Predicting the Need of Mechanical Ventilation in Guillain-Barré Patients Using Machine Learning Algorithms with Relevant Features

摘要: Guillain-Barre Syndrome (GBS) is an autoimmune neurological disorder characterized by a fast evolution. Almost third of patients with this condition presents breathing difficulty and need mechanical device to assist them. We aim at creating diagnostic model the for ventilation in GBS. use experimentation real dataset that contains clinical, serological, nerve conduction tests data. In dataset, 41 out total 122 required ventilation. JRip, SVM (Support Vector Machines) linear kernel C4.5 are used create predictive models. examine whether selecting relevant variables through filter methods makes possible increase accuracy model. The analyzed are: symmetrical uncertainty, chi squared information gain. An accurate was obtained after experimentation.