深度学习方法的比较使用胸部X射线图来预测临床恶化:回顾性观察研究
Mahmudur Rahman1, Jifan Gao2, Kyle A Carey3
1Department of Medicine, University of Wisconsin-Madison, 610 Walnut St, Madison, WI, 53792, United States, 1 608-262-9564.
JMIR AI
|July 3, 2025
概括
使用胸部放射图的深度学习模型可以预测住院患者的临床恶化. DenseNet121模型在识别有重症监护室转移或死亡风险的患者方面表现出卓越的表现.
科学领域:
- 医学成像分析分析 医学成像分析
- 医疗保健中的人工智能
- 临床信息学是一种临床信息学.
背景情况:
- 早期发现临床恶化对于改善患者的治疗结果至关重要.
- 目前的早期预警系统主要使用结构化数据,忽视了其他预测方式.
- 胸部X射线图,通常在恶化期间进行,可以提供有价值的预测信息.
研究的目的:
- 将各种计算机视觉模型和数据增强技术进行比较和验证,以预测使用胸部X射线图的临床恶化.
- 评估不同深度学习架构在识别高风险患者中的有效性.
主要方法:
- 成年患者的回顾性观察性研究,提升了早期预警分数 (eCART).
- 包括在48小时内进行的胸部X射线扫描的患者.
- 五种计算机视觉模型 (VGG16,DenseNet121,Vision Transformer,ResNet50,Inception V3) 和四种数据增强方法的比较.
主要成果:
- 在胸部X射线图上预训练的DenseNet121模型,经过组图正常化和高斯噪声增强,实现了最高的预测歧视 (AUROC 0.734,AUPRC 0.414).
- 视觉变压器模型显示了最低的歧视 (AUROC 0.598).
结论:
- 胸部X射线图具有基于深度学习的临床恶化预测的巨大潜力.
- 与其他架构相比,DenseNet121表现出卓越的性能.
- 历史图规范化和随机高斯噪声增强可能会提高DenseNet121和VGG16.6的性能.
相关概念视频
01:30Assessment of Airway, Skin Color, and Use of Accessory Muscles
1.2K
A thorough assessment of respiratory health is paramount in clinical settings to identify and manage respiratory distress and ensure adequate oxygenation. This article elaborates on the critical aspects of respiratory evaluation, including airway assessment, skin color examination, and the observation of accessory muscle use, which are integral to effectively diagnosing and managing patients with respiratory conditions.
Introduction
The initial evaluation of a patient's respiratory system...
Introduction
The initial evaluation of a patient's respiratory system...
1.2K
01:30Radiological Investigation I: X-ray and CT
428
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...
428