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

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 () and Bailin Niu ()
Additional contact information
Li Zhu: Nanchang University, School of Information Engineering, Jiangxi Provincial Key Laboratory of Advanced Signal Processing and Intelligent Communications
Zengtian Chen: Nanchang University, School of Information Engineering, Jiangxi Provincial Key Laboratory of Advanced Signal Processing and Intelligent Communications
Hong Zhang: Chongqing University, Department of Laboratory Medicine, Chongqing Center for Clinical Laboratory, Chongqing Academy of Medical Sciences, Chongqing General Hospital, School of Medicine
Hongjun Chen: Nanchang University, School of Information Engineering, Jiangxi Provincial Key Laboratory of Advanced Signal Processing and Intelligent Communications
Lanqi Liu: Nanchang University, School of Medicine
Wei Yu: The Affiliated Lihuili Hospital of Ningbo University, Ningbo Institute of Innovation for Combined Medicine and Engineering (NIIME)
Kai Wu: Nanchang University, School of Information Engineering, Jiangxi Provincial Key Laboratory of Advanced Signal Processing and Intelligent Communications
Yijin Chen: The Affiliated Lihuili Hospital of Ningbo University, Ningbo Institute of Innovation for Combined Medicine and Engineering (NIIME)
Xingyu Tao: Chongqing University, Department of Intensive Care Medicine, Chongqing Emergency Medical Center, Chongqing University Central Hospital, School of Medicine
Zefeng Yu: Nanchang University, School of Information Engineering, Jiangxi Provincial Key Laboratory of Advanced Signal Processing and Intelligent Communications
Linhui Shi: The Affiliated Lihuili Hospital of Ningbo University, Department of Intensive Care Medicine
Jialian Wang: Chongqing University, Department of Intensive Care Medicine, Chongqing Emergency Medical Center, Chongqing University Central Hospital, School of Medicine
Fan Zhang: Qilu Hospital of Shandong University, Department of Gastroenterology
Jiaying Shen: The Affiliated Lihuili Hospital of Ningbo University, Ningbo Institute of Innovation for Combined Medicine and Engineering (NIIME)
Fen Liu: Nanchang University, School of Medicine
Chongke Hu: The Affiliated Lihuili Hospital of Ningbo University, Ningbo Institute of Innovation for Combined Medicine and Engineering (NIIME)
Yangguang Ren: Kunming Medical University, College of Life Science and Laboratory Medicine
Tzu-Ming Liu: University of Macau, Institute of Translational Medicine, Faculty of Health Sciences & Ministry of Education Frontiers Science Center for Precision Oncology
Yang Luo: Chongqing University, Department of Laboratory Medicine, Chongqing Center for Clinical Laboratory, Chongqing Academy of Medical Sciences, Chongqing General Hospital, School of Medicine
Fei Guo: The Affiliated Lihuili Hospital of Ningbo University, Ningbo Institute of Innovation for Combined Medicine and Engineering (NIIME)
Bailin Niu: Chongqing University, Department of Intensive Care Medicine, Chongqing Emergency Medical Center, Chongqing University Central Hospital, School of Medicine

Nature Communications, 2025, vol. 16, issue 1, 1-15

Abstract: Abstract 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.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.nature.com/articles/s41467-025-65365-z Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:16:y:2025:i:1:d:10.1038_s41467-025-65365-z

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-025-65365-z

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().