A horse walks into the vet's office

The figures don't lie. Technological advancements are making it increasingly easy to accurately measure horses' movements. Gait analysis has been widely used on humans for some time now.

A horse walks into a vet's office... If the patient in this joke were human, it would go like this: ‘Doctor, my right lower leg hurts’. Horses can't talk though. ‘That makes it hard to determine the cause if, for example, they have become lame’, observes Filipe Serra Bragança, a researcher at the Faculty of Veterinary Medicine. ‘The animal's owner may be able to talk, but that doesn't necessarily mean they can adequately pinpoint the problem. The horse's owner may believe the lameness is in the front left leg even though it later turns out to be in the right hind leg.’ It's up to the vet to make a diagnosis, but that's not always easy, as Serra Bragança points out. ‘A team of the world's best vets might examine a horse and still not agree on the exact source and severity of a lameness.’

That's why we need gait analysis to objectively determine the source of the pain. Recent years have seen major advances in this area, as Serra Bragança enthusiastically explains. ‘We now have access to a few systems to help us study movement in horses; for example, Utrecht University and two of our partners have developed the Equimoves measurement system to enable quicker and better diagnostics.’

Horses and Parkinson's sufferers

Human medicine has been somewhat more successful in applying these technologies. In fact, it works pretty much the same way it does with animals, explains Jaap van Dieën. Van Dieën is a professor and head of the Movement Sciences Department at VU Amsterdam. ‘We use camera systems to record people's movements, just as we do with horses. In most cases, we also film them while walking on the treadmill so that we can measure a large number of steps. These days, we also measure the forces exerted by the feet, which gives a very complete picture. In terms of technology development, they're currently working hard to make the system more accessible to a broad group of users. For example, they've started using cheaper sensors instead of cameras. We're also seeing a trend towards gathering big data on a lot of people with a wide range of conditions. After all, if you can link a patient to a large database, you might only need three sensors instead of 20.’

The clinical aspects of our research into lameness are very subjective. That kind of subjectivity is also at play in human medicine.

When it comes to horses, most movement analysis is currently conducted on professional racehorses. Applications in human medicine are a lot broader. Van Dieën is mainly focused on clinical applications. ‘For example, we use movement analysis to study cerebral palsy patients. We also use it to analyse falling issues among the elderly, people who have had a stroke or patients suffering from Parkinson's disease. We're trying to find out how healthy people manage to avoid falling and figure out what's going wrong in these patient groups. Could we eventually recognise the problem early on and intervene on time? Other than that, it's always difficult to adjust medication levels accurately.

If you can tell whether the medication is at the right level from a movement pattern (more quickly than you could on the basis of subjective observations), you can make the necessary adjustments. We're also focused on early diagnostics. Parkinson's tends to be discovered at a very late stage, whereas people may have had vague symptoms for years and early treatment could probably have achieved much better results. We hope movement patterns will ultimately allow us to detect diseases or problems in the early stages. That sort of knowledge can then also be applied to horses.’ Early diagnosis would also be a welcome development in equestrian medicine, Serra Bragança agrees. ‘Unfortunately, we're still years behind human medicine when it comes to gait analysis. We've basically just started using the technology on horses, so I think there's huge potential across the full spectrum of equestrian sports: from riding school horses to professional athletes.’

Grand Prix dressage rider Jolanda de Pijper has been following the developments with great interest. De Pijper has international ambitions for her Bombay. ‘Bombay already knows all the exercises, so now we're just focusing on fine-tuning everything and strength training – we actually train five times a week. I'm terrified he might get injured: that would be back to square one. The earlier you detect and locate an injury, the better. That will obviously benefit his performance, but it's also good for the animal's wellbeing – as the rider, I'm also responsible for his health. New technologies might allow us to measure a horse's condition on a monthly basis. If a particular leg turns out to be weaker, you can work on it and draw up a training plan. 

Entrenched ideas

As Serra Bragança sees it, the greatest benefit of technology lies in greater objectivity. ‘The clinical aspects of our research into lameness are very subjective. As a result, everything that follows – including the medications used to treat the condition – tends to have insufficient scientific basis. The subsequent analysis to determine whether the situation has improved through medication is also subjective.’ In Van Dieën's view, this subjectivity also affects human medicine. ‘We tend to have very entrenched ideas about how things work. For example, clinicians invariably claim that differences in leg length are a huge problem that will eventually lead to back pain. However, when you start measuring, you find that people can often compensate those differences quite well and that there is no evidence of any link with back problems. Objective measurements are primarily a way of debunking entrenched views.’

Hands-on expert De Pijper believes dressage would also benefit from more objective methods. ‘It can be so hard to pinpoint exactly what's wrong with a horse, even with all those experts weighing in. In fact, all those different opinions may actually be the problem. If you put ten people in a room, they'll all have different opinions about an injury. If technology can help us reach a consensus more easily, that's definitely a bonus.’

I'm terrified he might get injured. The earlier you detect and locate an injury, the better

Combining knowledge

So could human medicine – despite its head start in this field – learn something from animal movement analysis? ‘Absolutely’, Van Dieën says. ‘When it comes to animals, there's more anatomical variation between the species; that tells us a lot about underlying and evolutionary principles. The restrictions on technology development are also a little less strict when it comes to research on animals.’ Serra Bragança: ‘I think we could collaborate more closely and gather and share ideas. We're basing a lot of our research decisions on existing knowledge on humans. If something works in humans, I'll try it too. With the help of our colleague Van Dieën, we have now also started to measure horses' movements using techniques that have already been applied to humans for some time.’ ‘Those are basically just scaled-down versions’, Van Dieën explains. ‘There's not really that much difference. It's obviously easier to give instructions to people, but it's definitely also doable with horses. Horses can be trained quite successfully, Serra Bragança agrees. ‘After all, they have been used to being led, trotted and lunged from an early age. We want to make the technology even more user-friendly, so that you can measure horses during training with a focus on prevention. We're already offering the option of movement analysis to the owners of our patients at the Academic Veterinary Hospital, but that still doesn't tell us how these horses move in their own environment. That's why we would like to be able to measure during training sessions, using the same big data approach they want to apply to people. We can then apply machine learning to optimise the horse's training routine.’

A horse walks into the vet's office... The fact that that joke set-up doesn't segue into a conversation isn't strictly a disadvantage, Van Dieën argues. ‘People can talk, that's true, but patients' descriptions of their own symptoms can also be misleading. After all, pain is also a psychological phenomenon, a matter of perception. If you ask a human patient to point out their pain, the answer doesn't necessarily tell the whole story.’

This is an article from Vetscience issue 11 (in Dutch).