A Platform For Sensitive Detection Of Magnetic Immunochromatography Strip

Point of care testing(POCT),is a novel detection method that doesn't require complicated sample processing procedures to get detection result,which means it can realize on-site detection and display detection result for patient immediately.POCT has great prospects with a short assay time and simple operation,as well as being cost-effective,which means it is very suitable for early diagnosis.Immunochromatography test strip(ICTS)is one of the most widely used format for POCT.With the prosperous development of nanotechnology,new types of material have been developed as labels for ICTS,and the demand of test devices for quantitative detection has been proposed.In this study,we developed an accurate and sensitive diagnostic platform called magnetic immunoassay reader(MIR)for rapidly detecting magnetic signals from magnetic nanoparticles(MNPs)on ICTSs.In addition,this platform can detect not only single target ICTSs,but multiplex targets simultaneously.The platform mainly comprised two components: a MIR apparatus and data processing server.The MIR apparatus was developed to quantify magnetic signals from magnetic probes on ICTSs and convert it to digital data.MIR mainly comprised a C-shaped ferromagnetic core to provide an external magnetic field and magnetize the MNPs,and two coils parallel to the strip to detect the signal induced by the magnetized MNPs.Besides,we designed ICTS cartridge to accommodate the strip and protect the strip from pollution when handled.The data processing server was designed to process the digital data for detection result.The server was designed based on.NET framework and C# language and including complete function,such as user login,patient information registration,automatic detection,diagnosis,print,data management and so on.In addition,Microsoft SQL Server was used to build the database of the system to make the management detection data conveniently.In the most critical part of the server,median filtering and wavelet denoising were applied to remove noises.Because of the complexity of the noise,the waveform of weak signal was still disturbed.Therefore,a support vector machine model was used to distinguish the weak positive samples from the blank samples,and then a custom waveform reconstruction method was used to restore the distorted waveform of the weak signal,which greatly improved the specificity and sensitivity of the system.Following,we selected two clinical markers,human chorionic gonadotropin(HCG)and myocardial infarction(cTnI,CKMB and Myo),to evaluate the analytical performance of the system.50 HCG urine samples and 59 myocardial infarction serum samples were detected.The quantitative detection range of HCG was 1-1000 mIU/mL and the limit of detection was 0.014 mIU/mL.For HCG detection,the sensitivity and specificity of the detection were both 100%.For the detection of three items of myocardial infarction,the detection values obtained by our work had good linear correlations with the standard values.The accuracy and repeatability of the system were tested and good results were obtained,which indicated that the detection system introduced in this paper had the potential to be applied in clinical detection.