Research On Signal Detection And Processing For Portable Microfluidic Chip System

Microfluidic chip technology is an important branch of micro total analysis system.It has attracted much attention because of its many advantages,such as high separation efficiency,low sample consumption,and low environmental pollution.After decades of development,it has been penetrated into the fields of life science,biomedicine,analytical chemistry and so on.At present,the conventional microfluidic chip signal detection equipments are mostly designed for the laboratory applications.There are some problems such as large volume,high cost,and limited operation range and so on.Therefore,it is of great significance to study a portable and versatile microfluidic chip detection system.In addition,because the microfluidic chip signal is weak and vulnerable to high frequency noise and low frequency baseline drift,hence,signal denoising and baseline correction are the key problems of microfluidic chip signal detection and analysis system.Firstly,the working principle of microfluidic chip signal detection was analyzed in this paper,some different detection methods of microfluidic chip signals were compared,and the characteristics of microfluidic chip signals and the source of noise were introduced.These lay a theoretical foundation for the detection and processing method of microfluidic chip signal.Secondly,the processing methods of microfluidic chip signal were researched based on wavelet transform.The signal of the microfluidic chip is susceptible to high frequency noise,an improved threshold function method for microfluidic chip signal denoising was proposed accordingly.The simulation results show that the denoising efficiency and peak feature retention ability are superior to,the exiting techniques,such as Savitzky-Golay,Fast Fourier Transform and soft threshold method.Aiming at the problem of baseline drift of microfluidic chip signal,wavelet transform was used to decompose the signal in different frequency bands,the energy ratio of wavelet coefficients of each scale was used to determine the baseline,and then the wavelet coefficients representing the baseline were zeroed.Thus,the baseline drift of microfluidic chip signal can be effectively corrected.Thirdly,a portable microfluidic chip signal detection system was studied and designed.In the hardware design,based on the principle of contactless conductivity detection,the excitation signal generation,receiving signal detection and processing circuit were designed with STM32 as a microcontroller,and the detection of weak microfluidic chip signal was realized.In the software design,based on the human-computer interaction mode of the combination of PC soft-ware and mobile phone APP,the microfluidic chip signal software analysis platforms were constructed to realize real-time monitoring and analysis of signals.Finally,the improved threshold function denoising method and baseline correction method have been successfully applied to the self-developed portable microfluidic chip signal detection system.Experimental results show that the proposed methods can not only effectively eliminate noise and correct baseline drift,but also the characteristics information of a signal were well maintained.Thus,this a powerful approach to improve the signal detection effect and analysis accuracy of portable microfluidic chip system.