Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Joakim Edin, Andreas Motzfeldt, Simon Flachs, Lars Maaløe

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.

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

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.

Paper Structure

This paper contains 21 sections, 3 equations, 3 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Methods
  4. Symphony for Medical Coding
  5. Data
  6. Public datasets
  7. ACI
  8. MDACE
  9. Proprietary datasets
  10. Emergency department (ED)
  11. Ambulatory (AMB)
  12. Neurology (NEURO)
  13. Experiments
  14. Evaluation setting
  15. Evaluation metrics
  16. ...and 6 more sections

Figures (3)

  • Figure 1: Overview of the reasoning workflow learned in Symphony for Medical Coding.
  • Figure 2: A sample user interface showing the evidence spans provided by Symphony for Medical Coding.
  • Figure 3: F1 performance as a function of code space size (log-scaled) from yuanReliableClinicalCoding2025azhengMedDCRLearningDesign2025 and experiments in this paper. Each point corresponds to a dataset with a distinct number of unique codes. Symphony demonstrates superior performance across label space complexity.