A machine learning approach using partitioning around medoids clustering and random forest classification to model groups of farms in regard to production parameters and bulk tank milk antibody status of two major internal parasites in dairy cows

Abstract

Fasciola hepatica and Ostertagia ostertagi are internal parasites of cattle compromising physiology, productivity, and well-being. Parasites are complex in their effect on hosts, sometimes making it difficult to identify clear directions of associations between infection and production parameters. Therefore, unsupervised approaches not assuming a structure reduce the risk of introducing bias to the analysis. They may provide insights which cannot be obtained with conventional, supervised methodology. An unsupervised, exploratory cluster analysis approach using the k–mode algorithm and partitioning around medoids detected two distinct clusters in a cross-sectional data set of milk yield, milk fat content, milk protein content as well as F. hepatica or O. ostertagi bulk tank milk antibody status from 606 dairy farms in three structurally different dairying regions in Germany. Parasite–positive farms grouped together with their respective production parameters to form separate clusters. A random forests algorithm characterised clusters with regard to external variables. Across all study regions, co–infections with F. hepatica or O. ostertagi, respectively, farming type, and pasture access appeared to be the most important factors discriminating clusters (i.e. farms). Furthermore, farm level lameness prevalence, herd size, BCS, stage of lactation, and somatic cell count were relevant criteria distinguishing clusters. This study is among the first to apply a cluster analysis approach in this context and potentially the first to implement a k–medoids algorithm and partitioning around medoids in the veterinary field. The results demonstrated that biologically relevant patterns of parasite status and milk parameters exist between farms positive for F. hepatica or O. ostertagi, respectively, and negative farms. Moreover, the machine learning approach confirmed results of previous work and shed further light on the complex setting of associations a between parasitic diseases, milk yield and milk constituents, and management practices.

Establishment and Validation of Fourier Transform Infrared Spectroscopy (FT–MIR) Methodology for the Detection of Linoleic Acid in Buffalo Milk

“…Chen et al [ 59 ] used RF to evaluate the most significant drivers of soil fungal diversity, including plant communities and soil physicochemical properties. Similarly, Andreas et al [ 60 ] utilized RF to identify tillage type as the most important factor affecting dairy cows with Fasciola hepatica, with higher-ranking variables yielding more accurate predictions than those with lower importance scores. The variable importance measure can be used by RF to select and order the spectral regions that are most predictive.…”

Perinatal mortality in German dairy cattle: Unveiling the importance of cow-level risk factors and their interactions using a multifaceted modelling approach

“…The described calculations suggested an optimal and achievable sample size of 250 farms per region [ 21 , 23 , 24 ]. Farm selection was stratified by administrative district and herd size, i.e.…”

Random forest classification as a tool in epidemiological modelling: Identification of farm-specific characteristics relevant for the occurrence of Fasciola hepatica on German dairy farms

“…A total number of 86,304 dairy cows (North: 24,980 cows; East: 49,936 cows South: 11,388 cows) on 765 farms (North: 253; East: 252; South: 260) were included in the study. Sample size calculation and farm selection are described in [ 28 , 29 , 31 ]. In brief, sample size calculation was based on the formula suggested for prevalence studies: where n is the sample size to be calculated, Z the level of confidence, P the assumed prevalence, and d the precision.…”

“…The aim of this work was the application of a RF approach to a multifactorial data set of dairy cow housing conditions, management practices, and production parameters to identify and rank relevant covariates for farm-level presence of F. hepatica. For this purpose, we were able to build upon previous work of our group using a cross-sectional data set [27][28][29]. The application of a machine learning approach increases the understanding of the on-farm network of factors affecting farm-level positivity for this important helminth.…”