Turn messy healthcare data into analytics-ready data.
pyaar harmonize is a growing suite of harmonizers that standardize the vocabularies healthcare data is built on, RxNorm, ICD-10, CPT, LOINC, SNOMED, and more. It starts where the mess usually does: drug names.
Built to plug into the Tuva Project
The open-source Tuva Project turns raw claims and clinical data into a common data model through data quality, vocabulary normalization, and analytics-ready data marts. pyaar harmonize builds the same normalization primitives as small, browser-first tools, aligned with Tuva's vocabularies and data-quality principles. The goal: to become a trusted partner in the Tuva ecosystem.
The harmonizers
One tool per vocabulary. Drug names is live now, the rest are on the way.
Drug Name Normalization
Map brand names, abbreviations, and misspellings to a single generic ingredient and RxCUI. Upload a CSV, get a standardized column back.
Open tool →NDC Crosswalk
Resolve raw National Drug Codes to ingredients and RxCUIs for claims and pharmacy data.
Diagnosis Normalization
Clean and validate diagnosis codes, map descriptions to codes, and bridge legacy ICD-9.
Procedure Codes
Standardize procedure and service codes across claims sources into one vocabulary.
Lab Harmonization
Align lab result names and units to LOINC so results are comparable across systems.
Clinical Terms
Normalize clinical concepts and conditions to SNOMED for downstream analytics.
Provider Identity
Resolve provider identifiers and taxonomy codes into clean, deduplicated entities.
Data Quality Checks
Profile input data for completeness, conformance, and plausibility before it hits your model.