Imagine seeing a digital reflection of your patient. Not just their chart, but a living, learning simulation of their physiology.

That’s the promise of digital twin models in medicine, and it’s closer to reality than you might think.

Until recently, this idea felt more like science fiction than actual science. But with advances in multimodal AI, continuous data streams, and simulation modeling, we’re closer than ever to building virtual versions of our patients that can predict responses before they happen.

A digital twin is a computational model of a patient’s body. It’s built from their genetics, labs, imaging, wearables, and lifestyle data and evolves over time. It allows clinicians to simulate interventions and forecast likely outcomes.

That’s a game-changer for personalized medicine. Instead of asking, “What’s the average response to this peptide or protocol?”, we can start asking, “What’s the likely response for this specific patient?”

While still in its early phases, digital twin technology has already shown promise in oncology and cardiology. As large language models (LLMs) get more sophisticated, they build out the technology needed to create functional digital twins with predictive capabilities that can forecast health outcomes (i.e., deliver healthcasts) over different time horizons:

These aren’t static dashboards. They’re dynamic simulations that learn from continuous streams of data: labs, imaging, wearables, and patient-reported metrics. The result is a living view of physiology that updates as the patient changes.

And this is only just the start. The growing market for digital twins (across a range of industries) is expected to be $73.5 billion by 2027. This technology will be all around us soon.

Early innovators like Jona are already demonstrating what’s possible, integrating gut microbiome data to model system-wide physiology in real time. These advances mark a shift from retrospective review to living prediction, where every data point refines the care you deliver next.

These tools are only as valuable as the humans who interpret them, and the clinicians that help patients build their “health portfolio" of data.

Predictive models can project risk trajectories, identify likely responders, or simulate the effect of combining lifestyle changes with pharmacologic interventions. But the real power lies in the way clinicians use these forecasts to align with patients, turning prediction into precision, and data into direction.

Here’s where you come in: Your role shifts from reactive problem-solver to strategic health planner. Think of yourself more like a wealth manager for wellbeing, guiding patients toward compounding health returns.

📙 Your Homework: While fully dynamic twins are still emerging, you can start integrating predictive thinking into your clinical workflow:

The most exciting frontier is already forming.

Today’s digital twins draw largely from hospital and registry data, but the next generation will merge insights from body composition scans, cognitive and lifestyle monitoring, and multi-omic platforms, creating continuous, personalized feedback loops between visits.

That’s where the promise becomes personal: a feedback system that continuously learns from each patient and informs your next decision, helping you intervene earlier, personalize more precisely, and keep patients progressing between appointments.

The takeaway: The future of clinical care isn’t reactive, it’s proactive. Digital twins turn intuition into prediction, and foresight into action.

This week on How I AI, we sat down with Leo Grady, Ph.D., the founder of Jona. The company uses AI to understand the individual’s gut microbiome, then provides insights tailored uniquely to the individual.

The process start with a lab test. Using the resulting data, Jona creates a digital twin of that patient’s gut. Then, Jona’s AI system matches that person’s individual data to the hundreds of thousands of published studies about what impacts the microbiome. By using study data and the digital twin to model how different changes — like going vegan or keto — would impact the individual, Jona saves people from the time and energy drain of trying different changes to find what works.

With Jona, Dr. Grady and his team aim to deliver on what so many clinicians dream about: parsing the thousands of studies coming out each month and pulling out what’s relevant to the individual patient.

With this kind of support, Dr. Grady says clinicians can better serve three core camps of patients:

All of this gets particularly powerful when this kind of care is applied over time. With the digital twin and retesting, providers can see how adjustments affect the individual. This allows them to better tailor their care.

Leo’s philosophy in building Jona:

📙 Your Homework: Pick one patient from each of the three camps above. Explore how you can use AI-backed tech to help them with their goal, whether that’s getting to the root of a specific condition, supporting their general wellness, or upping their game.

👉🏽 Learn more live: Join Leo Grady of Jona and Sunita Mohanty of Vibrant Practice in an upcoming webinar on 10/28 at 9am PT / noon ET on “AI Made Simple: What Clinicians Need to Know and How to Use it Today”.

Curious to test out Vibrant Practice? Our AI-powered platform is ready-made to create personalized care plans. Test out our sandbox today.

The Modern Clinician is written for functional, integrative, and longevity-focused physicians who want to scale their impact and deliver cutting-edge care.

If you liked this one, share it with a colleague! We appreciate you spreading the word.

To learn more about the why behind this newsletter, start with our first post introducing The Modern Clinician.