Sickness detection for feedlot cattle

The beef industry contends with significant economic losses due to illnesses such as bovine respiratory disease (BRD), with estimates surpassing $800-$900 million annually. Traditional methods of detection rely heavily on human observation, often leading to inaccuracies in diagnosis and treatment. Leveraging wearables and activity monitoring, this study introduces the HerdDogg® WelfareTag® cow tag, which employs advanced analytics and machine learning (ML) to identify deviations in cattle behavior indicative of health issues. Data from 1,141 animals across 10 facilities over nine months were collected and analyzed. Through time series analysis and ML classification, the model achieved a 77% true positive rate in detecting reduced movement associated with illness within one day, with a low 7% false positive rate. While periodic surveillance remains necessary, the model offers an objective and efficient method to complement traditional detection approaches, potentially reducing misdiagnosis rates and economic losses in the industry.

Illness in feedlot cattle is a significant source of economic burden in the beef industry. Bovine respiratory disease (BRD) alone is estimated to cost the beef cattle industry upwards of $800-$900 million per year in animal deaths, reduction of feed efficiency, and treatment costs.1 The identification of abnormalities in behavior displayed by cattle suffering from BRD and other health issues is a costly and inefficient process. This is especially true for large operations where employees do not have time or resources to closely examine every animal.

Traditional methods for detecting cattle that are infected with BRD, or any other illness, is to employ pen riders to survey the animals once or twice per day looking for subjective deviations from normal behavior and appearance. This can often be challenging, as unhealthy cattle tend to mask signs of weakness or illness in the presence of humans.2 Previous studies have shown that clinical diagnosis of BRD can be highly inaccurate. Up to 38% of diseased animals may go undiagnosed, while up to 37% of diagnosed animals that end up being treated may not in fact be suffering from disease.3

Previous research has also shown that changes in physical behavior accompany illnesses and disease in both beef and dairy cattle.4,5,6 Using wearables and activity monitoring to identify these behavioral changes in a timely manner is an alternative method for decreasing morbidity and mortality rates and minimizing economic losses. The HerdDogg® WelfareTag® brand cow tag, or WelfareTag, aids in the detection of health issues by utilizing advanced analytics and machine learning (ML) to detect changes in an individual animal’s activity relative to its baseline, as well as that of other animals belonging to the same facility or pen. Using an activity monitoring system combined with a machine learning classification model, we find that we’re able to accurately identify activity commonly seen in cattle that have health issues.

Individual animal activity is monitored using the WelfareTag, which contains an accelerometer that records movement data every six minutes. Data is collected in fixed intervals and uploaded to a cloud storage system over WiFi or a cellular data connection using strategically placed data gathering stations. In this study, we utilized data collected for 1,141 animals at 10 different facilities over a 9 month period from Aug 2023 - Apr 2024.

Figure 1. An example of the movement data for an individual animal (black) and the movement data for the facility’s animals (orange).

Data for all animals was first reviewed by subject-matter experts familiar with cattle activity. When available, treatment datasets provided by the facilities in this study were used as a reference to look for correlations between activity health issues. This approach is based on previous studies showing changes in behavior patterns for both beef and dairy cattle.4,5,6 When reviewing treatment data, we found that many of the animals with treatment records do not have noticeable changes in activity over time. On average this was the case for approximately 50% of the treatment datasets that were reviewed. In these cases, data was not labeled around the time of treatment. Additionally, animals that did not have treatment records but still displayed activity that deviated from the norm were labeled for model training.

Time series analysis is then performed using open source Python packages to extract statistical, temporal, and frequency-domain signals in the data which distinguish the movement of healthy cattle from those that are potentially sick.7,8 These signals are also aggregated over various groups of cattle, such as which pen they belong to. This additional baseline helps prevent false positive detections when the entire lot has periods of lower activity induced by external factors such as environmental or feed changes.

After all data transformations and feature engineering is complete, a machine learning (ML) classification model is trained over 90% of the data. A stratified k-fold cross-validation with 5 folds is used for sampling in order to account for the heavy imbalance of unhealthy to healthy labels in the training set. The remaining 10% of the data is used as a holdout set to validate against the trained model. This holdout set only contains data at times after the data it was trained on in order to simulate how this model is evaluated on live data in a production environment. This process is repeated as more data is acquired in order to have a model that is up-to-date on the latest trends within an individual group of cattle.

When evaluating the performance of the ML model, we use the following definitions:

• If a detection is within one day of labeled activity, that detection is classified as a true positive.
  • Consecutively labeled days are to be counted as a single label. Similarly, multiple detections around a set of consecutively labeled days will be counted as only one true positive. For example, if the dates 01/02, 01/03, and 01/04 are labeled, and detections come back for 01/01 and 01/03, only one true positive is recorded.

• If a day receives a true value for a detection and it is not within one day of a labeled activity, that detection is classified as a false positive.

• If a day receives a false value for a detection and it is not labeled, that detection is classified as a true negative.
• If a labeled health incident does not have any detections within one day of the labeled dates, that health incident is classified as a single false negative.

Using our activity monitoring system and our ML algorithm, we find that we’re able to detect labeled activity related to reduced movement in cattle within one day 77% of the time. This gives us a corresponding false negative rate of 23%. We are able to accomplish this detection rate while maintaining a false positive rate of only 7%. It is important to note also that this model has continuously improved over time with repeated training. We expect that this trend will continue as more data is added and additional seasonal effects can be captured.

Figure 2. An example of an algorithm detection on a sharp decline of an individual (blue) animal’s activity relative to others within the same premise (gold), followed by a detection from the ML algorithm (red).

Figure 3. Model performance as a function of the number of months used as training data.

With a false negative rate of 23%, some periodic surveillance of cattle by pen riders is still necessary to catch health issues which are not detected by the ML model. This is in part because there were cases where an unhealthy animal has been treated, but there were no noticeable changes in the accelerometer data being collected. Some of these cases can possibly be attributed to animals which do not show signs of lethargy, but instead only display more visible symptoms such as nasal discharge or coughing. However, it is also possible that some of the treated animals did not show anomalies in their data because they were incorrectly pulled by a pen rider to be treated. In the treatment datasets that we reviewed, we saw that on average around 50% of the treatments given did not have any visible changes in movement activity. This further supports the claim from previous studies that clinical diagnosis is incorrectly given to healthy calves up to 37% of the time.³ The low false positive rate of this model therefore provides an opportunity to remove some of the subjectivity in traditional pen riding methods which may be responsible for misdiagnosis.

With a true positive rate of 77% we find evidence that our ML algorithm is picking up on the same patterns in the data that are labeled by subject-matter experts. This allows larger facilities with thousands of cattle to more accurately and efficiently allocate resources directly to cattle which are showing signs of behavioral changes related to health issues. From the previous research showing that up to 38% of diseased animals may go undiagnosed, we see an opportunity to use this solution to alert to behavioral changes in movement related to disease that may be missed by pen riders.³ In the cases where these changes in activity are related to diseases such as BRD, this means that we can also prevent potentially devastating economic impact from increased death loss.

1. Chirase, N., & Greene, L. (2001). Dietary zinc and manganese sources administered from the fetal stage onwards affect immune response of transit stressed and virus infected offspring steer calves. Animal Feed Science and Technology, 93(3–4), 217–228. https://doi.org/10.1016/s0377-8401(01)00277-2
   
2. Timsit, E., N. Bareille, H. Seegers, A. Lehebel, and S. Assié. 2011. “Visually Undetected Fever Episodes in Newly Received Beef Bulls at a Fattening Operation: Occurrence, Duration, and Impact on Performance 1,2.” Journal of Animal Science 89 (12): 4272–80. https://doi.org/10.2527/jas.2011-3892
3. White, B. J., & Renter, D. G. (2009). Bayesian estimation of the performance of using clinical observations and harvest lung lesions for diagnosing bovine respiratory disease in post-weaned beef calves. Journal of Veterinary Diagnostic Investigation, 21(4), 446–453. https://doi.org/10.1177/104063870902100405
4. Richeson, J. T., Lawrence, T. E., & White, B. J. (2018). Using advanced technologies to quantify beef cattle behavior.  Translational animal science, 2(2), 223–229. https://doi.org/10.1093/tas/txy004.
5. Dittrich, I., Gertz, M., & Krieter, J. (2019). Alterations in sick dairy cows' daily behavioural patterns. Heliyon, 5(11), e02902.  https://doi.org/10.1016/j.heliyon.2019.e02902
6. Smith, J. L., Vanzant, E. S., Carter, C. N., & Jackson, C. B. (2015). Discrimination of healthy versus sick steers by means of continuous remote monitoring of animal activity. American Journal of Veterinary Research, 76(8), 739-744. Retrieved May 21, 2024, from https://doi.org/10.2460/ajvr.76.8.739
7. TSFEL: Time Series Feature Extraction Library. SoftwareX 11 (2020). https://doi.org/10.1016/j.softx.2020.100456.
8. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat Methods 17, 261–272 (2020). https://doi.org/10.1038/s41592-019-0686-2