Once a user has built a cohort of synthetic subjects (SynthPerson™), the second stage in the platform builds a full (synthetic) Personal Health Record (SynthPHR™) for each subject.

When generating the PHR records, the platform utilizes scientifically-valid Controlled Terminologies (including SNOMED-CT CORE, LOINC, and CDISC’s own Controlled Terminology) as well as Real-World Data sources (including from the FDA, the European Medicines Agency, and the Center for Medicare and Medicaid Services’ (“CMS”) National Provider Identifier.

There are versions of many controlled terminologies for languages other than English that can be easily plugged into the platform. The platform is then able to generate PHRs in multiple languages.

Finally, the platform outputs the PHR records in multiple formats, including: CSV, JSON, and SQLite. Additional output formats (FHIR, HL7, OMOP) can be easily added to the platform.

Figure 1: Synthetic Personal Record (SynthPHR™) overview below illustrates the core capabilities of the synthetic PHR generation stage.

The software platform is divided into “Functional Blocks” for ease of development and re-use.

We’ll now analyze each functional block’s capabilities.

Table 1 below summarizes the functional blocks used to generate Synthetic Personal Health Records.


Functional Block

Purpose: Randomly assigns to each subject


Generates records of each Synthetic Person


SNOMED-CT CORE-defined medical conditions


FDA-approved medical devices


EMA- or FDA-approved pharmaceutical products

Lab Results

LOINC-defined lab results


SNOMED-CT CORE-defined procedures


Current Medicare-registered healthcare providers


Vital signs




Our in-house team will be delighted to give you an in-depth view of the power and capabilities of the TrialTwin platform.