Construction Of Novel Serum Non-label Diagnostic Technique Based On Raman Spectroscopyand Multivariate Statistical Analysis And Its Application In Coronary Heart Disease

Objective:Coronary heart disease(CHD)has always been one of the most important cardiovascular diseases,which is caused by the damage and stenosis of the coronary arteries supplying blood to the myocardium and the myocardial ischemia,hypoxia or myocardial necrosis caused by blockage.In China prevalence and mortality are still in the rising stage,which has become a major public problem.The current common clinical diagnostic methods include serum cardiac markers,electrocardiogram,echocardiography,cardiac stress test and coronary angiography,but each has its own limitations.In this study,we use the Raman confocal system to detect serum samples of patients with coronary heart disease,and to explore its application value in the diagnosis of coronary heart disease.Method:1.Clinical case data analysis:119 serum samples were selected in this experiment,including 57 patients with coronary heart disease diagnosed by coronary angiography,and 62 healthy patients.The measurement data of two groups were expressed by X±SD,and the t-test was used,the count data of both sets of data were analyzed by chi-square test.2.Using the microscopic confocal Raman system to collect the serum Raman characteristics of each specimen,every sample was tested in 4 different positions,using LabSpec 6 for background removal,denoise,baseline calibration,area normalization and zeroing.In order to remove different degrees of thermal background or fluorescence in the spectrum,a more stable Raman signal is obtained.Then,the four spectral values??at different positions of each sample are averaged,and the same type of spectral results are also averaged.3.Orthogonal partial least squares discriminant analysis(OPLS-DA)was used for statistical processing and analysis using SIMCA-P14.1 analysis software.Result:1.The Raman mean peak spectrum of the coronary heart disease group and the healthy group are similar;2.At the displacements of 1081.20/cm,1102.01/cm,1127.76/cm,1337.66/cm,1603.83/cm and 1657.69/cm,the spectral vibration intensity of the coronary heart disease group is higher than that of the healthy group;the displacement of 1154.37/cm,1190.83/cm and 1518.30/cm,the spectral vibration intensity of the healthy group was significantly higher than that of the coronary heart disease group.The changes of these peaks corresponded to different biomolecules and metabolites,which may be beneficial for the screening of coronary heart disease by serum Raman spectroscopy;3.Using OPLS-DA model analysis for 89 training set samples,the cumulative interpretation rate of the model to the X matrix R~2X(cum)is 0.795,the cumulative interpretation rate of the model to the Y matrix R~2Y(cum)is 0.820,and the predictability of the model Q~2(cum)is 0.553.The total diagnostic accuracy was73.33%for 30 samples.Conclusions:The microscopic confocal Raman system combined with OPLS-DA analysis can distinguish the known samples quickly and conveniently,and obtain high accuracy against the prediction set.As an exploratory experimental method,it provides a new possibility for the detection of coronary heart disease,and is expected to develop into a new clinical auxiliary diagnostic method for coronary heart disease.