Patient Digital Twins in clinical studies

Something changed in clinical research: regulators opened the door to digital twins. The European Medicines Agency has issued its first qualification opinion on an AI-based trial methodology, and the FDA has published guidance for the use of AI in drug development. What was a research idea is becoming accepted practice.

Most of the industry's attention has gone to statistical twins — models that predict how a patient would have fared untreated, so that trials can run with smaller control arms. Industry analyses point to meaningful reductions in trial cost and duration, and in rare diseases, where every enrolled patient counts, the case is stronger still.

We look at the same technology from the patient's side.

A clinical study asks a lot of its participants: strict schedules, questionnaires, site visits, months of commitment. When participants drop out, data gaps appear, statistical power suffers, and timelines slip. A Patient Digital Twin accompanies each participant through the study — explaining every step in plain language, checking in between visits, capturing patient-reported outcomes as they happen rather than as visit-day recollections, and alerting the study team early when something needs attention.

The result sponsors care about: fewer dropouts, better adherence, cleaner and more continuous data — and answers that arrive sooner.

And because clinical research runs on evidence, everything the twin does is traceable end to end. Our engine was built in pharmaceutical research, where every data point must stand up to an audit years later — word for word.

If you are designing a study and wondering what a companion twin would change for your participants and your data, we would be glad to explore it with you.

Talk to us →