使用胸部放射学深度学习模型查患者错误识别错误:一项七位读者研究
Kiduk Kim1, Kyungjin Cho2, Yujeong Eo3
1Department of Convergence Medicine, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-Ro 43-Gil, Songpa-Gu, Seoul, 05505, Republic of Korea.
Journal of imaging informatics in medicine
|September 11, 2024
概括
深度学习模型可以准确地从配对的胸部X射线中识别患者,匹配人类专家的性能. 这项技术提供了一种可靠的方法,以防止放射学中患者的错误识别.
科学领域:
- 放射学 放射学是一门学科.
- 人工智能的人工智能
- 医疗成像医学成像
背景情况:
- 在医学成像中错误识别患者存在重大风险.
- 从配对的胸部X射线图 (CXRs) 中准确识别患者对于诊断完整性至关重要.
- 现有的患者识别方法可能容易出现人为错误.
研究的目的:
- 评估深度学习 (DL) 模型在从配对的CXR中识别患者的性能.
- 将DL模型的诊断准确性与人类放射学专家进行比较.
- 与经验丰富的放射科医生相比,评估基于DL的患者识别的非劣势.
主要方法:
- 开发并验证了使用240,004个CXRs的大数据集的深度学习模型.
- 利用多个验证数据集 (内部,CheXpert,胸部ImaGenome) 代表不同的种群.
- 在读者研究中,与初级住院医生,高级住院医生和经过董事会认证的放射科医生进行了模型性能比较.
主要成果:
- 在所有数据集中,SimChest深度学习模型展示了优越的患者识别性能 (AUC范围:0.933-0.999).
- 放射科医生达到0.900的平均精度,性能随经验增加 (0.874到0.935).
- 与人类专家相比,SimChest实现了非劣质性能 (P=0.015),平均准确率为0.904.
结论:
- 深度学习模型,特别是SimChest,可以有效地从配对的CXR中识别患者.
- DL模型在从CXR中识别患者时,为人类专家提供非劣质的性能.
- 这项技术有望通过查错误识别来提高患者的安全性.
相关概念视频
01:30Radiological Investigation I: X-ray and CT
218
Radiological investigations, including X-rays and computed tomography (CT) scans, are critical for diagnosing and evaluating various medical conditions. These imaging techniques provide valuable insights into the body's internal structures, aiding in the detection of abnormalities, assessment of disease progression, and development of treatment strategies. This article delves into two primary radiological investigations, chest X-rays and CT scans, outlining their purpose, procedures, and...
218
01:25Errors occurring during blood pressure monitoring
615
Blood pressure monitoring is a crucial clinical procedure in diagnosing and managing various cardiovascular conditions. Despite its significance, the accuracy of blood pressure measurements can be compromised by multiple factors, potentially leading to either falsely high or low readings. These inaccuracies are critical as they can significantly impact patient care. So, it is vital to understand these challenges deeply and adopt strategic approaches to minimize errors.
Several factors...
Several factors...
615