Skip to main content
Skip to navigation

This site is archival. Please visit the current MIZZOU magazine site for up-to-date content.

Getting a leg up

An MU veterinarian invents a device to diagnose horse lameness

MU horse

Veterinary medicine student Emily Bond jogs with Gunner, who wears two acclerometers and a gyroscope to diagnose lameness. As Gunner runs, the lightweight devices on his head, rump and right front foot transmit information to a computer with software that helps pinpoint the problem. Photo by Rob Hill

When one of us two-legged patients goes to the doctor with, say, a sore elbow, we give our clinicians some critical, if basic, information. We point to the joint. We say, “It’s this one, doc.” 

But horses, with the possible exception of Mister Ed, are less forthcoming. That leaves veterinarians to figure out the problem using their own two eyes and whatever training and intuition they can muster. Diagnosing and treating subtle lameness is the most common task equine vets face day in, day out. “Even small injuries really affect how well horses can do their work,” says Kevin Keegan, BS Ag ’78, DVM ’83, who directs MU’s E. Paige Laurie Equine Lameness Program. 

Horses in the 21st century likely work as one of the nation’s 3.9 million mounts for pleasure riding or 2.7 million animals that compete in shows. The economic impact of recreational riding and showing horses is $60.6 billion a year, according to the American Horse Council. If you throw in horses employed in racing and other pursuits such as workers on farms, ranches and police beats, the impact could be as high as $102 billion. 

With stakes that high, horse owners have their own lobby in Washington, D.C., and equine vets spend lots of time watching lame horses trot back and forth to gather clues about which joint or muscle or tendon or ligament is the culprit. 

There’s always a catch

But there’s a fatal flaw with this scenario, Keegan says: Focusing a trained eye on a horse with subtle lameness isn’t always a reliable way to start the diagnostic process. Keegan noticed the problem long ago and, with colleagues, has been conducting research on it since joining the College of Veterinary Medicine faculty in 1990. Keegan recently went to market with diagnostic hardware and accompanying software that could be a boon to equine practitioners. More on the new technology shortly. 

In the meantime, it’s worth sketching the path Keegan has broken over the past 19 years in turning his observation into research questions and then turning his research results into a useful and marketable product. More and more faculty members at MU are breaking their own trails in markets ranging from nanotechnology to antacids. When startup companies do well, the researchers, their investors and the university make money. Plus, society benefits from new products and services — many of which improve quality of life. In fiscal 2008 alone, MU brought in $6.2 million in licensing fees from projects like Keegan’s. 

It all started with an observation: It’s tough to pick up subtle lameness. “I have always noticed that, in horses with mild to even moderate lameness, there is a significant amount of disagreement between veterinarians about whether a horse is lame and in which leg, and I always thought it would be better to have a more objective system.” The next step was to confirm this observation and get some idea of how unreliable the human eye is for this task.

Horses abandoned in increasing numbers

Kevin Keegan works hard to ensure that veterinarians are better equipped to diagnose and care for horses. But as the number of abandoned horses in Missouri and the U.S. are on the rise, MU faculty members Robin Hurst-March and husband Zac March help ensure that unwanted animals receive the care they need to survive.

In September 2007, the United States Court of Appeals upheld the state of Illinois’ decision to ban the slaughter of horses for human consumption. At that time, Illinois was home to the last remaining slaughterhouse in the country. The court’s decision effectively ended the slaughter of horses for food in the United States.

“This [court ruling], plus the tough economic times, has lead to desperate actions on the part of some horse owners, such as simply abandoning their unwanted horses in national parks or along roadways,” says Hurst-March, professor of biological science. She and March, who is director of information technology for the College of Veterinary Medicine, founded and run Out 2 Pasture, an organization that rescues lame or injured racehorses. “We work for one of the most recognized rescue organizations, and this surge in horses whose owners can no longer afford to care for them has us (as well as most) busting at capacity.”

In addition, Hurst-March says, this “perfect storm” of circumstances is difficult for horse rescue operations around the country. In these tough economic times, donations to horse rescues have plummeted, which means that many horse rescues will no longer accept a horse without some kind of financial support from the owner to help offset costs.

Keegan started by conducting one of the field’s first big studies. He made videotapes of lame and healthy horses running on a treadmill, and he sent them to experienced vets and vets fresh out of school. None of the respondents knew ahead of time which horses were supposed to be injured and which were supposed to be sound, and none knew that Keegan repeated horses on the tape. It turned out that experts agreed with themselves very well, better than vets fresh out of school. But they did not agree with each other — “well below what is acceptable,” Keegan says. “That didn’t sit well in the veterinary community.” 

Vets pointed out that, in practice, they don’t diagnose videotaped horses running in a straight line on treadmills. Plus, there’s more to assessing an injury than watching horses run. Fair enough. In response, Keegan and colleagues designed a more robust study, this time not using videotape but live horses. This study also included more horses and a fuller assessment of lameness. For instance, horses trotted not only straight but also in a circle, and vets put them through other standard examinations. Even when performing full, live lameness evaluations, there was still significant disagreement among experts. When asked which leg was not lame, vets agreed about half of the time, Keegan says. But strictly by chance they would have agreed 20 percent of the time anyway. In the end, the message was the same as in the first study: The standard procedure doesn’t cut it. 

Now Keegan and his colleagues were ready to start looking in earnest for a better way. “In 1993, we moved into a new facility [MU’s Veterinary Medical Teaching Hospital], and the college purchased state-of-the-art motion analysis equipment.” Keegan put reflective spheres on various parts of horses and filmed them with high-speed cameras as they ran on the treadmill. The cameras picked up how the spheres moved and fed the data into software that created moving 3-D images of the horses’ strides. “The human eye samples about 15 times a second, but the cameras captured 120 frames a second. We just started filming many different horses, some we knew were sound and some we knew were lame.” 

This exercise gave Keegan a wealth of data — everything from stride length to joint angles. “But I realized that I still didn’t know the best thing to measure,” he says. So he called on Marjorie Skubic, associate professor of electrical and computer engineering. She helped him sift through the mountains of data and look for useful patterns. This data-mining unearthed what turned out to be a breakthrough finding.

Some veterinarians look for clues by observing the distance a certain leg swings to the side or how much a joint flexes. But Keegan’s data pointed elsewhere. “We found that the pattern of how the torso moves up and down was key,” he says. Not surprisingly, the general pattern of vertical movement is tied to how much other parts of the horse rise and fall as well. 

This new factor — vertical movement — was important but still too general to be useful to practicing veterinarians. So, Keegan then enlisted the expertise of another engineer, Frank Pai, C.W. LaPierre Professor of Mechanical and Aerospace Engineering at the University of Missouri. Pai took data analysis methods he had developed for finding cracks in airplanes and adapted them for the stride data. Pai’s analysis gave Keegan the ability to use motion analysis to objectively measure lameness in horses. 

Even though Keegan and Pai had succeeded with this finer-grained analysis, the results only helped diagnose horses that could be brought to the college, trained to run on a treadmill, decked out with reflective spheres, filmed with high-speed cameras and then analyzed. That’s too costly, time consuming and cumbersome. Keegan needed a quick, easy and affordable way to get the data, but he wasn’t sure what to do next. 

Then in the winter of 2000, Keegan went to a conference and presented the latest work he had done with Pai. “It was a bioengineering conference. I believe I was the only equine veterinarian there,” he says. “After the talk, a Japanese engineer, Yoshiharu Yonezawa, came up to me and said he had an idea for how to get the data without a treadmill. We had a little trouble communicating. There was a slight language barrier. So, we just sat down and tried to communicate, mostly by drawing on the hotel’s napkins. About six months later, he sent me a package of electronics — two sensors for feet, one for the head and another cell phone-like device that sat on the horse’s back.” 

It worked! Keegan could quickly strap the sensors on a horse, set the creature trotting, and the sensors would radio data to his computer. Since then, he has refined the device. For instance, one of the foot sensors turned out to be unnecessary; all the sensors have gotten smaller and the software for picking up lameness has improved. The device was an excellent aid for live evaluation of lameness in horses.

Then in 2006, Keegan hit another roadblock: money. Until then, he had been funding his project with small grants from foundations, MU and the College of Veterinary Medicine. This sort of seed money is available for research, but once the device worked and he started refining it as a commercial product, all Keegan’s labors came under the heading of development. The seed money for research dried up, and he had to look elsewhere. 

Help on the way

Enter Jake Halliday, director of the Missouri Innovation Center, an affiliate of MU that has provided business assistance to faculty entrepreneurs since 1986. The center helps faculty members move their ideas from lab (or barn) to market. Halliday’s MBA class in the Trulaske College of Business worked up a market analysis and business plan for Keegan’s company, Equinosis, and helped him present it to Centennial Investors, a Columbia-based group that ventures capital on startup companies. Centennial put up $330,000 to sustain the company through its startup phase. The money goes toward manufacturing the hardware, developing software, patent expenses, license payments to MU and general corporate purposes. Halliday also hooked Keegan up with The Incubation Factory in St. Louis, which provides management experience and other support to accelerate a product’s path to the marketplace. 

Ownership of Equinosis is shared roughly equally by three groups — the inventors, Centennial Investors and The Incubation Factory. The Missouri Innovation Center also has a minor ownership stake. 

A licensing agreement gives MU an up-front fee and reimbursement for its investment to patent the technology. MU receives product sales royalties, one-third of which go to the inventors.

“The process of technology transfer and product commercialization can be win-win-win for all involved,” Halliday says. “The inventors are rewarded for their innovation; the university and the public obtain a return on their investment in research infrastructure; investors are rewarded for supporting high-risk startup companies; and the community gains a new company that over time will create high paying jobs and contribute to the tax base and quality of life in the area.”

Bottom lines

The Lameness Locator system is priced in the $10,000 to $20,000 range, roughly the cost of an ultrasound machine, which is standard in veterinary practices. 

And the prospects? The company’s business plan projects worldwide sales of around $30 million. If that pans out, the company could be valued at about
$90 million after five years.

Of course, it’s the big if that keeps inventors and investors up at night. “A lot of private practitioners, especially the old timers, are skeptical of what comes out of universities,” Keegan says. “But word is getting out. And students will grow up with this in veterinary school; they’ll want to have it in their practices.” 

Share your comments with Mizzou magazine at

Note: If published, feedback may be edited for length, style and clarity.