims to analyze patterns of the swine industry network and identify key factors responsible for between-farm contacts at the French scale. The analysis confirms the topological stability of the network over time while highlighting the important roles of companies, types of farm, farm sizes, outdoor housing systems and batch-rearing systems. Both approaches revealed to be complementary and very effective to understand the drivers of the network. Results of this study are promising for future developments of epidemiological models for livestock diseases. This study is part of the One Health European Joint Programme: BIOPIGEE.
- 7 April 2022
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 17 (4) , e0266457
Open AccessComplex network analysis to understand trading partnership in French swine production
The circulation of livestock pathogens in the pig industry is strongly related to animal movements. Epidemiological models developed to understand the circulation of pathogens within the industry should include the probability of transmission via between-farm contacts. The pig industry presents a structured network in time and space, whose composition changes over time. Therefore, to improve the predictive capabilities of epidemiological models, it is important to identify the drivers of farmers’ choices in terms of trade partnerships. Combining complex network analysis approaches and exponential random graph models, this study aims to analyze patterns of the swine industry network and identify key factors responsible for between-farm contacts at the French scale. The analysis confirms the topological stability of the network over time while highlighting the important roles of companies, types of farm, farm sizes, outdoor housing systems and batch-rearing systems. Both approaches revealed to be complementary and very effective to understand the drivers of the network. Results of this study are promising for future developments of epidemiological models for livestock diseases. This study is part of the One Health European Joint Programme: BIOPIGEE.
- Open Access
How mechanistic modelling supports decision making for the control of enzootic infectious diseases