AI Beyond Artificial Insemination: Wyoming Scientist Revolutionizes Cattle Heart Disease Detection
Table of Contents
A University of Wyoming researcher is pioneering the use of artificial intelligence to combat congestive heart failure in cattle, a growing threat to the beef industry with important economic implications.
For decades, “AI” in the cattle industry has meant artificial insemination. But for Chase Markel,a PhD student at the University of Wyoming,the acronym now represents a groundbreaking approach to animal health: artificial intelligence. A lifelong cattle producer and Wyoming native, Markel is leveraging the power of machine learning to improve the efficiency and accuracy of diagnosing a debilitating heart condition in beef cattle.
Markel’s research focuses on congestive heart failure, a condition often linked to pulmonary hypertension – also known as high-altitude disease or brisket disease – which is becoming increasingly prevalent in finishing beef cattle. “It’s a tricky disease, and there’s no easy solution,” Markel explains. “The more that you learn, the more that you realize you don’t know.”
His work suggests that the economic impact of subclinical cases – where animals survive but experience reduced performance – may actually outweigh the losses from direct mortality. “The main economic impact of this disease likely is not just death loss, but more so the loss of these production efficiencies in terms of live animal growth performance, carcass quality, and then sensory attributes of
Evaluating Heart Images for Risk Factors Associated with Congestive Heart Failure.
The foundation of this AI model was a massive dataset of manually scored heart images. Markel painstakingly analyzed nearly 1,000 images taken from commercial processing plants in Nebraska and Colorado, using a 1-5 scoring system developed by Tim Holt, a professor at Colorado State University and a key collaborator. In Holt’s system, a score of 1 indicates a normal heart, while a score of 5 resembles a “deflated volleyball.”
The resulting data was then fed into the AI model, training it to replicate the scoring process. So far, the experiment has yielded notable results: the model has achieved 92% accuracy in identifying the correct score for previously unseen images.
Markel is already planning to expand the dataset to 15,000 images, recognizing that variations in heart size, shape, and appearance – influenced by factors like plant processing techniques and animal breed – require a more robust training set. “there’s all types of shapes and sizes of hearts,” he says. “Things look very different in different plants with different lighting and different animals.”
Addressing Subjectivity and Expanding Applications
While the 92% accuracy rate is promising, Markel acknowledges the model’s limitations. The initial training relied solely on his own subjective assessments. To improve the model’s reliability, he plans to incorporate scoring data from other researchers.
Despite these challenges, Markel views his work as a prosperous “proof of concept.” “Hopefully someday somebody can actually use this in the industry,” he says. The immediate application lies in improving efficiency within meat-processing plants. Currently, these facilities excel at collecting data relevant to their bottom line, but often miss crucial facts about subclinical conditions.
Markel believes his image classification system – or similar AI-powered tools – could help plants pinpoint which carcasses are adding value and identify potential disease-related factors impacting quality. He also envisions future iterations benefiting Wyoming producers directly, leveraging the wealth of data they already collect.
“Producers themselves collect data every day,” he points out. “They might do it on the back of a piece of notebook paper, but that data has a lot of value.” Markel advocates for incorporating AI models and machine learning into conventional data analysis methods to unlock new insights and improve profitability for producers.
“As researchers, we need to start incorporating those tools into our research and try to help build that technology so producers and people out in the industry can actually utilize those tools and help improve their bottom line,” he concludes.
UWagNews
