Metraipolis › archive › dc_1bf1f2e4a71e
RAG-TRACE: Tracing Measurement Evidence and Diagnostic Completeness in AI-Generated ECG Reports — arXiv:2611.01409
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.01409
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T18:22:49.711Z
- SHA-256 of the archived text
e80cdc319fcd651e802bf563e4064111a2198e7dd6af78f285fbd0d1ee18d758- Stable URL of this page
https://metraipolis.com/documento/dc_1bf1f2e4a71e
How to cite
RAG-TRACE: Tracing Measurement Evidence and Diagnostic Completeness in AI-Generated ECG Reports — arXiv:2611.01409 — arXiv (en), https://arxiv.org/abs/2611.01409 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1bf1f2e4a71e (Metraipolis archive dc_1bf1f2e4a71e, sha256 e80cdc319fcd651e802bf563e4064111a2198e7dd6af78f285fbd0d1ee18d758)
BibTeX
@misc{dc1bf1f2e4a71e,
title = {RAG-TRACE: Tracing Measurement Evidence and Diagnostic Completeness in AI-Generated ECG Reports — arXiv:2611.01409},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.01409},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1bf1f2e4a71e; sha256 e80cdc319fcd651e802bf563e4064111a2198e7dd6af78f285fbd0d1ee18d758},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.01409. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Purpose: Retrieval-augmented generation (RAG) can introduce electrocardiogram (ECG) criteria that lack patient-specific measurement support. This study evaluates the reliability of retrieved diagnostic criteria in AI-generated ECG reports. Methods: We analyzed 1,083 human-validated PTB-XL patient records using a library of 47 criteria cards, including 33 without numeric verification rules. Language models generated structured reports with and without retrieval. Criterion applicability was evaluated using Marquette 12SL measurements, with independent assessment using Uni-G and comparison against expert-validated diagnostic labels. Results: Keyword retrieval introduced untestable or measurement-disconfirmed criteria in 46.8% of GPT-4.1-mini reports. A deterministic verification rule withdrew 647 criterion applications, including 639 without numeric checks, while preserving 268 of 269 applications independently supported by Uni-G. Restricting retrieval to 14 testable criteria reduced the initial unsupported-criterion rate to 13.8%. Under matched instructions and an identical output schema, retrieval increased unsupported-criterion reports from 17.4% to 31.1%, while hypertrophy recall remained unchanged at 62.6%. Conclusion: These findings identify a fundamental limitation of retrieval-based clinical report generation: retrieved criteria do not necessarily provide valid patient-specific evidence. Measurement-based verification can reduce unsupported criterion applications, but diagnostic validity requires independent assessment. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2611.01409 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-03T18:10:07.925Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).