Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
Assessing Completeness of Clinical Histories Accompanying Imaging Orders Using Adapted Open-Source and Closed-Source Large Language Models
12
Zitationen
9
Autoren
2025
Jahr
Abstract
= .31) despite being a smaller model. Using Mistral-7B, 26.2% (12 803 of 48 942) of unannotated clinical histories were found to contain all five elements. Conclusion An easily deployable fine-tuned open-source LLM (Mistral-7B), rivaling GPT-4 Turbo in performance, could effectively extract clinical history elements with substantial agreement with radiologists and produce a benchmark for completeness of a large sample of clinical histories. The model and code will be fully open-sourced. © RSNA, 2025
Ähnliche Arbeiten
Refinement and reassessment of the SERVQUAL scale.
1991 · 3.967 Zit.
Radiobiology for the Radiologist.
1974 · 3.502 Zit.
ACR Thyroid Imaging, Reporting and Data System (TI-RADS): White Paper of the ACR TI-RADS Committee
2017 · 2.445 Zit.
Accuracy of Physician Self-assessment Compared With Observed Measures of Competence
2006 · 2.331 Zit.
Technology as an Occasion for Structuring: Evidence from Observations of CT Scanners and the Social Order of Radiology Departments
1986 · 2.257 Zit.