Development of a new clinical mastitis detection method for automatic milking systems

Abstract

This study investigated the potential for accurate detection of clinical mastitis (CM) in an automatic milking system (AMS) using electronic data from the support software. Data from cows were used to develop the model, which was then tested on 2 independent data sets, 1 with 311 cows (same farm but from a different year) and 1 with 568 cows (from a different farm). In addition, the model was used to test how well it could predict CM 1 to 3 d before actual clinical diagnosis. Logistic mixed models were used for the analysis. Twelve measurements were included in the initial model before a backward elimination, which resulted in the following 6 measurements being included in the final model: quarter-level milk yield (MY; kg), electrical conductivity (EC; mS/cm), average milk flow rate (MF; kg/min), occurrence of incompletely milked quarters in each milking session (IM; yes or no), MY per hour (MYH; kg/h), and EC per hour (ECH; mS/cm/h) between successive milking sessions. The other 6 measurements tested but not included in the final model were peak milk flow rate (kg/min), kick-offs (yes or no) in each milking session, lactation number, days in milk (d), blood in milk (yes or no), and a calculated mastitis detection index used by DeLaval (DelPro software; DeLaval International AB, Tumba, Sweden). All measurements were assessed to determine their ability to detect CM as both individual variables and combinations of the 12 above-mentioned variables. These were assessed by producing a receiver operating characteristic curve and calculating the area under the curve (AUC) for each model. Overall, 9 measurements (i.e., EC, ECH, MY, MYH, MF, IM, peak flow rate, lactation number, and mastitis detection index) had significant mastitis detection ability as separate predictors. The best mastitis prediction was possible by incorporating 6 measurements (i.e., EC, ECH, MY, MYH, MF, and IM) as well as the random cow and quarter effects in the model, resulting in 90% sensitivity and 91% specificity with excellent AUC (0.96). Assessment of the model was found to produce robust results (AUC >0.9) in different data sets and could detect CM with reductions in sensitivity and specificity with increasing days before actual diagnosis. This study demonstrated that improved mastitis status prediction can be achieved by using multiple measurements, and new indexes based on that are expected to result in improved accuracy of mastitis alerts, thereby improving the detection ability and utility on farm.

Modeling cow somatic cell count using sensor data as input to generalized additive models

“…2a). Contrary to our results, a previous study by Khatun et al (2018) showed that MDi performed no better than did conductivity at the quarter level in a mastitis detection model, and MDi was excluded from their final model. The differences in the results might be due to the nature of the different outcomes studied (SCC vs. clinical mastitis defined as veterinary treatment), model choice (nonlinear vs. linear model), difference in milking interval, or available data (one vs. two herds).…”

Suitability of somatic cell count, electrical conductivity, and lactate dehydrogenase activity in foremilk before versus after alveolar milk ejection for mastitis detection

“…We further evaluated our hypothesis using ROC assessment, which is a useful tool for assessing performance in predicting clinical mastitis (Khatun et al, 2018). In the ROC analysis, the mastitis diagnostic test in this study may be considered excellent (AUC > 0.9) compared with those used in other studies (AUC ≤ 0.73; Norberg et al, 2004;Mollenhorst et al, 2010;Petzer et al, 2017).…”

“…Construction of ROC curves was performed using the pROC package in R (version 3.4.4; Robin et al, 2011). The ROC assessment graphically illustrates the diagnostic test to present sensitivity (Se) versus the complement of specificity (1 − Sp) for varying cut points (Hanley and McNeil, 1982;Khatun et al, 2018). The cut points are determined for different probabilities (or linear predictors) of the fitted GLMM.…”

A Review of Welfare Indicators of Indoor-Housed Dairy Cow as a Basis for Integrated Automatic Welfare Assessment Systems

“…Such systems could provide a good basis for developing a broader integrated system that should then be tested on the farm. For instance, one commercial system that integrates different milk parameters (Del Pro software, DeLaval International AB, Tumba, Sweden) was found to have a high reliability in detecting clinical mastitis [119]. By adding to such a system other detection techniques-e.g., those that focus on indicators of lameness or heat stress-one could make the first steps in developing a more inclusive cow welfare assessment system.…”