Detection of estrus and mastitis: Field performance of a model

A new detection model ('IMAG model')for estrus and mastitis in dairy cows w as tested on four farms during several years. Such a test is necessary beca use information is lacking about the performance of detection models under field conditions. The test gave insight into the field performance of the I MAG model and the results were compared with the results of older models an d with the results predicted by experts. Sensor data of milk yield, milk te mperature, electrical conductivity of milk and animal activity were the inp uts for the IMAG model. The IMAG model is based on time series analysis com bined with a Kalman filter This structure yields cow-dependent model parame ters and combines data of different sensors. Results were compared with the manufacturers model (supplied with the sensors), based only on exponential smoothing on data from one sensor The sensor equipment differed between fa rms. The sensitivity (percentage of estruses detected)for estrus varied fi- om 63 to 80%, depending on the threshold used. Specificity (non-estruses no t detected as estrus) varied from 94 to 98%. The sensitivity for clinical m astitis varied from 55 to 80%, depending on the threshold used. The specifi city for mastitis varied from 94 to 99%. Significant differences existed be tween forms, in sensitivity for estrus and mastitis. The applied equipment could only partially explain the differences in estrus and mastitis detecti on results between farms. No relation between stage of lactation and activi ty level was found, although a lower activity level in the first period of lactation might be expected. The main conclusion is that a detection model can give good results, but only if the equipment is working properly. The n ew model outperformed the manufacturer's model.