Parameter selection for and implementation of a web-based decision-support tool to predict extubation outcome in premature infants

Abstract

Background: Approximately 30% of intubated preterm infants with respiratory distress syndrome (RDS) will fail attempted extubation, requiring reintubation and mechanical ventilation. Although ventilator technology and monitoring of premature infants have improved over time, optimal extubation remains challenging. Furthermore, extubation decisions for premature infants require complex informational processing, techniques implicitly learned through clinical practice. Computer-aided decision-support tools would benefit inexperienced clinicians, especially during peak neonatal intensive care unit (NICU) census.

Methods: A five-step procedure was developed to identify predictive variables. Clinical expert (CE) thought processes comprised one model. Variables from that model were used to develop two mathematical models for the decision-support tool: an artificial neural network (ANN) and a multivariate logistic regression model (MLR). The ranking of the variables in the three models was compared using the Wilcoxon Signed Rank Test. The best performing model was used in a web-based decision-support tool with a user interface implemented in Hypertext Markup Language (HTML) and the mathematical model employing the ANN.

Results: CEs identified 51 potentially predictive variables for extubation decisions for an infant on mechanical ventilation. Comparisons of the three models showed a significant difference between the ANN and the CE (p = 0.0006). Of the original 51 potentially predictive variables, the 13 most predictive variables were used to develop an ANN as a web-based decision-tool. The ANN processes user-provided data and returns the prediction 0-1 score and a novelty index. The user then selects the most appropriate threshold for categorizing the prediction as a success or failure. Furthermore, the novelty index, indicating the similarity of the test case to the training case, allows the user to assess the confidence level of the prediction with regard to how much the new data differ from the data originally used for the development of the prediction tool.

Conclusion: State-of-the-art, machine-learning methods can be employed for the development of sophisticated tools to aid clinicians' decisions. We identified numerous variables considered relevant for extubation decisions for mechanically ventilated premature infants with RDS. We then developed a web-based decision-support tool for clinicians which can be made widely available and potentially improve patient care world wide.

Figure 1

Example of an Algorithm for Ventilator Management (adapted from Carlo and Martin, 1986).

Figure 2

Schematic representation of the prediction tool. Data entered by the user are delivered through the Internet to the ANN housed on an Apache server. The ANN that was programmed in MATLAB calculates the prediction, which is again returned through the Internet to the user and displayed in the Internet browser.

Figure 3

Variables Relevant for Outcome Prediction.

Figure 4

Data Entry Page for Decision-Support Tool. Thirteen variables are required for calculation of the prediction. Additional variables are requested for statistical purposes and future fine-tuning of the prediction tool. Once all required fields are completed, the information is submitted to the ANN by clicking the appropriate button.

Figure 5

Results page of decision-support tool. The ANN returns the prediction score along with the novelty index. For categorization of the prediction into success or failure, a table with threshold values and the appropriate sensitivity/specificity pairs is provided. The novelty index indicates the level of confidence in the prediction.