Scholarlink

Background <p>Ovarian cancer patients requiring intensive care unit (ICU) admission face particularly grave prognosis, yet current prognostic models rely on static baseline characteristics and generic severity scores, neglecting the rich temporal dynamics of vital signs that may better capture physiological deterioration patterns.</p> Objectives <p>To develop and validate an interpretable Long Short-Term Memory (LSTM) neural network integrating hourly vital signs and static features for predicting 28-day mortality in ICU-admitted ovarian cancer patients.</p> Methods <p>This retrospective multicenter study utilized MIMIC-IV (2008–2022) and SICdb (2013–2021) databases. Adult ovarian cancer patients with first ICU admission, length of stay ≥ 24&#xa0;h, and complete hourly measurements of six vital signs (blood pressures, heart rate, respiratory rate, oxygen saturation) were included. MIMIC-IV was split into training (<i>n</i> = 269) and internal validation (<i>n</i> = 115) sets, while SICdb (<i>n</i> = 154) served as external validation. A dual-pathway LSTM network integrated temporal and static features (age, BMI, SOFA score). Model performance was assessed using AUROC, calibration metrics, and compared against traditional ICU scores and five static-only machine learning baselines. Five-fold cross-validation and 100 repeated temporal ablation trials assessed robustness. SHapley Additive exPlanations (SHAP) analysis quantified feature importance and temporal dynamics.</p> Results <p>The LSTM model achieved AUROC of 0.845 (95%CI: 0.789–0.901) in training, 0.785 (95%CI: 0.655–0.915) in internal validation, and 0.767 (95%CI: 0.617–0.917) in external validation, significantly outperforming SOFA (AUC: 0.683–0.733), SAPS II (AUC: 0.743), SAPS 3 (AUC: 0.678), and OASIS (AUC: 0.659 − 0.569). Five-fold cross-validation confirmed stability (mean AUROC: 0.785 ± 0.018). Negative predictive values exceeded 94% across all cohorts. The LSTM achieved AUC improvements of 0.10–0.17 over static baselines, with repeated permutation trials confirming significant temporal contributions (all <i>P</i> &lt; 0.001). Temporal ablation identified the 12–24&#xa0;h window as most critical (ΔAUC = 0.062). SHAP analysis revealed static features dominated prediction, while respiratory rate and diastolic blood pressure were leading temporal contributors.</p> Conclusions <p>This interpretable LSTM model accurately predicts mortality in ICU-admitted ovarian cancer patients by capturing temporal vital sign dynamics, significantly outperforming traditional severity scores and enabling personalized risk stratification for clinical decision-making.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Temporal deep learning for 28-day mortality prediction in critically ill ovarian cancer patients: a multicenter development and validation study using hourly vital signs

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月29日

查看全文
  • Chengling Wang,
  • Li Liu,
  • Jianguo Hu,
  • Xingshu Li,
  • Xingchuan Li,
  • Qikun Zhu,
  • Liang Yang,
  • Ke Pu,
  • Renlan Li

摘要

Background

Ovarian cancer patients requiring intensive care unit (ICU) admission face particularly grave prognosis, yet current prognostic models rely on static baseline characteristics and generic severity scores, neglecting the rich temporal dynamics of vital signs that may better capture physiological deterioration patterns.

Objectives

To develop and validate an interpretable Long Short-Term Memory (LSTM) neural network integrating hourly vital signs and static features for predicting 28-day mortality in ICU-admitted ovarian cancer patients.

Methods

This retrospective multicenter study utilized MIMIC-IV (2008–2022) and SICdb (2013–2021) databases. Adult ovarian cancer patients with first ICU admission, length of stay ≥ 24 h, and complete hourly measurements of six vital signs (blood pressures, heart rate, respiratory rate, oxygen saturation) were included. MIMIC-IV was split into training (n = 269) and internal validation (n = 115) sets, while SICdb (n = 154) served as external validation. A dual-pathway LSTM network integrated temporal and static features (age, BMI, SOFA score). Model performance was assessed using AUROC, calibration metrics, and compared against traditional ICU scores and five static-only machine learning baselines. Five-fold cross-validation and 100 repeated temporal ablation trials assessed robustness. SHapley Additive exPlanations (SHAP) analysis quantified feature importance and temporal dynamics.

Results

The LSTM model achieved AUROC of 0.845 (95%CI: 0.789–0.901) in training, 0.785 (95%CI: 0.655–0.915) in internal validation, and 0.767 (95%CI: 0.617–0.917) in external validation, significantly outperforming SOFA (AUC: 0.683–0.733), SAPS II (AUC: 0.743), SAPS 3 (AUC: 0.678), and OASIS (AUC: 0.659 − 0.569). Five-fold cross-validation confirmed stability (mean AUROC: 0.785 ± 0.018). Negative predictive values exceeded 94% across all cohorts. The LSTM achieved AUC improvements of 0.10–0.17 over static baselines, with repeated permutation trials confirming significant temporal contributions (all P < 0.001). Temporal ablation identified the 12–24 h window as most critical (ΔAUC = 0.062). SHAP analysis revealed static features dominated prediction, while respiratory rate and diastolic blood pressure were leading temporal contributors.

Conclusions

This interpretable LSTM model accurately predicts mortality in ICU-admitted ovarian cancer patients by capturing temporal vital sign dynamics, significantly outperforming traditional severity scores and enabling personalized risk stratification for clinical decision-making.