Scholarlink

Objectives <p>To create and test a locally adapted large language model (LLM) for automated scoring of radiology requisitions based on the reason for exam imaging reporting and data system (RI-RADS), and to evaluate its performance based on reference standards.</p> Materials and methods <p>This retrospective, double-center study included 131,683 radiology requisitions from two institutions. A bidirectional encoder representation from a transformer (BERT)-based model was trained using 101,563 requisitions from Center 1 (including 1500 synthetic examples) and externally tested on 18,887 requisitions from Center 2. The model’s performance for two different classification strategies was evaluated by the reference standard created by three different radiologists. Model performance was assessed using Cohen’s Kappa, accuracy, F1-score, sensitivity, and specificity with 95% confidence intervals.</p> Results <p>A total of 18,887 requisitions were evaluated for the external test set. External testing yielded a performance with an F1-score of 0.93 (95% CI: 0.912–0.943); κ = 0.88 (95% CI: 0.871–0.884). Performance was highest in common categories RI-RADS D and X (F1 ≥ 0.96) and lowest for rare categories RI-RADS A and B (F1 ≤ 0.49). When grouped into three categories (adequate, inadequate, and unacceptable), overall model performance improved [F1-score = 0.97; (95% CI: 0.96–0.97)].</p> Conclusion <p>The locally adapted BERT-based model demonstrated high performance and almost perfect agreement with radiologists in automated RI-RADS scoring, showing promise for integration into radiology workflows to improve requisition completeness and communication.</p> Key Points <p><Emphasis Type="BoldItalic">Question</Emphasis> <i>Can an LLM accurately and automatically score radiology requisitions based on standardized criteria to address the challenges of incomplete information in radiological practice</i>?</p> <p><Emphasis Type="BoldItalic">Findings</Emphasis> <i>A locally adapted BERT-based model demonstrated high performance (F1-score 0.93) and almost perfect agreement with radiologists in automated RI-RADS scoring across a large, multi-institutional dataset</i>.</p> <p><Emphasis Type="BoldItalic">Clinical relevance</Emphasis> <i>LLMs offer a scalable solution for automated scoring of radiology requisitions, with the potential to improve workflow in radiology. Further improvement and integration into clinical practice could enhance communication, contributing to better diagnoses and patient care</i>.</p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Applying large language model for automated quality scoring of radiology requisitions using a standardized criteria

  • 期刊论文

  • 发表日期：

    2025年8月20日

查看全文
  • Raşit Eren Büyüktoka,
  • Murat Surucu,
  • Pelin Berfin Erekli Derinkaya,
  • Zehra Hilal Adibelli,
  • Ali Salbas,
  • Ali Murat Koc,
  • Asli Dilara Buyuktoka,
  • Yalcın Isler,
  • Mehmet Alperen Ugur,
  • Elif Isiklar

摘要

Objectives

To create and test a locally adapted large language model (LLM) for automated scoring of radiology requisitions based on the reason for exam imaging reporting and data system (RI-RADS), and to evaluate its performance based on reference standards.

Materials and methods

This retrospective, double-center study included 131,683 radiology requisitions from two institutions. A bidirectional encoder representation from a transformer (BERT)-based model was trained using 101,563 requisitions from Center 1 (including 1500 synthetic examples) and externally tested on 18,887 requisitions from Center 2. The model’s performance for two different classification strategies was evaluated by the reference standard created by three different radiologists. Model performance was assessed using Cohen’s Kappa, accuracy, F1-score, sensitivity, and specificity with 95% confidence intervals.

Results

A total of 18,887 requisitions were evaluated for the external test set. External testing yielded a performance with an F1-score of 0.93 (95% CI: 0.912–0.943); κ = 0.88 (95% CI: 0.871–0.884). Performance was highest in common categories RI-RADS D and X (F1 ≥ 0.96) and lowest for rare categories RI-RADS A and B (F1 ≤ 0.49). When grouped into three categories (adequate, inadequate, and unacceptable), overall model performance improved [F1-score = 0.97; (95% CI: 0.96–0.97)].

Conclusion

The locally adapted BERT-based model demonstrated high performance and almost perfect agreement with radiologists in automated RI-RADS scoring, showing promise for integration into radiology workflows to improve requisition completeness and communication.

Key Points

Question Can an LLM accurately and automatically score radiology requisitions based on standardized criteria to address the challenges of incomplete information in radiological practice?

Findings A locally adapted BERT-based model demonstrated high performance (F1-score 0.93) and almost perfect agreement with radiologists in automated RI-RADS scoring across a large, multi-institutional dataset.

Clinical relevance LLMs offer a scalable solution for automated scoring of radiology requisitions, with the potential to improve workflow in radiology. Further improvement and integration into clinical practice could enhance communication, contributing to better diagnoses and patient care.

Graphical Abstract