AI-powered digital twins of individual patients

Patient digital twins forecast how an individual’s health may change over time

We invent, train, and deploy generative models capable of generating accurate, comprehensive forecasts of a patient’s health over time under relevant scenarios—which we call their digital twin.

“Digital Twins are simulated clinical records that share baseline data with actual subjects and comprehensively model their outcomes under standard-of-care.”

Modeling Disease Progression in Mild Cognitive Impairment and Alzheimer's Disease with Digital Twins

We are pioneering generative AI research

Our AI research centers on deep learning-based generative models for multivariate stochastic processes and associated clinical applications. In other words, we aim to simulate ‘what if?’ scenarios for individual patients so that we can computationally compare potential health outcomes.

Incredible data products

Data is the bedrock of machine learning. We’re building a strong foundation of highly curated datasets consisting of individual, longitudinal, research-consented patient data across our target indications.

State-of-the-art ML models

Our research is at the cutting edge of generative AI for multivariate time-series. We build and deploy new types of machine learning models for forecasting potential health outcomes for individual patients.

Rigorous clinical applications

We develop novel methods for using the predicted health outcomes from patients’ digital twins to improve clinical research. The outcome: efficient processes that determine the best treatments for each patient.

A patient’s digital twin forecasts their potential future health outcomes

The human body is too complex to simulate from the atom-up. We’re taking a top-down approach, leveraging artificial intelligence trained on highly curated historical datasets to forecast health outcomes so that we can accelerate medical innovation today.

A patient’s digital twin is computationally created with generative AI; 
it is not a matched patient from an external cohort.
A patient’s digital twin provides a probabilistic forecast of their specific health outcomes; 
it cannot be used like data from a new patient.
Clinical trials that incorporate digital twins are randomized with concurrent control groups;
they do not use synthetic control groups.


AI-generated digital twins of patients will transform healthcare

Support precision medicine

We envision a future where patients’ digital twins help doctors create more accurate diagnoses and more personalized treatment plans.

Reduce human experimentation

Patients’ digital twins will predict relative treatment effects, revealing how different treatments compare to each other with less patient experimentation.

Calculate precise predictions

An AI-powered future will enable a more efficient, ethical, and reliable healthcare system.


AI-powered clinical trials leverage patients’ digital twins to maintain power with smaller control groups

TwinRCTs use our regulatory qualified methods to incorporate patients’ digital twins into clinical trials, enabling trial designs with smaller control groups that reach enrollment targets faster than traditional designs. 

Treatment group
Control group - Significant reduction

We work with innovators in pharma and biotech to accelerate clinical drug development today