On the prediction of SAV transmission among Norwegian aquaculture sites.

Pancreas Disease (PD) is a viral disease that affects Atlantic salmon (Salmo salar) in Norwegian, Scottish and Irish aquaculture. It is caused by salmonid alphavirus (SAV) and represents a significant problem in salmonid farming. Infection with SAV leads to reduced growth, mortality, product downgrading, and has a significant financial impact for the farms. The overall aim of this study is to evaluate the effect of various factors on the transmission of SAV and to create a predictive model capable of providing an early warning system for salmon farms within the Norwegian waters. Using a combination of publicly available databases, specifically BarentsWatch, and privately held PCR analyses a feature set consisting of 11 unique features was created based on the input parameters of the databases. An ensemble model was developed based on this feature set using XG-Boost, Ada-Boost, Random Forest and a Multilayer Perceptron. It was possible to successfully predict SAV transmission with 94.4% accuracy. Moreover, it was possible to predict SAV transmission 8 weeks in advance of a 'PD registration' at individual aquaculture salmon farming sites. Important predictors included well boat movement, environmental factors, proximity to sites with a 'PD registration' and seasonality.

Short communications: an ear-attached accelerometer detects effects of regrouping on lying, rumination, and activity times in calves.

The objective of this study was to use a sensor-based accelerometer (ACC) to identify changes in lying, rumination, and activity times in weaned calves during the moving and regrouping process. Overall, 270 healthy Holstein calves (from approximately 16 regrouping events) at the age of approximately 4 months were enrolled and equipped with an ear-attached ACC (SMARTBOW, Smartbow GmbH/ Zoetis LLC). Sensor data were recorded for 5 d before (d -5) until 4 d after moving and regrouping (d 4). The day of regrouping was defined as d 0. Acceleration data were continuously processed by specific algorithms (developed by SMARTBOW) for lying, rumination, and activity. Lying, rumination, and activity times were averaged from d -5 to d -3 to generate a baseline value for each parameter. Parameters on d 0 to d 4 after regrouping were compared to this baseline. All parameters showed significant changes compared with the baseline at d 0. Significant decreases in rumination and inactive times were seen up to d 2. Lying time was significantly lower until d 3. The study results indicate that the ACC can be used to monitor the disruptive effects of regrouping on lying and rumination behaviors. Further research is necessary to elucidate how these changes have an impact on health, performance, and welfare and to evaluate how to reduce these potentially negative effects.

Erratum to "Use of a real-time location system to detect cows in distinct functional areas within a barn" (JDS Commun. 2:217-222).

[This corrects the article DOI: 10.3168/jdsc.2020-0050.].

Monitoring selected behaviors of calves by use of an ear-attached accelerometer for detecting early indicators of diarrhea.

One of the most important diseases in calves worldwide is neonatal calf diarrhea (NCD), which impairs calf welfare and leads to economic losses. The aim of this study was to test whether the activity patterns of calves can be used as early indicators to identify animals at risk for suffering from NCD, compared with physical examination. We monitored 310 healthy female Holstein-Friesian calves on a commercial dairy farm immediately after birth, equipped them with an ear tag-based accelerometer (Smartbow, Smartbow GmbH), and conducted daily physical examinations during the first 28 d of life. The Smartbow system captured acceleration data indicative of standing and lying periods and activity levels (active and inactive), shown as minutes per hour. We categorized calves as diarrheic if they showed fecal scores of ≥3 on a 4-point scale on at least 2 consecutive days. Incidence of diarrhea was 50.7% (n = 148). A mixed logistic regression model showed that lying [odds ratio (OR) = 1.19], inactive (OR = 1.14), and active (OR = 0.92) times, 1 d before clinical identification of diarrhea (d -1), were associated with the odds of diarrhea occurring on the subsequent day. Receiver operating characteristics curve showed that lying time at d -1 was a fair predictor for diarrhea on the subsequent day (area under curve = 0.69). Average lying time on d -1 was 64.8 min longer in diarrheic calves compared with their controls. Median lying and inactive times decreased, and active time increased with age over the study period. The 24-h pattern of behavior indices based on the output of the Smartbow system followed periods of resting and active times, and showed that between 2200 h and 0600 h, calves spent the greatest percentage of time lying and inactive. These results showed that the accelerometer system has the potential to detect early indicators associated with NCD. In future studies, additional data for the development and testing of calf- and event-specific algorithms (e.g., for detecting milk intake, playing behavior) should be collected, which might further improve the early detection of diarrhea in calves.

Use of a real-time location system to detect cows in distinct functional areas within a barn.

Automated sensor-based monitoring of cows has become an important tool in herd management to improve or maintain animal health and welfare. Location systems offer the ability to locate animals within the barn for, for example, artificial insemination. Furthermore, they have the potential to measure the time cows spend in important areas of the barn, which might indicate need for improvement in the management of the herd or individuals. In this study, we tested the sensor-based real-time location system (RTLS) Smartbow (SB, Smartbow GmbH) under field conditions. The objectives of this study were (1) to determine the accuracy of the system to predict the location of the cow and the agreement between visual observations and RTLS observations for the total time spent by cows in relevant areas of the barn and (2) to compare the performance of 2 different algorithms (Alg1 and Alg2) for cow location. The study was conducted on a commercial Austrian dairy farm. In total, 35 lactating cows were video recorded for 3 consecutive days. From these recordings, approximately 1 h was selected randomly each day for every cow (3 d × 35 cows). Simultaneously, location data were collected and classified by the RTLS system as dedicated to the alley, feed bunk, or cubicle on a 1-min resolution. A total of 6,030 paired observations were derived from visual observations (VO) and the RTLS and used for the final data analysis. Substantial agreement of categorical data between VO and SB was obtained by Cohen's kappa for both algorithms (Alg1 = 0.76 and Alg2 = 0.78). Similar results were achieved by both algorithms throughout the study, with a slight improvement for Alg2. The ability of the system to locate the cows in the predefined areas was assessed, and the results from Alg2 showed sensitivity, specificity, and positive predictive value of alley (74.0, 91.2, and 76.9%), feed bunk (93.5, 86.2, and 89.1%), and cubicle (90.5, 83.3, and 95.4%) and an overall accuracy of 87.6%.The correlation coefficient (r) between VO and SB for the total time cows spent (within 1 h) in the predefined areas was good to strong (r = 0.82, 0.98, and 0.92 for alley, feed bunk, and cubicle, respectively). These results show the potential of the system to automatically assess total time spent by cows in important areas of the barn for indoor settings. Future studies should focus on evaluating 24-h periods to assess time budgets and to combine technologies such as accelerometers and location systems to improve the performance of behavior prediction in dairy cows.

Comparison of behavioral patterns of dairy cows with natural estrus and induced ovulation detected by an ear-tag based accelerometer.

Dairy farms face many challenges and changes. With increasing herd sizes and fewer farmers or employees per cow, new strategies to maintain or improve reproductive management are required. One of the major challenges is to detect cows in estrus and to estimate the perfect time for artificial insemination (AI). Several estrus and ovulation synchronization programs with timed AI as well as estrus detection aids, e.g., tail-paint, pedometer, accelerometer, and others are available. A combination of ovulation synchronization programs and technical solutions, however, has rarely been tested. This study was designed to gain insights into behavioral patterns of cows subjected to an Ovsynch program and to test if behavioral data could be used to optimize the timing of insemination within an Ovsynch program. In this study, we used an ear-tag based 3D-accelerometer system (SMARTBOW, Smartbow GmbH, Weibern, Austria) to generate data of behavioral patterns, i.e., rumination and activity. In Part 1 of this study, behavioral patterns during the peri-estrus period were compared between cows with physiological estrus and cows subjected to an Ovsynch protocol. On the day before estrus and on the day of estrus/AI, cows with natural estrus showed a clear drop in rumination and "inactivity" and an increase in "high activity", based on an algorithm of the accelerometer system, whereas, cows in the Ovsynch protocol showed only minor changes in behavioral patterns. In Part 2, we analyzed behavioral patterns between synchronized cows that became pregnant after AI and synchronized cows that remained open. As a result, no differences were detected between these two Ovsynch groups before AI. Thus, in this study we found no evidence that behavioral patterns can be used to improve conception rates within an Ovsynch protocol.

Evaluation of an ear-attached accelerometer for detecting estrus events in indoor housed dairy cows.

Precision dairy farming technologies have tremendous potential to improve and support farmers in herd management decisions, particularly in reproductive management. Nowadays, estrus detection in cows is challenging and several supporting tools are available. In this study, a 3D-accelerometer integrated into an ear-tag (SMARTBOW, Smartbow GmbH, Weibern, Austria) was used for the detection of cows in estrus. Movement pattern based on accelerometer data were analyzed and processed by algorithms and machine learning, resulting in estrus alerts. For the evaluation of the system, reproductive performance data of 579 estrus events of multiparous cows were used to retrospectively evaluate the accuracy of estrus alerts generated by the accelerometer-based system and the overall performance of the system. Estrus events were classified as 'gold standard' events, if an estrus followed by AI resulted in pregnancy, and as 'recorded estrus' events, if two estrus events with an interval of 18-25 d were in the herd records, independent of whether estrus was followed by AI or pregnancy. In total, 316 'gold standard' events were matched with estrus alerts generated by the accelerometer-based system, resulting in a sensitivity of 97%. Furthermore, 263 'recorded estrus' events were compared with correct or incorrect estrus alerts by the system. Sensitivity, specificity, positive and negative predictive values, accuracy, and error rate for 'recorded estrus' events were 97%, 98%, 96%, 94%, 96%, and 2%, respectively. In summary, the SMARTBOW system is suitable for an automated detection of estrus events of multiparous cows in indoor housed dairy cows.

Evaluation of an ear-tag-based accelerometer for monitoring rumination in dairy cows.

The objective of this study was to evaluate the ear-tag-based accelerometer system Smartbow (Smartbow GmbH, Weibern, Austria) for detecting rumination time, chewing cycles, and rumination bouts in indoor-housed dairy cows. For this, the parameters were determined by analyses of video recordings as reference and compared with the results of the accelerometer system. Additionally, we tested the intra- and inter-observer reliability as well as the agreement of direct cow observations and video recordings. Ten Simmental dairy cows in early lactation were equipped with 10-Hz accelerometer ear tags and kept in a pen separated from herd mates. A total mixed ration was fed twice a day via a roughage intake control system. During the study, cows' rumination and other activities were directly observed for 20 h by 2 trained observers. Additionally, cows were video recorded for 19 d, 24 h a day. After exclusion of unsuitable videos, 2,490 h of cow individual 1-h video sequences were eligible for further analyses. Out of this, one hundred 1-h video sequences were randomly selected and visually and manually classified by a trained observer using professional video analyses software. Based on these analyses, half of the data was used for development (based on data of 50-h video analyses) and testing (based on data of additional 50-h video analyses) of the Smartbow algorithms, respectively. Inter- and intra-observer reliability as well as the comparison of direct against video observations revealed in high agreements for rumination time and chewing cycles with Pearson correlation coefficients >0.99. The rumination time, chewing cycles, as well as rumination bouts detected by Smartbow were highly associated (r > 0.99) with the analyses of video recordings. Algorithm testing revealed in an underestimation of the average ± standard deviation rumination time per 1-h period by the Smartbow system of 17.0 ± 35.3 s (i.e., -1.2%), compared with visual observations. The average number ± standard deviation of chewing cycles and rumination bouts was overestimated by Smartbow by 59.8 ± 79.6 (i.e., 3.7%) and by 0.5 ± 0.9 (i.e., 1.6%), respectively, compared with the video analyses. In summary, the agreement between the Smartbow system with video analyses was excellent. From a practical and clinical point of view, the detected differences were negligible. However, further research is necessary to test the system under various field conditions and to evaluate the benefit of incorporating rumination data into herd management decisions.