An intelligent detection method for premature rupture of membranes

Abstract

The present invention relates to the field of liquid detection technology, and specifically to an intelligent detection method for premature rupture of membranes. First, based on the ion intensity deviation and ion intensity change of the base peak diagram and the total ion current diagram in time sequence, the chromatographic data change characteristics during the detection process are analyzed, and the sample component richness of the pad secretions detected in this test is comprehensively analyzed in combination with the ion intensity change; then, the chromatographic data of each biomarker are further analyzed to evaluate the chromatographic separation completeness of the biomarker when interfered by other costs, and to amplify the accuracy of the biomarker with higher chromatographic separation completeness in evaluating the sample contamination characteristics; finally, by analyzing the mass spectrometry data characteristics of the biomarker, the contamination interference degree of the sample is calculated in combination with the chromatographic separation completeness and the sample component richness, thereby improving the detection accuracy of the biomarker while taking into account the contamination interference.

Description

Technical Field

[0001] The present invention relates to the technical field of liquid detection, and in particular to an intelligent detection method for premature rupture of membranes. Background Art

[0002] Premature rupture of membranes (PROM) refers to the rupture of the fetal membranes before labor begins, resulting in the leakage of amniotic fluid. PROM is typically detected by monitoring biomarkers in sanitary pad secretions. Traditionally, pH test strips are used for biomarker detection, which typically detects a single biomarker. However, when a biomarker is interfered with by other bodily fluids, such as vaginal inflammation or blood contamination, the test results may be affected. To ensure accuracy, existing technologies utilize mass spectrometry to detect various biomarkers in sanitary pad secretions extracted from sanitary pads.

[0003] However, in actual applications, the secretion samples on the panty liner may contain other liquids such as blood and urine, and the blood, urine and other samples also contain a large amount of high-abundance interferents such as salt and protein, which will cause different ion interference effects, resulting in errors in the mass spectrometry results and reducing the accuracy of biomarker detection; therefore, the existing technology of directly monitoring biomarkers of panty liner secretions through mass spectrometry technology without considering contamination interference has low accuracy. Summary of the Invention

[0004] In order to solve the technical problem of low accuracy of the existing method of directly monitoring biomarkers of pad secretions through mass spectrometry without considering interference, the purpose of this application is to provide an intelligent detection method for premature rupture of membranes. The technical solution adopted is as follows:

[0005] The first aspect of the present application provides an intelligent method for detecting premature rupture of membranes, comprising:

[0006] The physiological characteristics of each pregnancy test of the mother and the corresponding pad secretions are separated into components to obtain the corresponding mass spectra, base peak maps, total ion current maps, and extracted ion current chromatograms of each biomarker;

[0007] Determine the richness of the sample components for each test based on the physiological characteristic information during pregnancy and the ion intensity deviation and ion intensity change of the base peak graph and the total ion current graph in time sequence;

[0008] The completeness of the chromatographic separation of each biomarker in each test was determined based on the peak morphology distortion and signal intensity distribution in the extracted ion current chromatogram of each biomarker;

[0009] The sample contamination interference degree of each test is determined based on the richness of the sample components, the suppression of the spectral peaks of various biomarkers in the mass spectrum, and the corresponding chromatographic separation completeness; and premature rupture of membranes is detected based on the sample contamination interference degree.

[0010] Furthermore, the process of obtaining the sample component richness includes:

[0011] Using a deep learning model, the corresponding secretion volume index is determined based on all the physiological characteristics of the mother during pregnancy during each test.

[0012] Determining the corresponding ion abundance according to the ion intensity deviation between the total ion current graph and the base peak graph at each time point;

[0013] The degree of ion fluctuation at each time point is determined based on the relative deviation of ion intensity changes between each time point and adjacent time points;

[0014] Determine an adjustment coefficient at each time point based on the product of the ion abundance and the ion fluctuation degree; and use the ratio of the adjustment coefficient to the corresponding ion intensity in the total ion current map at the corresponding time point as a weighting coefficient;

[0015] The richness of the sample components in each test is determined based on the product of the mean value of the weighted coefficients at all time points and the secretion volume index.

[0016] Furthermore, the process of obtaining the secretion volume index includes:

[0017] All the physiological characteristics of the mother during pregnancy during each test are input into the trained fully connected neural network model, and the secretion volume index of each test is output.

[0018] Furthermore, the process of obtaining the ion richness includes:

[0019] The difference between the corresponding ion intensity in the total ion current map and the corresponding ion intensity in the base peak map at each time point is normalized to determine the ion richness at each time point.

[0020] Furthermore, the process of obtaining the ion fluctuation degree includes:

[0021] The difference between the derivative value at each time point in the total ion current graph and the derivative value at the previous time point is used as the first reference difference; the difference between the derivative value at each time point in the total ion current graph and the derivative value at the next time point is used as the second reference difference; and the degree of ion fluctuation at each time point is determined based on the mean between the first reference difference and the second reference difference.

[0022] Furthermore, the process of obtaining the completeness of the chromatographic separation includes:

[0023] Determine all merged chromatographic peaks based on the distribution of maximum points and inflection points in the extracted ion current chromatogram;

[0024] Determine the corresponding drift distortion characteristic parameter according to the product between the time length of the corresponding interval of each merged chromatographic peak and the absolute value of the skewness;

[0025] The local separation characteristic value of each merged chromatographic peak is determined based on the ratio between the signal maximum value in each merged chromatographic peak and the drift distortion characteristic parameter; the chromatographic separation completeness of each biomarker in each detection is determined based on the average of the local separation characteristic values ​​of all merged chromatographic peaks.

[0026] Furthermore, the process of obtaining the merged chromatographic peaks includes:

[0027] In the extracted ion current chromatogram, the inflection point closest to each maximum point and the inflection point closest to each maximum point are used as adjacent inflection points of each maximum point; and the local signal segment corresponding to the interval between the two adjacent inflection points corresponding to each maximum point is used as the initial chromatographic peak;

[0028] Merge the initial chromatographic peaks that are adjacent in time sequence to determine all merged chromatographic peaks.

[0029] Furthermore, the process of obtaining the sample contamination interference degree includes:

[0030] Determining a corresponding weighted interference degree of separation according to the product of the number of maxima corresponding to each biomarker in the mass-to-charge ratio interval of the mass spectrum and the chromatographic separation completeness;

[0031] Determine the local variation value of each mass-to-charge ratio based on the average of the differences between the ion intensity of each mass-to-charge ratio and the ion intensities of the two adjacent mass-to-charge ratios; determine the corresponding interval intensity deviation value based on the average of the local variation values ​​of all mass-to-charge ratios of each biomarker in the mass-to-charge ratio interval of the mass spectrum;

[0032] Determining a corresponding ionization suppression effect value according to the product of the spectral peak area corresponding to each biomarker in the mass-to-charge ratio interval of the mass spectrum and the intensity deviation value of the interval;

[0033] Determining a reference interference degree for each biomarker according to a ratio between the separation weighted interference degree and the ionization suppression effect value;

[0034] The sample contamination interference degree of each test was determined by performing positive correlation normalization based on the product of the mean of the reference interference degree of all biomarkers and the abundance of the sample components.

[0035] Furthermore, the process of detecting premature rupture of membranes according to the sample contamination interference degree includes:

[0036] When the sample contamination interference degree is greater than a preset interference threshold, the mass spectrum detection process of the biomarker is optimized, and premature rupture of membranes is detected based on the mass spectrum of the biomarker obtained by the optimized detection process.

[0037] Furthermore, the optimized detection process includes:

[0038] During sample pretreatment, solid-phase extraction was used to remove high-abundance interfering proteins, and isotope-labeled internal standards were added. Hydrophilic interaction chromatography was used in the chromatographic conditions, and a volatile buffer with low ionic strength was selected. Dynamic exclusion and multiple reaction monitoring modes were used in mass spectrometry parameter adjustment. The desolvation temperature was increased to 500°C.

[0039] In a second aspect, the present application provides an intelligent detection system for premature rupture of membranes, the system comprising:

[0040] The data acquisition and preprocessing module is used to separate the physiological characteristics of each pregnancy test of the mother and the corresponding sanitary pad secretions to obtain the corresponding mass spectrum, base peak diagram, total ion current diagram and extracted ion current chromatogram of each biomarker;

[0041] A first determination module is used to determine the richness of the sample components of each test based on the physiological characteristic information of pregnancy and the ion intensity deviation and ion intensity change of the base peak graph and the total ion current graph in time sequence;

[0042] The second determination module is used to determine the completeness of the chromatographic separation of each biomarker in each test based on the chromatographic peak morphology distortion and signal intensity distribution of each biomarker in the extracted ion current chromatogram;

[0043] The premature rupture of membranes detection module is used to determine the sample contamination interference degree of each test based on the richness of the sample components, the suppression of the spectral peaks of various biomarkers in the mass spectrum, and the corresponding chromatographic separation completeness; and perform premature rupture of membranes detection based on the sample contamination interference degree.

[0044] In a third aspect, the present application provides a computer device comprising a memory and a processor. The memory is configured to store computer program code, and the processor is configured to call and execute the computer program code from the memory to perform the method of the first aspect or any embodiment of the first aspect of the present application.

[0045] In a fourth aspect, the present application provides a computer program product, comprising a computer program code. When the computer program code is executed, the method of the first aspect or any embodiment of the first aspect of the present application is performed.

[0046] In a fifth aspect, the present application provides a computer-readable storage medium, which stores computer program code. When the computer program code is executed, it performs the method of the first aspect of the present application or any embodiment of the first aspect.

[0047] This application has the following beneficial effects:

[0048] This application first analyzes the change characteristics of the chromatographic data during the detection process based on the ion intensity deviation and ion intensity change in the time series of the base peak diagram and the total ion current diagram, and comprehensively analyzes the sample component richness of the pad secretions detected in this test in combination with the ion intensity change, providing a basis for subsequent sample contamination feature analysis; then the chromatographic data of each biomarker are further analyzed to evaluate the chromatographic separation completeness of the biomarker when interfered by other costs, which can amplify the accuracy of biomarkers with higher chromatographic separation completeness in evaluating sample contamination characteristics and avoid optimizing the detection process only for chromatography or mass spectrometry; finally, by analyzing the mass spectrometry data characteristics of the biomarker, the contamination interference degree of the sample is calculated in combination with the chromatographic separation completeness and the sample component richness, and the sample with higher contamination interference degree is retested after optimizing the detection process, thereby improving the detection accuracy of the biomarker while taking into account the contamination interference. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0050] Figure 1 A flow chart of an intelligent method for detecting premature rupture of membranes provided by one embodiment of the present invention;

[0051] Figure 2 A structural diagram of an intelligent premature rupture of membranes detection system provided by one embodiment of the present invention;

[0052] Figure 3 The present invention provides a schematic diagram of a computer device structure according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the following is a detailed description of the specific implementation method, structure, characteristics and effects of an intelligent detection method for premature rupture of membranes proposed in accordance with the present invention, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment, and the specific features, structures or characteristics in one or more embodiments may be combined in any suitable form. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0054] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0055] The specific scheme of the intelligent detection method for premature rupture of membranes provided by the present invention is described in detail below with reference to the accompanying drawings.

[0056] This application embodiment provides an intelligent detection method for premature rupture of membranes. Figure 1 , which shows a flow chart of an intelligent method for detecting premature rupture of membranes provided by one embodiment of the present invention, the method comprising:

[0057] Step S101: Separate the physiological characteristic information of each pregnancy test of the parturient and the corresponding sanitary pad secretions to obtain the corresponding mass spectrum, base peak map, total ion current map and extracted ion current chromatogram of each biomarker.

[0058] First, the physiological characteristics of pregnancy during each examination of the mother are collected through the medical database; in a specific implementation of an embodiment of the present invention, the physiological characteristics of pregnancy include estrogen and progesterone levels, vaginal secretion pH value, blood sugar and gestational diabetes screening results, which can be adjusted according to the type of data collected by the mother during the physical examination or inspection in the specific implementation environment, and the collected physiological characteristics of pregnancy are all data that the mother has authorized, and will not be further elaborated here.

[0059] Furthermore, considering that the panty pad secretions are easily subject to cross-interference caused by other liquids such as urine, blood and other secretions, which affects the accuracy of the test paper type detection method in detecting a single biomarker; therefore, in order to detect the biomarker more accurately, the present application first uses chromatography-mass spectrometry technology to perform high-precision detection and processing on the secretions extracted from the panty pad; in a specific implementation method of an embodiment of the present invention, a panty pad containing a multi-layer absorbent core is first used to collect the panty pad secretions of the parturient for each test, the panty pad secretions on the panty pad are extracted into a dedicated container through a straw or a test strip sampling tool, and the secretions are dissolved in a buffer solution, i.e., 0.9% physiological salt. Water, filter the sample to remove large particles of impurities such as pad fibers; then remove impurities by centrifugation at 3000 rpm and 5 minutes, and extract the supernatant; then remove impurities by liquid-liquid extraction, and separate the biomarkers in the amniotic fluid; then enzymatically treat the sample to optimize the detection sensitivity. The implementer can also use derivatization treatment instead of enzymatic treatment process; in a specific implementation of an embodiment of the present invention, the biomarkers include: fetal fibronectin, placental α-microglobulin-1, insulin-like growth factor binding protein-1 and soluble intercellular adhesion molecule-1, which can be adjusted according to the specific implementation environment.

[0060] The processed samples were further separated by a liquid chromatography system using a reversed-phase chromatographic column (such as Phenomenex Gemini C18, 100mm×2.1mm, 3μm) at a column temperature of 40 degrees Celsius to improve separation efficiency and reduce injection pressure; the mobile phase configuration was: phase A: 0.1% formic acid + 4mmol / L ammonium acetate aqueous solution (pH3.0) to enhance ionization efficiency; phase B: acetonitrile (containing 0.1% formic acid); a gradient elution method with a flow rate of 0.4mL / min was used to separate the various components in the secretions; the split samples were then ionized by electrospray ionization, and the full scan mode was used to enable the mass spectrometry system to obtain the corresponding mass spectrum (MS), base peak chromatogram (BPC), total ion current chromatogram (TIC) and extracted ion chromatogram (EXC) of each biomarker. chromatogram, EIC graph); wherein the abscissa of the MS graph is the mass-to-charge ratio of the ion, and the ordinate is the ion intensity; the ordinate of the BPC graph is the ion intensity of the strongest ion peak; the ordinate of the TIC graph is the total intensity of all ion signals; the ordinate of the EIC graph is the ion intensity corresponding to the corresponding biomarker; the abscissa of the BPC graph, TIC graph and EIC graph is the retention time of the ion in the chromatogram; it should be noted that the specific acquisition process of the MS graph, BPC graph, TIC graph and EIC graph, the reverse chromatographic column, the gradient elution method and the electrospray ionization method are all technical means well known to those skilled in the art and are not further defined or elaborated here.

[0061] Step S102: Determine the component richness of each sample detected based on the physiological characteristic information of pregnancy and the ion intensity deviation and ion intensity change of the base peak graph and the total ion current graph in time sequence.

[0062] To analyze the interference of other bodily fluid components in the mass spectrometry data of sanitary pad secretions, it is generally necessary to assess the physiological factors of the mother herself, that is, to analyze the physiological characteristics of pregnancy. In addition, during the sampling process of sanitary pad samples, the extracted samples may be obtained from a mixture of multiple secretions or bodily fluids. These bodily fluids contain highly abundant proteins (such as albumin and mucin) and metabolites (such as urea and creatinine). These matrices (matrix refers to components other than the analyte in the sample) may significantly interfere with the analysis of the analytes and affect the accuracy of the analytical results. To analyze the matrix effect interference in this mass spectrometry data, it is first necessary to analyze the richness of the sample components. The greater the richness of the sample components, the more abundant the ion species that can be ionized, and the correspondingly more severe the contamination interference. The ion intensity deviation between the base peak and the total ion current chromatogram can generally reflect the richness of the ion species to a certain extent. Therefore, this application determines the richness of the sample components for each test based on the physiological characteristics of pregnancy and the ion intensity deviation and ion intensity changes in the base peak and total ion current chromatogram over time.

[0063] Preferably, in a specific implementation of the embodiment of the present invention, the process of obtaining the sample component richness includes:

[0064] A deep learning model uses all of the maternal physiological characteristics of pregnancy at each test to determine the corresponding secretion volume index. This process involves inputting all of the maternal physiological characteristics of each test into a trained fully connected neural network model, which then outputs the secretion volume index for each test. Because data influencing other body fluid components is often multimodal, a comprehensive assessment of maternal physiological status using a single analytical method is difficult. Therefore, a neural network approach is used to assess maternal physiological status.

[0065] In a specific implementation of an embodiment of the present invention, the neural network training process is as follows: select the physiological characteristic information of pregnancy that affects the accuracy of secretions collected by maternal pads, that is, the estrogen and progesterone levels, vaginal secretion pH value, blood sugar and gestational diabetes screening results in the embodiment of the present invention, collect a large number of physiological characteristic information sets during pregnancy, and use manual scoring to score each physiological characteristic data of pregnancy in combination with the above characteristics. The scoring values ​​are: 0.1, 0.2, 0.3, ... 1, a total of 10 score types, corresponding to 10 physiological condition levels respectively. The scoring result is the secretion volume index. The higher the secretion volume index, the better the maternal pad is at this time. The more types and amounts of secretions that the pad may extract; the classification network structure is a 5-layer fully connected neural network. In other embodiments, it can be set to a neural network of other structures. This embodiment does not specifically limit it. The loss function is a cross-entropy function; the labeled data set is divided into a training set and a validation set in a ratio of 7:3, and the training set data is input into the classification network for training. The gradient descent method is used for training until the loss function converges and the classification network training is completed; according to the physiological characteristics of pregnancy in the most recent prenatal examination report and other physical examination results of the parturient, the physiological characteristics of pregnancy are input into the trained neural network, and the secretion volume index of the current parturient is output. The secretion volume index of the parturient can reflect the types and amounts of secretions that may be contained in the way of sampling through the pad under her own physiological condition. The larger the value, the more it is necessary to improve the ion separation in the pre-processing process of mass spectrometry detection.

[0066] Further, the corresponding ion richness is determined based on the ion intensity deviation between the total ion current graph and the base peak graph at each time point; in a specific implementation method of an embodiment of the present invention, the process of obtaining the ion richness includes: normalizing the difference between the corresponding ion intensity in the total ion current graph at each time point and the corresponding ion intensity in the base peak graph to determine the ion richness at each time point.

[0067] Chromatographic data can separate different compounds and use signal intensity to reflect the relative abundance of compounds. The BPC diagram can reflect the maximum ion intensity data at different times of the separation process. When the sample is interfered by body fluids such as urine or blood, a large number of small molecule metabolites (urea, creatinine) will be introduced, causing the biomarker signal to be masked by noise, which in turn causes the overall ion intensity at each time point in the BPC diagram to remain low. TIC records the total intensity of all ion signals entering the mass spectrometer at each time point during the chromatographic separation process and their retention time in the chromatogram. During the separation process, the greater the relative intensity of all ion signals generated at the same time point and the lower the intensity of the single ion with the highest signal intensity, the less the single ion that conducts is present in the currently ionized ions. Therefore, it can be said that the ion species ionized at the current moment are rich. Therefore, the greater the overall ion richness at each time point, the greater the richness of the sample components.

[0068] Considering the presence of highly abundant non-target compounds in a sample, their elution can result in sudden increases or decreases in signal, i.e., frequent numerical fluctuations in certain areas of the TIC plot. This phenomenon often occurs when a sample contains multiple highly abundant components, which may elute sequentially during chromatography and produce significant signal changes. Therefore, it is also necessary to assess the compositional richness of the sample in conjunction with the data fluctuation characteristics in the TIC plot. Therefore, the degree of ion fluctuation at each time point is further determined based on the relative deviation of ion intensity changes between each time point and adjacent time points.

[0069] In a specific implementation of an embodiment of the present invention, the process of obtaining ion richness includes: the process of obtaining the degree of ion fluctuation includes: taking the difference between the derivative value of each time point in the total ion current diagram and the derivative value of the previous time point as the first reference difference; taking the difference between the derivative value of each time point in the total ion current diagram and the derivative value of the next time point as the second reference difference; determining the degree of ion fluctuation at each time point based on the mean between the first reference difference and the second reference difference; so that the greater the overall degree of ion fluctuation at each time point, the more significant the signal change reflected, and the greater the corresponding sample component richness.

[0070] The ion richness and ion fluctuation degree are further combined, and the adjustment coefficient of each time point is determined according to the product between the ion richness and the ion fluctuation degree; so that the adjustment coefficient is positively correlated with the sample component richness; the ratio between the adjustment coefficient and the corresponding ion intensity in the total ion current diagram at the corresponding time point is used as the weighting coefficient; because the lower the single ion intensity with the highest signal intensity at each time point in the TIC diagram, the richer the corresponding ion species are, so on the basis of the adjustment coefficient, the corresponding ion intensity in the total ion current diagram at the corresponding time point is combined for ratio processing, so that the obtained weighting coefficient more accurately characterizes the sample component at each time point;

[0071] The secretion volume index reflects the types and quantities of secretions that may be contained through panty liner sampling under one's own physiological conditions. Therefore, based on the overall size of the weighted coefficients at all time points, the secretion volume index is combined to comprehensively characterize the sample component richness, which characterizes the number of secretion component types. The embodiment of the present invention determines the sample component richness of each test based on the product of the mean of the weighted coefficients at all time points and the secretion volume index.

[0072] In a specific implementation of the embodiment of the present invention, the process of obtaining the sample component richness is expressed by the formula: ;in, For the The richness of sample components detected; For the Secretion volume index of the second test; For the The number of time points that correspond to each other in the total ion current graph and base peak graph of each detection; For the The total ion current graph and base peak graph of the detection correspond to the The corresponding ion intensity in the total ion current at each time point; For the The total ion current graph and base peak graph of the detection correspond to the The corresponding ion intensity in the base peak graph at each time point; It is a linear normalization function. The normalization method can be adjusted according to the specific implementation environment, such as the sigmoid function. For the The total ion current graph and base peak graph of the detection correspond to the The degree of ion fluctuation at each time point; For the The total ion current graph and base peak graph of the detection correspond to the Ion abundance at each time point; For the The total ion current graph and base peak graph of the detection correspond to the The adjustment coefficient at each time point; For the The total ion current graph and base peak graph of the detection correspond to the The weighting coefficient of each time point.

[0073] In a specific implementation of the embodiment of the present invention, the process of obtaining the ion fluctuation degree is expressed by the formula: ;in, For the The total ion current graph and base peak graph of the detection correspond to the The first reference difference at each time point; For the The total ion current graph and base peak graph of the detection correspond to the The second reference difference at each time point; it should be noted that when the corresponding time point is the first time point, the corresponding second reference difference is used as the degree of ion fluctuation; when the corresponding time point is the last time point, the corresponding first reference difference is used as the degree of ion fluctuation.

[0074] It should be noted that the time interval between two adjacent time points in the embodiment of the present invention is set to 1 minute, which can be adjusted according to the specific implementation environment. In order to ensure that the calculation results are meaningful, when performing fractional operations in the embodiment of the present invention, when the denominator is 0, it is necessary to add a parameter adjustment factor greater than 0 to the denominator to prevent the denominator from being 0. The value of the parameter adjustment factor is set by the implementer according to the actual situation, and this application does not impose any special restrictions.

[0075] Step S103: determining the completeness of the chromatographic separation of each biomarker in each detection based on the chromatographic peak morphology distortion and signal intensity distribution of each biomarker in the extracted ion current chromatogram.

[0076] In addition to analyzing the richness of sample components, it is also necessary to analyze the impact of interfering components such as other body fluids on the ionization of biomarkers during the actual ionization process. Therefore, it is necessary to analyze the interfering components and various biomarkers in combination with the characteristics of chromatographic data; among the interfering components that may be contained in maternal pads, urine contains a large amount of small-molecule organic acids and inorganic salts, which may affect the stability of the stationary phase of the chromatographic column (for example, producing high concentrations of sodium ions (about 130-260 mmol / L in urine) to compete with ion pair reagents). In the chromatographic data, this feature is reflected in the abnormal chromatographic peak morphology of the biomarker in the EIC diagram, with peak widening or reduced peak height, and the appearance of target peak retention time drift characteristics. Among them, the EIC diagram is the chromatogram of the parent ion of a specific mass-to-charge ratio, that is, the extracted ion current chromatogram corresponding to each biomarker.

[0077] In addition to the interference of salt in urine, phospholipids in blood may also interfere with the ionization process of biomarkers. When phospholipids co-elute with biomarkers, they may cause peak distortion, such as shoulder peaks (e.g., splitting of the PAMG-1 peak into two peaks). Therefore, the chromatographic separation of biomarkers can be further evaluated by their chromatographic peak characteristics.

[0078] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the completeness of chromatographic separation includes:

[0079] Based on the distribution of maximum points and inflection points in the extracted ion current chromatogram, all merged chromatographic peaks are determined. The process of obtaining merged chromatographic peaks includes: in the extracted ion current chromatogram, the inflection point closest to and closest to each maximum point are used as the adjacent inflection points of each maximum point; the local signal segment corresponding to the interval between the two adjacent inflection points corresponding to each maximum point is used as the initial chromatographic peak; each initial chromatographic peak adjacent in time sequence is merged to determine all merged chromatographic peaks. By merging chromatographic peaks, it is avoided that the shoulder peak is divided into two peaks when selecting the chromatographic peak, which affects the interference analysis of phospholipids in the blood on the ionization process;

[0080] The drift distortion characteristic parameter corresponding to each merged chromatographic peak is determined by multiplying the time length of the interval corresponding to each merged chromatographic peak by the absolute value of the skewness. The local separation characteristic value of each merged chromatographic peak is determined by the ratio between the maximum signal in each merged chromatographic peak and the drift distortion characteristic parameter. The chromatographic separation completeness of each biomarker in each test is determined by taking the average of the local separation characteristic values ​​of all merged chromatographic peaks. The time length of the interval corresponding to the merged chromatographic peak, i.e., the peak width, can reflect the retention time drift characteristics of the biomarker chromatographic peak. The larger the corresponding peak width, the more pronounced the retention time drift characteristics.

[0081] In addition, the absolute value of skewness represents the peak shape distortion characteristics of the corresponding biomarker merged chromatographic peak. The larger the absolute value of the skewness, the more obvious the peak shape distortion characteristics of the corresponding merged chromatographic peak. Therefore, the larger the drift distortion characteristic parameter (the product of the time length of the corresponding interval of each merged chromatographic peak and the absolute value of the skewness), the greater the corresponding interference and the lower the completeness of the chromatographic separation. The maximum signal value in the merged chromatographic peak of the biomarker can reflect the ionization efficiency of the corresponding biomarker. Low peak height is usually caused by matrix suppression. Therefore, the larger the overall maximum signal value in each merged chromatographic peak, the more complete the corresponding chromatographic separation and the higher the completeness of the chromatographic separation. In addition, the presence of multiple peaks in the biomarker EIC plot indicates that the target compound and matrix interferences (such as metabolites and salts) co-elute due to similar retention times, or that different adsorption-desorption kinetics on the chromatographic column lead to peak splitting. Therefore, further adjustment of the weighted mean calculation results can make the characterization of chromatographic separation completeness more accurate.

[0082] In a specific implementation of the embodiment of the present invention, the process of obtaining the completeness of chromatographic separation is expressed by the formula: ;in, For the The first test Completeness of chromatographic separation of the biomarkers; For the The first test The number of merged chromatographic peaks in the extracted ion chromatograms of the biomarkers; For the The first test Extracted ion chromatogram of the biomarker The maximum value of the signal of the merged chromatographic peak; For the The first test Extracted ion chromatogram of the biomarker The time length of each merged chromatographic peak; For the The first test Extracted ion chromatogram of the biomarker The absolute value of the skewness of the combined chromatographic peaks; it should be noted that the calculation of skewness is a technical means well known to those skilled in the art and will not be further limited or elaborated here; For the The first test Extracted ion chromatogram of the biomarker Drift distortion characteristic parameters of the merged chromatographic peaks; For the The first test Extracted ion chromatogram of the biomarker The local separation characteristic values ​​of the merged chromatographic peaks.

[0083] Step S104: Determine the sample contamination interference degree for each test based on the sample component richness, the suppression of the spectral peaks of various biomarkers in the mass spectrum, and the corresponding chromatographic separation completeness; perform premature rupture of membranes test based on the sample contamination interference degree.

[0084] Premature rupture of membranes (PROM) has multiple biomarkers. In addition to analyzing the biomarker separation characteristics in chromatographic data, it is also necessary to analyze the actual biomarker information in the mass spectrometry data and assess the degree of contamination interference in the current monitoring process. For different biomarkers, the more complete the chromatographic separation, the more consistent the representation in the mass spectrometry data should be with the actual sample performance. However, even with complete chromatographic separation of biomarkers, high-concentration components in the matrix (such as urea and creatinine in urine, or phospholipids in blood) can still inhibit or enhance the ionization efficiency of the biomarker by competing for charge or altering the droplet surface tension. For example, in electrospray ionization, phospholipids or high-concentration salts can reduce the ionization efficiency of the biomarker by coating the ion source surface, resulting in signal attenuation. In the mass spectrometry data, this is reflected as a significant decrease in the peak area corresponding to the biomarker and a weaker main peak signal. Furthermore, samples with higher component richness may also experience a greater suppression of biomarker ionization efficiency. Therefore, when the overall completeness of the chromatographic separation of various biomarkers is greater, the sample component richness is greater, and the peak suppression in the mass spectrum is more significant, indicating that the sample is subject to stronger interference during this detection process and the corresponding contamination interference is greater.

[0085] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the sample contamination interference degree includes:

[0086] The corresponding separation weighted interference degree is determined based on the product of the number of maxima corresponding to each biomarker in the mass-to-charge ratio interval of the mass spectrum and the chromatographic separation completeness. First, the chromatographic separation completeness can characterize the credibility of the mass spectrometry data, so it is equivalent to a weight. Specifically, when the chromatographic separation is complete, if contamination interference is still detected, such as ion suppression or overlapping peaks, it means that the contamination interference is more hidden and the impact is more serious. At this time, the influence of the interference is amplified by the chromatographic separation completeness. The number of mass spectrum peak maxima can reflect the peak splitting or overlapping interference of the biomarker in the mass spectrum. The larger the corresponding mass spectrum peak maximum, the more interfering peaks of other interfering substances in the mass-to-charge ratio interval corresponding to the biomarker in the mass spectrum, and the more significant the contamination interference. Therefore, after weighting by the chromatographic separation completeness, the separation weighted interference degree can more accurately characterize the contamination interference suffered by each biomarker.

[0087] The local variation value of each mass-to-charge ratio is determined based on the average of the differences between the ion intensity of each mass-to-charge ratio and the ion intensities of the two adjacent mass-to-charge ratios. The corresponding interval intensity deviation value is determined based on the average of the local variation values ​​of all mass-to-charge ratios in the mass-to-charge ratio interval of the mass spectrum of each biomarker. The corresponding ionization suppression effect value is determined based on the product of the corresponding spectral peak area in the mass-to-charge ratio interval of the mass spectrum of each biomarker and the interval intensity deviation value.

[0088] The interval intensity deviation value measures the fluctuation of ion intensity within the mass spectrum peak. The smaller the overall local change value in the corresponding biomarker mass-to-charge ratio interval, the smoother the peak shape, and the biomarker signal may be uniformly suppressed or covered due to matrix interference, that is, the smaller the interval intensity deviation value, the more serious the corresponding contamination interference; the peak area reflects the degree of suppression of the biomarker ionization efficiency. The smaller the corresponding peak area, the higher the degree of suppression of the ionization efficiency and the more serious the contamination interference; therefore, the smaller the ionization suppression effect value obtained by multiplying the peak area and the interval intensity deviation value, the more serious the corresponding contamination interference.

[0089] Combined with the corresponding correlation, the reference interference of each biomarker is determined according to the ratio between the separation weighted interference and the ionization suppression effect value; the larger the reference interference, the more serious the interference to the corresponding biomarker; considering that the richness of sample components characterizes the degree of interference from the dimension of the number of sample components, the product of the mean of the reference interference of all biomarkers and the sample component richness is finally used for positive correlation normalization to determine a more accurate sample contamination interference for each test.

[0090] In a specific implementation of the embodiment of the present invention, the process of obtaining the sample contamination interference degree is expressed by the formula: ;in, For the The degree of sample contamination interference during the test; For the The richness of sample components detected; For the The first test Completeness of chromatographic separation of the biomarkers; For the The first test The number of maxima corresponding to a biomarker in the mass-to-charge ratio interval of the mass spectrum; For the The first test The weighted interference of the separation of the biomarkers; For the The first test The peak area corresponding to the biomarker in the mass-to-charge ratio interval of the mass spectrum; For the The first test The interval intensity deviation value corresponding to the biomarker in the mass-to-charge ratio interval of the mass spectrum; For the The first test Ionization suppression effect value of the biomarker; For the The first test Reference interference of the biomarker; is a hyperbolic tangent function, and the normalization method, such as linear normalization, can be adjusted according to the specific implementation environment.

[0091] In a specific implementation of the embodiment of the present invention, the process of obtaining the interval intensity deviation value includes: ;in, For the The first test The interval intensity deviation value corresponding to the biomarker in the mass-to-charge ratio interval of the mass spectrum; For the The first test The number of mass-to-charge ratios corresponding to the biomarker in the mass-to-charge ratio interval of the mass spectrum; For the The first test The first biomarker in the mass-to-charge ratio interval of the mass spectrum The absolute value of the difference between the ion intensity of a mass-to-charge ratio and the ion intensity of the previous mass-to-charge ratio; For the The first test The first biomarker in the mass-to-charge ratio interval of the mass spectrum The absolute value of the difference between the ion intensity of one mass-to-charge ratio and the ion intensity of the next mass-to-charge ratio; For the The first test The first biomarker in the mass-to-charge ratio interval of the mass spectrum The local change value of the mass-to-charge ratio; when the mass-to-charge ratio does not exist before, the corresponding As the local change value; when there is no mass-to-charge ratio after the mass-to-charge ratio, the corresponding As a local change value; it should be noted that the acquisition of the mass-to-charge ratio interval corresponding to each biomarker in the mass spectrum is a technical means well known to those skilled in the art and will not be further elaborated here.

[0092] Finally, the pollution interference degree of each detection is calculated, and finally, premature rupture of membranes detection is performed according to the pollution interference degree. Preferably, in some possible implementations of the embodiments of the present invention, the process of performing premature rupture of membranes detection according to the sample pollution interference degree includes:

[0093] When the sample contamination interference is greater than the preset interference threshold, the mass spectrometry detection process of the biomarker is optimized, and premature rupture of membranes is detected based on the mass spectrometry of the biomarker obtained by the optimized detection process; wherein, the process of the optimized detection process includes: in the sample pretreatment process, solid phase extraction is used to remove high-abundance interfering proteins, and isotope-labeled internal standards are added to correct extreme value effects; hydrophilic interaction chromatography is used in the chromatographic conditions, and the gradient elution time can also be extended to improve the separation of polar metabolites, and a volatile buffer with low ionic strength is selected to reduce salt interference; dynamic exclusion and multiple reaction monitoring modes are used in mass spectrometry parameter adjustment to reduce non-target ion interference; the desolvation temperature is increased to 500°C to enhance the dissociation specificity of the target. In a specific implementation of an embodiment of the present invention, the preset interference threshold is set to 0.75, which can be adjusted according to the specific implementation environment. The greater the sample contamination interference, the more serious the sample contamination, and therefore the more necessary it is to optimize the mass spectrometry detection process; after obtaining the mass spectrum of the biomarker obtained by the optimized detection process, the premature rupture of membranes in the parturient is analyzed by the biomarkers in the corresponding mass spectrometry data. It should be noted that the process of analyzing the premature rupture of membranes in the parturient by the biomarkers in the corresponding mass spectrometry data is a technical means well known to those skilled in the art and will not be further elaborated here.

[0094] In summary, an intelligent detection method for premature rupture of membranes first analyzes the change characteristics of chromatographic data during the detection process based on the ion intensity deviation and ion intensity change of the base peak graph and total ion current graph in time series, and comprehensively analyzes the sample component richness of the pad secretions detected in this test in combination with the ion intensity change, providing a basis for subsequent sample contamination feature analysis; then the chromatographic data of each biomarker are further analyzed to evaluate the chromatographic separation completeness of the biomarker when interfered by other costs, which can amplify the accuracy of biomarkers with higher chromatographic separation completeness in evaluating sample contamination characteristics and avoid optimizing the detection process only for chromatography or mass spectrometry; finally, by analyzing the mass spectrometry data characteristics of the biomarker, the contamination interference degree of the sample is calculated in combination with the chromatographic separation completeness and the sample component richness, and the sample with higher contamination interference degree is retested after optimizing the detection process, thereby improving the detection accuracy of the biomarker while taking into account the contamination interference.

[0095] This application also provides an intelligent detection system for premature rupture of membranes. Figure 2 , which shows a structural diagram of an intelligent premature rupture of membranes detection system provided by an embodiment of the present invention. The system includes: a data acquisition and preprocessing module 201, a first determination module 202, a second determination module 203 and a premature rupture of membranes detection module 204.

[0096] The data acquisition and preprocessing module 201 is used to separate the physiological characteristics of each pregnancy test of the mother and the corresponding sanitary pad secretions to obtain the corresponding mass spectrum, base peak diagram, total ion current diagram and extracted ion current chromatogram of each biomarker;

[0097] The first determination module 202 is used to determine the richness of the sample components in each test based on the physiological characteristics of pregnancy and the ion intensity deviation and ion intensity change in the base peak graph and the total ion current graph in time sequence;

[0098] The second determination module 203 is configured to determine the completeness of the chromatographic separation of each biomarker in each test based on the chromatographic peak morphology distortion and signal intensity distribution of each biomarker in the extracted ion current chromatogram;

[0099] The premature rupture of membranes detection module 204 is used to determine the sample contamination interference degree of each test based on the sample component richness, the suppression of the spectral peaks of various biomarkers in the mass spectrum, and the corresponding chromatographic separation completeness; and perform premature rupture of membranes detection based on the sample contamination interference degree.

[0100] It should be noted that the system provided in the above embodiment is merely illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the intelligent detection system for premature rupture of membranes provided in the above embodiment and the intelligent detection method for premature rupture of membranes provided in the embodiment are based on the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0101] The present application also provides a computer device. Figure 3 , which shows a schematic diagram of the structure of a computer device provided by an embodiment of the present invention, the computer device includes a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302, wherein when the processor 302 executes the computer program 303, the computer device can execute any one of the intelligent detection methods for premature rupture of membranes introduced above.

[0102] An embodiment of the present application also provides a computer program product. When the computer program product is run on a computer device, the computer device can execute any one of the intelligent detection methods for premature rupture of membranes described above.

[0103] An embodiment of the present application also provides a computer-readable storage medium, which stores computer program code. When the computer program code runs on a computer device, the computer device can execute any of the intelligent detection methods for premature rupture of membranes introduced above.

[0104] In the embodiments provided in the present application, it should be understood that the provided computer devices, computer program products and computer-readable storage media are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the methods provided above and will not be repeated here.

[0105] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0106] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. An intelligent detection method for premature rupture of membranes, characterized in that: The method comprises: The physiological characteristics of each pregnancy test of the mother and the corresponding pad secretions are separated into components to obtain the corresponding mass spectra, base peak maps, total ion current maps, and extracted ion current chromatograms of each biomarker; Determine the richness of the sample components for each test based on the physiological characteristic information during pregnancy and the ion intensity deviation and ion intensity change of the base peak graph and the total ion current graph in time sequence; The completeness of the chromatographic separation of each biomarker in each test was determined based on the peak morphology distortion and signal intensity distribution in the extracted ion current chromatogram of each biomarker; Determine the sample contamination interference degree for each test based on the sample component richness, the suppression of spectral peaks of various biomarkers in the mass spectrum, and the corresponding chromatographic separation completeness; and perform premature rupture of membranes detection based on the sample contamination interference degree; The process of obtaining the sample component richness includes: Using a deep learning model, the corresponding secretion volume index is determined based on all the physiological characteristics of the mother during pregnancy during each test. Determining the corresponding ion abundance according to the ion intensity deviation between the total ion current graph and the base peak graph at each time point; The degree of ion fluctuation at each time point is determined based on the relative deviation of ion intensity changes between each time point and adjacent time points; Determine an adjustment coefficient at each time point based on the product of the ion abundance and the ion fluctuation degree; and use the ratio of the adjustment coefficient to the corresponding ion intensity in the total ion current map at the corresponding time point as a weighting coefficient; Determine the component richness of each sample tested according to the product between the mean of the weighted coefficients of all time points and the secretion volume index; The process of obtaining the sample contamination interference degree includes: Determining a corresponding weighted interference degree of separation according to the product of the number of maxima corresponding to each biomarker in the mass-to-charge ratio interval of the mass spectrum and the chromatographic separation completeness; Determine the local variation value of each mass-to-charge ratio based on the average of the differences between the ion intensity of each mass-to-charge ratio and the ion intensities of the two adjacent mass-to-charge ratios; determine the corresponding interval intensity deviation value based on the average of the local variation values ​​of all mass-to-charge ratios of each biomarker in the mass-to-charge ratio interval of the mass spectrum; Determining a corresponding ionization suppression effect value according to the product of the peak area corresponding to each biomarker in the mass-to-charge ratio interval of the mass spectrum and the intensity deviation value of the interval; Determining a reference interference degree for each biomarker based on a ratio between the separation weighted interference degree and the ionization suppression effect value; The sample contamination interference degree of each test was determined by performing positive correlation normalization based on the product of the mean of the reference interference degree of all biomarkers and the abundance of the sample components.

2. The intelligent detection method for premature rupture of membranes according to claim 1, characterized in that: The process of obtaining the secretion volume index includes: All the physiological characteristics of the mother during pregnancy during each test are input into the trained fully connected neural network model, and the secretion volume index of each test is output.

3. The intelligent detection method for premature rupture of membranes according to claim 1, characterized in that: The process of obtaining the ion richness includes: The difference between the corresponding ion intensity in the total ion current map and the corresponding ion intensity in the base peak map at each time point is normalized to determine the ion richness at each time point.

4. The intelligent detection method for premature rupture of membranes according to claim 1, characterized in that: The process of obtaining the ion fluctuation degree includes: The difference between the derivative value at each time point in the total ion current graph and the derivative value at the previous time point is used as the first reference difference; the difference between the derivative value at each time point in the total ion current graph and the derivative value at the next time point is used as the second reference difference; and the degree of ion fluctuation at each time point is determined based on the mean between the first reference difference and the second reference difference.

5. The intelligent detection method for premature rupture of membranes according to claim 1, characterized in that: The process of obtaining the completeness of the chromatographic separation comprises: Determine all merged chromatographic peaks based on the distribution of maximum points and inflection points in the extracted ion current chromatogram; Determine the corresponding drift distortion characteristic parameter according to the product between the time length of the corresponding interval of each merged chromatographic peak and the absolute value of the skewness; The local separation characteristic value of each merged chromatographic peak is determined based on the ratio between the signal maximum value in each merged chromatographic peak and the drift distortion characteristic parameter; the chromatographic separation completeness of each biomarker in each detection is determined based on the average of the local separation characteristic values ​​of all merged chromatographic peaks.

6. The intelligent detection method for premature rupture of membranes according to claim 5, characterized in that: The acquisition process of the merged chromatographic peaks includes: In the extracted ion current chromatogram, the inflection point closest to each maximum point and the inflection point closest to each maximum point are used as adjacent inflection points of each maximum point; and the local signal segment corresponding to the interval between the two adjacent inflection points corresponding to each maximum point is used as the initial chromatographic peak; Merge the initial chromatographic peaks that are adjacent in time sequence to determine all merged chromatographic peaks.

7. The intelligent detection method for premature rupture of membranes according to claim 1, characterized in that: The process of detecting premature rupture of membranes according to the sample contamination interference degree includes: When the sample contamination interference degree is greater than a preset interference threshold, the mass spectrum detection process of the biomarker is optimized, and premature rupture of membranes is detected based on the mass spectrum of the biomarker obtained by the optimized detection process.

8. The intelligent detection method for premature rupture of membranes according to claim 7, characterized in that: The optimized detection process includes: During sample pretreatment, solid-phase extraction was used to remove high-abundance interfering proteins, and isotope-labeled internal standards were added. Hydrophilic interaction chromatography was used in the chromatographic conditions, and a volatile buffer with low ionic strength was selected. Dynamic exclusion and multiple reaction monitoring modes were used in mass spectrometry parameter adjustment. The desolvation temperature was increased to 500°C.

Images