Evaluation of calving indicators measured by automated monitoring devices to predict the onset of calving in Holstein dairy cows

Abstract

Dystocias are common in dairy cows and often adversely affect production, reproduction, animal welfare, labor, and economics within the dairy industry. An automated device that accurately predicts the onset of calving could potentially minimize the effect of dystocias by enabling producers to intervene early. Although many well-documented indicators can detect the imminence of calving, research is limited on their effectiveness to predict calving when measured by automated devices. The objective of this experiment was to determine if a decrease in vaginal temperature (VT), rumination (RT), and lying time (LT), or an increase in lying bouts (LB), as measured by 3 automated devices, could accurately predict the onset of calving within 24, 12, and 6 h. The combination of these 4 calving indicators was also evaluated. Forty-two multiparous Holstein cows housed in tie-stalls were fitted with a temperature logger inserted in the vaginal cavity 7±2 d before their expected calving date; VT was recorded at 1-min intervals. An ear-attached sensor recorded rumination time every hour based on ear movement while an accelerometer fitted to the right hind leg recorded cow position at 1-min intervals. On average, VT were 0.3±0.03°C lower, and RT and LT were 41±17 and 52±28 min lower, respectively, on the calving day compared with the previous 4 d. Cows had 2±1 more LB on the calving day. Of the 4 indicators, a decrease in VT≥0.1°C was best able to predict calving within the next 24 h with a sensitivity of 74%, specificity of 74%, positive and negative predictive values of 51 and 89%, and area under the curve of 0.80. Combining the indicators enhanced the performance to predict calving within the next 24, 12, and 6 h with best overall results obtained by combining the 3 devices for prediction within the next 24 h (sensitivity: 77%, specificity: 77%, positive and negative predictive values: 56 and 90%, area under the curve: 0.82). These results indicate that a device that could simultaneously measure these 4 calving indicators could not precisely determine the onset of calving, but the information collected would assist dairy farmers in monitoring the onset of calving.

The utility of sensor technology to support reproductive management on dairy farms

“…A decrease in "rumination" prior to calving was observed in this study, consistent with other studies (Bar and Solomon, 2010, Bucher and Sundrum, 2014, Clark, et al, 2015, Ouellet, et al, 2016, Pahl, et al, Saint-Dizier and Chastant-Maillard, 2015, Schirmann, et al, 2013). In the current study, this decrease was about 15 minutes per hour (the difference between the case and control dataset during the hour in which calving started).…”

“…The sensitivity of the "DTC + sensor" model increased around 20 percentage points at a specificity of approximately 99%, when evaluated on an hourly basis using a three-hour time window. The model performance in this study, as measured in AUC values on a separate test dataset (Table 2), was higher than previous findings of Ouellet et al (2016). This was unexpected because the latter study lacked an independent validation and had a smaller dataset (32 cows) (Ouellet, et al, 2016).…”

“…Other studies have also reported activity to increase during the 24 hours before calving (Clark, et al, 2015, Miedema, et al, 2011b, Saint-Dizier and Chastant-Maillard, 2015, Titler, et al, 2015. For "temperature", a decrease of about 3°C was observed during the hour in which calving started, whereas other studies have reported decreases in the range of 0.2-0.5 °C (Burfeind, et al, 2011, Ouellet, et al, 2016. The sensor in the current study measured ear temperature, which is more sensitive to environmental influences than core body temperature (Gonzalezjimenez and Blaxter, 1962).…”

“…Such sensors could potentially be used as an alternative to visual observation for onset and progression of calving , Ouellet, et al, 2016, Rutten, et al, 2017, Titler, et al, 2015. For instance, the walking behaviour and postures of cows prior to calving have been analyzed with video camera's (Cangar, et al, 2008).…”

“…Another example is temperature, which has been shown to decrease before the start of calving and this can be monitored with sensors (Burfeind, et al, 2011, Saint-Dizier andChastantMaillard, 2015). Lastly, changes in rumination, lying time and lying bouts determined by sensors have been associated with the start of calving (Bar and Solomon, 2010, Bucher and Sundrum, 2014, Ouellet, et al, 2016, Rutten, et al, 2017, Titler, et al, 2015.…”