In this month's Science Corner, the cutting edge of technology is to the fore. We look at whether automation can both benefit horse welfare and reduce labour, if machine learning can help vets more objectively grade scopes, and if a smartphone app can accurately map back movement in a trotting horse. Plus, how beneficial are fly rugs really?
Each month, we break down five recently published scientific studies that could affect the way we breed, train, and race Thoroughbreds. We explain what the research found and what it means for you.
You can click the title of each paper to read it in full.
#1. Would automation actually be better for feeding hay?
What’s it about?
This two-phase study, conducted by researchers from the University of Turin and the Universita di Bologna, looked at the effects of automated feeding devices (AFDs) on behaviour and wastage versus manually feeding hay. In the first phase, groups of horses were evaluated on their behaviour when faced with either feeding strategy, and in the second phase, individually housed horses were fed with AFDs utilising varying rates of how long they stayed open.
Key findings:
In the group setting, the AFDs led to more residual hay after feeding, compared to no residual hay for ground feeding, and social hierarchy strongly influenced who actually got to access the AFDs.
In the individual setting, short opening schedules for the AFDs led to more stress behaviours than the longer schedules and more residual hay. The medium opening schedule appeared to strike the best balance between regulating intake and mitigating stress behaviours.
What it means for you:
Automation technology is very common in other livestock settings, but less so in horses. This study offers some food for thought on the suitability of using automated feeders to reduce labour in a stud or stable environment. It's not a complete write-off as an idea, but it definitely requires some careful consideration.
#2. Machine learning shows promise to simplify grading scopes
What’s it about?
Thousands of hours each year is dedicated to analysing scopes, and despite a new grading scale being released this year, it is still a very subjective exercise. This paper presented at the 46th Bain Fallon Memorial Lectures this year by Dr Josephine Hardwick and team looks at whether machine learning can be used to help standardise grading laryngeal function.
Veterinary consensus graded 4605 post-sale scope videos before the same dataset was analysed using a machine learning pipeline, which used part of the dataset as training.
Key findings:
When looking at fresh scopes it had not seen in training, the machine learning model matched the veterinary consensus for grade 52% of the time and matched consensus for a "pass/fail" mark 82% of the time.
The distribution of quantitative features on a scope video had a lot of overlap between grades, which shows how hard it can be to separate the grades out.
What it means for you:
The work presented in this paper demonstrates that there is promise to machine learning being able to help veterinarians come to more objective conclusions in the future. When a scope can make or break a sale, it could be very useful to have a completely objective measurement of a video to give more buyers and vendors confidence.
#3. Fly rugs keep flies OFF!
What’s it about?
The fly rug is a common piece of kit to keep troublesome warm weather visitors from bothering horses, but how good of a job do they actually do? Three University of Tennessee researchers have put it to the test, alternating fly rug wearing on five horses over a 10-week period while routinely measuring cortisol levels, fly avoidance behaviours, heart rate, and temperature.
Key findings:
Fly rug use significantly reduced fly avoidance behaviours.
Mean cortisol levels were lowest with the rugs off and heart rate was highest with them on, but body temperature was significantly lower with the rugs on.
Horse wearing a fly sheet as part of the study | Image courtesy University Of Tennessee
What it means for you:
The researchers acknowledge their small sample size means this study doesn't carry a lot of weight, however the significant reduction in fly avoidance behaviours does point to the rugs doing what they are supposed to do: keeping flies off!
#4. Smartphone tech moving strides closer to accurately identifying causes of pain
What’s it about?
In the followup to a study published - and shared in Science Corner - earlier this year, researchers at the University of Copenhagen have once again tested how well smartphone assymetry detection app RealHorse® can quantify thoracolumbar flexion-extension range of motion in the hopes it can help with diagnosing the cause and location of equine back pain.
Key findings:
When RH was compared to Qualisys®, the trial-level agreement between the two systems was high for horses trotting in a straight line, and a little lower when looking at the same horses on a straight line at stride-level.
However, there was moderate variability when looking at horses moving on a circle.
Smartphone data collection setup | Image courtesy of University of Copenhagen
What it means for you:
There is promise to using RH to objectively assess back movement, but more work needs to be done to quantify the correct decision thresholds. If this could be really improved, then it could serve as another tool in a vet's diagnostic toolkit to help assess and pinpoint lameness.
#5. Unwanted bacteria in the reproductive tract is becoming less sensitive to ceftiofur
What’s it about?
When breeding is such a time sensitive endeavour, every positive swab is a costly setback. In this study, a team from Murdoch University retrospectively examined reproductive tract cultures covering 767 isolates from mares in Western Australia, and compiled an antibiogram of what bacteria they found, their frequency, and what antibiotic each bacteria was most susceptible to.
Key findings:
Escherichia coli (E. coli) was the most common bacteria culturd and appeared in 36% of the isolates, while Streptococcus appeared in 31%.
Gram-negative isolates were most susceptible to high importance antimicrobials ceftiofur and enrofloxacin, however the susceptability of both Gram-positive and Gram-negative isolates to ceftiofur decreased by 10 points or more (79% to 69% and 91% to 75%) across the five years of isolats studied.
Reproductive tract isolate antibiogram | Image courtesy University of Murdoch
What it means for you:
What is probably most striking is the decrease in how good of a job ceftiofur can do in clearing up bacteria in the reproductive tract. This serves as a reminder to be careful with how we use antibiotics and antimicrobials - we need the ones we have to work well for as long as possible, and not just in mares. Good vets and farms will do sensitivity tests before administering antibiotic treatment.
