Art or science?

For most people, medicine will fall into the latter category. But, while medical training gives the hard and fast facts, there is an element of creative thinking involved: an amalgamation of previous patient stories, watching and learning.

“So it’s an art then?”

More so than you might think, which is why it seems almost ludicrous to introduce machines into the world of medicine. Despite its backdrop of science, machine learning – which, for some people, equals robots – surely can’t have a place in the medical world; surely these robots can’t take the place of a doctor?

Actually, they can. And they are. But not in the way you might think.

Rather than robots wheeling awkwardly around GP offices and dishing out medication, machine learning is actually being integrated into the profession in more nuanced ways. Think about things that are time-consuming but can, when it comes down to it, be put on autopilot.

Things like making more accurate diagnoses, finding better treatments, and hunting down cost-effective ways to prevent side-effects and illnesses.

It all comes down to machine learning.

What Exactly Is Machine Learning?

This popular branch of artificial intelligence (AI) can take mammoth amounts of data and analyze it quickly and accurately.

Basically, it has the ability to do things it would take the mere human mind a long time to execute in a successful way.

This is precisely why it’s slotting in nicely in the medical world.

It can tackle large quantities of data from hospitals, medical records, and research and pare it down into actionable results much, much quicker than a human could. And, in a profession where time is such a vital factor, this could be a life-saver – literally.

Today, modern medicine is heavily reliant on continuous studies, with reams and reams of fresh information emerging almost every day. When you consider this, it makes sense that machines might just be better equipped to keep up with and analyze data than the – let’s face it – relatively limited and time-starved human mind.

But let’s get one thing straight here: when we talk about machine learning, we’re not talking robots in the sci-fi movie sense.

Rather, it might be better to liken it to scalable insight.

This is how it works: AI technologies collect the amassed knowledge of physicians, as well as the learned experiences from dealing with thousands of patients, and scales that information to levels where it can populate studies and serve up answers that might have taken a much longer time to get to.

What Does Machine Learning Mean for the Medical Profession?

Vice president of Watson Health at IBM, Steve Harvey, sums it up nicely: “The way artificial intelligence starts to really impact what’s going on in healthcare is to be able to start cloning all the expert knowledge, so now all of a sudden you get access to all types of care, anywhere.”

It’s kind of like bringing together every physician in the world’s collective knowledge and making one super-human doctor.

And accessing that super-human doctor quickly and accurately is becoming a much-coveted skill in the medical profession.

Physicians need to be able to get hold of information on disease symptoms and new drugs immediately.

“Doctors are realizing that if they want to make sense of massive amounts of data, machine learning is a way of allowing them to learn from that data,” says Francesca Dominici, professor of biostatistics at the Harvard T.H. Chan School of Public Health.

Let’s take a look at it in action.

The APOLLO Program at the University of Texas

At the University of Texas’ MD Anderson Cancer Center, the APOLLO program is using machine learning technologies to parse through genetic data. As a result, it guides doctors towards treatments that are best suited to each and every individual patient.

Cogito and Mental Health

It’s not just physical ailments that machine learning is helping to tackle. Numerous start-ups are developing apps that detect symptoms of depression, bipolar disorder, and other mental health illnesses.

Take Cogito, for example.

This mental health app is built on machine learning technologies and aims to monitor social media activity and phone calls to detect and draw out communication patterns in sufferers of depression. These insights will be able to predict when a sufferer is likely to have a depressive episode.

So, Is Machine Learning the Future of the Medical World?

While it seems like machine learning is here to stay in the medical profession, there are still some kinks that need working out.

Like with any new technology, it’s not perfect yet, and it’s not even widely available.

We have tons of amazing AI tools working their magic out there, like Siri, self-driving cars, and Google Translate, but it has actually had very little impact in the medicine industry so far – and rightly so, maybe, when it really is a matter of life and death.

I mean, can physicians really rely on machines to get it right?

Well, if you take into account that fact that the prescriptions doctors give for blood thinners are only accurate 67% of the time and that cardiologists miss out a whopping 250,000 out of 300,000 people who will die suddenly each year, it kind of puts things into perspective. Doctors are faced with such huge amounts of data that one more piece could tip them over the edge.

But we’re still in the very early days.

A lot of the software is cumbersome, particularly when it comes to gathering the knowledge of doctors and physicians in the first place. But for people working in the medical world, having a way to collect and analyze such large amounts of data can only mean good things.

Yes, the human mind is glorious in many ways, but there is always the chance it might miss something – and, of course, it can’t see into the future.

Already, new technologies have emerged that are able to predict just how aggressive a disease is in a patient and which treatments will work best for that particular strain.

When you look at it this way, it’s easy to see how machine learning will become an integral part of diagnoses and medical research.

It’s not man versus machine here, it’s the human mind working alongside machinery – two minds are better than one, right? Even if one of those “minds” conjures up images of dystopian futures.

We’re still in the early stages of integrating machine learning into the medical world – and there’s still a long way to go – but there are already important parts of the profession that it’s helping.

By utilizing both the greatness of the human mind, the years of training that physicians undergo, and the power of machine learning, we might just see some of the biggest improvements and steps forward in human health take place over the next few years.