Capabilities
Currently isn't another data pipe. It's a multi-tenant platform with scope-based access, PHI isolation by design, and a pipeline that produces both FHIR R4 and OMOP CDM 5.4 from a single ingestion.
Architecture
Every user sees a different product — purpose-built for their role, powered by the same infrastructure underneath.
Full system oversight for Currently operators. Manage all HIEs, providers, patients, billing, CRM, CMS, and infrastructure.
Network-level administration for Health Information Exchanges. Oversee providers, approve registrations, view cross-org analytics.
Day-to-day data operations for healthcare organizations. Upload files, track processing, browse records, export results.
A dedicated portal for patients to access and manage their own health records.
Pipeline
Every file processed by Currently produces both FHIR R4 bundles and OMOP CDM 5.4 tables — no separate projects, no duplicate work.
Auto-detect format from file content. Parse HL7 segments (ADT, ORM, ORU, MDM) or extract CCDA clinical sections. No manual format selection.
Normalize demographics, dates, codes, and names. Optional NLP extraction from clinical notes. Full transformation logging for audit trails.
Patient, Encounter, Observation, Condition, Medication, Procedure, DiagnosticReport, Practitioner, Organization — with relationship preservation.
person, visit_occurrence, condition_occurrence, drug_exposure, measurement, observation, procedure_occurrence — with Athena vocabulary resolution.
Per-record completeness scores, per-source mapping rates, and quality metrics. Documented ~92.5% terminology mapping rate on real HIE data.
FHIR bundles as JSON, OMOP tables as CSV, bulk NDJSON, and optional push to Google Cloud Healthcare FHIR/HL7v2 stores and BigQuery.
Security
Security isn't a feature we added — it's how the databases are structured.
Protected Health Information lives in a dedicated database with its own access controls. The platform database contains only de-identified and anonymized data. Shared UUIDs link records across both without exposing PHI.
Every request resolves a session with scope, permissions, and data filter before any business logic runs. No role string checks in application code — just session.can flags derived from Clerk.
Queries are automatically filtered by the user's scope. HIE users see their network. Providers see their org. Patients see their records. Platform admins see everything. One codebase, four lenses.
Clerk handles auth with MFA support. All data mutations sync to both databases. Transformation logs track every AI-assisted change. Role changes flow through audit trails.
Integration
Connect any data source and export to any format. Currently adapts to your infrastructure — not the other way around.
Get started
Start processing HL7 messages, ensure data quality, and enable FHIR-based interoperability — all from one platform.