Objectives were to assess reproductive and productive outcomes associated with failure of clinical cure in dairy cows diagnosed with metritis following antimicrobial therapy. This retrospective cohort study included data from 3 experiments performed in 5 dairies. Metritis was characterized by the presence of watery, fetid, reddish-brownish vaginal discharge within 21 DIM (study d 0). Cows not diagnosed with metritis (i.e., cows may have had other diseases postpartum; NoMT; n = 1,194) were paired based on lactation number and calving date. All cows with metritis received antimicrobial therapy (ampicillin or ceftiofur). Clinical cure was evaluated on d 10 based on vaginal discharge score, and cows were categorized as cured (MTC; n = 1,111) or not cured (MTnoC; n = 299). Purulent vaginal discharge (28 ± 3 or 32 ± 3 DIM), cytological endometritis (35 ± 3 or 39 ± 3 DIM), and estrous cyclicity (50 ± 3 and 64 ± 3, 36 ± 3 and 50 ± 3, or 37 ± 5 and 51 ± 5 DIM) were evaluated in subgroups of cows. Proportions of cows with purulent vaginal discharge and cytological endometritis were greatest for MTnoC (91.7 and 91.4%), intermediate for MTC (74.0 and 73.3%), and smallest for NoMT (38.1 and 36.4%). Proportion of cyclic cows was smaller for MTnoC compared with MTC and NoMT (62.0, 71.0, and 71.0%). Pregnancy per artificial insemination following first service was smaller for cows with metritis compared with their counterparts with no metritis (NoMT = 28.1, MTC = 26.1, MTnoC = 22.0%). Pregnancy loss tended to be greater for MTnoC compared with MTC (NoMT = 11.5, MTC = 11.1, MTnoC = 18.4%). Hazard of pregnancy by 300 DIM was smallest for MTnoC, intermediate for MTC, and greatest for NoMT. Death by 60 DIM (3.9, 1.1, and 0.6%) and removal from herd by 300 DIM (26.3, 17.4, and 15.4%) were greatest for MTnoC compared with MTC and NoMT, respectively. Milk production among multiparous cows was smaller for MTnoC compared with MTC and NoMT in the first 10 mo postpartum, whereas MTC produced less milk compared with NoMT only during the first 2 mo postpartum (NoMT = 42.0 ± 0.22, MTC = 40.6 ± 0.28, MTnoC = 37.7 ± 0.54 kg/d). Failure of clinical cure was not associated with milk yield in primiparous cows (NoMT = 35.2 ± 0.31, MTC = 33.9 ± 0.31, MTnoC = 35.0 ± 0.52 kg/d). Cows diagnosed with metritis that do not undergo clinical cure by 10 d of onset of antimicrobial therapy have impaired reproductive performance, reduced milk production, and increased risk of leaving the herd.

Animals' individual behavior is commonly monitored by live or video observation by a person. This can be labor intensive and inconsistent. An alternative is the use of machine learning-based computer vision systems. The objectives of this study were to 1) develop and optimize machine learning model frameworks for detecting, tracking and classifying individual behaviors of group-housed broiler chickens in continuous video recordings; and 2) use an independent dataset to evaluate the performance of the developed machine leaning model framework for individual poultry behaviors differentiation. Forty-two video recordings from 4 different pens (total video duration = 1,620 min) were used to develop and train multiple models for detecting and tracking individual birds and classifying 4 behaviors: feeding, drinking, active, and inactive. The optimal model framework was used to continuously analyze an external set of 11 videos (duration = 326 min), and the second-by-second behavior of each individual broiler was extracted for the comparison of human observation. After comparison of model performance, the YOLOv5l, out of 5 detection models, was selected for detecting individual broilers in a pen; the osnet_x0_25_msmt17, out of 4 tracking algorithms, was selected to track each detected bird in continuous frames; and the Gradient Boosting Classifier, out of 12 machine learning classifiers, was selected to classify the 4 behaviors. Most of the models were able to keep previously assigned individual identifications of the chickens in limited amounts of time, but lost the identities throughout an examination period (≥4 min). The final framework was able to accurately predict feeding (accuracy = 0.895) and drinking time (accuracy = 0.9) but subpar for active (accuracy = 0.545) and inactive time (accuracy = 0.505). The algorithms employed by the machine learning models were able to accurately detect feeding and drinking behavior but still need to be improved for maintaining individual identities of the chickens and identifying active and inactive behaviors.

Our objectives were to compare the effects of ceftiofur crystalline free acid (CCFA) and ampicillin trihydrate (AMP) treatments of cows diagnosed with metritis on uterine health, behavior, reproductive, and productive responses. A controlled randomized clinical trial was designed. Metritis was defined as vaginal discharge (VD) = 5 (fetid, watery, red/brown) within 21 d in milk (DIM) and rectal temperature (RT) <39.5°C, whereas VD = 5 and RT ≥39.5°C was defined as puerperal metritis. On the day of diagnosis (d 0), cows were paired by parity and severity of metritis (metritis vs. puerperal metritis) and assigned randomly to the AMP and CCFA treatments. Cows enrolled in the AMP (n = 308) treatment were moved to a nonsalable-milk pen, where they were treated once daily for 5 d, and were moved back to their original pen 72 h after the last treatment (d 7). Cows enrolled in the CCFA (n = 310) treatment remained in their original pen and received 2 treatments of CCFA, 72 h apart. Rectal temperature was measured daily from d 0 to 6 and on d 11. Vaginal discharge was evaluated on d 4, 6, and 11 to assess cure. Cure was defined as the absence of treatment with additional antimicrobial before experiment d 11, VD <5, and RT <39.5°C. Cows were examined at 28 ± 3 DIM for purulent VD (PVD) and at 35 ± 3 DIM for cytological endometritis. Pregnancy was diagnosed at 40 ± 3 and 60 ± 7 d after first and second artificial inseminations. Cure of metritis did not differ between treatments on d 11 (AMP = 64.6 ± 3.1, CCFA = 63.5 ± 3.1%). Cows treated with AMP had greater RT from experiment d 1 to 6 compared with cows treated with CCFA (AMP = 39.1 ± 0.02, CCFA = 39.0 ± 0.02°C). Cows in the AMP treatment had greater prevalence of PVD at 28 ± 3 DIM (AMP = 82.6 ± 2.3, CCFA = 74.4 ± 2.7%) and tended to have greater prevalence of cytological endometritis at 35 ± 3 DIM (AMP = 77.8 ± 6.2 vs. CCFA = 61.7 ± 7.5%) than CCFA-treated cows. Treatment did not affect the hazard of pregnancy among multiparous cows; however, among primiparous cows, CCFA treatment reduced the hazard of pregnancy and increased the median days to pregnancy (AMP = 145 vs. CCFA = 169 d). Finally, average daily milk yield up to 14 wk postpartum was not affected by treatment (AMP = 38.0 ± 0.4, CCFA = 37.5 ± 0.4 kg). We conclude from the current experiment that CCFA was more effective in reducing RT and improving uterine health of metritic cows; however, the improved hazard of pregnancy of primiparous cows treated with AMP is important and warrants further investigation.

Objectives of the current experiment were to evaluate the effects on reproductive performance of adding an automated estrous detection (AED) device to the reproductive management of Holstein cows. In addition, we aimed to determine whether the AED device would equally influence the reproductive performance of cows receiving artificial insemination and those receiving embryo transfer. Cows (n = 1,244) were observed daily for estrous detection by trained farm personnel with the aid of a tail head mounting device. Cows (n = 641) enrolled in the AED treatment were fitted with a neck-mounted automated estrus detection device, and cows enrolled in the control (CTRL, n = 603) treatment were not fitted with the AED device. Cows not detected in estrus by 54 ± 3 DIM received 1 injection of PGF2α, and those not detected in estrus by 68 ± 3 DIM were enrolled in an ovulation synchronization protocol. The hazard of first service tended to be affected by treatment [CTRL: referent, AED: adjusted hazard ratio (AHR) = 1.11, 95% confidence interval (CI) = 0.98, 1.25], and we determined a tendency for the AED treatment to reduce the proportion of cows receiving the first service at fixed time (39.7 ± 5.7 vs. 45.5 ± 5.8%). After the first service, high-producing cows enrolled in the AED treatment were more likely to be pregnant (19.1 ± 2.4 vs. 31.8 ± 3.0%), but no differences between the AED and CTRL treatments were observed among low-producing cows (24.9 ± 2.9 vs. 24.7 ± 2.9%). The hazard of second service was affected by treatment (CTRL: referent, AED: AHR = 1.22, 95% CI = 1.01, 1.47), and the AED treatment reduced the proportion of cows receiving the second service at fixed time compared with the CTRL treatment (15.7 ± 2.2 vs. 21.9 ± 2.5%). After the second service, the AED treatment tended to increase the proportion of cows pregnant (31.0 ± 3.0 vs. 24.9 ± 2.6%) and tended to reduce the proportion of cows that lost pregnancy between the first and second pregnancy exams (9.7 ± 3.3 vs. 16.7 ± 4.6%). The interaction between treatment and type of service did not affect percentage of cows pregnant and pregnancy loss to the first and second services. The hazard of pregnancy was affected by the interaction between treatment and milk yield, in that cows with above the median milk yield in the AED treatment became pregnant at a faster rate than CTRL cows, but no difference between treatments was observed among cows with milk yield below the median. Cows with above the median milk yield had shorter duration of estrus and were less likely to have activity peak at estrus ≥89 (0 = minimum, 100 = maximum). The current experiment suggests that AED devices may improve reproductive performance not only by increasing service rates but also by improving accuracy of estrous detection. According to the current experiment, high-producing cows may benefit most from the addition of AED devices to reproductive management.

The objectives of the 2 studies reported herein were to validate the accuracy of an automated monitoring device (AMD) to detect side lying, resting, activity, rumination, eating, walking, and panting in nonlactating and lactating dairy cows. Additionally, we aimed to determine whether the total time per cow-state recorded by the AMD within a 30-min interval corresponds to the total time per cow-state recorded simultaneously by visual observation. Study personnel (n = 2) observed pregnant nonlactating Holstein cows (n = 10) for 30 min in the morning and 30 min in the afternoon for 6 consecutive days and recorded continuously each cow-state. In study 2, study personnel (n = 2) observed lactating Holstein cows (n = 10) for 30 min in the morning and 30 min in the afternoon for 6 consecutive days. In both studies, cow-state was recorded every second, and within 1 min, the most prevalent cow-state was considered to be the behavior presented by the cow during that interval. Using the observer as the gold standard, test characteristics were calculated for the minute-by-minute interval analyses. For the 30-min interval analyses, the concordance correlation coefficient (pc) and the coefficient of determination (R2) between the total minutes for each cow-state recorded by the observer and the AMD were calculated. In study 1, for the minute-by-minute interval analyses, test characteristics were high for rumination (≥90.1%) and eating (≥73.8%), moderate for resting (≥62.9%), but negligible for medium activity (≥17%). For the 30-min interval analyses, the correlations between the total time of visual observations compared with the total time recorded by AMD for rumination (R2 = 0.97, pc = 0.98) and eating (R2 = 0.91, pc = 0.94) were very high, for resting (R2 = 0.77, pc = 0.79) was high, and for medium activity (R2 = 0.41, pc = 0.41) was low. In study 2, for the minute-by-minute interval analyses, test characteristics were high for rumination (≥79.4%), eating (≥74.2%), and resting (≥73.0%), but they were low for panting (≥31.3%) and negligible for medium activity (≥22.2%). For the 30-min interval analyses, the correlations were similar to study 1 (rumination: R2 = 0.85, pc = 0.91; eating: R2 = 0.95, pc = 0.97; resting: R2 = 0.84, pc = 0.90; medium activity: R2 = 0.44, pc = 0.57; and panting: R2 = 0.21, pc = 0.42). In summary, the AMD used in this study provided accurate data regarding resting, rumination, and eating of pregnant nonlactating and lactating Holstein cows.