Research On The Method For Measuring The Length Of Root Canal Based On Multifrequency Impedance Ratio Method

Oral health is considered as one of the top ten standards for human health by the World Health Organization.With the rapid development of the social economy,people pay more and more attention to oral health.Root canal treatment is the most fundamental and effective method for dealing with oral problems.The key to the success of root canal treatment is the accurate measurement of the working length of the root canal.The use of an electronic apex locator(EAL)is the most common method for measuring root canal length with high accuracy.However,due to the complexity of the measurement environment,including the shape of the root canal,the size of the apical foramen and the remnant of the root canal,the measurement accuracy is greatly affected,so that there is a certain measurement error in the measurement.Doctors often need to use the radiograph for comprehensive judgment.Besides that,in the domestic market,imported products are dominating.High prices and severe technical barriers make the development of EAL almost stagnant.Aiming at the above problems,this paper designs and implements a prototype of EAL,and proposes a method for measuring the length of root canal based on multifrequency impedance ratio method combined with machine learning to improve the accuracy and stability when measurement conditions change.The main research contents of this paper include:1.The design and implementation of the hardware system of EAL.The EAL proposed in this paper uses the embedded processor LPC1857 as the CPU to control the direct digital synthesizer to generate the sinusoidal signal of sweep frequency as the signal source.It uses the circuit of signal acquisition and the module of serial communication to complete the acquisition and transmission of impedance.2.Software design and programming debugging of EAL.The program design adopts functional modular design and layered design.The functional modules are correspond to the respective hardware requirements,and are encapsulated into APIs for the main function to perform logical calls.3.Algorithm design for determining the position of the apical foramen.The prediction model is established,and the effective features of the model are selected by feature selection.Through the performance comparison between the linear regression model and the neural network model,and the performance comparison between different optimization algorithms,the final model of determining the position of the apical foramen is determined.4.Design and implementation of the experiment.Through the measurement of multiple sets of root canals,the results show that the prototype of EAL in this research has an accuracy up to 92.65% and strong robustness after using the method based on multifrequency impedance ratio method combined with machine learning.