A Decision Support System based on Support Vector Machine for Diagnosis of Periodontal Disease

Background and Objective: Early and accurate diagnosis of many diseases is crucial in the process of their treatment.On the other hand, the presence of many unknown variables leads to more complicated decision making. In recent years, various machine learning methods such as neural networks, fuzzy logic and support vector machines (SVM) have been developed to solve complex medical decision problems. Therefore, this study aimed to design a supportive vector-based decision-making support system to diagnosis various periodontitis diseases.Materials and Methods: Data were collected from 300 patients referring to the Periodontics department of Hamadan Dental School between 2016 and 2018. Among these patients, 160 had Gingivitis disease, and 140 had periodontitis. In the designed classification model, 11 variables such as age, sex, smoking, gingival index, plaque index and so on were used as input and output variable shows the individual s health status and include three states (Gingivitis, Localized Periodontitis, Generalized Periodontitis). Also, the performance of different Kernel functions was evaluated.Finding: Out of 105 patients with gingivitis in the training data set, SVM classification model based on radial kernel function was able to correctly identify 93 patients and 12 patients mistakenly diagnosed periodontitis. Also, out of 95 patients with periodontitis in the training data set, the radial function SVM classification model was able to correctly identify 85 patients and 10 patients have diagnosed gingivitis mistakenly. Using different kernel functions in the design of the SVM classification model showed that the radial kernel function has the best performance with an accuracy of 84% for the test set and 89% for the train data set.Conclusion: Designing accurate decision-making support systems facilitates and accelerates diagnostic processes. The system designed in this study will help less experienced dentists and young residents in making decisions for diagnosis of periodontal disease

Keywords:

Periodontics , Support Vector Machine , Classification , Diagnosis , Decision Support System