MTS: Improving Thoracic Surgery Based on Clustering and Classification Techniques

Abstract

Thoracic and cardiothoracic surgery is one of the most dangerous diseases that may face a big problem. When performing such operations, cardiothoracic surgery is a medical field that specializes in surgery for diseases that affect the rib cage and heart. The major challenge is to address outliers and missing values problems for predicting thoracic disease and improving its performance. In this paper, we proposed an effective technical model to improve the performance of thoracic surgery. The current authors applied two experiments. The first experiment includes the application of the clustering task as one of the unsupervised machine learning methods. The supervised techniques were applied for classification tasks in the second experiment. Using these two experiments, we got good results, and it improved the performance of chest surgeries, where the highest accuracy and F1 reached 86.17% and 92.56%, respectively. Therefore, our model is considered an effective proposal and outperforms its peers.