ChatGPT与成人峡部裂型腰椎滑脱症NASS临床指南的对比分析。
A comparative analysis between ChatGPT versus NASS clinical guidelines for adult isthmic spondylolisthesis.
作者信息
Lin Che Chung Justin, Krzyż Ewa Zuzanna, Tsai Sung Huang Laurent, Wang Ying-Chih, Chang Chia-Wei, Lin Tung Yi, Fu Tsai Sheng
机构信息
Department of Orthopaedic Surgery, Chang Gung Memorial Hospital, Keelung Branch, Taiwan.
School of Medicine, Chang Gung University, Taoyuan, Taiwan.
出版信息
N Am Spine Soc J. 2025 Feb 22;22:100599. doi: 10.1016/j.xnsj.2025.100599. eCollection 2025 Jun.
BACKGROUND
Isthmic spondylolisthesis is a prevalent condition often diagnosed in adults, especially those with low back pain. The main objective of this study was to evaluate the clinical viability of ChatGPT 3.5 and 4.0 by assessing its capacity to produce recommendations consistent with NASS's Evidence-based Clinical Guidelines for adult isthmic spondylolisthesis.
METHODS
To achieve the purpose of this study, we used the 2014 NASS Evidence-Based Clinical Guideline for Multidisciplinary Spine Care and presented its 31 questions to ChatGPT 3.5 and ChatGPT 4.0 separately, evaluating their responses for appropriateness and consistency with the guidelines.
RESULTS
ChatGPT 3.5 and ChatGPT 4.0 demonstrated concordance rates with the NASS guidelines of 45% and 42%, respectively, with ChatGPT 3.5 showing higher accuracy (91%) for questions with definitive recommendations and both versions showing lower concordance (20%) for questions with no direct recommendations.
CONCLUSIONS
Future enhancements should focus on enabling ChatGPT to better reflect the latest evidence and clinical complexities, especially concerning issues that involve medical terms.
背景
峡部裂性腰椎滑脱是一种常见病症，常在成年人中被诊断出来，尤其是那些患有腰痛的人。本研究的主要目的是通过评估ChatGPT 3.5和4.0生成与美国国家脊柱疾病协会（NASS）关于成人峡部裂性腰椎滑脱的循证临床指南一致的建议的能力，来评估其临床可行性。
方法
为实现本研究目的，我们使用了2014年NASS多学科脊柱护理循证临床指南，并将其31个问题分别呈现给ChatGPT 3.5和ChatGPT 4.0，评估它们的回答与指南的适宜性和一致性。
结果
ChatGPT 3.5和ChatGPT 4.0与NASS指南的一致率分别为45%和42%，ChatGPT 3.5对于有明确建议的问题显示出更高的准确率（91%），而两个版本对于没有直接建议的问题显示出较低的一致性（20%）。
结论
未来的改进应集中于使ChatGPT能够更好地反映最新证据和临床复杂性，特别是涉及医学术语的问题。
Zotero 插件