Design Of Software System And Algorithm For Multifunctional Cellular Microphysiometer

Cell-based biosensors are special electronic device that transfer non-electrical signals of cells in vitro into easy detecting electrical quantities, it consists of two levels of transducer composition, the first level transducer is the biological living cells, the secondary is cells attached semiconductor devices. According to different parameters, the common secondary transducers include microelectrode array (Microelectrodes Array, MEA), cell impedance sensor (ECIS), light addressing cell potential sensor (LAPS), etc. Traditional cell physiological parameter testing instruments only test single cell physiological parameters and can’t fully reveal the mechanisms of cell physiological activity change, In this study, we developed a set of multifunctional cellular Microphysiometer based on the integrated chip. The main work of my studies was to design the software system for multifunctional cellular Microphysiometer.The software system consists of three-cell-parameters testing module including MEA, ECIS, and LAPS, sample flow control module, cell detection environment control module and data processing module. Cell physiological parameters testing module can record and display experiment data, including cell extracellular action potential detected by the MEA, cell impedance detected by the EICS and photocurrent detected by the LAPS in real time. These testing parameters can directly or indirectly reflect electrophysiology, cell growth, adhesion, proliferation, apoptosis and metabolism. The system provides a flow control interface for operator to manage the sample flow control module which is mainly responsible for sample injection and liquid waste cleaning. Meanwhile, cell testing environment control module can monitor environment temperature, humidity and CO2concentration in real time, provide alarm function and ensure the smooth progress of the cells. Furthermore, Data processing module provides online data processing, including data playback, cell metabolic rate calculation, cell index calculation and IC50fitting. Software system was designed using C#language based on.NET Framework4.0, which can support Windows series of operating system. WPF, with higher visual quality, is the software designer. The system adopts the hierarchical structure to facilitate subsequent upgrade.In addition, intelligent identification algorithms were studied. Through the study and comparison of different processing algorithm, we designed algorithms with low complexity, specific include signal denoising, spike detection and classification. Optimized the algorithm of automatic searching LPAS working point and realized the transform from the cell impedance to cell index and IC50.At last, a series of cell experiments were carried out to verify the function and stability of the software system. Three different response signals from mouse taste cells were distinguished by applying the improved spike detection and classification algorithms. The result shows that. Discrete wavelet transform, outperformed than the principle component analysis method. The improved automatic searching working point algorithm was proved better than the original algorithm in accuracy and execution time through the ion scan experiment.