Saved in:
Bibliographic Details
Main Authors: Liu, Jinghui, Nguyen, Anthony
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.17755
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913137657118720
author Liu, Jinghui
Nguyen, Anthony
author_facet Liu, Jinghui
Nguyen, Anthony
contents Clinical coding maps clinical documentation to standardized medical codes, an essential yet time-consuming administrative task that could benefit from automation. Current models on ICD coding are typically optimized for codes from a specific ICD version. However, in reality, ICD systems evolve continuously, and different versions are adopted across time periods and regions. Moreover, ICD coding suffers from the long-tail problem, and rare code performance can be a bottleneck for developing implementable models. We examine whether it is viable to train version-independent models by combining data annotated in different ICD versions, which may help address these challenges. We add ICD-9 data to the training of a modified label-wise attention model for ICD-10 prediction, and find that despite the version mismatch, adding ICD-9 yields a 27% increase in micro F1 for 18K rare ICD codes compared to training on ICD-10 alone. On 8K frequent ICD-10 codes, the multi-version training also substantially improves macro metrics, with far fewer model parameters.
format Preprint
id arxiv_https___arxiv_org_abs_2605_17755
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Bridging the Version Gap: Multi-version Training Improves ICD Code Prediction, Especially for Rare Codes
Liu, Jinghui
Nguyen, Anthony
Computation and Language
Artificial Intelligence
Clinical coding maps clinical documentation to standardized medical codes, an essential yet time-consuming administrative task that could benefit from automation. Current models on ICD coding are typically optimized for codes from a specific ICD version. However, in reality, ICD systems evolve continuously, and different versions are adopted across time periods and regions. Moreover, ICD coding suffers from the long-tail problem, and rare code performance can be a bottleneck for developing implementable models. We examine whether it is viable to train version-independent models by combining data annotated in different ICD versions, which may help address these challenges. We add ICD-9 data to the training of a modified label-wise attention model for ICD-10 prediction, and find that despite the version mismatch, adding ICD-9 yields a 27% increase in micro F1 for 18K rare ICD codes compared to training on ICD-10 alone. On 8K frequent ICD-10 codes, the multi-version training also substantially improves macro metrics, with far fewer model parameters.
title Bridging the Version Gap: Multi-version Training Improves ICD Code Prediction, Especially for Rare Codes
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2605.17755