Prediction of death in burn patients with septic shock with Pseudomonas aeruginosa using machine learning based techniques

BACKGROUND AND OBJECTIVESPseudomonas aeruginosa is an opportunistic bacterial pathogen and important cause of wound infection and sepsis in burn patients. In this study, using machine learning (ML) a new and valuable step has been taken to predict the risk of death in burn patients with septic shock infected with P. aeruginosa.MATERIALS AND METHODSIn this study, data from the records of burn patients with septic shock infected with P. aeruginosa, including demographic information (such as age, gender), percentage of burns and other clinical variables of these patients in the hospital of Rasht province, will use to predict the risk of death using two ML methods, SVM (Support Vector Machine) and ANN (Artificial Neural Network). The data will divide into two training and testing categories. AUC index (Area under curve) will used to evaluate the accuracy of the models.RESULTS AND DISCUSSIONIn this research, a new and valuable step in predicting the risk of death due to P. aeruginosa sepsis in burn patients and also identifying antibiotic resistance in patients with septic shock which infected with P. aeruginosa.CONCLUSIONThe advances achieved in the field of artificial intelligence (AI) today witness different applications in diagnostic/preventive medical sciences. Such research will be of great help to the decision-makers in the field of treatment and health and will greatly reduce the costs of treatment and diagnosis.

کلیدواژه ها:

Pseudomonas aeruginosa ، Burn Patients ، Machine Learning ، Artificial Intelligence ، Data Mining