Key results
The company
CodaMetrix
codametrix.comAI-powered medical coding and revenue cycle automation for health systems.
Result highlights
- 70% reduction in manual coding
- 60% reduction in coding-related denials
The story
A healthcare AI startup spun out of Mass General Brigham processes 300,000 patient encounters daily for health systems representing over $180 billion in net patient revenue.
Medical coding requires selecting accurate billing codes from 100,000 evolving options based on inconsistent doctor notes, where errors cause immediate payment denials. Disjointed legacy systems forced engineers to move massive datasets manually and lacked the governance controls required for sensitive patient information.
The company centralized its coding automation on Databricks to ingest clinical notes, predict codes, and integrate results directly with EHR systems like Epic. MLflow manages model training and serving, while Unity Catalog enforces fine-grained access controls to meet strict HIPAA standards. Spark Declarative Pipelines streamline data processing, allowing the team to leverage spot instances for cost-effective scaling.
Scope & timeline
- 40% operational efficiency improvement
Quotes
“Traditional approaches to augment coder workflow won’t address the scale and quality issues faced by health systems. What they need is a platform that can act autonomously to take a case from the end of the clinical workflow through to submitting a claim to a payer. You send us your encounters, and not only do we make predictions on what the ICD (International Classification of Diseases) and CPT (Current Procedural Terminology) codes should be, but we actually send them straight to billing.”