小波变换在神经细胞传感器信号去噪中的应用

摘要

将神经元培养在半导体传感器芯片表面,构建出能够对细胞电生理特性进行长时程无损测量的神经细胞传感器,具有广阔的生物医学应用前景.由于生物信号非常微弱,采集到的信号往往难以满足实际检测的需要.本研究针对培养在光寻址电位传感器(LAPS)芯片表面的PC12细胞的胞外电生理信号,采用小波变换方法对其进行去噪处理.通过小波变换将信号分解为不同层次的小波系数,得到每一层的阈值.并根据每一层系数特点,按阈值进行分别处理,得到新的小波系数,最后根据该系数,重构了信号.对去噪后的信号进行频谱分析,发现有效信号频率集中在小于2 kHz的范围内,信噪比得到提高,表明小波变换是神经细胞传感器信号去噪的有效方法.%Neural cell based biosensors, with neurons cultured on the semiconductor sensor chip,are suitable for long-term and non-invasive measurement of cell electrophysiological properties. They have broad prospects for biomedical applications. For biological signal is very weak,the collected signal is often difficult to satisfy the needs of the actual detection. In this study,wavelet de-noising was used to deal with extracellular electrophysiological signals of PC12 cells cultured on the surface of Light Addressable Potentiometric Sensors (LAPS). Signal was decomposed into wavelet coefficients at different levels by wavelet transform, obtaining threshold of every level. Based on the characteristics of coefficients in different levels, we dealt with them according to the threshold,acquiring new wavelet coefficients,and finally reconstructed the signal. It is found that the effective frequency is focused in the range of less than 2 kHz by the spectral analysis of the de-noised signal. The signal to noise ratio is improved,indicating wavelet transform is an effective way to de-noise the signal for neural cell based biosensors.