A Time-resolved Fluorescence Immunoassay System Based On Multichannel Scaled Photon-counting Technology

Time-resolved fluorescence immunoassay, which uses anthanide complexes oflong time constant as a marker to mark the antibody or antigen, is one of the immunemarkers technologies. Rare earth elements, which is of long fluorescence timeconstant, large wavelength difference between the emitted and excitation light, makesthe time-resolved fluorescence immunoassay with high sensitivity and signal-to-noiseratio. Currently, time-resolved fluorescence immunoassay technology is widely usedin clinical community. However, most instruments are manufactured abroad, whichrestricts the promotion of the technology. Therefore, the localization of the instrumentwill promote the development of the domestic industry.This thesis designs a time-resolved fluorescence immunoassay system based onmultichannel scaled photon-counting technology. The composition of the systemincludes a light source, modulation, detection, data acquisition and data processingmodule. A modulated LED is used to get the excitation light pulse. The modulationand data acquisition module accomplished with FPGA can set the pulse width of lightsource. This thesis designs a FPGA-based multichannel scaled photon-counter withpre-accumulator. By setting channel width, the system can detect differentfluorescence signals, and achieve real-time imaging. Compared with the traditionalanalysis system, the system with high counting rate can not only set some parametersin counter by custom, but also get the light intensity and time constant of thefluorescence signals. The system reliability is increased.This thesis presents an improved Prony algorithm to process the two-componentfluorescence signals with five parameters and analyses the effect of the gate width,noise, light intensity ratio and lifetime ratio of the two components to the calculationresult.Experimental results show that the time-resolved fluorescence immunoassaysystem based on multichannel scaled photon-counting has high accuracy and linearityin the detection of reagents. By applying the improved Prony algorithm totwo-component fluorescence signals which has the baseline drift and more than tentimes difference in time constant, a reasonable accuracy is achieved.