Skip to main content

Comparison of IGS 2 score and a new neural network biomarker score performance in predicting 1-year mortality

Introduction

The aim of this study was to compare the IGS 2 score with a predictive model built using clinical characteristics and biomarkers associated with 1-year mortality from procalcitonin, C-reactive protein, troponin I (TnI) and NT-Pro-BNP.

Methods

We conducted a prospective observational cohort study in a polyvalent ICU. We calculated the IGS 2 severity score in the first 24 hours. We dosed NT-Pro-BNP, TnI, procalcitonin and C-reactive protein at admission to the ICU. To test the biomarker model, the patient database was divided into training (70%) and test (30%) databases. A neural network model was trained from variables associated with 1-year mortality. All statistics were done using R v.2.15.1. We used the pROC package to plot and compare ROC curves.

Results

Lower age, NT-Pro-BNP and TnI were significantly associated with 1-year survival. We built a neural network model using the training database. We used the test database to compare our model with the IGS 2 score. The difference between the AUC of the biomarker score and the IGS 2 score was statistically significant (Figure 1).

Figure 1
figure1

Comparison of area under ROC curves (AUC) of the biomarker score and IGS 2.

Conclusion

Our new neural network model built using age, NT-PRO-BNP and TnI succeeded the internal-validation processes and predicts 1-year mortality better than IGS 2.

Author information

Affiliations

Authors

Corresponding author

Correspondence to C Romdhani.

Rights and permissions

This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and Permissions

About this article

Cite this article

Romdhani, C., Labbene, I., Amor, M.B. et al. Comparison of IGS 2 score and a new neural network biomarker score performance in predicting 1-year mortality. Crit Care 17, P480 (2013). https://doi.org/10.1186/cc12418

Download citation

Keywords

  • Predictive Model
  • Severity Score
  • Neural Network Model
  • Score Performance
  • Procalcitonin