Detection of dental plaque using enhanced K-means and silhouette methods

Abstract

Detection of Dental plaque is important for the doctors, the victims as well as the researchers. Plaque is a Microbial Biofllm that continuously forms on the teeth surface, later it reacts with food materials that contains higher concentration of sugars, starches and then releases acids. Finally it attacks the tooth enamel causing gingivitis and other diseases if proper treatment is not given. Though there were dental plaque indices existing but they could not produce better results. The existing methodology mainly depends on the visual Capacity of the dentist to isolate or for revealing the areas of plaque. So to overcome the problem, this paper proposed a solution for getting the results of dental plaque without human intervention using enhanced K-Means unsupervised clustering algorithm on HSI (hue, saturation and intensity) color model. Subsequently, silhouette method is applied on the enhanced K-Means algorithm to evaluate the quality of clustering. The experimental results are better than the traditional approach.