Interpretable machine learning applied to on‐farm biosecurity and porcine reproductive and respiratory syndrome virus

Abstract

Effective biosecurity practices in swine production are key in preventing the introduction and dissemination of infectious pathogens. Ideally, on-farm biosecurity practices should be chosen by their impact on bio-containment and bio-exclusion, however quantitative supporting evidence is often unavailable. Therefore, the development of methodologies capable of quantifying and ranking biosecurity practices according to their efficacy in reducing disease risk have the potential to facilitate better informed choices of biosecurity practices.Using survey data on biosecurity practices, farm demographics, and previous outbreaks from 139 herds, a set of machine learning algorithms were trained to classify farms by porcine reproductive and respiratory syndrome virus status, depending on their biosecurity practices and farm demographics, to produce a predicted outbreak risk. A novel interpretable machine learning toolkit, MrIML-biosecurity, was developed to benchmark farms and production systems by predicted risk, and quantify the impact of biosecurity practices on disease risk at individual farms.Quantifying the variable impact on predicted risk 50% of 42 variables were associated with fomite spread while 31% were associated with local transmission. Results from machine learning interpretations identified similar results, finding substantial contribution to predicted outbreak risk from biosecurity practices relating to: the turnover and number of employees; the surrounding density of swine premises and pigs; the sharing of haul trailers; distance from the public road; and farm production type. In addition, the development of individualized biosecurity assessments provides the opportunity to better guide biosecurity implementation on a case-by-case basis. Finally, the flexibility of the MrIML-biosecurity toolkit gives it the potential to be applied to wider areas of biosecurity benchmarking, to address biosecurity weaknesses in other livestock systems and industry relevant diseases.

Modelling and assessing additional transmission routes for porcine reproductive and respiratory syndrome virus: Vehicle movements and feed ingredients

“…Additionally, the on‐farm model parameters were oversimplified, since we have based those estimations through historical records of PRRSV outbreaks, in which farms with fewer infections were considered to have better biosecurity levels. Although there are several ways that the current version of PigSpread model can be expanded, the inclusions of specific on‐farm biosecurity practices and infrastructure (e.g., present of cleaning and disinfection stations) could not only improve model calibration but also analyze the role of individualized and combined biosecurity on PRRSV dissemination (Sykes et al., 2021). Another important limitation of our modelling work was the lack of information about the vaccination strategies used by each farm, which could have contributed to the probability of new PRRSV outbreaks (Galvis et al., 2022).…”

Modelling and assessing additional transmission routes for porcine reproductive and respiratory syndrome virus: vehicle movements and feed ingredients

“…Although there are several ways that the current version of PigSpread model can be expanded, the inclusions of specific on-farm biosecurity practices and infrastructure (e.g. present of cleaning and disinfection stations) could not only improve model calibration, but to analyze the role of individualized and combined biosecurity on PRRSV dissemination (Sykes et al, 2021). Another important limitation of our modelling work was the lack of information about the vaccination strategies used by each farm, which could have contributed to the probability of new PRRSV outbreaks (Galvis et al, 2021).…”

A comparison of explainable artificial intelligence methods in the phase classification of multi-principal element alloys

“…In this paper, in addition to SHAP, we will focus on two other instance-level post-hoc model interpretability approaches, namely breakDown (BD) analysis and Ceteris-Paribus (CP) profiles, that have received little attention in the materials science community. The BD method, similar to the SHAP method, is also based on the variable attribution principle that decomposes the prediction of each individual observation into particular variable contributions 30 , 31 . Unlike the SHAP values, the BD values provide order specific explanations of variables’ contributions in a greedy way 5 .…”