Could This Computer Help You Beat Cancer?
Nov. 22, 2022 – The 1960s marked the arrival of computers in medicine. Expensive, cumbersome hunks of plastic and metal that could (maybe) get test results to a doctor faster. The 1980s saw the first real difference-making functions computers could offer – clinical, financial, administrative – and in 1991, the Institute of Medicine published the first manifesto on what electronic health records could (and would) be.
Since then, we’ve seen computer breakthroughs across all areas of medicine, with artificial intelligence, virtual reality, and telemedicine brought to the fore. But something else is brewing that not a lot of people know about yet: Quantum computing, a completely new type of computing that has already begun to advance everything from drug development and disease identification to the security of electronic records.
“Think of it as transitioning from getting light through fire and candles and now having electricity, and there’s a light bulb that is lighting it all,” says Lara Jehi, MD, Cleveland Clinic’s chief research information officer.
What Is Quantum Computing?
Classical computers (aka binary computers), which are the foundation of today’s devices, including artificial intelligence and machine learning, work by using information known as bits. These appear as 0 or 1 (sometimes defined as off/on or false/true).
Quantum computers, on the other hand, use quantum bits known as qubits. And yes, the definition of “quantum” – as in: very, very small – applies.
International Business Machines, more commonly known as IBM, is currently leading this new tech. A common misconception about quantum computers is that they are “a next evolution of computers that will get faster,” says Frederik Flöther, PhD, life sciences and health care lead with IBM Quantum Industry Consulting. Instead, he wants us to look at quantum computing as something completely new “because it is fundamentally a different hardware, a different software, not just an evolution of the same.”
How does it work differently from existing computers? Quantum computing deals in nature. Therefore, qubits have to be based on the natural world. What does that mean? Nobel Prize-winning physicist Richard Feynman was famously quoted as saying, “Nature isn’t classical, dammit, and if you want to make a simulation of nature, you’d better make it quantum mechanical, and by golly it’s a wonderful problem, because it doesn’t look so easy.”
Nature, says Jehi, doesn’t work in black and white or fit into boxes.
“We have to convert it to zeros and ones because that’s what computers speak,” she explains. But quantum computing uses the principles of quantum mechanics. “It’s exactly how nature works, because it is based on the fundamental unit of everything in nature, which is atomic structure.”
Very, very small indeed. And that’s why quantum computing could be game-changing tech in medicine.
“Quantum computers can be used to represent a bunch of different solutions to a problem all at the same time, and then collapse down to the optimal solution, the one that actually works,” says Tony Uttley, president and chief operating officer with Quantinuum, a collaboration between Cambridge Quantum and Honeywell Quantum Solutions that is working to drive the future of quantum computing. “And the reason it does that is because of some fabulous properties of quantum physics.”
Establishing a Quantum Computing Beachhead
Scientists around the globe are studying quantum computers and looking into how they can harness this technology to make some big gains in the medical world.
IBM has created the IBM Quantum Network and is partnering with different organizations, from startups to Fortune 500 companies, to develop and test technology in various settings. One of these partnerships with the Cleveland Clinic is set to establish the “Discovery Accelerator,” focused on advancing health care through high-performance computing on the hybrid cloud, quantum computing technologies, and artificial intelligence.
Many people around the country are now using this technology on existing computers by tapping into the cloud, but with limited qubit access. IBM has researchers in places like Germany and Japan working on quantum computers and will be installing the country’s first of IBM’s next-generation 1,000+ qubit quantum systems on the Cleveland Clinic campus, which they are planning to use to help further investigate quantum computing’s many predicted benefits.
But what are those benefits?
Drug Discovery and Development
Quantum chemistry is one main area quantum computing is poised to help.
“The immediate application of that would be in drug discovery,” says Jehi. When scientists make drugs, they sit in a lab and develop different chemical formulas for what might constitute that drug.
“But for us to really know if it’s going to work, we need to be able to imagine how that chemical composition will translate into a structure,” she says.
Even in their most powerful form, today’s supercomputers are slow in their ability to change this chemical formula on paper to a simulation of what the chemical compound will look like. And in many cases, they can’t do this type of analysis.
“So, we end up making the drugs without knowing exactly how they’re going to look, which is not really the optimal way of creating a drug you expect to work” explains Jehi. “It’s a waste of time creating compounds that aren’t going to have any effect.”
Quantum computers will allow researchers to create and see these molecular structures and know how they bind and interact with the human body. In effect, they’ll know if a potential drug will work before ever having to physically make it.
Because of its differences from classic computing, quantum computers are not limited in their ability to simulate how different compounds can appear. Being able to simulate the compounds that drugs are made of can lead to a faster discovery of medications to treat a wide range of conditions.
Eventually, this technology could assist with disease analysis, working on a molecular level to allow computers/AI to contemplate, for example, cancer molecules and gain a deeper understanding of how they function.
Jehi says quantum computing can also be used to study things like chronic illnesses. These are conditions that people must live with and manage, and how a person is feeling in this instance can vary day-to-day, based on things like what a person is eating, the weather, or medications they are taking.
“There are so many different possibilities for what could change a patient’s trajectory in one way versus another,” says Jehi.
She stresses that if we have a group of patients, and we’ve captured everything that’s happened to them along their disease journey, it’s very challenging to mimic what that group looks like, and then study the effects of these different interventions on it using traditional computing.
“It just gets way too complicated, and the computers that we have can’t keep up with analyzing the effects of the different possibilities. It gets jumbled up,” Jehi says.
But quantum computing can offer quantum machine learning, meaning you use this special quantum ability to handle different simulations and different possibilities.
Cleveland Clinic, for instance, is looking at how some patients who undergo general surgeries have heart complications after their procedures.
“It would be transformative if we could identify ahead of time who is at highest risk of having a heart attack after surgery, as so we could take care of those people better,” she says.
The clinic’s current data set includes records for 450,000 patients, and current AI/machine learning makes sifting through this very slow and complex. The clinic is using machine learning approaches to create a synthetic data set, a smaller group that is a replica of the much larger one. Quantum technology could improve and speed this analysis to produce models that better perform.
“Imagine you go get a CT scan,” says Uttley. “There are already AI solutions that you can run that set of images through and ask, ‘Does this look like something that would be cancer?’” This existing technology, he explains, works well on things that are typical and have been identified before, because that’s how machine learning works. If AI has seen something 100,000 times, it can often find something else that looks like it.
But today’s classical computers aren’t equipped to identify something unfamiliar. “Those are places where quantum computers can be much better at thinking of images and being able to say, ‘I can detect rare cancers or rare conditions that you don’t have a huge library of things that look like that,’” Uttley says.
This is also where researchers can use a quantum computer to be able to figure out what things could look like.
“The beauty of quantum computing is that it is a bias formation in quantum physics, this more probabilistic design. And so you can take advantage of that probabilistic design to help them think about this,” Uttley says.
How Far Out Are We?
Uttley says we’re in an emergent era of quantum computing. Quantum computers exist and that’s a big deal, but a lot of this technology is still in fairly early stages.
“It’s a little bit like we’re at the beginning of the internet and saying, how are things going to play out,” he explains.
Right now, companies like Quantinuum are striving to perform computations on both a quantum and classic computer, compare the results, and say, “We’re getting the same answer.”
“So, this is the era where we’re able to build trust and say these quantum computers are actually working correctly,” Uttley explains.
In the future, he says, we can possibly imagine something like a quantum MRI that is able to understand your body in a way that transmits that data to a quantum computer to detect what’s wrong, and be able to tell the difference between cancerous and non-cancerous. That will allow faster treatments and tailoring them to specific patient populations.
“What we’re doing today might seem slightly less sexy than that, but is maybe even equally important,” says Uttley.
This is using quantum computers to make the best encryption keys that can be made. The medical community, which is already using quantum computing to execute this, is excited about this being a better means of keeping patient data as secure as possible.
In June, Quantinuum launched InQuanto, which is quantum computing software that is allowing computational chemists, who, until now, only had classical computers at their fingertips. The move created an opportunity to start thinking about the problems that they worked on and what they would do with a quantum computer. As quantum computers become higher-performing over the years, Uttley says the software will go from tasks like isolating one molecule to solving larger problems.
“That will happen over this next decade, where I think we’ll see the first kind of real use cases come out in the next likely 2 to 3 years,” he says. For now, this technology will likely be used in tandem with classical computers.
Uttley says that progress in the quantum world and medicine will continue to grow at a slow and steady pace, and in years to come, we’ll likely see things start to click and then eventually, this to take off “full force.”