Artificial intelligence presents boundless opportunities to transform human health. Already, it’s aiding in the discovery of new drugs and helping to create better, faster medical diagnostics. It’s not difficult to imagine a time in the near future when A.I. is able to reliably predict and intercept disease before symptoms ever arise.  

However, with so much hype surrounding A.I., it’s easy to forget that we’ve only scratched the surface in understanding what’s possible with machine learning, and how it may be applied to health care, says Guna Rajagopal, Ph.D., Global Head of Computational Sciences, Discovery Sciences, at Janssen Research & Development.

Dr. Rajagopal leads a team of data scientists who are using high-performance computing to assist in the creation of new drugs and health care products. He’s also currently working with Johnson & Johnson Innovation, JLABS to evaluate nominees to the Artificial Intelligence for Drug Discovery QuickFire Challenge, which will award up to $100,000 in grants and one year of JLABS residency to individuals or teams with the best ideas for using artificial intelligence to advance health care.

With the excitement of the QuickFire Challenge selections looming (expect an announcement in mid-December), we spoke with Dr. Rajagopal to learn more about how A.I. is changing the future of medicine.

Q. Artificial intelligence means different things to different people. How do you define A.I., especially as it relates to health care?

A. A.I. is the science of building and programming a machine that’s able to imitate human cognition. The machine can learn from experience and generalize, which is where the intelligence part comes in. Most of us already interact with A.I. in our daily lives, whether it’s Amazon giving us personalized suggestions of products we might like, to real-time alerts of potential credit card fraud. Regardless of the industry, the purpose of A.I. is to guide humans to make informed decisions based on enormous amounts of data. As it pertains to health care, this data may include an individual’s genetic and environmental factors, activity trackers and bio-sensors, blood samples and electronic health records. The amount of health and genomic data that we’re generating is growing exponentially every year. While our A.I. tools are improving by the day, we also have to recognize that big-data research is still very much a nascent field. We have a lot to learn.

Q. What are some of the biggest transformations that A.I. will bring to health care?

A. It’s impossible to predict all the ways that A.I. will change health and medicine over time, but I believe it will be truly fundamental and sweeping. I’ll defer to the famous quote from American futurist Roy Amara: “We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.” This is true of A.I., just as it has been for other transformational technologies, such as GPS. Can you imagine life without GPS? I can’t. Yet, in the early days, GPS was expensive and difficult to deploy, no one foresaw all the ways it would simplify our lives and help society. If you look at the impact of A.I. for health care, you have to consider not only the how it relates to the development of new medicines and faster clinical decision-making, but also elements such as pharmacy and supply chain — how we optimize our resources to get the right products to patients most efficiently.

Q. ‘Precision medicine’ just may be the health care buzzword of the decade. How might A.I. impact the development of drugs that treat patients in a more precise way?

A. We’ve only begun to realize the benefits of precision medicine to treat disease, with notable early successes in cancer. The underlying idea is that we can use patients’ personalized information, such as genetic or molecular profiles, to determine what treatment approaches will work best for them as individuals. As we move forward, our success in advancing precision medicine will depend not just on collecting and storing vast datasets on diverse patient populations, but also on our ability to develop sophisticated machine-learning algorithms that can mine this data to answer specific health care questions. How do we find the “signals” in the noise that lead to actionable insights? What are the genetic variants that matter? Why is it that disease manifests itself in one patient, but not in another? These are extremely complex questions. We see A.I. as a tool to help us analyze these factors and bring clarity to patients earlier in the process.    

Q. What about preventative health? Does A.I. have a role to play there?

A. Of course. Disease prevention is the holy grail. Technologies that enable early disease detection and interception will be truly game-changing, as we have seen that patient outcomes improve when disease is detected earlier. With A.I., when we are looking at large sets of patient data, we’re searching not just for markers of disease, but also for markers of health. These insights are extremely valuable in the context of drug development. 

Q. In the latest QuickFire Challenge for Johnson & Johnson Innovation, JLABS, you’ve put a call out for entrepreneurs and innovators that are using artificial intelligence to advance drug discovery and development in some way. What do you hope comes as a result of this type of competition?

Even though Johnson & Johnson is among the world’s largest health care companies, we realize that we can’t do it all on our own. Our goal is to create an ecosystem of innovation around A.I. in health care, partnering with the best and thinking holistically about how solutions can be applied to various aspects of drug development and patient care. If you think end-to-end about our health care system — from the lab where drugs are created, to the patient’s bedside — the one thing that ties it all together is data. I’m optimistic about what A.I. can do to bring meaning to our growing pools of information, aiding in our ability to intercept, diagnose and treat disease.