An Intelligent Decision Support System for the Surgical Preoperative Phase: An Approach Based on Machine Learning

In recent years, Artificial Intelligence (AI) has experienced an unprecedented resurgence of interest thanks to significant technological advances and is beginning to impose itself in various fields. Artificial Intelligence techniques and in particular machine learning are emerging in the medical field, particularly in surgery. In this context, managers have noted that the operating room is one of the most decisive resources of a hospital structure and that improving its operational efficiency and patient safety is a major priority, in particular the preoperative phase, deterministic phase which anticipates the other phases. Through this paper, we will present a new approach, which is based initially on a machine learning algorithm, it is: Random Forest, going to another algorithm: Support Vector Machine (SVM). These models will be used in a decisive surgical phase, through the analysis of the predictive data of the assessments, in order to propose a system of aid for the preoperative decision of the patients. A comparative study between these two models will be retained, based on experimental phases, to better strengthen our approach. We will be using a fame dataset; it is an Institutional Review Board approved dataset, including relevant predictors. While using a development environment dedicated to the field of Artificial Intelligence. An important phase will also be reserved for the evaluation of the performance of each model during the prediction of the preoperative results. During the research path, we will draw a considerable connection which argues the choice of this pair (Random Forest, SVM). This idea gives us the inspiration for the reflection toward a perspective of a hybrid model based on these last two models.