Scholarlink

Objective <p>To assess the ability of large language models (LLMs) to accurately simplify lumbar spine magnetic resonance imaging (MRI) reports.</p> Materials and methods <p>Patients who underwent lumbar decompression and/or fusion surgery in 2022 at one tertiary academic medical center were queried using appropriate CPT codes. We then identified all patients with a preoperative ICD diagnosis of lumbar spondylolisthesis and extracted the latest preoperative spine MRI radiology report text. The GPT-4 API was deployed on deidentified reports with a prompt to produce translations and evaluated for accuracy and readability. An enhanced GPT prompt was constructed using high-scoring reports and evaluated on low-scoring reports.</p> Results <p>Of 93 included reports, GPT effectively reduced the average reading level (11.47 versus 8.50, <i>p</i> &lt; 0.001). While most reports had no accuracy issues, 34% of translations omitted at least one clinically relevant piece of information, while 6% produced a clinically significant inaccuracy in the translation. An enhanced prompt model using high scoring reports-maintained reading level while significantly improving omission rate (<i>p</i> &lt; 0.0001). However, even in the enhanced prompt model, GPT made several errors regarding location of stenosis, description of prior spine surgery, and description of other spine pathologies.</p> Conclusion <p>GPT-4 effectively simplifies the reading level of lumbar spine MRI reports. The model tends to omit key information in its translations, which can be mitigated with enhanced prompting. Further validation in the domain of spine radiology needs to be performed to facilitate clinical integration.</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

Assessing the ability of large language models to simplify lumbar spine imaging reports into patient-facing text: a pilot study of GPT-4

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年9月9日

查看全文
  • Rushmin Khazanchi,
  • Austin R. Chen,
  • Parth Desai,
  • Daniel Herrera,
  • Jacob R. Staub,
  • Matthew A. Follett,
  • Mykhaylo Krushelnytskyy,
  • Hanna Kemeny,
  • Wellington K. Hsu,
  • Alpesh A. Patel,
  • Srikanth N. Divi

摘要

Objective

To assess the ability of large language models (LLMs) to accurately simplify lumbar spine magnetic resonance imaging (MRI) reports.

Materials and methods

Patients who underwent lumbar decompression and/or fusion surgery in 2022 at one tertiary academic medical center were queried using appropriate CPT codes. We then identified all patients with a preoperative ICD diagnosis of lumbar spondylolisthesis and extracted the latest preoperative spine MRI radiology report text. The GPT-4 API was deployed on deidentified reports with a prompt to produce translations and evaluated for accuracy and readability. An enhanced GPT prompt was constructed using high-scoring reports and evaluated on low-scoring reports.

Results

Of 93 included reports, GPT effectively reduced the average reading level (11.47 versus 8.50, p < 0.001). While most reports had no accuracy issues, 34% of translations omitted at least one clinically relevant piece of information, while 6% produced a clinically significant inaccuracy in the translation. An enhanced prompt model using high scoring reports-maintained reading level while significantly improving omission rate (p < 0.0001). However, even in the enhanced prompt model, GPT made several errors regarding location of stenosis, description of prior spine surgery, and description of other spine pathologies.

Conclusion

GPT-4 effectively simplifies the reading level of lumbar spine MRI reports. The model tends to omit key information in its translations, which can be mitigated with enhanced prompting. Further validation in the domain of spine radiology needs to be performed to facilitate clinical integration.