For a while now, it seems that we’re unable to make decisions all by ourselves. Faced with the large amount of information around us, we tend to choose what the majority chooses – even though we’re not exactly sure the reasons behind our choice.
This, of course, is not a new problem. But nowadays, it’s an interesting one. Because everything’s trending and, more often than not, we feel like some of these trends belong together. We’ve warned you once before that chocolate on pizza tastes funny. In other words: it’s good to research combinations, before deciding to combine.
When it comes to the possibility of a sacred technological marriage between AI and VR, scientists have been researching the combination for years. Let’s sift through their conclusions.
The Evolution of AI
Artificial intelligence has become incredibly popular in the modern world – and there are plenty of reasons for that.
First of all, even though your brain is a magical place and AI has still long way to go before emulating it, and even though humans are still way smarter than machines, computers process data much, much faster. That’s because human brains and “machine brains” work differently and it’s safe to say that although computers have problems distinguishing a fluffy dog from a KFC meal, they have no problem solving complex mathematical equations in a second or predicting outcomes based on input/output pairings.
This powerful process is so simple you can start exploring it yourself today. First, you’ll need some relevant data and proper algorithms; then you just feed the program with the data and the program uncovers the pattern below it and is able to make predictions about future inputs based on this discovered pattern.
Interestingly enough, even experts consider this magical. The more powerful the computers are, the less able they are to understand the way the algorithms work. Simply put, computers have started learning by themselves, and are able to improve in real-time, growing ever more helpful. Half a century ago, this was not the case.
Even back then, AI was fascinating, but it was also fairly straightforward. For one, it was not exactly AI: it was human intelligence masked under thousands and thousands of lines of code. It was the programmers who did the predicting and who already knew all the outcomes. No AI-based program was able to improve without a human intervention, simply due to the fact that it didn’t work in the manner AI works nowadays. To say that AI has evolved would be a major understatement.
AI and Healthcare
A simple example may illustrate this.
As, unfortunately, most of us know, brain tumours are often diagnosed too late because the first symptoms people experience are ignored mostly because they are too mild and unclear and may not indicate a serious problem. Medicine has progressed, but doctors still don’t know why tumours arise in the first place; it’s even more baffling when it comes to young and healthy people. Before AI arrived, the only thing we could do was shrug our shoulders and helplessly repeat “well, what can you do”. But smart computers are here and they have given us more than a glimpse of hope.
Deploying AI in oncology means using smart programs to find underlying patterns in large datasets – quickly and accurately, in a manner impossible for the human brain. Uncovering a pattern from past inputs means developing a predictive model for future inputs. In plain terms: it means uncovering the reasons which lead to tumours and the symptoms which may help in early detection.
Google’s DeepMind is, once again, a leader in the field. A while ago, it started developing an AI software able to distinguish between healthy and cancerous tissues by feeding the software with CT and MRI scans from former University of London patients. A recently published white paper outlines the initial successes:
“Metastasis detection is currently performed by pathologists reviewing large expanses of biological tissues. This process is labour intensive and error-prone.  At 8 false positives per image, we detect 92:4% of the tumours, relative to 82:7% by the previous best automated approach. For comparison, a human pathologist attempting exhaustive search achieved 73:2% sensitivity.  Our approach could considerably reduce false negative rates in metastasis detection.”
Radiologists, oncotherapeutists, and surgeons have so much to look forward to: with the help of AI, they may be able to make better prognoses, develop adequate treatments, and even prevent tumours in patients which exemplify certain risk factors.
Augmented Reality (AR) and Virtual Reality (VR)
When it comes to deciding the importance of new technologies, it’s always useful to reflect on how the world around us has changed in their presence. In the case of AR and VR, it’s safe to say that the change has been profound; in fact, Mark Zuckerberg himself has stated that with VR, the future itself has arrived.
AR and VR are related, but different concepts. AR means adding virtual components into the real world, and VR creating an entirely artificial one. In other words, VR replaces the real world with a simulated one, and AR merely adds simulated elements into the real world. Either way – a true heaven for gamers! And it shouldn’t surprise no one that both VR and AR were first used in the gaming industry.
Nowadays, however, VR and AR are moving beyond. And scientists use the gaming experiences to advance them.
From Gaming to Healing
For instance, in games such as “Gnomes and Goblins”, players interact with various AI agents in a sophisticated VR world. Scientists have quickly realized that similar games can be employed in psychotherapy to help children with communication issues/disabilities, or even in psychotherapy to help people overcome some traumas. Because in either case, the problem is the real world. And we’ve come to a stage when we can improve it – or change it altogether.
Medicine has already profited from VR and AR in many different ways: VR and AR make possible for surgeons to train before a complicated operation, help patients recover from complex injuries, and even allow youngsters to experience how older people feel.
Combining AI and VR/AR
Combining VR and AI is only a matter of time. With AI programs able to analyse immense amounts of data and finding patterns human eyes are unable to uncover, and VR capable of visualizing these patterns, medicine – and humanity – has a lot to look forward to.
We will soon be able to identify better many illnesses which are currently enigmatic or rare and doctors will be able to understand the genesis and the nature of some serious diseases without even having real-life patients.
As time goes by, medicine will probably become a much more personalized affair. AI programs will be able to monitor the health of a single individual in real-time – and report of possible issues – and doctors may even have the opportunity of diagnosing problems virtually, through simulated replicas of patients. Patients, on the other hand, will benefit from these simulations by learning from them how to behave when a certain problem arises or how to treat an injury on a daily basis – without even leaving their own houses.
Ever since we realized its benefits, AI has become ubiquitous. It has been implemented in many fields of human endeavour – with almost all others on the waiting list. The same is true with VR/AR: it’s long time since they were used exclusively in the gaming industry.
Combining AI and VR may revolutionize medicine. So much so – that even surgeons may become redundant in a not so distant future. In the near future, nonetheless, they will be able to train on simulated bodies and identify diseases better with the use of AI software. Patients will benefit as well: AI/VR will help them acquire the right treatment for difficult injuries. And there is so much more.
Kevin Kelly, a futurist and the author of The Inevitable: Understanding 12 Technological Forces That Will Shape Our Future is convinced that “just as AI is essential to make VR work, VR will propel advancements in AI”. He addresses the issue in more practical terms as well:
“I don’t imagine that the VR companies will manufacture the AI. I think they’re going to be purchasing it from AI companies, just as they’ll be purchasing electricity to run their service. But they will become a huge customer of AI.”