With a spike in equine fatalities at Churchill Downs this spring, racehorse safety is once again in the spotlight. Dr. Michael Hardy, longtime regulatory veterinarian and executive director of the Racing Medication and Testing Consortium, recently outlined two systems that could use new technology to help prevent injuries.
Speaking at this week’s Track Superintendent Field Day at Horseshoe Indianapolis, Hardy provided a short update on the use of StrideSafe at Churchill Downs. As we reported at the end of 2022, officials at Churchill had planned to conduct a study using the sensor system well before national media began monitoring a spate of breakdowns at the facility.
StrideSafe had its first significant test in New York, where it was placed on every runner starting in summer 2021. The sensor system is attached to a saddle towel and measures concussion and acceleration across all four of a horse’s limbs. Its goal is to detect a departure in the horse’s typical way of going which could signal an impending injury. The system provides each horse a green, yellow, or red rating for its performance, with green ratings indicating no problem, red indicating a serious departure from normal motion and yellow indicating a potential problem.
In New York, the system had been used observationally and retrospectively, meaning officials recorded each horse’s rating and whether or not the horse suffered an injury in-race after the races had been run. Officials indicated the system was able to correctly identify 90 percent of fatal injuries that happened on the circuit in the time it was used, meaning that 90 percent of horses who suffered breakdowns got red ratings in that final race.
Churchill’s goal for their study of StrideSafe was to test the system in action by providing trainers and veterinarians with information about their horses’ yellow or red ratings after a race to allow them to do follow-up investigation to look for any physical issues. The project involves cooperation from Kentucky Horse Racing Commission veterinarians and will include data analysis from Dr. Warwick Bayly of Washington State University.
Support our journalism
If you appreciate our work, you can support us by subscribing to our Patreon stream. Learn more.
While the research at Churchill is ongoing, Hardy said initial feedback from trainers and veterinarians has been positive. Trainer Dale Romans recently told The Blood-Horse about a horse in his barn whose StrideSafe data suggested a possible abnormality. The horse underwent a PET scan, which turned up bony changes that suggested a condylar fracture could have been imminent.
Hardy said Churchill officials plan to continue using StrideSafe on runners at Ellis Park, a track the company also owns.
Video analysis may also be part of the next generation of injury prevention. At Keeneland, Hardy said the track safety team has 17 cameras in various places around the racing surface. These allow veterinarians to pull up imagery of a horse they may have concerns about to get a history on how that horse has been moving in the mornings, even if those veterinarians weren’t actively watching the horse in real time. This can also be useful in retrospective looks at a horse who may have suffered a serious injury in a race, to help veterinarians figure out what to look at for warning signs going forward.
Hardy also outlined another system in development by 1/ST Racing which uses video technology to identify abnormal motion from a horse. 1/ST recognized that given the national shortage of equine veterinarians, there may not be enough eyes watching horses for gait abnormalities, especially in the morning. The organization has installed a huge number of sensors and cameras around its Santa Anita facility to monitor both racing and training and the company is working on a series of models to collect data points from videos of moving horses. The developing system uses 15 models to collect and analyze 350,000 data points per race, studying the way at which different joints and limbs on the horse are interacting with the ground. The system can use AI to generate a 360-degree view of each horse as it moves across the track.
Hardy said the system has encountered its challenges – the size of a track can complicate cameras’ ability to get views on all horses, lighting can be a challenge, and busy periods in the morning can also obstruct views of some horses as they move around each other. Also, existing modeling technology was evolved to study human movement, not necessarily to detect changes in equine motion.
Currently, 1/ST is working to input information from veterinary observation to the artificial intelligence system to help it “learn” what problematic motion looks like. The full system is in place at Santa Anita and Hardy said it will be coming to other 1/ST tracks soon.
“The idea is the system will automatically identify high-risk workouts,” said Hardy. “The relentless pursuit to do better is the concept here, and it certainly takes a collective effort.”
The company is also developing an app that will share sensor information with owners and trainers after collection to help them make strategic conditioning and veterinary decisions.
Autor Natalie Voss