Decision Model to Predict the Implant Success

摘要

In this work we propose to build a set of binary logistic models that could assess the probability of success or no success in oral rehabilitation process taking into account some genetic factors, individual habits clinical and non-clinical factors.The study was conducted in a retrospective evaluation and consisted of 155 subjects undergoing oral rehabilitation in the Northern region of Portugal. We evaluated multiple factors in the construction of binary logistic regression models. We have chosen the model that gave statistically better discriminating power between success and failure, through the value of area under the ROC curve. The model that reveals better performance was Model 4, with AUC = 0.789 and a 95% confidence interval [0.715; 0.863].