PMID- 32664417
OWN - NLM
STAT- MEDLINE
DCOM- 20210324
LR  - 20210324
IS  - 1424-8220 (Electronic)
IS  - 1424-8220 (Linking)
VI  - 20
IP  - 14
DP  - 2020 Jul 10
TI  - Using Sensor Data to Detect Lameness and Mastitis Treatment Events in Dairy Cows: 
      A Comparison of Classification Models.
LID - 10.3390/s20143863 [doi]
LID - 3863
AB  - The aim of this study was to develop classification models for mastitis and 
      lameness treatments in Holstein dairy cows as the target variables based on 
      continuous data from herd management software with modern machine learning 
      methods. Data was collected over a period of 40 months from a total of 167 
      different cows with daily individual sensor information containing milking 
      parameters, pedometer activity, feed and water intake, and body weight (in the 
      form of differently aggregated data) as well as the entered treatment data. To 
      identify the most important predictors for mastitis and lameness treatments, 
      respectively, Random Forest feature importance, Pearson's correlation and 
      sequential forward feature selection were applied. With the selected predictors, 
      various machine learning models such as Logistic Regression (LR), Support Vector 
      Machine (SVM), K-nearest neighbors (KNN), Gaussian Naive Bayes (GNB), Extra Trees 
      Classifier (ET) and different ensemble methods such as Random Forest (RF) were 
      trained. Their performance was compared using the receiver operator 
      characteristic (ROC) area-under-curve (AUC), as well as sensitivity, block 
      sensitivity and specificity. In addition, sampling methods were compared: Over- 
      and undersampling as compensation for the expected unbalanced training data had a 
      high impact on the ratio of sensitivity and specificity in the classification of 
      the test data, but with regard to AUC, random oversampling and SMOTE (Synthetic 
      Minority Over-sampling) even showed significantly lower values than with 
      non-sampled data. The best model, ET, obtained a mean AUC of 0.79 for mastitis 
      and 0.71 for lameness, respectively, based on testing data from practical 
      conditions and is recommended by us for this type of data, but GNB, LR and RF 
      were only marginally worse, and random oversampling and SMOTE even showed 
      significantly lower values than without sampling. We recommend the use of these 
      models as a benchmark for similar self-learning classification tasks. The 
      classification models presented here retain their interpretability with the 
      ability to present feature importances to the farmer in contrast to the "black 
      box" models of Deep Learning methods.
FAU - Post, Christian
AU  - Post C
AUID- ORCID: 0000-0001-5746-7830
AD  - Physiology Unit, Institute of Animal Science, University of Bonn, 53115 Bonn, 
      Germany.
FAU - Rietz, Christian
AU  - Rietz C
AD  - Department of Educational Science, Faculty of Educational and Social Sciences, 
      University of Education Heidelberg, 69120 Heidelberg, Germany.
FAU - Buscher, Wolfgang
AU  - Buscher W
AUID- ORCID: 0000-0002-7212-7639
AD  - Livestock Technology Section, Institute for Agricultural Engineering, University 
      of Bonn, 53115 Bonn, Germany.
FAU - Muller, Ute
AU  - Muller U
AD  - Physiology Unit, Institute of Animal Science, University of Bonn, 53115 Bonn, 
      Germany.
LA  - eng
GR  - 2815710515/Bundesministerium fur Ernahrung und Landwirtschaft/
PT  - Journal Article
DEP - 20200710
PL  - Switzerland
TA  - Sensors (Basel)
JT  - Sensors (Basel, Switzerland)
JID - 101204366
SB  - IM
MH  - Animals
MH  - Bayes Theorem
MH  - Cattle
MH  - Dairying
MH  - Female
MH  - Lameness, Animal/*diagnosis/*therapy
MH  - *Machine Learning
MH  - Mastitis/*diagnosis/*therapy
MH  - Milk
MH  - Sensitivity and Specificity
PMC - PMC7411665
OTO - NOTNLM
OT  - classification
OT  - lameness
OT  - machine learning
OT  - mastitis
OT  - sensor data
COIS- The authors declare no conflict of interest.
EDAT- 2020/07/16 06:00
MHDA- 2021/03/25 06:00
PMCR- 2020/07/01
CRDT- 2020/07/16 06:00
PHST- 2020/04/30 00:00 [received]
PHST- 2020/07/01 00:00 [revised]
PHST- 2020/07/09 00:00 [accepted]
PHST- 2020/07/16 06:00 [entrez]
PHST- 2020/07/16 06:00 [pubmed]
PHST- 2021/03/25 06:00 [medline]
PHST- 2020/07/01 00:00 [pmc-release]
AID - s20143863 [pii]
AID - sensors-20-03863 [pii]
AID - 10.3390/s20143863 [doi]
PST - epublish
SO  - Sensors (Basel). 2020 Jul 10;20(14):3863. doi: 10.3390/s20143863.