Download PDF
Thromb Haemost
DOI: 10.1055/a-2806-3756
Original Article: Coagulation and Fibrinolysis

Practical Machine Learning Model for Early and Accurate Prediction of Disseminated Intravascular Coagulation Before Its Progression to an OVERT Stage

Authors

  • Yutaka Umemura

    1   Department of Traumatology and Acute Critical Medicine, The University of Osaka Graduate School of Medicine, Osaka, Japan
    2   Division of Trauma and Surgical Critical Care, Osaka General Medical Center, Osaka, Japan

  • Masataka Fujimoto

    3   School of Medicine and Health Sciences, College of Medical Sciences, University of Tsukuba, Tsukuba, Japan
    4   MeDiCU, Inc., Osaka, Japan

  • Takahiro Kinoshita

    4   MeDiCU, Inc., Osaka, Japan

  • Satoshi Fujimi

    2   Division of Trauma and Surgical Critical Care, Osaka General Medical Center, Osaka, Japan

  • Kazuma Yamakawa

    5   Department of Emergency and Critical Care Medicine, Osaka Medical and Pharmaceutical University, Takatsuki, Osaka, Japan

  • Jun Oda

    1   Department of Traumatology and Acute Critical Medicine, The University of Osaka Graduate School of Medicine, Osaka, Japan

Funding Information This study was supported by a Grant from Japan Society for the Promotion of Science; Grant-in-Aid for Scientific Research C, 24K12133.
Further Information
   Permissions and Reprints


Graphical Abstract

Abstract

Background

In patients with sepsis, anticoagulant therapy is expected to have maximal efficacy when administered before the development of sepsis-induced overt disseminated intravascular coagulation (DIC). This therapeutic strategy requires a valid method for real-time, early prediction of sepsis-induced DIC, which is likely to progress to an overt stage. We aimed to develop machine learning prediction models for overt DIC, based on the International Society on Thrombosis and Haemostasis criteria, using a large-scale electronic medical record database.

Methods

This multi-center, retrospective, observational study included adult patients with sepsis without overt DIC at day 1. The outcome was the development of overt DIC 24 hours after a certain time point. Predictive features were baseline data and time series data within 7 days postadmission. We constructed three separate models (minimum, compact, and full models), each incorporating increasingly comprehensive sets of features. Prediction models were constructed with machine learning algorithms, including eXtreme Gradient Boosting (XGBoost) and gradient boosting machine (GBM), and evaluated on a randomly selected 20% test set.

Results

Among 7,532 patients with sepsis, 766 developed overt DIC within 7 days. XGBoost and GBM achieved the highest prediction accuracy. The full, compact, and minimum models on the test set exhibited area under the receiver operating characteristic curve values of 0.914, 0.884, and 0.851, respectively. The full XGBoost model achieved an area under the precision–recall curve of 0.295; at 80% recall, its precision was 14.4%.

Conclusion

The machine learning model exhibited high accuracy in predicting overt DIC at a clinically reliable level.

Keywords

blood coagulation disorders - prediction - disseminated intravascular coagulation - machine learning - sepsis

Data Availability Statement

The data used in the analysis of this study cannot be shared due to the terms of the data provider.


Supplementary Material



Publication History

Received: 24 August 2025

Accepted after revision: 04 February 2026

Accepted Manuscript online:
10 February 2026

Article published online:
18 February 2026

© 2026. Thieme. All rights reserved.

Georg Thieme Verlag KG
Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany