Imagine having the knowledge to prevent a catastrophic event from happening prematurely. In horse racing, this can be an opportunity.
A new study examines whether the same systems used to help players choose a winning racehorse can provide the data needed to protect that same racehorse from injury.
Previous efforts to predict racehorse injuries have not been much better than tossing a coin. In the past injuries in horse racing have been treated as a binary result – the horse is either wounded or unharmed.
But injuries, largely as a result of bone damage, have a gradual onset.
The injury can develop over weeks or months, unless it is due to a traumatic event (such as a horse crashing into a fence), so it rarely occurs on the day it is observed. We know this because the majority of catastrophic injuries in racehorses show evidence of existing bone damage.
This damage accumulates during training and competition over time due to repeated loads on the musculoskeletal system. With each step made by a horse galloping at a moderate speed, loads of up to four tons were measured on the surface of the joint.
The bone can only withstand a limited number of these loads, and steps taken at faster speeds lead to greater loads.
Often, the initial detection of an injury is when the horse is damaged or shows signs of lameness, indicating that the threshold for bone damage has already been reached.
But instead of waiting for this trauma to become apparent, we realized that a way was needed to measure horses’ response to exercise and competition loads.
The dawn of the data-driven era
So what if there was a way to measure the onset of an injury using already established data collection systems? As it happened, there is.
It all started in 2010 when Tasmania’s main racing body, Tasracing, partnered with StrideMASTER, a start-up technology company developing training monitoring systems for the racing industry.
They developed one of the first breed-Daily timekeeping systems using GPS and accurate data from a motion sensor. The original purpose of this data was for real-time sectional time and racehorse position data, which were then intended for use as a broadcasting and betting product.
I worked with Tasracing before and in 2016, when I returned to Australia, we reconnected.
Because Tasracing prides itself on being data-driven, it is this approach that has provided us with a potential solution.
With access to biometric data – such as racehorse speed, stride length and stride frequency – we now have tools to measure changes in horses’ careers that can signal injury before it’s too late.
A horse that slows down is a horse to watch
Initially, our study compared StrideMASTER’s biometric data with other racial field information and veterinary findings provided by Tasracing between 2011 and 2016.
Using a statistical method that had not been used before in this setting, we first modeled the changes in stride characteristics on successive starts of the race, and then modeled the number of starts of the race before injury.
Finally, the two models were integrated into what is called a “joint model” to determine whether the observed changes in stride characteristics predicted injury.
And they did.
For horses at the beginning of their careers, those injured during the start of the race slow down their speed and reduce their stride length faster than six races before.
My colleague, Professor Chris Wheaton, noted that although we expected to see changes in speed and stride in races leading to injury, the fact that we had seen these changes so long before was surprising. Yes, we thought we could see changes, maybe one or two races, but not six – that’s pretty amazing.
For horses that have suffered an injury earlier in their careers – if we look only at the first few races for which there is still insufficient data on their normal performance on the pitch – additional monitoring during training may be needed to predict these early career injuries.
The risk of injury to the horse increases by 18 percent when it reduces its speed by 0.1 meters per second, or by 11 percent when it shortens its stride length by only 10 centimeters in the last part of each race.
And our findings are retained even after taking into account factors affecting speed and stride, such as the distance of the race and the conditions of the track.
Although these changes may sound small, within a single horse that has its own unique stride, they are a sign that something may be wrong.
This may be due to the fact that they cannot withstand the load, experience pain or otherwise compensate for the physiologically accumulated bone damage.
Veterinary advice should be sought at this time.
Established systems used in a meaningful way
Our study published in Veterinary magazine for horsesis just the dawn of the way data-based research can improve the well-being and safety of the racing industry by using new and meaningful ways to redirect data.
Our findings also demonstrate the enormous potential for identifying and preventing injuries in racehorses before they become catastrophic.
Competitive authorities must take the lead and insist on the wider application of motion sensor technology, which can monitor both training and competition. This will cover elements for which we are not yet clear – for example, whether the same deceleration and shortening step also occurs during exercise.
Similar comparison systems are now used in other jurisdictions biometric datathere will be more information, which will lead to improved algorithms, improved predictions and ultimately greater safety for horses.
Adelene SM Wong et al, Changes in the speed and stride characteristics of purebreds in successive starts of the race and their relationship to musculoskeletal injuries, Veterinary magazine for horses (2022). DOI: 10.1111 / evj.13581
University of Melbourne
Quote: Study of biometric data that could predict a racehorse injury (2022, June 20), retrieved on June 20, 2022 from https://phys.org/news/2022-06-exploring-biometric- racehorse-injury.html
This document is subject to copyright. Except for any fair transaction for the purpose of private research or study, no part may be reproduced without written permission. The content is provided for informational purposes only.