Generating analytics-ready healthcare data requires a data factory.

A fully integrated platform for systematically transforming raw healthcare data sources into quality-tested, analytics-ready healthcare data.

Data Factory Portal

A single view into data pipeline orchestration, data lineage, and data quality analytics

Connectors

Pre-built data pipelines for transforming raw data sources into a common format

Data Quality Intelligence

Systematic testing to uncover (i) atomic-level data problems (e.g., invalid bill_type_code) and (ii) analytics-level problems (e.g., patient with ESRD missing dialysis visits)

Master Patient Index

Probabilistic creation of a master ID for linking patients across disparate data sources

Semantic Normalization

AI-assisted mapping of local problem, lab, and medication terms to standard terminologies

Geo-coding

Map patient and provider addresses to latitude and longitude for geospatial analytics

Core Data Model

Common data model for unifying all claims and clinical data into a consolidated longitudinal patient record

Reference Terminology Sets

Atomic code sets and ontologies that form the basis of healthcare data (e.g., ICD-10, SNOMED, RxNorm, LOINC, HCPCS, etc.)

Reference Datasets

Includes social determinants, NPPES, specialty and taxonomy, practice affiliations, and death data

Measures

Common and complex measures related to cost, quality, utilization, and outcomes.

Groupers

Common and complex groupers from diagnoses and procedures to panel attribution and ED classification.

Care Pathways

Code sets that define diseases and the relevant lab tests, treatments, and common complications related to each underlying disease.

Risk Models

Supervised AI models for risk identification / stratification and risk adjustment.