Intelligent blood smear quality detection method for clinical laboratory

Abstract

The invention relates to the technical field of spectrum detection, in particular to an intelligent blood smear quality detection method for a clinical laboratory, which comprises the following steps: acquiring overall spectrum data of a blood smear and spectrum signals of detection positions of different areas in the blood smear; determining a baseline drift fluctuation section and a normal fluctuation section of the spectral signal of each detection position; performing primary correction on the spectral signal of the detection position and the overall spectral data to obtain a corrected spectral signal and first corrected overall spectral data; carrying out secondary correction on the first corrected overall spectrum data by utilizing peak values of baseline drift fluctuation sections adjacent to the current baseline drift fluctuation section in front and back in the corrected spectrum signals of the detection positions and a peak value of the current baseline drift fluctuation section, so as to obtain final second corrected overall spectrum data; and detecting the quality of the blood smear by using the second corrected overall spectral data. In this way, the detection precision and reliability of blood smear quality detection are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of spectral detection, and particularly relates to an intelligent detection method for the quality of blood smears used in a clinical laboratory. Background Art

[0002] A blood smear is made by evenly smearing a blood sample on a glass slide to form a thin film. After staining, the cell morphology, quantity, distribution and other characteristics are observed under a microscope. The quality of the blood smear determines the accuracy of the test. A blood smear of higher quality can directly and accurately reflect the blood quality or physical condition of the patient, while a blood smear of poor quality will affect the detection accuracy and efficiency, and cannot effectively use the test results for the analysis of the patient's disease. To ensure the quality of the patient's blood smear and improve the detection efficiency, it is necessary to detect the quality of the blood smear.

[0003] In some scenarios, due to problems such as a thick smear, cell stacking in the area, and baseline drift caused by the property transformation phenomenon during the chemical change of blood in the blood smear detection process, first-order derivative calculation of spectral signals is often used. Numerical methods such as the difference method or Savitzky-Golay filter are usually used to smooth the data and calculate the derivative. Subsequently, the part with baseline drift is identified through the derivative result and corrected. The first-order derivative method is often used to correct the baseline drift problem caused by a thick smear, cell stacking in the area, and the property transformation phenomenon during the chemical change of blood in the blood smear detection process. Although it can eliminate the baseline drift, the first-order derivative method will amplify the high-frequency noise in the spectral signal, resulting in an increase in signal background interference. In addition, this method will also weaken the spectral characteristic peaks, especially in the cell stacking area, which is likely to cause the loss of key biochemical information in the cell stacking area, resulting in relatively low detection accuracy and reliability for detecting the quality of blood smears. Summary of the Invention

[0004] In order to solve the technical problems of relatively low detection accuracy and reliability in detecting the quality of blood smears in the prior art, the purpose of the present invention is to provide an intelligent detection method for the quality of blood smears used in a clinical laboratory.

[0005] To solve the above technical problems, the specific technical solutions adopted are as follows: An embodiment of the present invention provides an intelligent detection method for the quality of blood smears in a clinical laboratory, including: obtaining the overall spectral data of a blood smear and the spectral signals at each detection position in different regions of the blood smear; determining the baseline drift fluctuation section and the normal fluctuation section of the spectral signal at each detection position according to the signal intensity of the extreme point of the spectral signal at each detection position in different regions and the baseline position where the extreme point is located; based on the baseline position of the normal fluctuation section within the detection position and the position where the peak of the normal fluctuation section is located, performing a primary correction on the spectral signal and the overall spectral data of the detection position to obtain a corrected spectral signal and a first corrected overall spectral data; using the peak of the baseline drift fluctuation section adjacent to the current baseline drift fluctuation section before and after in the corrected spectral signals of each detection position and the peak of the current baseline drift fluctuation section to perform a secondary correction on the first corrected overall spectral data to obtain the final second corrected overall spectral data; using the second corrected overall spectral data to detect the quality of the blood smear.

[0006] Optionally, determining the baseline drift fluctuation section and the normal fluctuation section of the spectral signal at each detection position according to the signal intensity of the extreme point of the spectral signal at each detection position in different regions and the baseline position where the extreme point is located includes: using the baseline position where the interval extreme point of the spectral signal at each detection position in each region is located as the starting and ending positions of the fluctuation, and using the spectral curve section between the starting and ending positions of the fluctuation as the fluctuation section of the spectral signal at each detection position; based on the second baseline position at the peak point of the fluctuation section adjacent to the current fluctuation section and the first baseline position of the current fluctuation section, determining the baseline drift parameter at the peak point of the current fluctuation section; based on the first signal intensity at the peak point of the fluctuation section adjacent to the current fluctuation section and the second signal intensity at the peak point of the current fluctuation section, determining the peak signal drift parameter at the peak point of the current fluctuation section; determining the first ratio between the baseline drift parameter and the peak signal drift parameter as the spectral curve drift information of the current fluctuation section; in the case where the spectral curve drift information of a continuous plurality of fluctuation sections is positive, determining the continuous plurality of fluctuation sections as the baseline drift fluctuation section, and vice versa as the normal fluctuation section.

[0007] Optionally, based on the second baseline position at the peak point of the fluctuation section adjacent to the current fluctuation section and the first baseline position of the current fluctuation section, determining the baseline drift parameter at the peak point of the current fluctuation section includes: calculating the first difference between the second baseline position and the first baseline position at the peak point of the subsequent fluctuation section adjacent to the current fluctuation section; determining the second ratio between the first difference and the first baseline position as the baseline drift parameter.

[0008] Optionally, determining a peak signal drift parameter at the peak point of the current fluctuation segment based on a first signal intensity at the peak point of a fluctuation segment adjacent to the current fluctuation segment and a second signal intensity at the peak point of the current fluctuation segment includes: calculating a second difference between the first signal intensity and the second signal intensity at the peak point of the subsequent fluctuation segment adjacent to the current fluctuation segment; determining a third ratio between the second difference and the second signal intensity as the peak signal drift parameter.

[0009] Optionally, based on the baseline position of a normal fluctuation segment within the detection position and the position where the peak of the normal fluctuation segment is located, initially correcting the spectral signal of the detection position and the overall spectral data to obtain a corrected spectral signal and a first corrected overall spectral data includes: determining a base-peak relative coefficient between the baseline position within the normal fluctuation segment and the position where the peak of the normal fluctuation segment is located based on the baseline position of the normal fluctuation segment and the position where the peak of the normal fluctuation segment is located; determining a baseline drift calibration coefficient according to a first time interval at both ends of the normal fluctuation segment, a second time interval at both ends of the baseline drift fluctuation segment, and the base-peak relative coefficient; using the baseline drift calibration coefficient to correct the baseline position of the baseline drift fluctuation segment to obtain a corrected baseline position of the baseline drift fluctuation segment; correcting the spectral signal of the corresponding detection position based on the corrected baseline position of the baseline drift fluctuation segment at each detection position and initially correcting the overall spectral data of the blood smear to obtain a corrected spectral signal and a first corrected overall spectral data.

[0010] Optionally, determining a base-peak relative coefficient between the baseline position within the normal fluctuation segment and the position where the peak of the normal fluctuation segment is located based on the baseline position of the normal fluctuation segment and the position where the peak of the normal fluctuation segment is located includes: calculating a third difference between the position where the peak of each normal fluctuation segment is located and the baseline position of the normal fluctuation segment, and superimposing the third differences of each normal fluctuation segment to obtain a first superimposed value; determining a fourth ratio between the third difference of each normal fluctuation segment and the first superimposed value as the base-peak relative coefficient between the baseline position within the normal fluctuation segment and the position where the peak of the normal fluctuation segment is located.

[0011] Optionally, determining a baseline drift calibration coefficient according to a first time interval at both ends of the normal fluctuation segment, a second time interval at both ends of the baseline drift fluctuation segment, and the base-peak relative coefficient includes: calculating a fifth ratio between the second time interval and the mean of the first time interval, and calculating the absolute value of the fourth difference between the fifth ratio and a predetermined value; calculating a first product between the absolute value of the fourth difference and the base-peak relative coefficient, and performing a normalization process on the first product to obtain the baseline drift calibration coefficient.

[0012] Optionally, using the peak values of the baseline drift fluctuation segments adjacent to the current baseline drift fluctuation segment before and after in the corrected spectral signals at each detection position and the peak value of the current baseline drift fluctuation segment, perform a secondary correction on the first corrected overall spectral data to obtain the final second corrected overall spectral data, including: using the peak values of the baseline drift fluctuation segments adjacent to the current baseline drift fluctuation segment before and after in the corrected spectral signals at each detection position and the peak value of the current baseline drift fluctuation segment, determine the change trends of the adjacent baseline drift fluctuation segments before and after; determine the identity of the change trends between different detection positions according to the change trends of the adjacent baseline drift fluctuation segments at different detection positions; group the detection positions based on the identity of the change trends to obtain multiple detection groups; based on the change trends of the adjacent fluctuation segments at different detection positions in each detection group and the positional relationship between different detection positions, determine the curve extension slope values between different detection positions in each detection group; use the detection group corresponding to the minimum value among the curve extension slope values as the analysis detection group for the blood smear; use the relative thicknesses at each detection position in the analysis detection group to determine the thickness weights of each detection position; perform a secondary correction on the corrected baseline positions based on the thickness weights to obtain the relative baseline positions of each detection position in the analysis detection group; perform a secondary correction on the first corrected overall spectral data based on the relative baseline positions of each detection position in the analysis detection group to obtain the final second corrected overall spectral data.

[0013] Optionally, determining the identity of the change trends between different detection positions according to the change trends of the adjacent baseline drift fluctuation segments at different detection positions includes: obtaining the second superimposed values of each detection position by superimposing the change trends of the adjacent baseline drift fluctuation segments at each detection position; calculating the second product between the second superimposed values of different detection positions; using the sign function to perform a sign processing on the second product to obtain the identity of the change trends between different detection positions.

[0014] Optionally, based on the change trends of the adjacent fluctuation segments at different detection positions in each detection group and the positional relationship between different detection positions, determining the curve extension slope values between different detection positions in each detection group includes: calculating the fifth difference between the second superimposed values of different detection positions, and calculating the sixth ratio between the fifth difference and the positional relationship between different detection positions; superimposing each sixth ratio to obtain the third superimposed value; calculating the fifth difference between the number of detection positions and a predetermined value, and determining the seventh ratio between the third superimposed value and the fifth difference as the curve extension slope value.

[0015] The present invention has the following beneficial effects: First, obtain the overall spectral data of the blood smear and the spectral signals at each detection position in different regions of the blood smear; and determine the baseline drift fluctuation segments and normal fluctuation segments of the spectral signals at each detection position according to the signal intensities of the extreme points of the spectral signals at each detection position in different regions and the baseline positions where the extreme points are located. Second, based on the baseline position of the normal fluctuation segment within the detection position and the position where the peak of the normal fluctuation segment is located, perform a primary correction on the spectral signal and the overall spectral data at the detection position to obtain a corrected spectral signal and a first corrected overall spectral data. Then, use the peaks of the baseline drift fluctuation segments adjacent to the current baseline drift fluctuation segment before and after in the corrected spectral signals at each detection position and the peak of the current baseline drift fluctuation segment to perform a secondary correction on the first corrected overall spectral data to obtain the final second corrected overall spectral data. Finally, use the second corrected overall spectral data to detect the quality of the blood smear.

[0016] In this way, the embodiment of the present invention can perform a primary correction on the spectral signals at each detection position and the overall spectral data according to the baseline drift conditions of the spectral signals at a single detection position in different regions of the blood smear, and then perform a secondary correction on the first corrected overall spectral data in combination with the baseline drift conditions and peak change conditions of the corrected spectral signals at each detection position in different regions. While eliminating baseline drift, it avoids the problems of amplifying high-frequency noise in the spectral signal and weakening spectral characteristic peaks by using the first derivative method, thereby avoiding the loss of key biochemical information in the cell stacking area, improving the reliability of spectral data correction, and improving the accuracy of the finally obtained spectral data, thus improving the detection accuracy and reliability of detecting the quality of the blood smear. Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0018] Figure 1 It is a flowchart of an intelligent detection method for the quality of blood smears used in a clinical laboratory provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of the division of fluctuation segments provided by an embodiment of the present invention; Figure 3 It is a schematic structural diagram of an intelligent detection device for the quality of blood smears used in a clinical laboratory provided by an embodiment of the present invention. Detailed Embodiments

[0019] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following specifically describes, with reference to the accompanying drawings and preferred embodiments, a method for intelligent detection of the quality of blood smears used in a clinical laboratory, including its specific implementation manner, structure, features, and effects. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0021] The following specifically describes the specific solution of a method for intelligent detection of the quality of blood smears used in a clinical laboratory provided by the present invention with reference to the accompanying drawings.

[0022] Embodiment 1: Please refer to Figure 1 , which shows a flowchart of a method for intelligent detection of the quality of blood smears used in a clinical laboratory provided by an embodiment of the present invention, including: Step S101, obtaining the overall spectral data of the blood smear and the spectral signals of each detection position in different regions of the blood smear.

[0023] Specifically, during the process of making the user's blood smear, whether it is manually made or automatically made, there are many unstable factors in the amount of blood droplets and the pushing process, resulting in the inability to batch produce many high-standard products. There may be phenomena such as being too thick, too thin, or uneven in thickness in the blood smear. Therefore, in the embodiment of the present invention, the blood smear made manually or automatically is placed on the spectrometer detection table for detection to obtain the overall spectral data of the blood smear and the spectral signals of each detection position in different regions of the blood smear.

[0024] More specifically, in the embodiment of the present invention, blood smears of patients are collected, and some unqualified blood smears are removed based on appearance detection, and the remaining blood smears are numbered according to the collection order. Then, a handheld Raman spectrometer is used to obtain the overall spectral data of the blood spectrum, realizing cloud-network-edge Internet of Things Raman monitoring and information retrieval, and performing preliminary preprocessing on the obtained spectral data to remove noise. Secondly, in the embodiment of the present invention, the blood smear is divided into multiple regions, such as the head region, the body region, and the tail region, and each detection position is evenly selected in each region and spectral detection is performed on each detection position.

[0025] Step S102: Determine the baseline drift fluctuation segments and normal fluctuation segments of the spectral signals at each detection position according to the signal intensities of the extreme points of the spectral signals at each detection position in different regions and the baseline positions where the extreme points are located.

[0026] Specifically, there may be cell overlapping phenomena in the overly thick regions of the blood smear, which may further lead to overlapping of blood cells, thus affecting the baseline shift of the blood smear spectrum and the quality detection accuracy. Therefore, in the embodiments of the present invention, based on the spectral signals at each detection position in different regions collected, the influence caused by the baseline shift due to cell overlapping in the spectral signals at each detection position in different regions is eliminated, the waveform correction of the overall spectral data is realized, the true spectral data of the blood smear is obtained, and the quality detection of the blood smear is completed. Among them, during the detection process of the blood smear spectral technology, the baseline shift is a regional baseline deviation caused by uneven thickness of the blood smear. And because biochemical component dynamic changes will occur after the blood leaves the body, baseline drift (i.e., baseline shift showing irregular fluctuations) may occur during the detection process. Therefore, through analysis, it can be seen that the change of single-coordinate spectral data is beneficial to the analysis of the overall baseline shift of the blood smear. Therefore, in the embodiments of the present invention, the accuracy of the spectral data at a single detection position is analyzed through the difference analysis of spectral data changes.

[0027] Further, as an optional embodiment of the present invention, determining the baseline drift fluctuation segments and normal fluctuation segments of the spectral signals at each detection position according to the signal intensities of the extreme points of the spectral signals at each detection position in different regions and the baseline positions where the extreme points are located includes: taking the baseline positions where the intermediate extreme points of the spectral signals at each detection position in each region are located as the starting and ending positions of the fluctuation, and taking the spectral curve segment between the starting and ending positions of the fluctuation as the fluctuation segment of the spectral signal at each detection position; based on the second baseline position at the peak point of the fluctuation segment adjacent to the current fluctuation segment and the first baseline position of the current fluctuation segment, determining the baseline drift parameter at the peak point of the current fluctuation segment; based on the first signal intensity at the peak point of the fluctuation segment adjacent to the current fluctuation segment and the second signal intensity at the peak point of the current fluctuation segment, determining the peak signal drift parameter at the peak point of the current fluctuation segment; determining the first ratio between the baseline drift parameter and the peak signal drift parameter as the spectral curve drift information of the current fluctuation segment; in the case where the spectral curve drift information of a continuous plurality of fluctuation segments is positive, determining the continuous plurality of fluctuation segments as the baseline drift fluctuation segments, and vice versa as the normal fluctuation segments.

[0028] Specifically, in the embodiments of the present invention, the thicker the blood smear, the more cells are present in the region, and the more serious the data distortion caused by cell stacking phenomenon. Therefore, when analyzing a single coordinate of the blood smear in the embodiments of the present invention, the spectral signals at each detection position in the volume region are taken as an example for analysis. In the embodiments of the present invention, taking the coordinate n of a detection position in the volume region as an example, its baseline position is denoted as . Obtain the spectral signal curve at coordinate n , and divide the fluctuation segments of the spectral sequence by the signal fluctuation at this detection position in the volume region. In the embodiments of the present invention, based on the change trend of the spectral signal intensity at the detection position and in combination with the positions of the extreme points, the waveband between two adjacent extreme points is taken as a fluctuation segment. Exemplarily, as Figure 2 shown, Figure 2 is a schematic diagram of the division of a fluctuation segment provided by an embodiment of the present invention. In the embodiments of the present invention, the initial position point of the spectrum at a certain detection position is used as the initial extreme point , and the baseline position of the extreme point is obtained along the change of the spectral signal at this detection position. The positions of the extreme points with the same direction of change (adjacent extreme points) are used as the starting and ending positions of the fluctuation, and then the fluctuation segment of the spectral signal is obtained.

[0029] Further, compare the spectral signal intensity at any end of the fluctuation segment with the signal intensity of the extreme point within the fluctuation segment . Then the larger value is the peak value. If the signal intensity at any end within the fluctuation segment , then readjust the position of the initial extreme point to (only for the initial extreme point). Continuously detect and then readjust the fluctuation segment based on the position of the adjusted initial extreme point. The signal extreme point within the adjusted fluctuation segment is used as the signal peak value .

[0030] Further, the baseline drift in the embodiments of the present invention is a smooth and continuous fluctuation on the spectral curve. In the embodiments of the present invention, the baseline drift phenomenon of the spectral curve is analyzed by the change value between spectral curves . Among them, first obtain the baseline drift parameter. As an optional embodiment of the present invention, based on the second baseline position at the peak point of the fluctuation segment adjacent to the current fluctuation segment and the first baseline position of the current fluctuation segment, determining the baseline drift parameter at the peak point of the current fluctuation segment includes: calculating the first difference between the second baseline position and the first baseline position at the peak point of the next fluctuation segment adjacent to the current fluctuation segment; determining the second ratio between the first difference and the first baseline position as the baseline drift parameter.

[0031] Specifically, in the embodiment of the present invention, taking the fluctuation segment m as the current fluctuation segment as an example, the embodiment of the present invention specifically calculates the baseline drift parameter by the following formula: In the above formula, is the baseline drift parameter at the peak point of the current fluctuation segment m. m is the number of the fluctuation segment, is the first baseline position at the peak point of the current fluctuation segment m. is the second baseline position at the peak point of the fluctuation segment adjacent to the current fluctuation segment m. is the baseline position offset at the peak point of the adjacent fluctuation segment, is the baseline drift parameter at the peak point of the current fluctuation segment m.

[0032] Furthermore, as an optional embodiment of the present invention, based on the first signal intensity at the peak point of the fluctuation segment adjacent to the current fluctuation segment and the second signal intensity at the peak point of the current fluctuation segment, determining the peak signal drift parameter at the peak point of the current fluctuation segment includes: calculating the second difference between the first signal intensity and the second signal intensity at the peak point of the next fluctuation segment adjacent to the current fluctuation segment; determining the third ratio between the second difference and the second signal intensity as the peak signal drift parameter.

[0033] Specifically, in the embodiment of the present invention, taking the fluctuation segment m as the current fluctuation segment as an example, the embodiment of the present invention specifically calculates the peak signal drift parameter by the following formula: In the above formula, represents the peak signal drift parameter at the peak point of the current fluctuation segment m. represents the second signal intensity at the peak point of the current fluctuation segment m. represents the first signal intensity at the peak point of the fluctuation segment m+1 adjacent to the current fluctuation segment m. is the difference in signal intensity at the peak point.

[0034] Furthermore, the embodiment of the present invention calculates the spectral curve drift information of the current fluctuation segment m by the following formula: In the above formula, represents the spectral curve drift information of the current fluctuation segment m compared to the adjacent fluctuation segment m+1. represents the peak signal drift parameter at the peak point of the current fluctuation segment m. is the baseline drift parameter at the peak point of the current fluctuation segment m.

[0035] Further, when there is a drift phenomenon at the baseline position, the corresponding absorption peak changes accordingly. When the baseline position and the signal intensity change in the same positive or negative direction (at this time is positive), and multiple consecutive fluctuation segments show the same performance, it indicates that there is a baseline drift problem in this part of the fluctuation segments, and it is determined as a baseline drift fluctuation segment; otherwise, it is a normal fluctuation segment.

[0036] Step S103: Based on the baseline position of the normal fluctuation segments within the detection position and the positions of the peaks of the normal fluctuation segments, perform an initial correction on the spectral signals of the detection position and the overall spectral data to obtain corrected spectral signals and first-corrected overall spectral data.

[0037] Specifically, in the embodiment of the present invention, the baseline of the baseline drift fluctuation segments with baseline drift is corrected based on the baseline of the non-baseline drift fluctuation segments (normal fluctuation segments). Among them, as an optional embodiment of the present invention, based on the baseline position of the normal fluctuation segments within the detection position and the positions of the peaks of the normal fluctuation segments, performing an initial correction on the spectral signals of the detection position and the overall spectral data to obtain corrected spectral signals and first-corrected overall spectral data includes: determining the base-peak relative coefficient between the baseline position within the normal fluctuation segments and the positions of the peaks of the normal fluctuation segments based on the baseline position of the normal fluctuation segments and the positions of the peaks of the normal fluctuation segments; determining the baseline drift correction coefficient according to the first time interval at both ends of the normal fluctuation segments, the second time interval at both ends of the baseline drift fluctuation segments, and the base-peak relative coefficient; using the baseline drift correction coefficient to correct the baseline position of the baseline drift fluctuation segments to obtain the corrected baseline position of the baseline drift fluctuation segments; based on the corrected baseline positions of the baseline drift fluctuation segments at each detection position, correcting the spectral signals of the corresponding detection positions and performing an initial correction on the overall spectral data of the blood smear to obtain corrected spectral signals and first-corrected overall spectral data.

[0038] Specifically, the embodiment of the present invention analyzes based on the normal fluctuation segments to obtain the base-peak relative coefficient between the baseline position and the fluctuation peak within the normal fluctuation segments. As an optional embodiment of the present invention, determining the base-peak relative coefficient between the baseline position within the normal fluctuation segments and the positions of the peaks of the normal fluctuation segments based on the baseline position of the normal fluctuation segments and the positions of the peaks of the normal fluctuation segments includes: calculating the third difference between the position of the peak of each normal fluctuation segment and the baseline position of the normal fluctuation segment, and superimposing the third differences to obtain a first superimposed value; determining the fourth ratio between the third difference of each normal fluctuation segment and the first superimposed value as the base-peak relative coefficient between the baseline position within the normal fluctuation segments and the positions of the peaks of the normal fluctuation segments.

[0039] Specifically, the embodiment of the present invention specifically calculates the base-peak relative coefficient using the following formula: In the above formula, represents the base-peak relative coefficient between the baseline position in the normal fluctuation segment and the position of the peak in the normal fluctuation segment. is the number of the normal fluctuation segment. is the total number of the normal fluctuation segments. is the position of the peak in the normal fluctuation segment. is the baseline position of the normal fluctuation segment. is the fluctuation difference between the position of the peak in the fluctuation segment and the baseline position.

[0040] Furthermore, when the baseline shifts, the spectral curve frequency (represented by the time interval between both ends of the fluctuation segment) is affected. Therefore, the baseline drift calibration coefficient is obtained by comparing the frequency changes of the baseline drift fluctuation segment and the normal fluctuation segment and the relative relationship between the fluctuation peaks. As an optional embodiment of the present invention, according to the first time interval between both ends of the normal fluctuation segment, the second time interval between both ends of the baseline drift fluctuation segment, and the base-peak relative coefficient, determining the baseline drift calibration coefficient includes: calculating the fifth ratio between the mean value of the second time interval and the mean value of the first time interval, and calculating the absolute value of the fourth difference between the fifth ratio and a predetermined value; calculating the first product between the absolute value of the fourth difference and the base-peak relative coefficient, and performing a normalization process on the first product to obtain the baseline drift calibration coefficient.

[0041] Specifically, the embodiment of the present invention calculates the baseline drift calibration coefficient by using the following formula: In the above formula, represents the baseline drift calibration coefficient. represents the second time interval between both ends of the baseline drift fluctuation segment, that is, the fluctuation frequency. represents the mean value of the fluctuation frequencies of the normal fluctuation segment, that is, the mean value of the first time interval between both ends of the normal fluctuation segment. is the difference between the fluctuation frequencies of the baseline drift fluctuation segment and the normal fluctuation segment, representing the deviation coefficient between the baseline drift fluctuation segment and the normal fluctuation segment. is the relative coefficient of the deviation coefficient and the baseline position, representing the overall difference between the baseline drift fluctuation segment and the normal fluctuation segment. The function represents a normalization function, which is used to perform a normalization process on the result to obtain the baseline drift calibration coefficient .

[0042] Furthermore, based on the obtained baseline drift calibration coefficient, the baseline position of the fluctuation segment with baseline drift is calibrated and corrected. The embodiment of the present invention specifically uses the following formula to correct the baseline position of the baseline drift fluctuation segment: In the above formula, is the corrected baseline position after correction of the baseline drift fluctuation segment m with baseline drift, is the baseline position before correction of the baseline drift fluctuation segment m. represents the baseline drift calibration coefficient.

[0043] Based on the above steps, the embodiments of the present invention process the data with baseline drift at each detection position in each region, obtain the corrected baseline position at each detection position, and ensure the accuracy of the baseline position. Further, the corrected baseline position is put into the Spectrum FL software model to correct the overall spectral data of the blood smear, and then the actual first corrected overall spectral data is obtained, and the peak of its spectral curve is the absorbance (representing the corrected absorbance).

[0044] Step S104: Use the peak values of the baseline drift fluctuation segments adjacent to the current baseline drift fluctuation segment before and after in the corrected spectral signals at each detection position and the peak value of the current baseline drift fluctuation segment to perform secondary correction on the first corrected overall spectral data to obtain the final second corrected overall spectral data.

[0045] Specifically, in the embodiments of the present invention, the blood tissues in each region on the blood smear all come from one patient, and the blood components are similar. When detecting the blood smear by spectral technology, the spectral signal intensities in different regions may be different, but the changing trends of the spectral signals are similar, that is, the changing trend of the spectral signal at adjacent optical wave wavelengths at one position is similar to the changing trend of the spectral curve signal intensity at another position. Furthermore, to ensure the data accuracy of the corrected spectral curves of the blood smear between different regions, the changing trend of the spectral curve can be detected by analyzing the changing trends of the spectral curves between different detection positions. The embodiments of the present invention obtain the first corrected overall spectral data after initial correction based on the operations of the above embodiments, and traverse the data at the remaining detection positions in the blood smear to achieve baseline drift correction of the spectra at each detection position of the blood smear.

[0046] Further, as an optional embodiment of the present invention, the peak values of the baseline drift fluctuation segments adjacent to the current baseline drift fluctuation segment before and after in the corrected spectral signals at each detection position and the peak value of the current baseline drift fluctuation segment are used to perform a secondary correction on the first corrected overall spectral data to obtain the final second corrected overall spectral data, including: using the peak values of the baseline drift fluctuation segments adjacent to the current baseline drift fluctuation segment before and after in the corrected spectral signals at each detection position and the peak value of the current baseline drift fluctuation segment to determine the change trend of the adjacent baseline drift fluctuation segments before and after; determining the identity of the change trends between different detection positions according to the change trends of the adjacent baseline drift fluctuation segments at different detection positions; grouping each detection position based on the identity of the change trends to obtain multiple detection groups; determining the curve extension slope value between different detection positions in each detection group based on the change trends of the adjacent fluctuation segments at different detection positions in each detection group and the positional relationship between different detection positions; taking the detection group corresponding to the minimum value in the curve extension slope values as the analysis and detection group of the blood smear; using the relative thickness at each detection position in the analysis and detection group to determine the thickness weight of each detection position; performing a secondary correction on the corrected baseline position based on the thickness weight to obtain the relative baseline position of each detection position in the analysis and detection group; performing a secondary correction on the first corrected overall spectral data based on the relative baseline position of each detection position in the analysis and detection group to obtain the final second corrected overall spectral data.

[0047] Specifically, the embodiment of the present invention extracts the spectral data between the corrected fluctuation segments and is based on the peak value of the corrected baseline drift fluctuation segment and combines the fluctuation segments on both sides of the corrected baseline drift fluctuation segment m 、 to determine the change trend of the adjacent baseline drift fluctuation segments before and after.

[0048] Further, the embodiment of the present invention specifically uses the following formula to calculate the change trend of the adjacent baseline drift fluctuation segments before and after: In the above formula, represents the change trend of the baseline drift fluctuation segments adjacent to the corrected baseline drift fluctuation segment m before and after. represents the signal intensity of the peak value of the corrected current baseline drift fluctuation segment m. represents the signal intensity of the peak value of the previous baseline drift fluctuation segment m - 1 adjacent to the corrected current baseline drift fluctuation segment m. represents the signal intensity of the peak value of the next baseline drift fluctuation segment m + 1 adjacent to the corrected current baseline drift fluctuation segment m. 、 They are respectively the ratios of the signal intensities of the peaks of the corrected current baseline drift fluctuation segment m to the signal intensities of the peaks of the adjacent fluctuation segments before and after. It is the ratio difference, indicating the changing trend between adjacent fluctuation segments. The larger the value, when it is positive, the overall fluctuation segment shows an increasing trend, and vice versa, it shows a decreasing trend.

[0049] Furthermore, the embodiments of the present invention traverse the changing trends of each detection position in each region, and determine the identity of the changing trends between different detection positions by obtaining the changing trends between the detection positions. As an optional embodiment of the present invention, determining the identity of the changing trends between different detection positions according to the changing trends of each adjacent baseline drift fluctuation segment of different detection positions includes: obtaining the superimposed value of the changing trends of each adjacent baseline drift fluctuation segment of each detection position to obtain the second superimposed value of each detection position; calculating the second product between the second superimposed values of different detection positions; using the sign function to perform sign processing on the second product to obtain the identity of the changing trends between different detection positions.

[0050] Specifically, the embodiments of the present invention use the following formula to calculate the identity of the changing trends between different detection positions: In the above formula, represents the identity of the changing trends between different detection positions. represents the superimposed value of the changing trends of each adjacent baseline drift fluctuation segment of the nth detection position. represents the superimposed value of the changing trends of each adjacent baseline drift fluctuation segment of the (n + 1)th detection position. is the product of the changing trends between detection positions, and uses the sign function to perform sign processing.

[0051] Furthermore, the embodiments of the present invention traverse all detection positions. When , the changing trends of the spectral curves of the corresponding two detection positions are the same, and then these two detection positions are divided into a detection group. For example, taking the current detection position as an example, if there are N detection positions among other detection positions that are with the current detection position, then these N detection positions and the current detection position are divided into a detection group. That is to say, each detection group includes at least two detection positions.

[0052] Further, as an optional embodiment of the present invention, based on the change trends of adjacent fluctuation segments at different detection positions in each detection group and the positional relationship between different detection positions, determining the curve extension slope value between different detection positions in each detection group includes: calculating the fifth difference between the second superimposed values of different detection positions, and calculating the sixth ratio between the fifth difference and the positional relationship between different detection positions; superimposing each sixth ratio to obtain a third superimposed value; calculating the fifth difference between the number of detection positions and a predetermined value, and determining the seventh ratio between the third superimposed value and the fifth difference as the curve extension slope value.

[0053] Specifically, the embodiment of the present invention calculates the curve extension slope value between different detection positions in each detection group using the following formula: Y = In the above formula, Y represents the curve extension slope value between different detection positions in the detection group. represents the superimposed value of the change trends of adjacent baseline drift fluctuation segments at the nth detection position in the detection group. represents the superimposed value of the change trends of adjacent baseline drift fluctuation segments at the (n + 1)th detection position in the detection group. represents the positional relationship between the nth detection position and the (n + 1)th detection position, that is, the distance between the two. represents the number of detection positions in the detection group.

[0054] Further, the embodiment of the present invention uses function (minimum value function) to screen the detection group where the minimum value of the curve extension slope value is located as the analysis detection group. That is .

[0055] Further, based on the spectral data within the above-screened analysis detection group, according to the absorbance obtained according to the above embodiment, based on the formula of absorbance: Beer-Lambert law , further obtain the relative thickness at detection position n, and further obtain the thickness weight ( is the minimum value to prevent the denominator from being 0).

[0056] Further, the embodiment of the present invention performs a secondary correction on the corrected baseline position based on the thickness weight, specifically using the formula for secondary correction, is the relative baseline position, is the thickness weight, is the corrected baseline position.

[0057] Further, the embodiment of the present invention takes the relative baseline position Bring the overall spectral data of the blood smear into the existing Spectrum FL software model for correction, and put the overall spectral data after secondary correction into the Spectral Residual model for residual analysis. Extract the residuals and retain the pathological signals through wavelet decomposition to reconstruct the final second-corrected overall spectral data.

[0058] Step S105: Detect the quality of the blood smear using the second-corrected overall spectral data.

[0059] Specifically, after obtaining the final second-corrected overall spectral data through the above steps in the embodiments of the present invention, bring the final second-corrected overall spectral data into the Raman Spectroscopy model. Detect the quality of the blood smear by analyzing the morphology of red blood cells and the integrity of the blood smear, and transmit the detection result to the tester.

[0060] The embodiments of the present invention can initially correct the spectral signals of each detection position and the overall spectral data based on the baseline drift of the spectral signals at a single detection position in different regions of the blood smear. Then, combine the baseline drift and peak change of the corrected spectral signals at each detection position in different regions to perform secondary correction on the first-corrected overall spectral data after the initial correction. While eliminating the baseline drift, it avoids the problems of amplifying high-frequency noise in the spectral signals and weakening spectral characteristic peaks by using the first derivative method, thereby avoiding the loss of key biochemical information in the cell stacking region, improving the reliability of spectral data correction, improving the accuracy of the finally obtained spectral data, and thus improving the detection accuracy and reliability of detecting the quality of the blood smear.

[0061] Embodiment 2: Corresponding to the intelligent detection method for the quality of blood smears in the inspection department provided in the above embodiment, based on the same technical concept, the embodiments of the present invention also provide an intelligent detection device for the quality of blood smears in the inspection department. This intelligent detection device for the quality of blood smears in the inspection department is used to execute the above intelligent detection method for the quality of blood smears in the inspection department. Figure 3 It is a schematic structural diagram of an intelligent detection device for the quality of blood smears in the inspection department provided by an embodiment of the present invention, as Figure 3 shown. The intelligent detection device for the quality of blood smears in the inspection department may vary greatly due to configuration or performance, and may include one or more processors 301 and a memory 302. The memory 302 is used to store computer programs that can run on the processor 301. The processor 301 is used to execute the programs stored in the memory 302 to implement the above Figure 1Each step in the method embodiments. Among them, the memory 302 can be transient storage or persistent storage. The application programs stored in the memory 302 can include one or more modules (not shown in the figure), and each module can include a series of computer-executable instructions for the intelligent detection device for the quality of blood smears in the inspection department.

[0062] Furthermore, the processor 301 can be set to communicate with the memory 302 and execute a series of computer-executable instructions in the memory 302 on the intelligent detection device for the quality of blood smears in the inspection department. The intelligent detection device for the quality of blood smears in the inspection department can also include one or more power supplies 303, one or more wired or wireless network interfaces 304, one or more input / output interfaces 305, and one or more keyboards 306.

[0063] Specifically in this embodiment, the intelligent detection device for the quality of blood smears in the inspection department includes a processor, a communication interface, a memory, and a communication bus; among them, the processor, the communication interface, and the memory complete mutual communication through the bus; the memory is used to store computer programs; the processor is used to execute the programs stored on the memory to implement the above Figure 1 Each step in the method embodiments, and has the beneficial effects of the above method embodiments. To avoid repetition, the embodiments of the present invention will not be described in detail here.

[0064] It should be noted that the intelligent detection device for the quality of blood smears in the inspection department provided by the embodiments of the present invention and the intelligent detection method for the quality of blood smears in the inspection department provided by the embodiments of the present invention are based on the same application concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned intelligent detection method for the quality of blood smears in the inspection department, and has the same or similar beneficial effects. The repeated parts will not be described again.

[0065] It should be noted that: the above-mentioned sequence of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0066] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments.

[0067] The embodiments of the present invention also propose a computer-readable storage medium. The computer-readable medium stores one or more programs, and when the one or more programs are executed by an electronic device including multiple application programs, the electronic device is caused to execute Figure 1The method disclosed in the illustrated embodiment realizes the functions and beneficial effects of each method in the foregoing method embodiments, which will not be elaborated herein.

[0068] Among them, the computer-readable storage medium includes read-only memory (ROM for short), random access memory (RAM for short), magnetic disk, optical disc, etc.

Claims

1. An intelligent detection method for the quality of blood smears used in the inspection department, characterized in that, Including: Obtaining the overall spectral data of the blood smear and the spectral signals at each detection position in different regions of the blood smear; Determining the baseline drift fluctuation segment and the normal fluctuation segment of the spectral signal at each detection position according to the signal intensity of the extreme point of the spectral signal at each detection position in different regions and the baseline position where the extreme point is located; Based on the baseline position of the normal fluctuation segment within the detection position and the position where the peak of the normal fluctuation segment is located, initially correcting the spectral signal at the detection position and the overall spectral data to obtain a corrected spectral signal and a first corrected overall spectral data; Using the peak of the baseline drift fluctuation segment adjacent to the current baseline drift fluctuation segment before and after in the corrected spectral signals at each detection position and the peak of the current baseline drift fluctuation segment to perform a secondary correction on the first corrected overall spectral data to obtain the final second corrected overall spectral data; Detecting the quality of the blood smear using the second corrected overall spectral data.

2. The intelligent detection method for blood smear quality in the inspection department according to claim 1, wherein The determining the baseline drift fluctuation segment and the normal fluctuation segment of the spectral signal at each detection position according to the signal intensity of the extreme point of the spectral signal at each detection position in different regions and the baseline position where the extreme point is located includes: Taking the baseline position where the interval extreme points in the spectral signal at each detection position in each region are located as the starting and ending positions of the fluctuation, and taking the spectral curve segment between the starting and ending positions of the fluctuation as the fluctuation segment of the spectral signal at each detection position; Based on the second baseline position at the peak point of the fluctuation segment adjacent to the current fluctuation segment and the first baseline position of the current fluctuation segment, determining the baseline drift parameter at the peak point of the current fluctuation segment; Based on the first signal intensity at the peak point of the fluctuation segment adjacent to the current fluctuation segment and the second signal intensity at the peak point of the current fluctuation segment, determining the peak signal drift parameter at the peak point of the current fluctuation segment; Determining the first ratio between the baseline drift parameter and the peak signal drift parameter as the spectral curve drift information of the current fluctuation segment; When the spectral curve drift information of a continuous plurality of fluctuation segments is positive, determining the continuous plurality of fluctuation segments as the baseline drift fluctuation segment, otherwise as the normal fluctuation segment.

3. The intelligent detection method for blood smear quality in the inspection department according to claim 2, characterized in that, The determining the baseline drift parameter at the peak point of the current fluctuation segment based on the second baseline position at the peak point of the fluctuation segment adjacent to the current fluctuation segment and the first baseline position of the current fluctuation segment includes: Calculating a first difference between the second baseline position at the peak point of the next fluctuation segment adjacent to the current fluctuation segment and the first baseline position; Determining the second ratio between the first difference and the first baseline position as the baseline drift parameter.

4. The intelligent detection method for blood smear quality in the inspection department according to claim 2, wherein The determining the peak signal drift parameter at the peak point of the current fluctuation segment based on the first signal intensity at the peak point of the fluctuation segment adjacent to the current fluctuation segment and the second signal intensity at the peak point of the current fluctuation segment includes: Calculating a second difference between the first signal intensity at the peak point of the next fluctuation segment adjacent to the current fluctuation segment and the second signal intensity; Determine that the third ratio between the second difference and the second signal strength is the peak signal drift parameter.

5. The intelligent detection method for the quality of blood smears used in the inspection department according to claim 1, wherein, Based on the baseline position of the normal fluctuation segment within the detection position and the position where the peak of the normal fluctuation segment is located, the primary correction of the spectral signal of the detection position and the overall spectral data includes: Based on the baseline position of the normal fluctuation segment and the position where the peak of the normal fluctuation segment is located, determine the base-peak relative coefficient between the baseline position within the normal fluctuation segment and the position where the peak of the normal fluctuation segment is located; According to the first time interval at both ends of the normal fluctuation segment, the second time interval at both ends of the baseline drift fluctuation segment, and the base-peak relative coefficient, determine the baseline drift calibration coefficient; Use the baseline drift calibration coefficient to correct the baseline position of the baseline drift fluctuation segment to obtain the corrected baseline position of the baseline drift fluctuation segment; Based on the corrected baseline position of the baseline drift fluctuation segment of each detection position, correct the spectral signal of the corresponding detection position and perform a primary correction on the overall spectral data of the blood smear to obtain a corrected spectral signal and a first corrected overall spectral data.

6. The intelligent detection method for the quality of blood smears used in the inspection department according to claim 5, characterized in that, The determination of the base-peak relative coefficient between the baseline position within the normal fluctuation segment and the position where the peak of the normal fluctuation segment is located based on the baseline position of the normal fluctuation segment and the position where the peak of the normal fluctuation segment is located includes: Calculate the third difference between the position where the peak of each normal fluctuation segment is located and the baseline position of the normal fluctuation segment, and superimpose the third differences to obtain a first superimposed value; Determine that the fourth ratio between the third difference of each normal fluctuation segment and the first superimposed value is the base-peak relative coefficient between the baseline position within the normal fluctuation segment and the position where the peak of the normal fluctuation segment is located.

7. The intelligent detection method for blood smear quality in the inspection department according to claim 5, characterized in that, The determination of the baseline drift calibration coefficient according to the first time interval at both ends of the normal fluctuation segment, the second time interval at both ends of the baseline drift fluctuation segment, and the base-peak relative coefficient includes: Calculate the fifth ratio between the second time interval and the mean of the first time intervals, and calculate the absolute value of the fourth difference between the fifth ratio and a predetermined value; Calculate the first product of the absolute value of the fourth difference and the base-peak relative coefficient, and perform a normalization process on the first product to obtain the baseline drift calibration coefficient.

8. The intelligent detection method for the quality of blood smears used in the inspection department according to claim 5, characterized in that, The secondary correction of the first corrected overall spectral data using the peaks of the baseline drift fluctuation segments adjacent to the current baseline drift fluctuation segment before and after and the peak of the current baseline drift fluctuation segment in the corrected spectral signal of each detection position to obtain the final second corrected overall spectral data includes: Using the peaks of the baseline drift fluctuation segments adjacent to the current baseline drift fluctuation segment before and after and the peak of the current baseline drift fluctuation segment in the corrected spectral signal of each detection position, determine the change trend of the adjacent baseline drift fluctuation segments before and after; Determine the identity of the change trends between different detection positions according to the change trends of the adjacent baseline drift fluctuation segments at different detection positions; Group each detection position based on the identity of the change trends to obtain multiple detection groups; Based on the change trends of adjacent fluctuation segments at different detection positions in each detection group and the positional relationship between different detection positions, determine the curve extension slope values between different detection positions in each detection group; Take the detection group corresponding to the minimum value among the curve extension slope values as the analysis and detection group of the blood smear; Use the relative thickness at each detection position in the analysis and detection group to determine the thickness weight of each detection position; Based on the thickness weight, perform a secondary correction on the corrected baseline position to obtain the relative baseline position of each detection position in the analysis and detection group; Based on the relative baseline position of each detection position in the analysis and detection group, perform a secondary correction on the first corrected overall spectral data to obtain the final second corrected overall spectral data.

9. The intelligent detection method for the quality of blood smears used in the inspection department according to claim 8, wherein, The determination of the identity of the change trends between different detection positions according to the change trends of adjacent baseline drift fluctuation segments at different detection positions includes: Determine the superposition value of the change trends of adjacent baseline drift fluctuation segments at each detection position to obtain the second superposition value of each detection position; Calculate the second product between the second superposition values of different detection positions; Use the sign function to perform sign processing on the second product to obtain the identity of the change trends between different detection positions.

10. The intelligent blood smear quality detection method for the inspection department according to claim 9, characterized in that, The determination of the curve extension slope values between different detection positions in each detection group based on the change trends of adjacent fluctuation segments at different detection positions in each detection group and the positional relationship between different detection positions includes: Calculate the fifth difference between the second superposition values of different detection positions, and calculate the sixth ratio between the fifth difference and the positional relationship between different detection positions; Superpose each sixth ratio to obtain the third superposition value; Calculate the fifth difference between the number of detection positions and a predetermined value, and determine that the seventh ratio between the third superposition value and the fifth difference is the curve extension slope value.

Images