The pork industry is an essential part of the global food system, providing a significant source of protein for people around the world. A major factor restraining productivity and compromising animal wellbeing in the pork industry is disease outbreaks in pigs throughout the production process: widespread outbreaks can lead to losses as high as 10% of the U.S. pig population in extreme years. In this study, we present a machine learning model to predict the emergence of infection in swine production systems throughout the production process on a daily basis, a potential precursor to outbreaks whose detection is vital for disease prevention and mitigation. We determine features that provide the most value in predicting infection, which include nearby farm density, historical test rates, piglet inventory, feed consumption during the gestation period, and wind speed and direction. We utilize these features to produce a generalizable machine learning model, evaluate the model’s ability to predict outbreaks both seven and 30 days in advance, allowing for early warning of disease infection, and evaluate our model on two swine production systems and analyze the effects of data availability and data granularity in the context of our two swine systems with different volumes of data. Our results demonstrate good ability to predict infection in both systems with a balanced accuracy of
Influenza spread on context-specific networks lifted from interaction-based diary data
Studying the spread of infections is an important tool in limiting or preventing future outbreaks. A first step in understanding disease dynamics is constructing networks that reproduce features of real-world interactions. In this paper, we generate networks that maintain some features of the partial interaction networks that were recorded in an existing diary-based survey at the University of Warwick. To preserve realistic structure in our artificial networks, we use a context-specific approach. In particular, we propose different algorithms for producing larger home, work and social networks. Our networks are able to maintain much of the interaction structure in the original diary-based survey and provide a means of accounting for the interactions of survey participants with non-participants. Simulating a discrete susceptible–infected–recovered model on the full network produces epidemic behaviour which shares characteristics with previous influenza seasons. Our approach allows us to explore how disease transmission and dynamic responses to infection differ depending on interaction context. We find that, while social interactions may be the first to be reduced after influenza infection, limiting work and school encounters may be significantly more effective in controlling the overall severity of the epidemic.
more »
« less
- NSF-PAR ID:
- 10211400
- Date Published:
- Journal Name:
- Royal Society Open Science
- Volume:
- 8
- Issue:
- 1
- ISSN:
- 2054-5703
- Page Range / eLocation ID:
- 191876
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
more » « less
Abstract $$85.3\%$$ $$58.5\%$$ $$58.7\%$$ $$72.8\%$$ $$74.8\%$$ Mycoplasma hyopneumoniae in the second system, respectively, using the six most important predictors in all cases. These models provide daily infection probabilities that can be used by veterinarians and other stakeholders as a benchmark to more timely support preventive and control strategies on farms. -
Background Internet data can be used to improve infectious disease models. However, the representativeness and individual-level validity of internet-derived measures are largely unexplored as this requires ground truth data for study. Objective This study sought to identify relationships between Web-based behaviors and/or conversation topics and health status using a ground truth, survey-based dataset. Methods This study leveraged a unique dataset of self-reported surveys, microbiological laboratory tests, and social media data from the same individuals toward understanding the validity of individual-level constructs pertaining to influenza-like illness in social media data. Logistic regression models were used to identify illness in Twitter posts using user posting behaviors and topic model features extracted from users’ tweets. Results Of 396 original study participants, only 81 met the inclusion criteria for this study. Of these participants’ tweets, we identified only two instances that were related to health and occurred within 2 weeks (before or after) of a survey indicating symptoms. It was not possible to predict when participants reported symptoms using features derived from topic models (area under the curve [AUC]=0.51; P=.38), though it was possible using behavior features, albeit with a very small effect size (AUC=0.53; P≤.001). Individual symptoms were also generally not predictable either. The study sample and a random sample from Twitter are predictably different on held-out data (AUC=0.67; P≤.001), meaning that the content posted by people who participated in this study was predictably different from that posted by random Twitter users. Individuals in the random sample and the GoViral sample used Twitter with similar frequencies (similar @ mentions, number of tweets, and number of retweets; AUC=0.50; P=.19). Conclusions To our knowledge, this is the first instance of an attempt to use a ground truth dataset to validate infectious disease observations in social media data. The lack of signal, the lack of predictability among behaviors or topics, and the demonstrated volunteer bias in the study population are important findings for the large and growing body of disease surveillance using internet-sourced data.more » « less
-
Close contacts between individuals provide opportunities for the transmission of diseases, including COVID-19. While individuals take part in many different types of interactions, including those with classmates, co-workers and household members, it is the conglomeration of all of these interactions that produces the complex social contact network interconnecting individuals across the population. Thus, while an individual might decide their own risk tolerance in response to a threat of infection, the consequences of such decisions are rarely so confined, propagating far beyond any one person. We assess the effect of different population-level risk-tolerance regimes, population structure in the form of age and household-size distributions, and different interaction types on epidemic spread in plausible human contact networks to gain insight into how contact network structure affects pathogen spread through a population. In particular, we find that behavioural changes by vulnerable individuals in isolation are insufficient to reduce those individuals’ infection risk and that population structure can have varied and counteracting effects on epidemic outcomes. The relative impact of each interaction type was contingent on assumptions underlying contact network construction, stressing the importance of empirical validation. Taken together, these results promote a nuanced understanding of disease spread on contact networks, with implications for public health strategies.more » « less
-
Close contacts between individuals provide opportunities for the transmission of diseases, including COVID-19. While individuals take part in many different types of interactions, including those with classmates, co-workers and household members, it is the conglomeration of all of these interactions that produces the complex social contact network interconnecting individuals across the population. Thus, while an individual might decide their own risk tolerance in response to a threat of infection, the consequences of such decisions are rarely so confined, propagating far beyond any one person. We assess the effect of different population-level risk-tolerance regimes, population structure in the form of age and household-size distributions, and different interaction types on epidemic spread in plausible human contact networks to gain insight into how contact network structure affects pathogen spread through a population. In particular, we find that behavioural changes by vulnerable individuals in isolation are insufficient to reduce those individuals’ infection risk and that population structure can have varied and counteracting effects on epidemic outcomes. The relative impact of each interaction type was contingent on assumptions underlying contact network construction, stressing the importance of empirical validation. Taken together, these results promote a nuanced understanding of disease spread on contact networks, with implications for public health strategies.more » « less
-
Kim, Andrew ; Agarwal, Sabrina (Ed.)Objectives: Pandemics have profoundly impacted human societies, but until rela- tively recently were a minor research focus within biological anthropology, especially within biocultural analyses. Here, we explore research in these fields, including molecular anthropology, that employs biocultural approaches, sometimes integrated with intersectionality and ecosocial and syndemic theory, to unpack relationships between social inequality and pandemics. A case study assesses the 1918 influenza pandemic's impacts on the patient population of the Mississippi State Asylum (MSA). Materials and Methods: We survey bioarchaeological and paleopathological litera- ture on pandemics and analyze respiratory disease mortality relative to sex, age, and social race amongst patient deaths (N = 2258) between 1912 and 1925. Logistic regression models were used to assess relationships between cause of death and odds of death during the pandemic (1918–1919). Results: Findings include substantial respiratory mortality during the pandemic, including from influenza and influenza syndemic with pneumonia. Older patients (40–59 years, 60+ years) had lower odds (p < 0.01) of dying from respiratory disease than younger patients, as did female patients compared to males (p < 0.05). Age pat- terns are broadly consistent with national and state trends, while elevated mortality amongst Black and/or African American patients may reflect intersections between gender roles and race-based structural violence in the Jim Crow South. Discussion: Future work in biological anthropology on past pandemics may benefit from explicit incorporation of biocultural frameworks, intersectionality, and ecosocial and syndemic theory. Doing so enables holistic analyses of interactions between social context, social inequality and pandemic outcomes, generating data informative for public health responses and pandemic preparedness.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1098/rsos.191876
