论文部分内容阅读
选择20名具有完整恒牙列的正常受试者,采集他们在正中(牙合)叩齿和磨牙时所产生的(牙合)音。对所采的(牙合)音建立自回归模型,依据模型系数计算(牙合)音的功率谱,分析其频率成分;并用模型系数建立Bayes'判别函数,对叩牙及磨牙产生的咬合音作识别。结果发现诸如累积功率频率,平均功率频率和谱峰频率等反映频率特征的指标,在二类(牙合)音中呈相似状态;然而Bayes'判别则有95％的识别符合率。结果提示,采用自回归模型的模式识别方法,在临床上诊断(牙合)音中有应用前景。
Twenty normal subjects with intact permanent dentition were selected and their occlusion sound was recorded during the occlusion and molars. Based on the model coefficients, the power spectrum of the occlusion sound was calculated and its frequency components were analyzed. The Bayes' discriminant function was established by using the model coefficients to determine the occlusion sound of the tapping teeth and molars For identification. The results showed that the indicators reflecting the frequency characteristics such as cumulative power frequency, average power frequency and spectral peak frequency were similar in the two types of occlusion sounds; however, Bayes' discriminant had 95% coincidence rate. The results suggest that the pattern recognition method using autoregressive model has clinical application in the diagnosis of occlusion.