Symphony for Medical Coding: A Next-Generation Agentic System for Scalable and Explainable Medical Coding

arXiv cs.AI / 4/1/2026

💬 OpinionSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper introduces “Symphony for Medical Coding,” an agentic system that converts free-text clinical narratives into standardized medical codes using clinical reasoning over guidelines rather than fixed-code prediction.
By giving the system direct access to coding guidelines and producing span-level evidence tied to supporting text, it aims to improve trust and explainability for safety-critical billing and reporting workflows.
The approach is designed to adapt across different coding systems without retraining for each new set of codes, addressing limitations of prior methods that require fixed label sets.
Evaluations on two public benchmarks and multiple real-world datasets across inpatient, outpatient, emergency, and subspecialty settings show state-of-the-art performance.
The authors position Symphony as a flexible, deployment-ready foundation for scalable automated medical coding in multiple healthcare contexts and regions (US and UK).

Abstract

Medical coding translates free-text clinical documentation into standardized codes drawn from classification systems that contain tens of thousands of entries and are updated annually. It is central to billing, clinical research, and quality reporting, yet remains largely manual, slow, and error-prone. Existing automated approaches learn to predict a fixed set of codes from labeled data, thereby preventing adaptation to new codes or different coding systems without retraining on different data. They also provide no explanation for their predictions, limiting trust in safety-critical settings. We introduce Symphony for Medical Coding, a system that approaches the task the way expert human coders do: by reasoning over the clinical narrative with direct access to the coding guidelines. This design allows Symphony to operate across any coding system and to provide span-level evidence linking each predicted code to the text that supports it. We evaluate on two public benchmarks and three real-world datasets spanning inpatient, outpatient, emergency, and subspecialty settings across the United States and the United Kingdom. Symphony achieves state-of-the-art results across all settings, establishing itself as a flexible, deployment-ready foundation for automated clinical coding.