A Novel neuro-fuzzy assessment index for orthodontics

Orthodontists have developed several occlusal indices during the past few decades to evaluate treatment success. These indices suffer from several limitations including a crisp decision making criterion that sometimes contradicts human intuition. The objective of this study is to develop an expert system that represent orthodontists´ visual perception in assessing patients, using adaptive neuro-fuzzy approach and improving the quality of earlier indices. We examined the reliability of our model by comparing the results with the opinion of a panel of expert orthodontists. Our index has shown a greater consistency with practitioners´ clinical judgments compare to previous indices. We conclude that utilization of neural network in this design has led to a superior performance.