Scholarlink

<p>Sepsis is a leading cause of hospital mortality, and its significant heterogeneity complicates prognosis and stratification. To address this challenge, we developed an explainable artificial intelligence prognostic model (SepsisFormer, a transformer-based neural network) and an automated risk-stratification tool (SMART) for sepsis. In a multi-center retrospective study of 12,408 sepsis patients, SepsisFormer achieved high predictive accuracy (AUC: 0.9301, sensitivity: 0.9346, and specificity: 0.8312). SMART (AUC: 0.7360) surpassed most established scoring systems. Seven coagulation-inflammatory routine laboratory measurements and patient age were identified to classify patients’ four risk levels (mild, moderate, severe, dangerous) and two subphenotypes (CIS1 and CIS2), each with distinct clinical characteristics and mortality rates. Notably, patients with moderate/severe levels or CIS2 derive more significant benefits from anticoagulant treatment. Our work, therefore, offers a set of simple, real-time executable tools for sepsis heterogeneity, demonstrating the potential to enhance sepsis clinical practice globally, particularly in resource-constrained healthcare settings.</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

Explainable AI unravels sepsis heterogeneity via coagulation-inflammation profiles for prognosis and stratification

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月24日

查看全文
  • Li Zhu,
  • Zengtian Chen,
  • Hong Zhang,
  • Hongjun Chen,
  • Lanqi Liu,
  • Wei Yu,
  • Kai Wu,
  • Yijin Chen,
  • Xingyu Tao,
  • Zefeng Yu,
  • Linhui Shi,
  • Jialian Wang,
  • Fan Zhang,
  • Jiaying Shen,
  • Fen Liu,
  • Chongke Hu,
  • Yangguang Ren,
  • Tzu-Ming Liu,
  • Yang Luo,
  • Fei Guo,
  • Bailin Niu

摘要

Sepsis is a leading cause of hospital mortality, and its significant heterogeneity complicates prognosis and stratification. To address this challenge, we developed an explainable artificial intelligence prognostic model (SepsisFormer, a transformer-based neural network) and an automated risk-stratification tool (SMART) for sepsis. In a multi-center retrospective study of 12,408 sepsis patients, SepsisFormer achieved high predictive accuracy (AUC: 0.9301, sensitivity: 0.9346, and specificity: 0.8312). SMART (AUC: 0.7360) surpassed most established scoring systems. Seven coagulation-inflammatory routine laboratory measurements and patient age were identified to classify patients’ four risk levels (mild, moderate, severe, dangerous) and two subphenotypes (CIS1 and CIS2), each with distinct clinical characteristics and mortality rates. Notably, patients with moderate/severe levels or CIS2 derive more significant benefits from anticoagulant treatment. Our work, therefore, offers a set of simple, real-time executable tools for sepsis heterogeneity, demonstrating the potential to enhance sepsis clinical practice globally, particularly in resource-constrained healthcare settings.