Wearables, AI And Your Personal Medical Avatar

Medical technology concept. 3D rendering.

getty

As we start to recognize the awesome power of large language models in the artificial intelligence age, we see that they work on the basis of ingesting data. So the vehicle for the analytics and insights that these engines bring is the raw data all around us. How do we harvest it, and bring it to the places where it can drive knowledge?

Throughout the last few years, the term “wearables” has gained quite a bit of traction. In so many cases, much of the core data has to do with a human body: how it is functioning, what is around it, how it moves, etc. Wearables, then, are the way to aggregate that information and make use of it. At the same time, changes in the cost of AI compute and advances in token parameterization, have led to more “edge computing,” where the data gets crunched closer to the point of origin.

All of this ushers us toward a new way to think about medicine.

In a variety of ways, interventions are based on clinicians knowing about the body’s vital signs and responses in real time: the more precise measurement of blood pressure, heart rate, sleep health, blood sugar, and much more.

Wearables are exceptionally useful here, as noted in this study abstract from the National Library of Medicine:

“Disease diagnosis and monitoring using conventional healthcare services is typically expensive and has limited accuracy. Wearable health technology based on flexible electronics has gained tremendous attention in recent years for monitoring patient health owing to attractive features, such as lower medical costs, quick access to patient health data, ability to operate and transmit data in harsh environments, storage at room temperature, non-invasive implementation, mass scaling, etc.”

Moore’s law has played a role, but the real catalyst is AI, the new ability of digital systems to “think” for us, given the right inputs.

Personalized Healthcare

With the rise in AI-based healthcare, there’s also a drive toward personalization: with so much individual data in play, often from wearables, the AI engines can make their own proclamations and “predictalytics” with regard to one unique person: the user.

I wanted to give this example from a new study from the Personalized Medicine Coalition called Personalized Medicine at FDA, that talks about gene therapy and more:

“Approval of five new gene or cell-based therapies,” write study authors. That’s pretty personal. “Gene and cell-based therapies promise to dramatically improve care for certain patients by genetically re-engineering a patient’s own cells to combat disease.”

Authors note that therapies approved in 2022 address rare genetic diseases including beta thalassemia, hemophilia B, and cerebral adrenoleukodystrophy, as well as “cancers including refractory multiple myeloma and certain types of non-muscle invasive bladder cancer.”

Daniel Kraft, an experienced doctor and computer science researcher, agrees. In covering the appeal and promise of personalized AI-driven healthcare with wearables and “otherables,” he cites the potential for prenatal genomic sequencing, and a new era of individualized data collection.

“How do we not have a failure of imagination?” he asks, suggesting that the national system in America needs to move from a model of “reimbursement-based medicine” to one of “innovation and evidence-based medicine,” and pointing out how human systems like the microbiome give AI the clues that it needs to help.

Down the Pike

“The future is coming a bit faster than we thought,” Kraft says, mentioning the need to upskill clinicians and helpers, and the weighty responsibilities that the medical community faces now. “Many of you don’t go to Dr. Google anymore. You go to Dr. GPT.”

This, he suggests, might have its downsides, although in general, Kraft seems upbeat about the possibilities.

“There’s a famous quote from Sam Altman, the founder of OpenAI, that GPT is like an e-bike for your mind,” he says, “but the dark side is, what if your kids never learn how to ride a regular bike, or never develop their biking muscles, or a medical student never develops a differential diagnosis by themselves? … but … I’m an optimist, and I think we don’t need to think just about generative AI, but (about) generative health, this idea that we’re into this agentic age where our avatars will be hyper personalized (according to) our age, our culture, our personality.”

Your Digital Twin

Kraft also points to the established practice of digital twinning, where a digital model emulates a physical, biological system, in order to show a range of state maps.

“You can have a digital twin at your heart of a patient of a population,” Kraft says. “’Good news, your digital twin is in excellent shape. Now, if we can only get you two to sync up’ … But now that we have these digital twins, we can understand that aging is not the same for everybody.”

And there’s more.

“Quantum Computing is coming next,” he adds. “We can leverage these technologies to create our personal avatars and agents … we’re able to use this augmented virtual reality now in more consumer based devices, which can enable a clinician, or a patient, or a consumer, to see the world in new ways, or a surgeon to see through a patient and get crowd-sourced knowledge, to do a faster, cheaper, more effective operation.”

All of that is exciting. It’s part of that sea change that we’re likely to see, where in the medical office of the future, doctors are taking wearable diagnostic tools from patients, plugging them into medical desktops, and getting much more visibility into home care and how patients are doing between visits.

“(New wearables) can measure almost every element of physiology and behavior,” Kraft says, “and that means we can measure our behaviors, which drive most of our chronic conditions and our shorter health spans. And what’s coming is quite incredible. Real time blood pressure, real time blood sugar, 24/7, from our wrists and beyond. … there are patches that can do an intensive care unit level of data, and allow you to do ‘hospital to home.’ You don’t even need to wear anything: the camera can do your … ‘medical selfie’ and pick up vital signs.”

Showing off items like a “GI pill” that can explore the body’s innards for diagnosis, Kraft enumerated various tools that will change our lives, including bionics for aging.

“You can now put on EXO boots … a full-on exoskeleton,” he says. “So if you’re 90 or 100, you can climb Mount Kilimanjaro as if you’re 50.”

That’s a real-life example of these devices promoting human achievements that will be remarkable. Kraft stresses the preventative and interventionary nature of these technologies.

“It’s a bit of a FICO score for your health every day, or a bit of a check engine light for your body, before you blow a gasket and have a critical issue, and end up in the emergency room,” he says. It’s an exciting time. It’s not about any one technology, not about AI or wearables or otherables, but how those exponential technologies are coming together and converging, and when we have that convergent, exponential mindset, we’ll realize that we’re kind of in the future today.”

This is a lot to think about, and it’s exciting, too. What kinds of problems can we solve, in aging, in disease prevention and remediation, in social and economic challenges around our health? The sky’s the limit, in so many ways. Check out what happens around the end of 2025, a banner year for AI.