Investigation on wear prediction model of dental restoration material based on ensemble learning

Abstract

The wear test was performed with TC4 alloy, by changing the normal bite force, sliding frequency and cycles, in artificial saliva. Taking the two results for testing samples and the others for training samples, the radial basis function (RBF) and multilayer perceptron (MLP) neural network model and least mean square (LMS) and K* model were built for predicting wear loss respectively. Then an ensemble learning model was built which integrated all the single models based on the weight determined by mean absolute error. Compared with the testing results, it was obtained that the error for ensemble learning model was between 3 and 4%. Also, its prediction error rate reduced over 50% for the tenth group data, which embodied good stability and high precision on predicting the wear loss for dental restorative material.

The accuracy of the prediction models for surface roughness and micro hardness of denture teeth

“…However, no research on the surface roughness and hardness prediction model of dental materials has been reported. There is only one study on wear prediction model of dental restoration material 19) . Since the resulting wear mechanisms differ according to the adjustment of the simulator, the study figured out that it is not correct to compare the wear results, unless the wear devices are identical.…”

“…Since the resulting wear mechanisms differ according to the adjustment of the simulator, the study figured out that it is not correct to compare the wear results, unless the wear devices are identical. Using machine learning method for wear prediction studies is preferred with an increasing trend in recent years 19) .…”

“…The usage of ANN for modeling composite materials is recommended rather than an in vitro study which is time consuming and expensive 18) . ANN is more successful when there is nonlinearity among the variables, mathematical formulas are not able to describe the function of various factors precisely 16,19,33) .…”

“…Any update about the weights of the neural network should be recorded simultaneously, so that it can learn in real time. It would then be effective for real time applications 19) .…”

“…The related literature reveals that ANN has been used for modeling purposes in the several fields of sciences. However, among dental researches, there is only one study 19) about the prediction of dental restoration material. They investigated the wear prediction model of titanium alloy and found that ensemble learning model had good stability and high precision on predicting the wear loss.…”