Wavelength selection for portable noninvasive blood component measurement system based on spectral difference coefficient and dynamic spectrum

Noninvasive and simultaneous quantitative analysis of multiple human blood components based on the grey analysis system

Noninvasive detection of human blood components is the dream of human beings and the goal of clinical detection. From the perspective of mathematical analysis, based on the grey analysis system, the principle of spectral chemical quantitative analysis and the solution method of multivariate linear equation, this paper pioneers the spectrum elimination method, and obtains a complete, high-precision, synchronous and noninvasive detection system for a variety of human blood components. The spectral elimination method applies the principle of elimination method in mathematics to the noninvasive quantitative analysis of human blood components by spectral method, reduces the influence of non-target components on the detection of target components, and improves the accuracy of noninvasive quantitative analysis of human blood components. To demonstrate the effectiveness of the method, taking the analysis of the contents of seven blood components (hemoglobin, red blood cell count, neutrophils, lymphocytes, monocytes, eosinophils and basophils) in blood as an example, fourteen models were established by two different methods. From the comparison of modeling results, it can be concluded that when the seven models established by spectral elimination method predict the corresponding seven components of all samples, the predicted correlation coefficients are more than 0.9500. The experimental results show that the spectral elimination method and non-invasive detection system proposed can predict the content of human blood components with high accuracy. This paper studies a high-precision, simultaneous and noninvasive quantitative analysis system of multiple human blood components for the first time, which not only makes great progress in the non-invasive chemical quantitative analysis of human blood components by spectroscopy, but also has great application value for clinical medical treatment and disease diagnosis.

Application of multi-wavelength dual-position absorption spectrum to improve the accuracy of leukocyte spectral quantitative analysis based on "M + N" theory

Leukocytes are the most important immune cells in human body, which are very important to maintain the immune function of human body. They can phagocytose foreign bodies and produce antibodies to resist the invasion of pathogens. Nowadays, the abuse of antibiotics is widespread, and the detection and analysis of leukocytes is very important for clinical diagnosis. It is of great medical significance to use chemical quantitative analysis method based on spectrum to realize the rapid and trace detection of leukocytes in clinic. It is the development direction of clinical detection in the future and provides a new way to improve the abuse of antibiotics. However, due to the influence of nonlinearity introduced by the measurement, the relationship between absorbance and concentration deviates from Lambert-Beer law, which leads to low measurement accuracy and restricts its development in clinical application. In order to improve the accuracy of spectral analysis, this paper with the guidance of "M + N" theory measured the transmission spectra of 392 whole blood samples under two different optical path lengths, and subtracts them to obtain the multi-band differential absorption spectra for modeling and prediction of leukocyte concentration. The following experiments were designed: using the transmission spectra measured at one position and the absorption spectra obtained by subtracting the transmission spectra measured at position 1 and position 2 as the input of modeling. Partial least squares (PLS) method was proposed in this paper for modeling and predicting the concentration of leukocyte. The experimental results show that the modeling results of dual-position absorption spectrum have been significantly improved and promoted compared with the modeling results of transmission spectrum in one position, and the calibration set correlation coefficient (R_C) values has increased by 57.92% to 0.864904, where the prediction set correlation coefficient (R_P) increased by 106.81% to 0.8502. The root mean square error of the calibration set (RMSEC) decreased by 40.01% from 3.1149 to 1.8686. The results suggest that modelling and analysing leukocytes with a multi-band dual-position absorption spectrum may reduce the influence of nonlinearity to a certain amount, significantly increase the model's prediction precision and accuracy, and obtain satisfactory results. This paper provides the possibility for rapid clinical micro-detection of leukocytes, as well as the ideas and directions for improving the accuracy of spectral quantitative analysis of components in complex solutions.

Noninvasive detection and analysis of human globulin based on dynamic spectrum

Noninvasive detection of blood components is the most ideal and effective method to prevent and detect many clinical diseases. However, the accuracy of noninvasive detection based on the spectrum is not always satisfactory. The influence of various interferences in measurement limits the accuracy of the analysis. The dynamic spectrum theory can theoretically eliminate the individual differences and measurement environment influence and improve measurement accuracy. The concentration of globulin is closely related to the status of the immune system, which is of great significance for clinical diagnosis. This paper improves the signal-to-noise ratio from all links of dynamic spectrum data processing to realize the noninvasive detection of globulin. Through reasonable pretreatment, extraction, quality evaluation, and variable screening, the valid information of the spectrum gets maximum utilization. Finally, using the partial least squares prediction model to predict globulin concentration. The results show that the model established by dynamic spectrum treated by this method has a good predictive performance for globulin. The correlation coefficient of the prediction set is 0.962, the root-mean-square error of the prediction set is only 1.058 g/L, the correlation coefficient of the calibration set is 0.996, and the root-mean-square error of the calibration set is 0.332 g/L. The experimental results show that reasonable data processing of dynamic spectrum can effectively improve the signal-to-noise ratio of the data, make the established model have good prediction accuracy and performance, and realize the high-precision prediction globulin. This paper provides a complete research idea and method for the noninvasive detection of blood components. It is hopeful to realize the noninvasive quantitative detection of trace components in blood.

Noninvasive Hemoglobin Level Prediction in a Mobile Phone Environment: State of the Art Review and Recommendations

BACKGROUND

There is worldwide demand for an affordable hemoglobin measurement solution, which is a particularly urgent need in developing countries. The smartphone, which is the most penetrated device in both rich and resource-constrained areas, would be a suitable choice to build this solution. Consideration of a smartphone-based hemoglobin measurement tool is compelling because of the possibilities for an affordable, portable, and reliable point-of-care tool by leveraging the camera capacity, computing power, and lighting sources of the smartphone. However, several smartphone-based hemoglobin measurement techniques have encountered significant challenges with respect to data collection methods, sensor selection, signal analysis processes, and machine-learning algorithms. Therefore, a comprehensive analysis of invasive, minimally invasive, and noninvasive methods is required to recommend a hemoglobin measurement process using a smartphone device.

OBJECTIVE

In this study, we analyzed existing invasive, minimally invasive, and noninvasive approaches for blood hemoglobin level measurement with the goal of recommending data collection techniques, signal extraction processes, feature calculation strategies, theoretical foundation, and machine-learning algorithms for developing a noninvasive hemoglobin level estimation point-of-care tool using a smartphone.

METHODS

We explored research papers related to invasive, minimally invasive, and noninvasive hemoglobin level measurement processes. We investigated the challenges and opportunities of each technique. We compared the variation in data collection sites, biosignal processing techniques, theoretical foundations, photoplethysmogram (PPG) signal and features extraction process, machine-learning algorithms, and prediction models to calculate hemoglobin levels. This analysis was then used to recommend realistic approaches to build a smartphone-based point-of-care tool for hemoglobin measurement in a noninvasive manner.

RESULTS

The fingertip area is one of the best data collection sites from the body, followed by the lower eye conjunctival area. Near-infrared (NIR) light-emitting diode (LED) light with wavelengths of 850 nm, 940 nm, and 1070 nm were identified as potential light sources to receive a hemoglobin response from living tissue. PPG signals from fingertip videos, captured under various light sources, can provide critical physiological clues. The features of PPG signals captured under 1070 nm and 850 nm NIR LED are considered to be the best signal combinations following a dual-wavelength theoretical foundation. For error metrics presentation, we recommend the mean absolute percentage error, mean squared error, correlation coefficient, and Bland-Altman plot.

CONCLUSIONS

We addressed the challenges of developing an affordable, portable, and reliable point-of-care tool for hemoglobin measurement using a smartphone. Leveraging the smartphone's camera capacity, computing power, and lighting sources, we define specific recommendations for practical point-of-care solution development. We further provide recommendations to resolve several long-standing research questions, including how to capture a signal using a smartphone camera, select the best body site for signal collection, and overcome noise issues in the smartphone-captured signal. We also describe the process of extracting a signal's features after capturing the signal based on fundamental theory. The list of machine-learning algorithms provided will be useful for processing PPG features. These recommendations should be valuable for future investigators seeking to build a reliable and affordable hemoglobin prediction model using a smartphone.

Time-resolved fluorescence spectroscopy based evaluation of stability of glucose oxidase

Glucose oxidase (GOx) is one of the most frequently used enzymes in a design of enzymatic biosensors and biofuel cells, which are novel electrical energy generation systems. Therefore, a better understanding of the mode of action of this enzyme is very important for further development of GOx-based sensors. In this research fluorescence properties of GOx in different acidic media have been estimated by the evaluation of redox states of active center that is flavine adenine dinucleotide (FAD). Steady-state fluorescence spectroscopy was applied to monitor the activity of GOx. A variation of pH has been invoked to gain a better understanding in the variations of GOx activity. The tendency of GOx activity to decrease over the time was determined, while increased intensity of the fluorescence band of GOx at 530 nm was associated with a decreased activity of the enzyme. The changes in fluorescence intensity of this band are caused by the dissociation of FAD from the enzyme. This process is not reversible, therefore, the decrease in the fluorescence intensity can be also associated with structural changes of the FAD during its reduction.

Improving the analysis accuracy of components in blood by SSP-MCSD and multi-mode spectral data fusion

In recent years, spectral quantitative analysis for blood components has been a research hotspot in biomedical engineering. But researches have been limited to the application of high-sensitivity spectroscopy instruments and the complexity of blood components-the overlapping of absorption curves for many components is severe. This has led to the difficulty in achieving satisfactory results when using spectroscopy to quantify components in blood. In order to enhance the model robustness and improve the model performance, this paper proposed a sample set partitioning strategy based on multi-component spatial distance (SSP-MCSD). Different from the other sample set partitioning strategies, which only consider the uniformity of the concentration distribution of the target component, this strategy also concerns to the concentration distribution of non-target components. The concentration of the target component and non-target components are used to construct a multi-dimensional space, and the Euclidean Distance of sample points in this space is used as the criterion to partition the sample set. At the same time, the spectra collected in multi-modes are fused for increasing the amount of information. So as to enhance the model robustness and to improve the analysis accuracy of the target components. In order to verify the effectiveness of this strategy, the serum of 101 volunteers was analyzed. Taking total protein in serum as the non-target component, the regression model for bilirubin concentration was established by transmission spectra, fluorescence spectra, and the joint spectra after fusion of the above two spectra, respectively. The experimental results showed that the prediction accuracy of the model established by SSP-MCSD combined with multi-mode spectral fusion is obviously higher than that of other methods. It can effectively improve the analysis accuracy of blood components.

Performance of calibration model with different ratio of sample size to the number of wavelength: Application to hemoglobin determination by NIR spectroscopy

Near infrared spectroscopy is widely used in composition analysis in fields of food, medicines, environment, and so on. The proportion of sample size and the wavelength used is very important for the performance of the calibration model. In this research, we explored the influence of ratio of sample size to the number of wavelength (SWR) on the performance of calibration model, with hemoglobin determination as an example. The results showed that RMSEC increases with the increase of SWR, when SWR is less than 0.5, namely the samples in the calibration set were less than half of the number of wavelengths used in establishing the calibration model, while RMSEP decreases with the increase of SWR. The calibration model was lack of reliability at this range for SWR. RMSEC and RMSEP tend to be stable when SWR value is greater than 0.9. However, in most cases, the samples size was limited, and wavelength selection was commonly used in practical spectroscopy analysis. In order to confirm that the effect of SWR were caused by both sample size and wavelength number, we also studied the performance of calibration model with different WSR. Wavelengths were selected by equidistant combination multiple linear regression (ECMLR) method. The conclusion from results were consistent with the previous part, namely when establishing calibration model, the number of wavelengths used should be less than the twice amount of samples in the calibration set to ensure the validity of the model. We recommend that wavelength selection part was indispensable for small sample size cases. This research can be important evidence and guide for other researches with spectroscopy methods.

Quantitative estimation of wastewater quality parameters by hyperspectral band screening using GC, VIP and SPA

Water pollution has been hindering the world's sustainable development. The accurate inversion of water quality parameters in sewage with visible-near infrared spectroscopy can improve the effectiveness and rational utilization and management of water resources. However, the accuracy of spectral models of water quality parameters is usually prone to noise information and high dimensionality of spectral data. This study aimed to enhance the model accuracy through optimizing the spectral models based on the sensitive spectral intervals of different water quality parameters. To this end, six kinds of sewage water taken from a biological sewage treatment plant went through laboratory physical and chemical tests. In total, 87 samples of sewage water were obtained by adding different amount of pure water to them. The raw reflectance (R_raw) of the samples were collected with analytical spectral devices. The R_raw-SNV were obtained from the R_raw processed with the standard normal variable. Then, the sensitive spectral intervals of each of the six water quality parameters, namely, chemical oxygen demand (COD), biological oxygen demand (BOD), NH₃-N, the total dissolved substances (TDS), total hardness (TH) and total alkalinity (TA), were selected using three different methods: gray correlation (GC), variable importance in projection (VIP) and set pair analysis (SPA). Finally, the performance of both extreme learning machine (ELM) and partial least squares regression (PLSR) was investigated based on the sensitive spectral intervals. The results demonstrated that the model accuracy based on the sensitive spectral ranges screened through different methods appeared different. The GC method had better performance in reducing the redundancy and the VIP method was better in information preservation. The SPA method could make the optimal trade-offs between information preservation and redundancy reduction and it could retain maximal spectral band intervals with good response to the inversion parameters. The accuracy of the models based on varied sensitive spectral ranges selected by the three analysis methods was different: the GC was the highest, the SPA came next and the VIP was the lowest. On the whole, PLSR and ELM both achieved satisfying model accuracy, but the prediction accuracy of the latter was higher than the former. Great differences existed among the optimal inversion accuracy of different water quality parameters: COD, BOD and TN were very high; TA relatively high; and TDS and TH relatively low. These findings can provide a new way to optimize the spectral model of wastewater biochemical parameters and thus improve its prediction precision.