文献速递儿科影像胸部DR

将成人AI模型应用于儿科胸部影像:可靠性研究

Applying Artificial Intelligence to Pediatric Chest Imaging: Reliability of Leveraging Adult-Based Artificial Intelligence Models.
PubMed 文献 · J Am Coll Radiol · 2023-07-17

本研究利用TorchXRayVision的成人训练AI算法,在公开儿科胸片数据集(n=5,856)上检测肺炎,以60%阈值二分类,对比数据集标签。结果显示敏感度79.83%、特异度67.66%、准确率76.54%,病毒性肺炎亚组PPV和F1下降。结论:成人模型可 adequate 检测1-5岁儿科肺炎,但专用儿科模型更优。

要点

  • 成人训练AI模型在儿科胸片肺炎检测中敏感度79.83%、特异度67.66%。
  • 病毒性肺炎亚组PPV降至69.9%、F1降至0.74,性能下降明显。
  • 专用儿科影像训练的模型表现更好,但借用成人模型可加速儿科AI研究。

关键影像表现/征象

  • AI算法判定肺炎3,923例(67.00%),正常1,933例(33.00%)。
  • 与标签对比:真阳性3,411例(58.25%),假阳性512例(8.74%),真阴性1,071例(18.29%),假阴性862例(14.72%)。

诊断标准与关键数值

  • 肺炎二分类阈值:AI输出概率60%。
  • 整体性能:敏感度79.83%,特异度67.66%,PPV 86.95%,NPV 55.41%,准确率76.54%,F1 0.83。
  • 细菌性肺炎亚组性能与整体一致;病毒性肺炎亚组PPV 69.9%,F1 0.74。

临床应用与阅片要点

  • 成人训练AI模型可用于儿科胸片肺炎初筛,但需注意病毒性肺炎假阳性增加。
  • 临床解读时应结合患儿年龄(1-5岁)及影像征象,AI结果不能替代放射科医生判断。

局限性

  • 研究使用公开数据集,标签可能存在噪声。
  • 仅评估单一成人训练算法,未与多中心或前瞻性数据验证。
  • 病毒性肺炎亚组样本量未报告,性能下降的普适性需进一步研究。

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

OBJECTIVE: The scarcity of artificial intelligence (AI) applications designed for use in pediatric patients can cause a significant health disparity in this vulnerable population. We investigated the performance of an adult-trained algorithm in detecting pneumonia in a pediatric population to explore the viability of leveraging adult-trained algorithms to accelerate pediatric AI research.

METHODS: We analyzed a publicly available pediatric chest x-ray dataset using an AI algorithm from TorchXRayVision. A 60% threshold was used to make binary predictions for pneumonia presence. Predictions were compared with dataset labels. Performance measures including true-positive rate, false-positive rate, true-negative rate, false-negative rate, sensitivity, specificity, positive predictive value (PPV), negative predictive value, accuracy, and F1-score were calculated for the complete dataset and bacterial and viral pneumonia subsets.

RESULTS: Overall (n = 5,856), the algorithm identified 3,923 cases with pneumonia (67.00%) and 1,933 (33.00%) normal cases. In comparison with the actual image labels, there were 3,411 (58.25%) true-positive cases, 512 (8.74%) false-positive cases, 1,071 (18.29%) true-negative cases, and 862 (14.72%) false-negative cases resulting in 79.83% sensitivity, 67.66% specificity, 86.95% PPV, 55.41% negative predictive value, and 76.54% accuracy and an F1-score of 0.83. Although the performance remained consistent in the bacterial pneumonia group, there was a significant decrease in PPV (69.9%) and F1-score (0.74) in the viral pneumonia group.

CONCLUSION: An adult-trained model adequately detected pneumonia in pediatric patients aged 1 to 5 years. Though models trained exclusively on pediatric images performed better, leveraging adult-based algorithms and datasets can expedite pediatric AI research.

来源:J Am Coll Radiol,查看原文。版权归原作者所有。

引用

  1. Morcos G, Yi PH, Jeudy J, et al. Applying Artificial Intelligence to Pediatric Chest Imaging: Reliability of Leveraging Adult-Based Artificial Intelligence Models. J Am Coll Radiol. 2023. PMID: 37467869. doi:10.1016/j.jacr.2023.07.004

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。