Equipped with sensors that mimic human skin and an AI system that learns your every move, could this new generation of safety wearable make back injuries a thing of the past?
Musculoskeletal disorders (MSD) are the most common occupational health problem in the world, but they are also one of the hardest to mitigate.
The world of work can be very physically demanding and outside of office ergonomics, industrial workers face the risk of manual handling injuries, sprains and repetitive strain injuries.
MSDs are injuries or pain in the body’s joints, ligaments, muscles, nerves, tendons and structures that support limbs, neck and back. The most common type of MSD is back pain, which can be caused by a variety of factors, including heavy lifting, repetitive motions and poor posture, all of which can be exacerbated by workplace practices and ergonomics – a lot of which we do without necessarily being aware of it.
Traditional approaches to preventing back pain in the workplace have focused on education and training programs and ergonomic assessments, as well as invasive monitoring technologies and the use of personal protective equipment (PPE), such as back belts. While these methods can be helpful, they all have their limitations and rely on workers actually using them successfully and comfortably, leaving a lot of room for human error.
In recent years, there has been growing interest in the use of technology to prevent back pain in the workplace. Wearable sensors, such as accelerometers and gyroscopes, can be used to track worker movements and provide real-time feedback on posture. This feedback is claimed to help workers identify and correct potentially harmful movements before they become repetitive and lead to injury.
While these may provide some benefits, even these wearable sensors only go so far because they require specific angle and rotation measurements to operate, which may not accurately reflect the real pressure and force being exerted on the back. However, what holds the greatest promise is the nature and relentlessness of tech advancements, that no matter the current limitations, a better option is probably not far away. This is the case with a new type of wearable device for back protection from a team that paired sensors with machine learning to produce an AI-powered safety wearable.
It’s got your back
Mimicry and real human behaviour is at the core of this new type of back device; rather than correct wearers according to a set of predetermined angles, it has a closer match to a skin-muscle system to help ‘learn’ how an individual user moves. In this way, it is hoped to become a sort of personal coach, guiding each wearer on how they can perform safer movements and prevent back injuries.
The device is called Backy. Developed by work safety tech firm TacnIQ, it contains quite unusual sensing technology; the array of connected sensors form a sort of network relaying information back and forth and to the user, in a way achieving similar sensitivity to human skin. “The human sense of touch is quite exquisite,” says Dr Benjamin Tee, part of the co-team behind Backy who developed the sensing technology. “Typically in the back area of your skin, you probably have a resolution of about 1-2cm, so we try to have around the same density of sensors.”
According to Tee, these sensors don’t just measure angles and rotations as with traditional wearables, they capture the real pressure and force being exerted on the user’s back. “Many different types of sensors being deployed on wearable devices to track posture tend to have a lot of noise,” he explains, referring to how lots of unnecessary data can block out the valuable information. “By mimicking an electronic version of skin, we are able to reduce a lot of noise and really provide highly accurate data.”
In the tech world there’s a saying, “garbage in, garbage out”, jokes Dr Harold Soh, the artificial intelligence development half of Backy’s co-team. “We have quality input,” he explains, “and then the AI’s part is really to translate that quality input into quality output. And the way to do that is essentially by predicting the posture.” The device’s AI engine is a sophisticated algorithm trained to recognise a wide range of postures and predict the risk of injury. “When the data comes in, what we try to do is predict whether it’s a good posture or a bad posture” says Soh. “We can then translate that kind of prediction into actionable information.”
But Backy doesn’t just observe – it acts. When the AI detects a potentially harmful movement, it triggers haptic vibrations, providing immediate feedback to the wearer and continuing a real-time feedback loop the team hopes will change worker movement habits over time. “It’s a behavioural changer,” says Aashish Mehta, CEO of TacnIQ. “It provides a strong element of strict awareness behind posture correction.”
But it’s not only the immediacy that’s important here, the system’s range of data collection and its feedback process is novel and could dramatically improve accuracy. By measuring real pressure and force, it provides a distinct assessment of injury risk to a degree the team says simply wasn’t available before, and it also delivers personalised guidance to each wearer. Added to this, being AI-powered, the software behind the tool is constantly learning and adapting, ensuring that the feedback remains relevant and effective over time.
Visibility on risk
Backy’s makers claim the device’s potential benefits extend beyond the individual worker to impact the entire workplace. By reducing the risk of back injuries, workers can focus on their tasks more freely and effectively, with the result of improved worker productivity and efficiency for the person, the team and the firm. Moreover, it may have a role to play in maintaining a culture of safety awareness and continuous improvement in workplaces if the real-time feedback does indeed help workers take ownership of their safety and make better decisions.
At the organisational level, Backy’s data collection capabilities could also support this more robust culture of safety by helping to identify bigger trends and patterns that may be contributing to back pain and injuries. This information could then be used to develop targeted interventions, such as ergonomic adjustments or process improvements, to reduce the risk of injury.
“The data allows managers to gain visibility, to say ‘whenever this bunch of workers are doing stuff at this time and at this location, they’re always doing bad pickups – why is that?’,” Soh explains. What the team have found is that sometimes workers are not intentionally acting poorly but the ergonomics of the workplace may actually force them to have to pick up items in a certain way.
“If [managers] go to look at the place, they might realise that this area just has everything set up in a bad way,” he explains. “All the pickup locations are on the floor and it’s really hard for people to operate, and so they need to change that.” He believes this visibility can lead to positive outcomes for both workers and businesses.
Having this data to really dig into trends has been transformational, says Mehta who explains it originated from an anomaly. “An automatic SMS alert was sent to the operational supervisor and they had to understand a bit more and identify and assess all the different tasks being done to figure out what the problem was and to change it,” he says. “It sounds very simple,” he adds, “but there was no data to be able to react to before – that’s what we are providing.”
Integration and personalisation
Adaptability is a key strength when it comes to changing people’s behaviour. When the team claimed that Backy acts as a personal coach, they mean the AI functionality lets it recognise and learn the behavioural and movement patterns of specific users, and then adjust its guidance for that user only. The intention is to circumvent worker resistance and instead maximise engagement. This has played a big part in the tool’s development and the feedback process includes elements of gamification to motivate workers and perhaps even generate a sense of healthy competition among workers. The thinking is, if workers can track their progress, they may compare scores with peers and strive for peak improvement.
Resistance is a real issue when companies invest in new technology, so making sure workers would actually want to use Backy took a lot of effort from the development team. According to Mehta, this effort was not in vain because user feedback suggests that workers actually like the idea of Backy sending them reminders to correct their posture because at the end of the day, protecting their backs is protecting their best interests. “You have to worry about your back to be able to just do the normal day to day tasks like picking up your children,” he says. “Your job shouldn’t ruin that future.”
Finally, successful integration into the workforce is vital, but so is integration into existing systems within the organisation. As such, it was essential that the device and its operating system would be not only able to merge with a firm’s safety protocols and softwares, but also be customisable for how the business as a whole operates. As such, the team made the data portion of the tool easy to format and integrate into various enterprise systems, so that the valuable insights would sit within existing workflows and reporting structures.
Force-sensing technology
While Backy’s technology represents a significant leap forward in workplace safety, as with any technology, it’s not a perfect solution and workers should still err on the side of caution. One limitation of Backy’s current iteration is its primary focus on the lower back. While this is a crucial area for preventing MSDs, other body parts, such as the shoulders and neck, are also susceptible to workplace injuries.
Mehta acknowledges this and says, “the biggest part to develop is measuring everything, right now we’re just focusing on the lower back.” However, he also reveals TacnIQ’s long-term ambitions are to eventually “map out the whole body”.
Another concern is the potential for worker complacency. Could Backy’s real-time feedback inadvertently make workers overly reliant on the device and less mindful of their own body’s signals? While this is a valid concern, the creators believe that the opposite is true. Mehta argues the wearable “provides a strong element of strict awareness behind posture correction”, encouraging workers to be more conscious of their movements and develop safer habits. Soon, it should become second nature.
Accounting for limitations is simply being realistic, and despite the minor drawbacks, the future for this AI-powered safety wearable is bright. Next, TacnIQ envisions a platform where force-sensing technology can be applied to a variety of applications beyond workplace safety. “We are going to provide a platform for others to be able to look into different industries,” says Mehta, hinting at the potential for Backy to revolutionise fields such as sports, healthcare, and even gaming. But for now, “we are going to focus on health and safety, of course.”
