Eric Topol aims to help people increase their healthspan.
Highlights:
- Dr. Eric Topol proposes using sensors, data, and AI to predict health risks early and intervene before disease develops.
- Current healthcare captures only snapshots of health (visits, tests, hospitalizations). Continuous data can provide a real-time, full picture of health.
- Integrating physiological, biological, environmental, and demographic data with multimodal AI can reveal patterns invisible to clinicians.
- The goal is living more years in good health, not simply extending lifespan.
- By automating repetitive tasks, AI could give physicians more time to listen and connect with patients, restoring the human side of medicine.
It was a few months before Steve Jobs introduced the first iPhone. In early 2007, “almost around these same dates I was at a conference organized by Qualcomm here in San Diego,” recalls Dr. Eric Topol, cardiologist, scientist, and author. In that room, one of the speakers explained that soon there would be a camera integrated into a smartphone connected to the internet. At that moment, he realized that medicine, as we knew it, could be completely transformed. Care, he thought, would no longer be limited to physical spaces made of bricks and cement. That intuition led him to become a pioneer in digital health research. “I had been thinking about it when we started to see the possibility of obtaining an internet-connected electrocardiogram in the late 1990s,” he recalls.
Even today, healthcare still operates in episodes. The consultation, imaging and laboratory tests, hospitalization, and even medical records are fragments of the health experience. “They offer a very limited view,” Topol says. It is like taking a snapshot of a moment rather than watching a full movie. For this reason, he has focused on studying how sensors and data can offer a more complete version of our health in real time, beyond the isolated episodes when we come into contact with the healthcare system.
The concept of “deep medicine” that he proposes aims to combine different types of data and use artificial intelligence (AI) to detect patterns invisible to the naked eye. The idea is to transform traditional medicine. Instead of acting when the problem is already present, anticipate the trajectory of risk. Detect early signals, predict what might happen, and intervene before disease develops. In essence, move from reactive medicine to anticipatory and proactive medicine while restoring a more person-centered form of care.
This integration is made possible through digital health, which applies technology to better understand and care for our bodies. Dr. Topol uses an analogy, a GPS for health. He calls it a “personal geographic information system.” Just as a GPS combines traffic data, satellite images, and routes to guide a destination, this system would integrate layers of information about each person. For instance, it would include physiological, anatomical, biological, demographic, and environmental data that we generate daily. By bringing all these layers together, we could gain a more complete view of health and detect changes in time, before they turn into disease.
We couldn’t even imagine this until the last two years. Now we have a completely new capability.
Eric Topol
For years, the volume and complexity of these data exceeded our ability to analyze them in an integrated way. Only recently have multimodal AI models risen to the challenge. For the first time, it is possible to analyze together the information from all the layers that compose this personal map. “We couldn’t even imagine this until the last two years,” Topol notes. The combination of multimodal and generative AI opens a new stage in which prediction and prevention can be conceived at scale. “Now we have a completely new capability,” he says.
What this capability enables, he explains, is the possibility of tracing “an arc at the individual level of a person’s health.” A line that describes risk and estimates the moment when a disease might manifest, something that “we have never been able to do before.”
One concrete example is the “foundational sleep model,” developed at Stanford by James Zou and his team. It was based on data from 65,000 people with continuous monitoring of brain activity, muscle activity, heart rate, blood pressure, and other physiological signals. From this integrated dataset, “they were able to predict 130 diseases with incredible accuracy,” Topol says.
For him, the value of these advances lies in what they can change inside the consultation room. Technology, he explains, should allow medicine to return to a human-centered relationship. Over time, healthcare has increasingly aligned with business logic. For example, spending fewer minutes with each patient and introducing screens that create a layer between the physician and the person. In that context, AI can already take over repetitive tasks and free up time for healthcare professionals. Time they can use to listen, interpret, and build trust.
AI is going to make us human again in medicine.
Eric Topol
At its core, medicine is about being human. This belief has always accompanied him, influenced by figures such as Dr. Kanu Chatterjee, a cardiologist at the University of California, San Francisco, and director of the Coronary Care Unit. “He has always been a medical hero to me,” Topol recalls. He admired both his diagnostic insight and the way he connected with patients, an example that shaped his entire career. During hospital rounds, Dr. Chatterjee would place his hands on the patient’s chest, listen carefully, and understand the situation as a whole.
The goal is not longevity, it is expanding the healthy lifespan.
Eric Topol
The ultimate goal of digital health is to optimize people’s well-being. “The goal is not longevity,” Topol says. What matters is “expanding the healthy lifespan.” Living longer does not necessarily mean living better. His most recent book, Super Agers, explores how this field of science is evolving and why there are reasons to trust future breakthroughs. “We can keep someone on life support for years, but that’s not really life,” he warns.
This distinction between living longer and living better also guides his current research. They are beginning a study focused on Alzheimer’s prevention. One group will receive close guidance to modify their lifestyle, supported by an Oura ring, a digital assistant, and human follow-up. Researchers will analyze data from sensors, biomarkers, and neural inflammation. “The idea is to test a very aggressive lifestyle improvement in people at extreme risk of Alzheimer’s,” explains Dr. Eric Topol.
If we want people to live more years in good health, we must also think about the reach of technology and who will be able to use it. Dr. Topol points out a significant risk that not everyone will have access. The populations that could benefit the most are those with the least access to healthcare and the shortest healthy lifespan, even within wealthy countries. Digital health could transform medicine, but its true impact will depend on how this gap is closed.
It’s really unfortunate that we need a global crisis or an existential threat to do something.
Eric Topol
Topol also expresses frustration at the slow pace with which medical innovations are adopted. “It’s really unfortunate that we need a global crisis or an existential threat to do something,” he says. Telemedicine, he notes, was technically ready, but it was only during the COVID-19 pandemic that it was implemented at scale. The same pattern repeats with other innovations that could reduce the burden of disease but take years to be incorporated. Even so, he remains convinced that we are facing a tremendous opportunity to strengthen primary prevention and intervene while people are still healthy. Perhaps, for the first time, we may have tools capable of anticipating risk, expanding healthy years of life, reducing the probability of developing multiple diseases, and at the same time giving physicians back the time to practice their profession with greater humanity.
