新生儿仰卧胸片气胸深度学习诊断模型的开发与评估
本研究回顾性纳入2011年1月至2024年12月单中心NICU新生儿,共648例气胸胸片与5511例对照胸片。采用ResNet-18模型结合知识蒸馏开发,八折交叉验证确定固定概率阈值。测试集AUC为0.975,敏感度87.6%,特异度95.3%,Grad-CAM右侧定位准确率91.3%,左侧67.6%。
要点
- 模型对新生儿仰卧胸片气胸筛查阴性预测值达98.5%,可辅助排除气胸
- 左侧气胸Grad-CAM定位准确率明显低于右侧,阅片时需重点关注左侧
- 存在新生儿暂时性呼吸增快或胎龄偏大时模型易误判,需结合临床
核心结论
- 深度学习模型对新生儿仰卧胸片气胸检测显示出良好的诊断性能,测试集AUC为0.975
- 潜在肺部异常可能影响模型表现,临床实施前需进行外部验证
研究设计与样本
- 回顾性单中心研究,纳入2011年1月至2024年12月NICU新生儿
- 含288名气胸新生儿648张胸片及3377名对照5511张胸片
- 采用ResNet-18架构,利用公开正常成人胸片进行预训练与知识蒸馏
- 按患者水平划分训练集、验证集和测试集,八折交叉验证确定固定概率阈值
关键影像表现/征象
- Grad-CAM定位与临床判定侧别一致:右侧92例中84例一致(91.3%)
- Grad-CAM定位与临床判定侧别一致:左侧71例中48例一致(67.6%),差异有统计学意义(P<0.001)
诊断标准与关键数值
- 测试集AUC为0.975(95%CI,0.961-0.986)
- 敏感度87.6%,特异度95.3%,准确率94.5%
- 阳性预测值68.5%,阴性预测值98.5%
- 新生儿暂时性呼吸增快与误分类独立相关(OR,1.88;95%CI,1.14-3.12;P=0.03)
- 胎龄与误分类独立相关(OR,1.10;95%CI,1.01-1.19;P=0.03)
临床应用与阅片要点
- 高阴性预测值(98.5%)支持将该模型作为仰卧位新生儿胸片气胸的筛查排除工具
- 左侧气胸定位准确率低于右侧,AI提示左侧可疑气胸时需更谨慎复核
- 合并新生儿暂时性呼吸增快或胎龄较大时模型易误判,报告时应结合临床资料综合判断
局限性
- 单中心回顾性研究设计
- 研究为内部测试集验证,缺乏外部验证
- 左侧气胸Grad-CAM定位准确率仅67.6%,明显低于右侧
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
BACKGROUND: Neonatal pneumothorax can progress rapidly, yet detection on supine chest radiographs remains challenging.
OBJECTIVE: To develop and evaluate an artificial intelligence model for detecting pneumothorax on supine neonatal chest radiographs.
MATERIALS AND METHODS: This retrospective single-center study included neonates admitted to the neonatal intensive care unit between January 2011 and December 2024. The dataset comprised 648 radiographs from 288 neonates with pneumothorax and 5,511 radiographs from 3,377 controls. A ResNet-18-based model was developed using publicly available normal adult chest radiographs for pre-training and knowledge distillation. Data were split at the patient level into training, validation, and test sets. Eight-fold cross-validation determined a fixed probability threshold applied for the test set. Gradient-weighted class activation mapping (Grad-CAM) was used for lung-level localization, and logistic regression identified factors associated with misclassification.
RESULTS: In the test dataset, the model achieved an area under the curve of 0.975 (95% confidence interval (CI), 0.961-0.986; sensitivity, 87.6%; specificity, 95.3%; accuracy, 94.5%; positive predictive value, 68.5%; negative predictive value, 98.5%). Grad-CAM localization corresponded to the clinically determined side in 84 of 92 right-lung instances (91.3%) and 48 of 71 left-lung instances (67.6%) (P<0.001). Transient tachypnea of the newborn and gestational age were independently associated with misclassification (OR, 1.88; 95% CI, 1.14-3.12; P=0.03 and OR, 1.10; 95% CI, 1.01-1.19; P=0.03, respectively).
CONCLUSION: The model showed promising diagnostic performance for detecting pneumothorax on supine neonatal chest radiographs. Underlying pulmonary abnormalities may influence model performance, and external validation is warranted before clinical implementation.
引用
- Sanmoto, Gao, Hitaka, et al. Development and evaluation of a deep learning model for computer-aided diagnosis of neonatal pneumothorax on chest radiographs. Pediatr Radiol. 2026. PMID: 42730927. doi:10.1007/s00247-026-06778-w
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。