Detection of clinical mastitis with sensor data from automatic milking systems is improved by using decision-tree induction.

The objective was to develop and validate a clinical mastitis (CM) detection model by means of decision-tree induction. For farmers milking with an automatic milking system (AMS), it is desirable that the detection model has a high level of sensitivity (Se), especially for more severe cases of CM, at a very high specificity (Sp). In addition, an alert for CM should be generated preferably at the quarter milking (QM) at which the CM infection is visible for the first time. Data were collected from 9 Dutch dairy herds milking automatically during a 2.5-yr period. Data included sensor data (electrical conductivity, color, and yield) at the QM level and visual observations of quarters with CM recorded by the farmers. Visual observations of quarters with CM were combined with sensor data of the most recent automatic milking recorded for that same quarter, within a 24-h time window before the visual assessment time. Sensor data of 3.5 million QM were collected, of which 348 QM were combined with a CM observation. Data were divided into a training set, including two-thirds of all data, and a test set. Cows in the training set were not included in the test set and vice versa. A decision-tree model was trained using only clear examples of healthy (n=24,717) or diseased (n=243) QM. The model was tested on 105 QM with CM and a random sample of 50,000 QM without CM. While keeping the Se at a level comparable to that of models currently used by AMS, the decision-tree model was able to decrease the number of false-positive alerts by more than 50%. At an Sp of 99%, 40% of the CM cases were detected. Sixty-four percent of the severe CM cases were detected and only 12.5% of the CM that were scored as watery milk. The Se increased considerably from 40% to 66.7% when the time window increased from less than 24h before the CM observation, to a time window from 24h before to 24h after the CM observation. Even at very wide time windows, however, it was impossible to reach an Se of 100%. This indicates the inability to detect all CM cases based on sensor data alone. Sensitivity levels varied largely when the decision tree was validated per herd. This trend was confirmed when decision trees were trained using data from 8 herds and tested on data from the ninth herd. This indicates that when using the decision tree as a generic CM detection model in practice, some herds will continue having difficulties in detecting CM using mastitis alert lists, whereas others will perform well.

[1]  W Steeneveld,et al.  Discriminating between true-positive and false-positive clinical mastitis alerts from automatic milking systems. , 2010, Journal of dairy science.

[2]  M. Rasmussen,et al.  Visual scoring of milk mixed with blood , 2005, Journal of Dairy Research.

[3]  Yoav Freund,et al.  Experiments with a New Boosting Algorithm , 1996, ICML.

[4]  M. Rasmussen,et al.  Visual scoring of clots in foremilk , 2005, Journal of Dairy Research.

[5]  G. H. Kroeze,et al.  Results of a multivariate approach to automated oestrus and mastitis detection , 1997 .

[6]  H. Hogeveen,et al.  Relationship between udder health and hygiene on farms with an automatic milking system. , 2010, Journal of dairy science.

[7]  Rik van der Tol,et al.  Using sensor data patterns from an automatic milking system to develop predictive variables for classifying clinical mastitis and abnormal milk , 2008 .

[8]  Tom M. Mitchell,et al.  Machine Learning and Data Mining , 2012 .

[9]  J. Ross Quinlan,et al.  Induction of Decision Trees , 1986, Machine Learning.

[10]  Ian H. Witten,et al.  Data mining: practical machine learning tools and techniques, 3rd Edition , 1999 .

[11]  H Hogeveen,et al.  Economic effects of bovine mastitis and mastitis management: A review , 2007, The Veterinary quarterly.

[12]  T. Larsen,et al.  Estimating degree of mastitis from time-series measurements in milk: a test of a model based on lactate dehydrogenase measurements. , 2007, Journal of dairy science.

[13]  H Hogeveen,et al.  Automatic detection of clinical mastitis is improved by in-line monitoring of somatic cell count. , 2008, Journal of dairy science.

[14]  H Hogeveen,et al.  Sensors and management support in high-technology milking. , 2003, Journal of animal science.

[15]  P. T. Johnstone,et al.  An automated in-line clinical mastitis detection system using measurement of conductivity from foremilk of individual udder quarters , 2009, New Zealand veterinary journal.

[16]  G. A. Mein,et al.  Performance evaluation of systems for automated monitoring of udder health: would the real gold standard please stand up? , 2008 .

[17]  D. McClish Analyzing a Portion of the ROC Curve , 1989, Medical decision making : an international journal of the Society for Medical Decision Making.

[18]  Henk Hogeveen,et al.  Automatic milking : a better understanding , 2004 .

[19]  Methods for estimating areas under receiver-operating characteristic curves: illustration with somatic-cell scores in subclinical intramammary infections. , 1999, Preventive veterinary medicine.

[20]  Joachim Krieter,et al.  Mastitis and lameness detection in dairy cows by application of fuzzy logic , 2009 .

[21]  Henk Hogeveen,et al.  Decision-tree induction to detect clinical mastitis with automatic milking , 2010 .

[22]  Henk Hogeveen,et al.  Data mining to detect clinical mastitis with automatic milking , 2010 .

[23]  Ian Witten,et al.  Data Mining , 2000 .

[24]  Henk Hogeveen,et al.  Performance evaluation of systems for automated monitoring of udder health: analytical issues and guidelines , 2008 .

[25]  H Hogeveen,et al.  Electrical conductivity of milk: ability to predict mastitis status. , 2004, Journal of dairy science.

[26]  C.J.A.M. de Koning,et al.  Milk quality on farms with an automatic milking system , 2004 .

[27]  Geoffrey I. Webb,et al.  MultiBoosting: A Technique for Combining Boosting and Wagging , 2000, Machine Learning.