Revenue-sensitive evaluation of AI-assisted ICD-10-CM coding and human-AI collaboration under dual DRG payment systems
Ming-I Chen, Ying-Lin Hsu, Yi-Hsin Chen · 2026
Read Revenue-sensitive evaluation of AI-assisted ICD-10-CM coding and human-AI collaboration under dual DRG payment systems by Ming-I Chen, Ying-Lin Hsu, Yi-Hsin Chen (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13507165.txt.
Description
Ming-I Chen, Ying-Lin Hsu, and Yi-Hsin Chen evaluate, in 2026 work, AI help for hospital diagnosis coding with an eye on payment, not just accuracy. Coding errors change what hospitals are paid under group-based payment systems in the United States and Taiwan. The team tested eleven models on intensive care records and built two money-aware scores for coding risk.
A fine-tuned language model led on standard accuracy, while general chat models used cold scored far worse in early tests. Money-aware scores exposed a wide gap between best and worst tuned models that plain metrics missed. Guiding human review by money risk at a fifth of cases captured most of the ideal gain. Live hospital trials are still needed.
Catalog record
| CREATOR | Ming-I Chen, Ying-Lin Hsu, Yi-Hsin Chen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 44073 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |