The Diagnostic Model of Neonatal Respiratory Distress Syndrome Based on Intelligent Algorithm

Abstract

The paper described a rapid decision support model for neonatal respiratory distress syndrome (NRDS), which was suitable for extensive neonatal related diseases for diagnose and identification rapidly. The available data, collected in No.307 hospital of PLA, was provided to several intelligent algorithms(artificial neural networks, random forests, support vector machines) to create a model for predicting the NRDS probability for newborns. It showed that prediction accuracy of the model for NRDS could reach up to 98.07% in test. We observed that predictions of the model are in agreement with the literature, demonstrating that model might be an important tool for supporting decision making in medical practice. Other feature of this method were the input parameters could be obtained easily in clinic and the implementation of the risk assessment could provide rapid decision support information for clinic.