Multi-level BRB Inference-Based Diagnosis for Large Intestinal Diseases

Large intestinal disease (LID) is a common digestive tract disease, which shows a high morbidity and mortality rate with the improvement of living materials and irregular living habits, seriously threatening human health and life. Therefore, it is necessary to diagnose LID promptly and rapidly in order to prevent further deterioration of the disease and thereby significantly reduce cancer and mortality rate. In this paper, we proposed a diagnostic model for LID based on a multi-level belief rule base (BRB) inference method. Specifically, the diagnosis of LID is performed by a multi-level BRB diagnostic model, and then the diagnostic model is optimized by Genetic Algorithm (GA) to improve the diagnostic accuracy. Finally, the feasibility of the model is verified by a real clinical dataset. Meanwhile, the proposed diagnostic model was compared with back-propagation neural network (BPNN) model and support vector machine (SVM) model. The experimental results show that the proposed method possesses higher accuracy for diagnosing LID.