GRANULAR COMPUTING BASED ON COMPLEX FUZZY SIMILARITY MEASURES IN DENTAL DIAGNOSIS SUPPORT

Abstract

The application of support techniques in disease examination is an important factor in reducing the overloading of doctors. The problem is howcome the technologies can increase the accuracy of diagnosing. In this paper, we develop an model integrating granular computing and complex fuzzy similarity measures.  Firstly, complex fuzzy similarity measures are used in order to evaluate the similarity degree among standard samples and the diagnosis samples. Then, the granular computing is applied to select the highest ability of the disease that the patients can be affected. The proposed model is implemented on dental dataset including X-ray images of wisdom teeth deviate. The experimental results show that the novel model gets higher accuracy than other related methods. This research supports to the dentists in wisdom teeth deviate diagnosing.