ICU 死亡预测中的生理轨迹与观测模式：测量条件变化下的可迁移性
Physiological Trajectories and Observation Patterns in ICU Mortality Prediction: Transportability Under Changing Measurement Conditions
Benjamin T. Pike
中文摘要
目标：区分生理轨迹与观测模式在 ICU 死亡预测中各自可迁移的预测价值，并评估测量时机是否独立于观测数量影响性能。材料与方法：使用 MIMIC-III、MIMIC-IV 和 eICU 在 48 小时预测院内死亡。嵌套表示逐步整合最新值快照（R1）、纵向生理摘要（R2）以及测量频率与时机特征（R3）。在 7,762 名 MetaVision 患者上拟合的模型，于 204 家医院的 21,682 名 MIMIC-IV 患者和 43,878 名 eICU 患者上评估。在保持保留观测数不变的情况下评估测量时机的影响。结果：生理轨迹改善了 eICU 的外部预测（R2-R1 对数损失差 -0.00922，95% CI -0.01148 至 -0.00704）。观测模式特征改善了本地开发的 eICU 模型（R3-R2 -0.00257，95% CI -0.00370 至 -0.00146），但从 MetaVision 迁移后不明显（-0.00012，95% CI -0.00206 至 +0.00208）。在 30 次源 bootstrap 重拟合中，R3 在 10 次中占优。在匹配观测数时，保留分布在整个窗口内的测量相比随机保留使对数损失降低：R3 为 0.03280，R2 为 0.00703。移除最后 12 小时相比数量匹配的随机移除使对数损失增加：R3 为 0.01947，R2 为 0.00822。在可用性降低时，本地 R3 增量从有益变为有害。讨论：生理轨迹迁移更一致，而观测模式特征对测量时机和可用性特别敏感。结论：临床预测模型不仅应在新的患者人群中评估，还应在提供其输入的测量条件下评估。
关键要点
- 01区分生理轨迹与观测模式的可迁移预测价值，并检验测量时机是否独立于观测数量影响性能。
- 02在 MIMIC-III 拟合、MIMIC-IV 与 eICU 外部验证，覆盖 204 家医院。
- 03生理轨迹在 eICU 外部预测中稳定改善；观测模式特征迁移后增益不明确。
- 04匹配观测数时，测量在时间窗内分散保留比随机保留带来更大增益。
- 05测量可用性降低时，本地观测模式特征的增益由有益转为有害。
解读
尚无解读。
原始英文摘要
arXiv:2610.42126v1 Announce Type: new Abstract: Objective: To distinguish the transferable predictive value of physiological trajectories from that of observation patterns in ICU mortality prediction, and to assess whether measurement timing affects performance independently of observation quantity. Materials and Methods: In-hospital mortality was predicted at 48 hours using MIMIC-III, MIMIC-IV, and eICU. Nested representations progressively integrated latest-value snapshots (R1), longitudinal physiological summaries (R2), and measurement frequency and timing features (R3). Models fitted on 7,762 MetaVision patients were evaluated on 21,682 MIMIC-IV patients and 43,878 eICU patients across 204 hospitals. The effect of measurement timing was assessed while holding the number of retained observations constant. Results: Physiological trajectories improved external prediction in eICU (R2-R1 log-loss difference -0.00922, 95% CI -0.01148 to -0.00704). Observation-pattern features improved locally developed eICU models (R3-R2 -0.00257, 95% CI -0.00370 to -0.00146), but not clearly after transfer from MetaVision (-0.00012, 95% CI -0.00206 to +0.00208). R3 was favoured in 10 of 30 source bootstrap refits. At matched observation counts, retaining measurements spread throughout the window reduced log loss relative to random retention by 0.03280 for R3 versus 0.00703 for R2. Removing the final 12 hours increased log loss relative to count-matched random removal by 0.01947 for R3 versus 0.00822 for R2. Under reduced availability, the local R3 increment changed from beneficial to harmful. Discussion: Physiological trajectories transferred more consistently, while observation-pattern features were particularly sensitive to the timing and availability of measurements. Conclusion: Clinical prediction models should be evaluated not only in new patient populations but also under the measurement conditions that supply their inputs.