Skip to main content

Dream team: how robotics and AI can combine to cure our ills

Better understanding of sophisticated analytical techniques such as machine learning holds the potential to dramatically improve the performance of robotic systems in sectors such as healthcare

Imagine a future where the twin strands of robotics and artificial intelligence come together to make life easier and more convenient for all.

In this distant utopia, smart factories with flexible production lines will be able to turn out highly-customized goods in the timeliest manner.

And in healthcare, meanwhile, previously unimagined examples of automation will provide the very best levels of treatment and care.

AI_MEDICAL_87323a975d3efe159675728999ac3d9af966361c.jpg

All sound too good to be true? Don’t you believe it. The recent confluence of several important technologies makes such scenarios far less fanciful than you might imagine.

The cause for this optimism stems from a growing appreciation of the complementary nature of robotics and AI. This combination holds tremendous potential across industrial sectors.

But before taking a look at some exciting applications, a bit of context is first required. Traditionally, robots have been considered as reactive agents of work, in that they require programming and can then follow a prescriptive set of commands. However, the application of AI – broadly characterised as the development of computer systems able to perform tasks normally requiring human intelligence – means robots could be made to ‘think’ for themselves. And this is where the excitement lies.

Specifically, AI comprises a sub-field called machine learning, which employs statistics and mathematical optimisation to enable the spotting of patterns through exposure to what can be huge amounts of data. Then, algorithms build a model from the accumulated data in order to make predictions or decisions without being explicitly programmed to perform the task.

AI_DEEP_LEARNING_7e0acb47131e6255534f8a4e42c8b744a816b4ba.jpg

Taken one step further, these algorithms can be built in layers of increasing complexity, inspired by the structure of neural networks in the brain. Known as deep learning, this technique can carry out an even more advanced form of advanced pattern recognition and therefore can be tasked with solving far greater types of problems.

Robotics applications

What does this mean in practice? Well, we’ve already seen the application of AI and its various sub-fields in the development of a host of applications, including virtual assistants such as Amazon Alexa and Apple’s Siri. Increasingly, though, it is the dual use of robotics and AI within industry that is causing ripples of excitement, across a wide range of different sectors.

Healthcare is one of the primary areas of potential, according to a recent report from PwC, which identifies several converging trends that support the adoption of AI. Firstly, ageing populations mean there is a global desire to drive down healthcare costs. Meanwhile, recent advances in sensors and wireless connectivity, along with the upgrading of IT systems, means much higher levels of health data are now available. And this has democratised access to data, enabling patients to play a more informed role in their own care than ever before.

These factors mean healthcare is perfectly positioned to take advantage of the increasing capability of robotics and AI, says PwC, leading to the adoption of more automated systems that can start to assist or make decisions that have a greater impact on individual lives. Ultimately, the power of AI could be used for examination, diagnosis, treatment and patient care, helping clinicians to speed-up their decision making and even perform certain tasks.

So what sorts of applications are starting to emerge? In terms of examination, a recent report from the IPPR policy thinktank suggests that AI-inspired automated systems could play a huge role in reducing the burden of repetitive and administrative tasks and freeing up of staff to spend more time on direct clinical care with patients. Among the many activities that should be carried out through digital technology include communicating medical notes, booking appointments and processing prescriptions.

The report also predicted a future in which robots and AI systems combine to assess, treat and support clinical practice. Someone arriving at hospital, for instance, may begin by undergoing digital triage in an automated assessment suite, with AI systems – including machine-learning – even able to make diagnoses of diseases such as pneumonia, breast and skin cancers, eye diseases and precursors to diabetes. 

AI_ROBTOICS_ed2aa9a6014690c309af307f239f60f4cc89f2de.jpg

Even more remarkable is the potential of using robots for surgery itself. Already, remotely-operated surgical robotic system comprising surgeons console, arms and monitoring systems and software are used to assist in minimally invasive procedures, such as tying knots, insert screws and make stitches with greater accuracy and dexterity than humans. The da Vinci robotic surgery system, made by Californian company Intuitive Surgical, is perhaps the best-known example of this technology, with more than 4,500 robots used worldwide.

Indeed, the global surgical robotics market was valued at $56,294 million in 2017 and is projected to reach $98,737 million by 2024, according to a report by Allied Market Research. And the range of applications is set to rapidly expand over the coming years, as patients become more comfortable with the concept of automated surgery. In the UK, for example, CMR Surgical recently unveiled its Versius robot, which it says is smaller and more flexible and versatile than existing robots, allowing it to perform a wider range of operations.

Machines that think

In the longer term, it is the use of robots enabled with AI that will really accelerate the pace of adoption of automated surgery. This remains a relatively new area of research, and the first of these devices have only recently come to prominence. Last year, for instance, plastic surgeons at Maastricht University Medical Centre performed the world’s first super-microsurgical intervention using ‘robot hands’, with a device capable of suturing 0.3 to 0.8mm blood vessels in a patient's arm. The surgical robot developed by Microsure, a spin-off of Eindhoven University of Technology and Maastricht University, was controlled by a surgeon whose hand movements were analysed and then converted into smaller and more precise movements, which the procedure performed by the robot’s tiny ‘hands’. The device also used AI to stabilise any tremors in the surgeon’s movements, making the procedure easier to perform. Going forward, it is hoped that surgeons will be able to use the AI-inspired robot during other types of complex microsurgical procedures, such as tissue reconstruction.

Once out of surgery, there is also a role for robotics in post-operative care. The IPPR report predicts a big future for so-called ‘bedside robots’, powered by AI-inspired voice recognition software, assisting patients with meals, transportation and rehabilitation. The robots and their digital systems will enhance communication with the patient’s family and friends, and biosensors will allow remote monitoring and alerting responses to clinical observations, such as potentially life-threatening conditions such as sepsis.

Beyond health services, so-called “care-bots” will empower people in old age, enabling better, longer, and more fulfilling lives, and improve social care, enabling people to remain more socially connected to friends and family.

Realising potential

It’s clear, then, that robotics and AI are having a marked impact across the healthcare industry – from diagnosis and treatment, through to recuperation and care. The question is: where next? Wouldn’t it make sense for such technology to be used by allied sectors, such as fitness and lifestyle? The market for wearables such as Fitbit has grown exponentially in recent years, and now as AI understanding grows and machine learning gets smarter, such devices could increasingly be used to look for predictable key markers that indicate health issues at formative stages, while advising on treatment options. In such cases, diagnosis might no longer made by medical specialists, but by AI spotting patterns learned from data analysis.

In short, robotics and AI promise enormous advances across a range of applications. And perhaps the true potential of these exciting complementary technologies will only be restricted by the limits of our imagination.

Favourite things are Family, Music and Judo. Also, I have the ability to retain and quote useless facts, something that pleases me but can annoy others. My engineering hero - Isambard Kingdom Brunel
DesignSpark Electrical Logolinkedin