DeltaForge parses FHIR resources, HL7 v2 messages, and X12 EDI documents into Delta tables. Query them all with SQL. No Spark cluster. Your data stays inside your network.
Common FHIR R4 and R5 resources parsed directly into Delta tables from JSON, XML, NDJSON, or RDF/Turtle. Clinical, medication, financial, and document resource types are covered.
Pipe-delimited HL7 v2 messages parsed segment by segment into Delta tables. ADT admission events, ORU lab results, and ORM orders each land in typed columns and are joinable with SQL.
Healthcare and supply-chain X12 documents parsed into Delta tables. Covers 837 claims, 835 remittances, 270/271 eligibility, 834 enrollment, 850 purchase orders, and more. EDIFACT and TRADACOMS also supported.
DeltaForge is installed and operated by you, so protected health information never leaves your environment. Run it on-premises, on VMs in your own AWS, Azure, or GCP account, or in a fully air-gapped network with no outbound connectivity required for query execution. For de-identified research extracts and limited data sets, built-in pseudonymisation replaces identifiers consistently across tables, and role-based access control and audit logging govern who can see what.
Healthcare formats and PHI handling, in practice
Three families on a shared SQL surface: FHIR R4 and R5 resources, HL7 v2 pipe-delimited messages, and X12 EDI documents. EDIFACT and TRADACOMS are also supported.
Yes. Each parser writes Delta tables with typed columns. A single SELECT can join FHIR Patient to HL7 v2 OBX observations to X12 837 claim lines on shared identifiers.
DeltaForge is installed and operated by the customer on their own infrastructure: on-premises, customer cloud VMs (AWS, Azure, GCP), or air-gapped environments.
No. DeltaForge has its own SQL engine that reads and writes the Delta Lake on-disk format directly. The same engine parses FHIR, HL7, and X12 and serves queries.
For the HL7 side, HL7 to SQL Without an Interface Engine walks through parsing ADT and ORU feeds. FHIR Analytics Without Spark covers resource flattening, and Query EDI Files with SQL shows 837 and 835 claims analysis end to end.
FHIR, HL7, and EDI parsed into Delta tables on your own infrastructure. Query across all three with SQL.