Incremental Diagnostic Value of Clinical Information for Large Language Models Across Multiple Organs: Retrospective Study
Jinqi Zhang, Xiaoyi Wang, Yanfeng Zhao, Xiaolin Cui, Liyun Zhao, Xinming Zhao, Hongmei Zhang, Zheng Zhu · 2026
Read Incremental Diagnostic Value of Clinical Information for Large Language Models Across Multiple Organs: Retrospective Study by Jinqi Zhang, Xiaoyi Wang, Yanfeng Zhao, Xiaolin Cui, Liyun Zhao, Xinming Zhao, Hongmei Zhang, Zheng Zhu (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/pmc-13519069.txt.
Description
Jinqi Zhang and seven co-authors ask how much extra clinical detail helps language models read medical scans. Models can draft the summary part of a radiology report from image findings alone. Whether adding the patient's story and lab values sharpens their judgment was not clear.
The team gathered reports for 301 patients with proven disease of the liver, lung, or breast from two hospitals. Three leading models each saw three levels of input: basic facts and findings, then the complaint or history, then key lab results. That design produced 2,709 model outputs. Judgments were scored against the proven diagnosis, for both cancer-versus-benign calls and exact disease calls. Radiologists' reports served as the human yardstick.
Catalog record
| CREATOR | Jinqi Zhang, Xiaoyi Wang, Yanfeng Zhao, Xiaolin Cui, Liyun Zhao, Xinming Zhao, Hongmei Zhang, Zheng Zhu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 53 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |