A method for correcting blood glucose near-infrared spectroscopy data based on blood absorption peaks

Abstract

The present invention mainly relates to the technical field of spectral data correction. In order to solve the problem that in near-infrared non-invasive blood glucose detection, the spectral data is offset due to the non-standard collection process, the uneven contact between the collection equipment and the person, and the change of the collection environment, which further affects the prediction ability of the spectral model and reduces the accuracy of blood glucose detection, the present invention provides a blood glucose near-infrared spectral data correction method based on the blood absorption peak, uses a near-infrared spectral device to collect blood spectral data, obtains the blood absorption peak, reconstructs the wavelength point of the spectral device according to the peak point of the blood absorption peak, and sets the peak point of the absorption peak as the correction point, reorganizes the wavelength point matrix according to the distribution of the wavelength point of the reconstructed spectral device, obtains the wavelength point matrix, uses the reconstructed spectral device to collect blood spectral data, writes the absorbance value of the correction point into the wavelength point matrix to calculate the matrix coefficient, and finally corrects the blood spectral data in combination with the wavelength point matrix coefficient.

Description

Technical Field

[0001] The invention relates to the technical field of spectrum data correction, and in particular to a method for correcting blood sugar near-infrared spectrum data based on blood absorption peaks. Background Art

[0002] Diabetes seriously threatens people's life and health, but there is no clinical cure for diabetes. In clinical practice, frequent blood sugar concentration testing and drug control are often used to maintain the blood sugar concentration of diabetic patients at a normal level. Scholars at home and abroad have invested a lot of energy and resources in blood sugar concentration detection technology and have achieved certain research results. The current blood sugar detection methods include three categories: non-invasive, minimally invasive and invasive. Among them, invasive and minimally invasive detection have higher accuracy and can meet the requirements of clinical blood sugar detection. However, both invasive and minimally invasive detection will bring pain and discomfort to diabetic patients. Therefore, the demand for non-invasive blood sugar detection has emerged.

[0003] Among many non-invasive blood glucose detection methods, the miniaturized near-infrared detection method has become the most widely used detection method due to its simple equipment, high efficiency and cost advantages. However, due to the limitations of its own hardware conditions, the spectral data stability of the miniaturized near-infrared spectrometer is weak and the accuracy is low. Especially in the actual application of miniaturized near-infrared non-invasive blood glucose detection, the spectral data acquisition process requires extremely precise, the acquisition process is not standardized, the equipment is not in uniform contact with the skin of the person being tested during acquisition, and the acquisition environment changes. These factors can easily cause the spectral data to shift, thereby affecting the prediction ability of the spectral model and reducing the detection accuracy of the non-invasive blood glucose detection technology. Therefore, how to implement a spectral data correction method to solve the problems of unstable spectral data and low accuracy in miniaturized near-infrared non-invasive blood glucose detection technology has become an urgent problem to be solved. Summary of the invention

[0004] The technical problems to be solved by the present invention are:

[0005] The present invention provides a method for correcting blood glucose near-infrared spectral data based on blood absorption peaks, which solves the problem that in near-infrared non-invasive blood glucose detection, the spectral data is offset due to non-standard collection process, uneven contact between collection equipment and people, and changes in the collection environment, which further affects the prediction ability of the spectral model and reduces the accuracy of blood glucose detection.

[0006] The technical solution adopted by the present invention to solve the above technical problems is as follows:

[0007] A method for correcting blood glucose near-infrared spectroscopy data based on blood absorption peaks comprises the following steps:

[0008] Step 1: Use near-infrared spectroscopy equipment to collect blood spectrum data and obtain blood absorption peaks;

[0009] Step 2: Reconstruct the wavelength point of the spectral device according to the peak point of the blood absorption peak, and set the peak point of the absorption peak as the reference correction point;

[0010] Step 3: Reorganize the wavelength point matrix according to the wavelength point distribution of the reconstructed spectrum device to obtain the wavelength point matrix;

[0011] Step 4: Use the reconstructed spectrum device to collect blood spectrum data, write the correction point absorbance value into the wavelength point matrix to calculate the matrix correction coefficient;

[0012] Step 5: Correct the non-invasively collected blood spectrum data in combination with the wavelength point matrix correction coefficient to complete the data correction.

[0013] Furthermore, the specific method of step 1 is: using a spliced ​​sensor to collect the spectrum of the blood sample, and the wavelength point with the maximum absorbance value in the spectrum is the wavelength point corresponding to the blood absorption peak.

[0014] Furthermore, according to the measurement band range λ1nm~λ2nm and resolution R of the selected spliced ​​near-infrared spectroscopic device, the number of wavelength points contained in the spliced ​​sensor is calculated. At the same time, according to the band range λ3nm~λ4nm and resolution R of the single sensor, the number of wavelength points X2 of the single sensor spectral device is calculated. The spectrum is reconstructed based on the number of wavelength points of the single sensor spectral device. The number of wavelength points contained in the spectral data collected and acquired by the reconstructed spectral device is the same as that of the single sensor spectral data.

[0015] Further, if the wavelength value corresponding to the blood absorption peak wavelength point is M1nm, and M1 / R is an integer, the specific method of step 2 is: use 4 reference correction points to correct the spectrum data, take the absorption peak as the center, take the wavelength point corresponding to the absorption peak as the reference correction point, and take X3 wavelength points to the left and right for spectrum reconstruction, wherein X3 is calculated as: X3=X2-4) / 2, and the reconstructed spectrum actually represents the data of X2 wavelength points, wherein the wavelengths of each wavelength point are [(M1-X3*R)nm, ..., (M1-2R)nm, (M1-R)nm, M1nm, M1nm, M1nm, M1nm, (M1+R)nm, (M1+2R)nm ..., (M1+X3*R)nm];

[0016] If the wavelength value corresponding to the blood absorption peak is M1nm, and M1 / R is not an integer, the specific method of step 2 is: round M1 to obtain a new M1 value, and use the wavelength point corresponding to the new M1 value as the reference correction point to perform the same calculation to obtain the wavelength value of each wavelength point of the reconstructed spectrum.

[0017] Furthermore, when (M1-X3*R)≤λ1, the reconstructed spectrum overflows to the left, and the leftmost wavelength of the reconstructed spectrum is selected as λ1. The wavelengths of each wavelength point after reconstruction are [λ1nm, …, (M1-2R)nm, (M1-R)nm, M1nm, M1nm, M1nm, (M1+R)nm, (M1+2R)nm…, (M1+X3*R)nm];

[0018] When (M1+X3*R ) ≥λ2, the reconstructed spectrum overflows to the right, and the rightmost wavelength of the reconstructed spectrum is selected as λ2. The wavelengths of each wavelength point after reconstruction are [(M1-X3*R)nm, …, (M1-2R)nm, (M1-R)nm, M1nm, M1nm, M1nm, (M1+R)nm, (M1+2R)nm, …, λ2nm].

[0019] Furthermore, the arrangement of the wavelength point matrix in step 3 is as follows: an m-row n-column matrix with the closest number of rows and columns is constructed according to the number of wavelength points in the reconstructed spectrum, and the four reference correction points of the reconstructed spectrum are written into the four diagonal points of the matrix, and are written into the matrix from left to right and from top to bottom in order from small to large wavelength values, and unfilled positions in the matrix are replaced by null values.

[0020] Furthermore, the calculation method of the matrix correction coefficient in step 5 is:

[0021] Write the absorbance values ​​of the four reference correction points into the wavelength point matrix, where P1, P2, P3, and P4 are the absorbance values ​​of the correction points at the upper left corner, upper right corner, lower left corner, and lower right corner of the wavelength point matrix. Calculate the absorbance values ​​of each wavelength point in the first row of the wavelength point matrix except the correction points, and obtain the absorbance values ​​as follows:

[0022]

[0023] Where i is the i-th wavelength point in the first row excluding the correction point.

[0024] Similarly, the absorbance values ​​of each wavelength point in the first column of the wavelength point matrix except the correction point are calculated, and the absorbance values ​​are obtained as follows:

[0025]

[0026] Where j is the jth wavelength point in the first column excluding the correction point.

[0027] Similarly, the absorbance values ​​of the remaining wavelength points in the wavelength point matrix are calculated in the same way, each wavelength point in the wavelength point matrix is ​​matched with its absorbance value one by one and the absorbance value is written into the wavelength point matrix.

[0028] Calculate the mean absorbance value P of the four correction points mean =(P1+P2+P3+P4) / 4, and then calculate the wavelength point matrix correction coefficient for each wavelength point in the wavelength point matrix one by one. Taking the m wavelength points in the first row of the wavelength point matrix as an example, the corresponding wavelength point matrix correction coefficients are:

[0029]

[0030] Among them, (α1,α2,…,α m ) are the wavelength point matrix correction coefficients corresponding to the m wavelength points in the first row of the wavelength point matrix.

[0031] Similarly, the wavelength point matrix correction coefficients corresponding to all wavelength points in the wavelength point matrix of the matrix can be calculated, and the null values ​​are not processed.

[0032] Furthermore, if the absorbance values ​​of the m wavelength points in the first row of the wavelength point matrix corresponding to the collected blood spectrum data are (P1, S1, S2, ..., P2), the wavelength point matrix correction coefficients corresponding to the m wavelength points in the first row of the wavelength point matrix are (α1, α2, α3 ..., α m ), combined with the wavelength matrix correction coefficient, the absorbance values ​​of the m wavelength points in the first row are corrected one by one. The absorbance values ​​of the m wavelength points in the first row after correction are (P1 / α1, S1 / α2, S2 / α3, ..., P2 / α m ).

[0033] Similarly, the absorbance values ​​of the remaining wavelength points are corrected one by one in combination with the wavelength point matrix correction coefficient. When the absorbance value data of all wavelength points are corrected, the corrected absorbance values ​​are matched with the wavelength points one by one. After the matching is completed, the new wavelength point matrix containing the corrected absorbance values ​​is reorganized again, and the square matrix of m rows and n columns is reorganized into a single-row matrix, and the null values ​​are removed. The single-row matrix is ​​then rearranged from left to right in the order of wavelength values ​​from small to large. The rearranged single-row matrix is ​​the corrected blood spectrum data.

[0034] Beneficial effects of the present invention:

[0035] The method for correcting near-infrared spectral data of blood glucose based on blood absorption peak described in the present invention reconstructs wavelength points based on blood absorption peaks, reduces the amount of spectral data while retaining the blood characterization capability, improves the efficiency of spectral analysis, and further corrects the blood spectral data in the near-infrared non-invasive blood glucose detection technology through multiple correction points, thereby solving the problems of easy deviation, weak stability and low accuracy of spectral data in the actual application of near-infrared non-invasive blood glucose detection, and can effectively improve the data stability and detection accuracy of miniaturized near-infrared non-invasive blood glucose detection technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 The present invention is a method flow chart of a method for correcting blood glucose near-infrared spectroscopy data based on blood absorption peaks.

[0037] Figure 2 It is the wavelength point matrix obtained after the wavelength point matrix is ​​reorganized in the embodiment. DETAILED DESCRIPTION

[0038] The present invention discloses a method for correcting blood glucose near-infrared spectroscopy data based on blood absorption peaks, such as Figure 1 As shown, the method first uses a near-infrared spectroscopy device to non-invasively collect blood spectral data to obtain a blood absorption peak, reconstructs the wavelength point of the spectral device according to the peak point of the blood absorption peak, and sets the peak point of the absorption peak as a correction point. Then, the wavelength point matrix is ​​reorganized according to the wavelength point distribution of the reconstructed spectral device to obtain a wavelength point matrix, and the reconstructed spectral device is used to collect blood spectral data. The absorbance value of the correction point is written into the wavelength point matrix to calculate the wavelength point matrix correction coefficient. Finally, the non-invasively collected blood spectral data is corrected in combination with the wavelength point matrix correction coefficient to complete the data correction.

[0039] Example:

[0040] In this embodiment, the miniaturized near-infrared spectroscopy device used is a multi-sensor splicing type, which includes 4 single sensors, and the band ranges are 800nm~1210nm, 1220nm~1630nm, 1640nm~2050nm, and 2060nm~2470nm. The splicing type multi-sensor miniaturized near-infrared spectroscopy device can expand the single sensor device that initially only covers the 410nm band range to cover 1670 nanometers, almost covering the entire near-infrared band range, effectively expanding the band range of the device, including as wide a band range as possible to accurately obtain the spectral absorption peak of blood. The wavelength point with the largest absorbance value in the band range of 800nm~2470nm is the blood absorption peak.

[0041] The resolution of the miniaturized splicing multi-sensor spectral device used in this embodiment is 10nm, and the band range is 800nm ​​to 2470nm. It can be calculated that the wavelength points contained in each multi-sensor spectral data are X1=1+(2470-800) / 10=168. The band range of a single sensor is 410nm, and the wavelength points contained in each single sensor spectral data are X2=1+410 / 10=42, the wavelength points of the spliced ​​multi-sensor spectral equipment are reconstructed, and the multi-sensor spectral data containing 168 wavelength points are reconstructed into spectral data containing 42 wavelength points. This can not only effectively reduce the amount of spectral data, but also use the conventional single-sensor spectral data modeling method for spectral modeling, effectively improving the universality of spectral data and spectral analysis efficiency.

[0042] The wavelength point reconstruction of the spectral device is based on the wavelength point of a single sensor. The number of wavelength points contained in the spectral data collected by the reconstructed spectral device is the same as that of the single sensor spectral data. According to the wavelength point matrix correction method proposed in this patent, 4 correction points are used to correct the non-invasively collected blood spectral data. The wavelength range corresponding to the peak point of the blood absorption peak is set to M1nm. If M1%10≠0, where % is the remainder symbol, it is rounded off to obtain a new M1 value, and the original M1 value is replaced. With the wavelength range M1nm as the center point, X3 wavelength points are taken to the left and right for spectral reconstruction, where X3 is calculated as follows:

[0043] X3=(X2-4) / 2=19

[0044] The reconstructed spectral data is actually represented as a set of absorbance values ​​at 42 wavelength points, where the wavelength ranges of each wavelength point are [(M1-190)nm, (M1-180)nm, …M1nm, M1nm, M1nm, M1nm, …(M1+180)nm, (M1+190)nm].

[0045] When (M1-190)≤800, the wavelength points of the reconstructed spectrum overflow to the left and the spectrum reconstruction cannot be performed. When the above situation occurs, the leftmost wavelength range of the reconstructed spectrum is selected as 800nm. In this case, the wavelength ranges of each wavelength point after reconstruction are [800nm, 810nm, …M1nm, M1nm, M1nm, M1nm, …1170nm, 1180nm].

[0046] When (M1+190)≥2470, the wavelength points of the reconstructed spectrum overflow to the right and the spectrum reconstruction cannot be performed. When the above situation occurs, the rightmost wavelength range of the reconstructed spectrum is selected as 2470nm. It can be further known that in this case, the wavelength ranges of each wavelength point after reconstruction are [2090nm, 2100nm, …M1nm, M1nm, M1nm, M1nm, …2460nm, 2470nm].

[0047] In this embodiment, each piece of blood spectrum data contains 42 wavelength points, of which 4 are correction points and 38 are to be corrected points. The 42 wavelength points need to be reorganized into a square matrix, such as Figure 2As shown, the most suitable reorganization method is a 6*7 square matrix. The square matrix contains 6 rows and 7 columns, which can completely cover the 42 wavelength points of the blood spectrum data. The correction points are fixed at the four diagonals of the square matrix, and the remaining wavelength points to be corrected are filled in the matrix in order from small to large according to the wavelength range to complete the wavelength point matrix reorganization. The obtained matrix is ​​the wavelength point matrix.

[0048] Write the absorbance values ​​of the four correction points into the wavelength point matrix, and set the absorbance values ​​of the four correction points to P1, P2, P3, and P4, respectively, where P1 is the absorbance value of the correction point in the upper left corner of the wavelength point matrix, P2 is the absorbance value of the correction point in the upper right corner of the wavelength point matrix, P3 is the absorbance value of the correction point in the lower left corner of the wavelength point matrix, and P4 is the absorbance value of the correction point in the lower right corner of the wavelength point matrix. Calculate the absorbance value of each wavelength point in the first row of the wavelength point matrix except the correction point, and obtain its absorbance value as follows:

[0049]

[0050] Where i is the i-th wavelength point in the first row excluding the correction point.

[0051] Similarly, the absorbance values ​​of each wavelength point in the first column of the wavelength point matrix except the correction point are calculated, and the absorbance values ​​are obtained as follows:

[0052]

[0053] Where j is the jth wavelength point in the first column excluding the correction point.

[0054] Similarly, the absorbance values ​​of the remaining wavelength points in the wavelength point matrix are calculated in the same way, each wavelength point in the wavelength point matrix is ​​matched with its absorbance value one by one and the absorbance value is written into the wavelength point matrix.

[0055] When calculating the wavelength point matrix correction coefficient, you need to first calculate the mean value P of the absorbance values ​​of the four correction points. mean =(P1+P2+P3+P4) / 4, and then calculate the wavelength point matrix correction coefficient for each wavelength point in the wavelength point matrix one by one. Taking the 7 wavelength points in the first row of the wavelength point matrix as an example, the corresponding wavelength point matrix correction coefficients are:

[0056]

[0057] Among them, (α1, α2, ..., α6, α7) are wavelength point matrix correction coefficients corresponding to the 7 wavelength points in the first row of the wavelength point matrix.

[0058] Similarly, the wavelength point matrix correction coefficients corresponding to all wavelength points in the (6*7) wavelength point matrix can be calculated.

[0059] Taking the 7 wavelength points in the first row of the wavelength point matrix as an example, the non-invasively collected blood spectrum data is corrected. If the absorbance values ​​of the 7 wavelength points in the first row of the wavelength point matrix corresponding to the non-invasively collected blood spectrum data are (P1, S1, S2, ... S5, P2), and the wavelength point matrix correction coefficients corresponding to the 7 wavelength points in the first row of the wavelength point matrix are (α1, α2, ... α6, α7), the absorbance values ​​of the 7 wavelength points in the first row are corrected one by one in combination with the wavelength point matrix correction coefficients, and the corrected absorbance values ​​of the 7 wavelength points in the first row are (P1 / α1, S1 / α2, S2 / α3, ... S5 / α6, P2 / α7).

[0060] Similarly, the absorbance values ​​of the remaining 35 wavelength points are corrected one by one in combination with the wavelength point matrix correction coefficient. After the absorbance value data of all wavelength points are corrected, the corrected absorbance values ​​are matched with the wavelength points one by one. After the matching is completed, the new wavelength point matrix containing the corrected absorbance values ​​is reorganized again, and the (6*7) square matrix is ​​reorganized into a (1*42) single-row matrix, and then the single-row matrix is ​​arranged from left to right in ascending order of wavelength value to obtain a reorganized single-row matrix. The reorganized single-row matrix is ​​the non-invasively collected blood spectrum data after the correction.

Claims

1. A method for correcting blood glucose near-infrared spectroscopy data based on blood absorption peaks, characterized in that: The following steps are involved: Step 1: Use a spliced ​​near-infrared spectrometer to collect spectral data of a blood sample and obtain a blood absorption peak; Step 2: Reconstruct the wavelength point of the spectral device according to the peak point of the blood absorption peak, and set the peak point of the absorption peak as the reference correction point; Step 3: reorganize the wavelength point matrix according to the wavelength point distribution of the reconstructed spectrum device to obtain the wavelength point matrix; the arrangement of the wavelength point matrix in step 3 is: construct an m-row n-column matrix with the closest number of rows and columns according to the number of wavelength points in the reconstructed spectrum, write the 4 reference correction points of the reconstructed spectrum into the 4 diagonal points of the matrix, and write them into the matrix from left to right and from top to bottom in the order of wavelength values ​​from small to large, and replace the unfilled positions in the matrix with null values; Step 4: Use the reconstructed spectrum device to collect blood spectrum data, write the correction point absorbance value into the wavelength point matrix to calculate the matrix correction coefficient; the specific method for obtaining the matrix correction coefficient is: write the absorbance values ​​of the four reference correction points into the wavelength point matrix, where P1, P2, P3, and P4 are the correction point absorbance values ​​at the upper left corner, upper right corner, lower left corner, and lower right corner of the wavelength point matrix, and calculate the absorbance value of each wavelength point in the first row of the wavelength point matrix except the correction point, and obtain its absorbance value as follows: Where i is the i-th wavelength point in the first row excluding the correction point; Similarly, the absorbance values ​​of each wavelength point in the first column of the wavelength point matrix except the correction point are calculated, and the absorbance values ​​are obtained as follows: Where j is the jth wavelength point in the first column excluding the correction point; Similarly, the absorbance values ​​of the remaining wavelength points in the wavelength point matrix are calculated by the same calculation method, and each wavelength point in the wavelength point matrix is ​​matched with its absorbance value one by one and the absorbance value is written into the wavelength point matrix; Calculate the mean absorbance value P of the four correction points mean =(P1+P2+P3+P4) / 4, and then calculate the wavelength point matrix correction coefficient for each wavelength point in the wavelength point matrix one by one. Taking the m wavelength points in the first row of the wavelength point matrix as an example, the corresponding wavelength point matrix correction coefficients are: Among them, (α1,α2,…,α m ) are respectively the wavelength point matrix correction coefficients corresponding to the m wavelength points in the first row of the wavelength point matrix; Similarly, the wavelength point matrix correction coefficients corresponding to all wavelength points in the wavelength point matrix of the matrix can be calculated, and the null values ​​are not processed; Step 5: Correct the blood spectrum data in combination with the wavelength point matrix correction coefficient.

2. The method for correcting blood glucose near-infrared spectrum data based on blood absorption peak according to claim 1, characterized in that: The wavelength point with the maximum absorbance value in the spectrum in step 1 is the wavelength point corresponding to the blood absorption peak.

3. The method for correcting blood glucose near-infrared spectrum data based on blood absorption peak according to claim 2, characterized in that: According to the selected measurement band range λ1nm~λ2nm and resolution R of the spliced ​​near-infrared spectroscopy device, the number of wavelength points contained in the spliced ​​sensor is calculated. At the same time, according to the band range λ3nm~λ4nm and resolution R of the single sensor, the number of wavelength points X2 of the single sensor spectral device is calculated. The spectrum is reconstructed based on the number of wavelength points of the single sensor spectral device. The number of wavelength points contained in the spectral data collected by the reconstructed spectral device is the same as that of the single sensor spectral data.

4. The method for correcting blood glucose near-infrared spectrum data based on blood absorption peak according to claim 3, characterized in that: If the wavelength value corresponding to the blood absorption peak wavelength point is M1nm, and M1 / R is an integer, the specific method of step 2 is: use 4 reference correction points to correct the spectrum data, take the absorption peak as the center, take the wavelength point corresponding to the absorption peak as the reference correction point, and take X3 wavelength points to the left and right for spectrum reconstruction, wherein X3 is calculated as follows: X3=(X2-4) / 2, and the reconstructed spectrum actually represents the data of X2 wavelength points, wherein the wavelengths of the various wavelength points are [(M1-X3*R)nm, ..., (M1-2R)nm, (M1-R)nm, M1nm, M1nm, M1nm, M1nm, (M1+R)nm, (M1+2R)nm ..., (M1+X3*R)nm]; If the wavelength value corresponding to the blood absorption peak is M1nm, and M1 / R is not an integer, the specific method of step 2 is: round M1 to obtain a new M1 value, and use the wavelength point corresponding to the new M1 value as the reference correction point to perform the same calculation to obtain the wavelength value of each wavelength point of the reconstructed spectrum.

5. The method for correcting blood glucose near-infrared spectrum data based on blood absorption peak according to claim 4, characterized in that: When (M1-X3*R)≤λ1, the reconstructed spectrum overflows to the left, and the leftmost wavelength of the reconstructed spectrum is selected as λ1. The wavelengths of each wavelength point after reconstruction are [λ1nm, …, (M1-2R)nm, (M1-R)nm, M1nm, M1nm, M1nm, (M1+R)nm, (M1+2R)nm …, (M1+X3*R)nm]; When (M1+X3*R)≥λ2, the reconstructed spectrum overflows to the right, and the wavelength at the rightmost end of the reconstructed spectrum is selected as λ2. The wavelengths of each wavelength point after reconstruction are [(M1-X3*R)nm,...,(M1-2R)nm,(M1-R)nm,M1nm,M1nm,M1nm,M1nm,(M1+R)nm,(M1+2R)nm,...,λ2nm].

6. The method for correcting blood glucose near-infrared spectrum data based on blood absorption peak according to claim 1, characterized in that: If the absorbance values ​​of the m wavelength points in the first row of the wavelength point matrix corresponding to the collected blood spectrum data are (P1, S1, S2, ..., P2), the wavelength point matrix correction coefficients corresponding to the m wavelength points in the first row of the wavelength point matrix are (α1, α2, ..., α m ), combined with the wavelength point matrix correction coefficient, the absorbance values ​​of the m wavelength points in the first row are corrected one by one, and the corrected absorbance values ​​of the m wavelength points in the first row are (P1 / α1, S1 / α2, S2 / α3, ..., P2 / α m ); Similarly, the absorbance values ​​of the remaining wavelength points are corrected one by one in combination with the wavelength point matrix correction coefficient. When the absorbance value data of all wavelength points are corrected, the corrected absorbance values ​​are matched with the wavelength points one by one. After the matching is completed, the new wavelength point matrix containing the corrected absorbance values ​​is reorganized again, and the square matrix of m rows and n columns is reorganized into a single-row matrix, and the null values ​​are removed. The single-row matrix is ​​then rearranged from left to right in the order of wavelength values ​​from small to large. The rearranged single-row matrix is ​​the corrected blood spectrum data.

Images