文献速递术前评估与围术期医学

选择偏倚对基于深度学习术中低血压预测模型性能的影响

The effect of selection bias on the performance of a deep learning-based intraoperative hypotension prediction model using real-world samples from a publicly available database.
PubMed 文献 · Br J Anaesth · 2025-05-22

本研究利用VitalDB公开数据集,评估选择偏倚对深度学习(DL)模型和仅用平均动脉压(MAP)模型预测术中低血压性能的影响。低血压事件定义为MAP<65 mm Hg持续1 min。在预测事件前5 min时,DL模型在无偏倚测试集与有偏倚测试集上的报警次数/小时分别为18.1 vs 10.8(P<0.001),阳性预测值分别为0.068 vs 0.937(P<0.001)。结论:两模型在无偏倚数据集上预测性能均变差,DL模型虽统计学优于MAP模型但临床意义不大。

要点

  • 选择偏倚会显著高估低血压预测模型的性能。
  • DL模型在无偏倚数据集上报警频繁且阳性预测值低。
  • 临床验证低血压预测模型时需注意数据集选择偏倚。

核心结论

  • DL模型和MAP模型在无偏倚数据集上预测性能均差于有偏倚数据集。
  • DL模型虽统计学优于MAP模型,但差异无临床意义。
  • 临床医生应考虑选择偏倚对低血压预测模型验证和临床性能的潜在影响。

研究设计与样本

  • 使用VitalDB开放数据集进行模型性能评估。
  • 低血压事件定义为MAP<65 mm Hg持续1 min。
  • 有偏倚数据集要求非低血压事件位于MAP>75 mm Hg持续>30 min的非低血压期中点,且与任何低血压事件间隔至少20 min;无偏倚数据集纳入所有样本,除非低血压事件已存在于输入段。
  • 比较DL模型和仅用MAP模型的每小时报警次数和阳性预测值。

主要发现

  • 预测事件前5 min,DL模型在有偏倚测试集上报警10.8次/小时,阳性预测值0.937;在无偏倚测试集上报警18.1次/小时,阳性预测值0.068(均P<0.001)。

临床怎么用

  • 临床选用低血压预测模型时应关注其验证数据集是否存在选择偏倚。
  • 不应仅依据有偏倚数据集上的高性能指标判断模型临床价值。

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)。

原文

BACKGROUND: There are models to predict intraoperative hypotension from arterial pressure waveforms. Selection bias in datasets used for model development and validation could impact model performance. We aimed to evaluate how selection bias affects the predictive performance of a deep learning (DL)-based model and a model using only mean arterial pressure (MAP) as input (MAP-only model). METHODS: We used the VitalDB open dataset. A hypotensive event was defined as a MAP <65 mm Hg for 1 min. For the 'biased dataset', 'non-hypotensive events' needed to be (a) at the centre of a 'non-hypotensive period' with a MAP of >75 mm Hg for more than 30 continuous minutes and (b) at least 20 min apart from any hypotensive event. For the 'unbiased dataset', all samples were included unless the hypotensive event was already in the input segment. The alarms per hour and positive predictive values were compared between the DL and MAP-only models. RESULTS: The DL model generally performed better than the MAP-only model. For the prediction of intraoperative hypotension 5 min before the event with the DL model, using the unbiased vs the biased testing dataset resulted in 18.1 vs 10.8 alarms per hour (P<0.001) and a positive predictive value of 0.068 vs 0.937 (P<0.001). CONCLUSIONS: Both the DL model and the MAP-only model demonstrated worse predictive performance when tested on the unbiased dataset compared with the biased dataset. Although the DL model statistically performed better than the MAP-only model, the difference between the two models was not clinically meaningful. Clinicians should consider the potential impact of selection bias on the validation and the clinical performance of hypotension prediction models. CLINICAL TRIAL REGISTRATION: NCT02914444.

来源:Br J Anaesth,查看原文。版权归原作者所有。

引用

  1. Yang HL, Celi LA, Lee H, et al. The effect of selection bias on the performance of a deep learning-based intraoperative hypotension prediction model using real-world samples from a publicly available database. Br J Anaesth. 2025. PMID: 40404499. doi:10.1016/j.bja.2025.03.024

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。