Period Area Extraction Method Of Dynamic Spectrum And Modeling Analysis

At present,wearable medical equipment is becoming more and more popular,this equipment requires not only the accuracy of measurement,it also has a very strict requirement for real-time measurement.But there are some differences in the computing speed of dynamic spectrum extraction methods.In order to improve the speed of dynamic spectrum extraction to meet the requirement of real-time,a new dynamic spectrum data extraction algorithm based on difference value extraction method is proposed in this paper.The single-cycle area at each wavelength of the logarithmic pulse wave was taken as the eigenvalue of the dynamic spectrum.This algorithm,which uses only addition and subtraction,can greatly increase the speed of operations.At the same time,a new sample automatic screening method based on the condition number of the sample itself is proposed.Then,this method is compared with the single edge extraction method and the difference value extraction method.The experimental results show that the modeling accuracy of this method is similar to that of single edge extraction method,the correlation coefficient is 0.8 and the root mean square error(RMSE)can be controlled at 10%.The modeling accuracy of the two methods is obviously better than difference value extraction method,which proves the feasibility of the method.But the extraction time of the new method is only one-seventh of the single-edge extraction method,which greatly improves the extraction time,providing more possibilities for real-time monitoring of mobile portable devices.An appropriate method for calibration set selection was very important for a robust quantitative model,especially for the noninvasive measurement of blood components.Partial Least Squares Regression(PLSR)is one of the most popular regression methods for establishing multivariate calibration models with pectroscopic data.However,the success of the PLSR model depends on the availability of a representative set.“M+N” theory provides a new idea of improving the model reliability of composition analysis,with M being the component information and N representing the outside disturbance.In this paper,a new calibration set selection method based on“M+N”theory is proposed.From the point of M elements,the method takes both the target component and non-targetcomponents into account.Dynamic spectrum(DS)was a noninvasive blood composition analysis method based on PPG.In this paper,we applied the new calibration set selection method on the prediction of hemoglobin by PLSR model with DS method.The total protein was regarded as the non-target component,which was the most important component in blood except hemoglobin.The experimental results show that compared with the random selection method,the new selection method can significantly improve the model accuracy.The correlation coefficient of the new selection method is increased by 8.03% and RMSEP is reduced by 15.41% than the selection method only considering hemoglobin concentration distribution.The experimental results verify the performance of the proposed calibration set selection method,which can guide the chemical composition analysis based on the spectrum to improve the prediction performance.