Artificial neural networks improve early outcome prediction and risk classification in out-of-hospital cardiac arrest patients admitted to intensive care

Abstract

Background: Pre-hospital circumstances, cardiac arrest characteristics, comorbidities and clinical status on admission are strongly associated with outcome after out-of-hospital cardiac arrest (OHCA). Early prediction of outcome may inform prognosis, tailor therapy and help in interpreting the intervention effect in heterogenous clinical trials. This study aimed to create a model for early prediction of outcome by artificial neural networks (ANN) and use this model to investigate intervention effects on classes of illness severity in cardiac arrest patients treated with targeted temperature management (TTM).

Methods: Using the cohort of the TTM trial, we performed a post hoc analysis of 932 unconscious patients from 36 centres with OHCA of a presumed cardiac cause. The patient outcome was the functional outcome, including survival at 180 days follow-up using a dichotomised Cerebral Performance Category (CPC) scale with good functional outcome defined as CPC 1-2 and poor functional outcome defined as CPC 3-5. Outcome prediction and severity class assignment were performed using a supervised machine learning model based on ANN.

Results: The outcome was predicted with an area under the receiver operating characteristic curve (AUC) of 0.891 using 54 clinical variables available on admission to hospital, categorised as background, pre-hospital and admission data. Corresponding models using background, pre-hospital or admission variables separately had inferior prediction performance. When comparing the ANN model with a logistic regression-based model on the same cohort, the ANN model performed significantly better (p = 0.029). A simplified ANN model showed promising performance with an AUC above 0.852 when using three variables only: age, time to ROSC and first monitored rhythm. The ANN-stratified analyses showed similar intervention effect of TTM to 33 °C or 36 °C in predefined classes with different risk of a poor outcome.

Conclusion: A supervised machine learning model using ANN predicted neurological recovery, including survival excellently, and outperformed a conventional model based on logistic regression. Among the data available at the time of hospitalisation, factors related to the pre-hospital setting carried most information. ANN may be used to stratify a heterogenous trial population in risk classes and help determine intervention effects across subgroups.

Keywords: Artificial intelligence; Artificial neural networks; Cerebral performance category; Critical care; Intensive care; Machine learning; Out-of-hospital cardiac arrest; Prediction; Prognostication.

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Fig. 1

ANN, artificial neural network. A schematic ANN with one input layer, two hidden layers and one single output layer. All nodes in the network are connected in resemblance to the human central nervous system. The input layers in our ANN consisted of variables (background, pre-hospital and/or admission data) whereas the output layer was the outcome variable Cerebral Performance Category (CPC) scale dichotomised into good (CPC 1–2) or poor (CPC 3–5) functional outcome

Fig. 2

Prediction performance. The prediction performance of long-term functional outcome is expressed as AUC in a ROC curve, by an ANN model using all 54 variables available on admission to intensive care. Of the 932 patients included in the study, 93 patients (10%) was randomly chosen and removed from the training set on which the ANN algorithm trained its prediction model. The trained ANN was then used to make a prediction of the outcome on the 93 patients earlier removed to represent the test set. The mean AUC for our ANN was 0.891, indicating an excellent performance to predict long-term outcome. AUC, area under the curve; ROC, receiver operating characteristics; ANN, artificial neural network

Fig. 3

Prediction performance in comparison. Comparison of the prediction performance of long-term outcome expressed as AUC in ROC curves, between our ANN model (blue) and the TTM risk score (green) from Martinell et al. The ANN model (AUC = 0.904) outperformed the TTM risk score (AUC = 0.839) significantly (p = 0.029) in a comparative analysis based on 80 patients (test set) from the TTM data set. Since the “TTM risk score” does not have a strategy for handling missing values, 13 patients were removed from the original test set with 93 patients when comparing the two models. The ANN AUCs in Figs. 2 and 3 differ for the same reason. AUC, area under the curve; ROC, receiver operating characteristics; ANN, artificial neural network; TTM, targeted temperature management

Fig. 4

Increased prediction performance when adding variables. The change in AUC during training (AUCCV) when adding one predictor at the time and running the optimization process each time. The predictive performance of the model (represented by the blue line and its corresponding CI in green area) initially increased rapidly, but then levelled out, gradually approaching the reference AUC (represented by the dotted line and its corresponding CI in the pink area) of the model using all 54 variables. After adding five variables, there was no significant difference between the two models regarding prediction performance, marked by a red X in the figure. AUC, area under the curve; CI, confidence interval

Fig. 5

Diagnostic odds ratio for the artificial neural network (ANN)-stratified risk groups The forest plot shows the logarithmic diagnostic odds ratio for five ANN-stratified risk groups of CPC score > 2 and its association to treatment with targeted temperature management at 33 °C and 36 °C. A diagnostic odds ratio > 1 implies a better functional outcome when treated with 36 °C compared to 33 °C. CPC, cerebral performance category