A deep learning-based heparin sodium detection method and system

Abstract

The application provides a heparin sodium detection method and system based on deep learning, which is suitable for the technical field of data processing. The method comprises the following steps: extracting features of heparin sodium detection spectrum signal information and heparin sodium detection electrochemical signal information to obtain heparin sodium detection spectrum signal feature information and heparin sodium detection electrochemical signal feature information; generating heparin sodium detection signal optimization feature information according to the heparin sodium detection spectrum signal feature information, the heparin sodium detection electrochemical signal feature information, a heparin sodium detection signal feature optimization iteration number and a heparin sodium detection signal feature optimization amplitude information; and performing classification processing on the heparin sodium detection signal optimization feature information to obtain heparin sodium detection information. The application is used for fully optimizing heparin sodium detection signals through the method of deep learning, so as to improve the accuracy, robustness and reliability of heparin sodium detection.

Description

Technical Field

[0001] This application belongs to the field of data processing technology, and in particular relates to a method and system for detecting heparin sodium based on deep learning. Background Technology

[0002] Heparin sodium, as an important anticoagulant drug in clinical practice, is widely used in thrombosis prevention, cardiovascular surgery, and other fields. The development of its detection technology directly affects drug quality control and clinical medication safety. Currently, heparin sodium detection technologies mainly include traditional methods such as high-performance liquid chromatography, spectrophotometry, and electrochemical analysis.

[0003] Traditional methods rely on expensive equipment and complex sample pretreatment, resulting in problems such as long detection cycles, high costs, and low sensitivity. Furthermore, the detection capabilities of instruments and equipment are insufficient for low-concentration heparin sodium signals, leading to limited detection sensitivity and making it difficult to meet the detection needs of trace samples, thus limiting the improvement of detection accuracy for heparin sodium. Summary of the Invention

[0004] In view of this, embodiments of this application provide a deep learning-based method and system for detecting heparin sodium, aiming to solve the problems of high cost, low accuracy, low effectiveness, and insufficient stability in the detection of low concentration heparin sodium in the prior art.

[0005] The first aspect of this application provides a deep learning-based method for detecting heparin sodium, comprising:

[0006] Acquire spectral signal information and electrochemical signal information for heparin sodium detection;

[0007] Feature extraction is performed on the heparin sodium detection spectral signal information and the heparin sodium detection electrochemical signal information to obtain heparin sodium detection spectral signal feature information and heparin sodium detection electrochemical signal feature information;

[0008] Based on the heparin sodium detection spectral signal feature information, the heparin sodium detection electrochemical signal feature information, the preset number of heparin sodium detection signal feature optimization iterations, and the preset heparin sodium detection signal feature optimization amplitude information, multiple heparin sodium detection signal optimization feature information are generated.

[0009] Based on the preset heparin sodium detection signal feature classification convergence threshold information, the optimized feature information of multiple heparin sodium detection signals is classified and processed to obtain heparin sodium detection information.

[0010] A second aspect of this application provides a deep learning-based heparin sodium detection system, comprising:

[0011] The heparin sodium detection signal information acquisition module is used to acquire heparin sodium detection spectral signal information and heparin sodium detection electrochemical signal information;

[0012] The heparin sodium detection signal feature information generation module is used to extract features from the heparin sodium detection spectral signal information and the heparin sodium detection electrochemical signal information to obtain heparin sodium detection spectral signal feature information and heparin sodium detection electrochemical signal feature information.

[0013] The heparin sodium detection signal optimization feature information generation module is used to generate multiple heparin sodium detection signal optimization feature information based on the heparin sodium detection spectral signal feature information, the heparin sodium detection electrochemical signal feature information, the preset heparin sodium detection signal feature optimization iteration number, and the preset heparin sodium detection signal feature optimization amplitude information.

[0014] The heparin sodium detection information generation module is used to classify and process multiple optimized feature information of heparin sodium detection signals based on preset heparin sodium detection signal feature classification convergence threshold information to obtain heparin sodium detection information.

[0015] A third aspect of this application provides a terminal device, the terminal device including a memory and a processor, the memory storing a computer program executable on the processor, the processor executing the computer program to implement the steps of the deep learning-based heparin sodium detection method described in the first aspect above.

[0016] A fourth aspect of this application provides a computer-readable storage medium, comprising: storing a computer program, wherein when executed by a processor, the computer program implements the steps of the deep learning-based heparin sodium detection method described in the first aspect above.

[0017] The beneficial effects of this application embodiment compared with the prior art are as follows: This application achieves multimodal complementarity of heparin sodium detection information by acquiring heparin sodium detection spectral signal information and heparin sodium detection electrochemical signal information, effectively overcoming the defects of single signal being susceptible to interference and incomplete detection information. By extracting features from heparin sodium detection spectral signal information and heparin sodium detection electrochemical signal information, it accurately captures subtle features in spectral and electrochemical signals, significantly improving feature characterization capabilities. By using preset heparin sodium detection signal feature optimization iteration times and preset heparin sodium detection signal feature optimization amplitude information, it deeply optimizes the extracted heparin sodium detection spectral signal feature information and heparin sodium detection electrochemical signal feature information, improving the adaptability to the needs of complex heparin sodium samples and variable heparin sodium detection environments, eliminating interference from various complex heparin sodium sample conditions and dynamically changing heparin sodium detection environments, and improving the accuracy, robustness, and effectiveness of heparin sodium detection information. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the implementation process of the deep learning-based heparin sodium detection method provided in Embodiment 1 of this application;

[0020] Figure 2 This is a schematic diagram of the implementation process of the deep learning-based heparin sodium detection method provided in Embodiment 2 of this application;

[0021] Figure 3 This is a schematic diagram of the implementation process of the deep learning-based heparin sodium detection method provided in Embodiment 3 of this application;

[0022] Figure 4 This is a schematic diagram of the implementation process of the deep learning-based heparin sodium detection method provided in Embodiment 4 of this application;

[0023] Figure 5 This is a schematic diagram of the implementation process of the deep learning-based heparin sodium detection method provided in Embodiment 5 of this application;

[0024] Figure 6 This is a schematic diagram of the implementation process of the deep learning-based heparin sodium detection method provided in Embodiment Six of this application;

[0025] Figure 7 This is a schematic diagram of the implementation process of the deep learning-based heparin sodium detection method provided in Embodiment 7 of this application;

[0026] Figure 8 This is a schematic diagram of the structure of the deep learning-based heparin sodium detection system provided in the embodiments of this application;

[0027] Figure 9 This is a schematic diagram of the terminal device provided in the embodiments of this application. Detailed Implementation

[0028] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0029] To illustrate the technical solution described in this application, specific embodiments are provided below.

[0030] Figure 1 The implementation flowchart of the deep learning-based heparin sodium detection method provided in Embodiment 1 of this application is shown, and is described in detail below:

[0031] Step S101: Obtain the heparin sodium detection spectral signal information and the heparin sodium detection electrochemical signal information.

[0032] In this embodiment, the heparin sodium detection spectral signal information refers to optical signal data related to the molecular structure or physicochemical properties of heparin sodium obtained through spectral analysis techniques. This can include ultraviolet-visible absorption spectral signals, infrared spectral signals, and fluorescence spectral signals. Specifically, the ultraviolet-visible absorption spectral signal refers to the absorption characteristics of heparin sodium molecules in the ultraviolet-visible light band (typically 200-800 nm), with different concentrations or purities of heparin sodium exhibiting specific absorption peaks. The infrared spectral signal refers to the vibrational absorption signal of functional groups in the heparin sodium molecule under infrared light irradiation, which can be used to reflect characteristic chemical bond information of the molecular structure. The fluorescence spectral signal refers to the intensity and peak position of the emission spectrum of heparin sodium or its derivatives when excited at a specific wavelength, which can be used as a detection signal. The heparin sodium detection electrochemical signal information refers to electrical signal data related to the electrochemical reaction of heparin sodium obtained through electrochemical analysis methods. This electrochemical signal information can include potential signals, current signals, and impedance signals. Among these, the potential signal refers to the change in electrode potential caused by ion exchange and coordination reactions between heparin sodium and the electrode surface, such as the sudden change in endpoint potential in potentiometric titration; the current signal can refer to the electrolytic current generated when heparin sodium undergoes a redox reaction on the electrode surface, such as the current versus voltage curve in voltammetry; the impedance signal can refer to the changes in resistance and capacitance caused by the interaction between heparin sodium molecules and the electrode interface in electrochemical impedance spectroscopy (EIS), which can be used to reflect molecular adsorption or reaction kinetics. Spectral signal information for heparin sodium detection can be obtained through ultraviolet-visible spectrophotometry, infrared spectroscopy, and fluorescence spectroscopy. Electrochemical signal information for heparin sodium detection can be obtained through potentiometric titration, voltammetry, and electrochemical impedance spectroscopy.

[0033] Step S102: Feature extraction is performed on the heparin sodium detection spectral signal information and the heparin sodium detection electrochemical signal information to obtain heparin sodium detection spectral signal feature information and heparin sodium detection electrochemical signal feature information.

[0034] In this embodiment, for the heparin sodium detection spectral signal information, the spectral signal can be arranged into a one-dimensional sequence according to wavelength or wavenumber. A sliding window is used to scan the sequence for local features. The data within each window is weighted and processed by an activation function to extract local features such as peak position, peak intensity, and peak width. As the layer deepens, the shallowly extracted local features are combined and abstracted to gradually form more representative high-level features, thereby obtaining the heparin sodium detection spectral signal feature information. For the heparin sodium detection electrochemical signal information, the importance weights of different time points or potential ranges in the signal can be calculated first. By summarizing and statistically analyzing the global information of the signal, the correlation between each part of the data and the characteristic reactions of heparin sodium is analyzed. Appropriate weights are assigned to different segments of the signal, highlighting key signal segments that reflect the redox reaction and ion exchange characteristics of heparin sodium, while suppressing noise and irrelevant signals, thereby extracting the heparin sodium detection electrochemical signal feature information.

[0035] Step S103: Based on the heparin sodium detection spectral signal feature information, the heparin sodium detection electrochemical signal feature information, the preset number of heparin sodium detection signal feature optimization iterations, and the preset heparin sodium detection signal feature optimization amplitude information, multiple heparin sodium detection signal optimization feature information are generated.

[0036] In this embodiment, both the preset number of iterations for optimizing the heparin sodium detection signal features and the preset amplitude information for optimizing the heparin sodium detection signal features can be set manually. The preset number of iterations for optimizing the heparin sodium detection signal features can be used as the loop termination condition. In each iteration, the current heparin sodium detection spectral signal features and heparin sodium detection electrochemical signal features are adjusted according to the preset amplitude information for optimizing the heparin sodium detection signal features. Specifically, an initial feature space can be constructed based on the spectral and electrochemical signal characteristics of heparin sodium detection. The step size is adjusted using a preset parameter of optimized amplitude information for heparin sodium detection signal characteristics. The weights of the feature information are randomly increased or decreased, or the combination methods are rearranged. A search is performed within the feature space to attempt to generate new combinations of heparin sodium detection signal features. Then, based on preset evaluation criteria, such as calculating the interclass distance and information gain of the new feature combination for heparin sodium and interfering signals, the ability of the new feature combination to distinguish heparin sodium from interfering substances is quantitatively evaluated. If the new feature combination outperforms the original combination in key indicators such as discrimination, the feature information is updated accordingly; otherwise, the original combination is retained. This iterative update process terminates with a preset number of iterations for optimizing heparin sodium detection signal features. Through multiple iterations, several optimized heparin sodium detection signal features that perform better in distinguishing heparin sodium characteristics are gradually selected, effectively improving the accuracy and effectiveness of the features for heparin sodium detection.

[0037] Step S104: Based on the preset heparin sodium detection signal feature classification convergence threshold information, classify and process the optimized feature information of multiple heparin sodium detection signals to obtain heparin sodium detection information.

[0038] In this embodiment, the similarity or distance metric between multiple optimized features of heparin sodium detection signals can be calculated first to determine the degree of correlation between them. Features with high similarity are grouped into one category. During the classification process, the degree of difference between features within a category and the degree of separation between categories are continuously monitored. When the difference within a category is less than the preset heparin sodium detection signal feature classification convergence threshold and the difference between categories reaches the distinguishable standard, the classification result is considered to have converged stably, and the classification adjustment is stopped. Based on the classification result, the concentration, purity, and other attributes of heparin sodium in the sample are determined, and the heparin sodium detection information is output, completing the entire detection and analysis process.

[0039] The deep learning-based heparin sodium detection method provided in this application acquires heparin sodium detection spectral signal information and heparin sodium detection electrochemical signal information, achieving multimodal complementarity of heparin sodium detection information. This effectively overcomes the shortcomings of single signals being susceptible to interference and incomplete detection information. By extracting features from the heparin sodium detection spectral signal information and heparin sodium detection electrochemical signal information, subtle features in the spectral and electrochemical signals are accurately captured, significantly improving feature representation capabilities. Through preset heparin sodium detection signal feature optimization iterations and preset heparin sodium detection signal feature optimization amplitude information, the extracted heparin sodium detection spectral signal feature information and heparin sodium detection electrochemical signal feature information are deeply optimized, improving the adaptability to the needs of complex heparin sodium samples and variable heparin sodium detection environments. This eliminates interference from various complex heparin sodium sample conditions and dynamically changing heparin sodium detection environments, improving the accuracy, robustness, and effectiveness of heparin sodium detection information.

[0040] Figure 2 The flowchart illustrating the implementation of the deep learning-based heparin sodium detection method provided in Embodiment 2 of this application is shown. The difference between this method and Embodiment 1 is that step S102 specifically includes:

[0041] Step S201: Time alignment and filtering are performed on the heparin sodium detection spectral signal information and the heparin sodium detection electrochemical signal information to obtain the heparin sodium detection spectral signal denoising information and the heparin sodium detection electrochemical signal denoising information.

[0042] In this embodiment, time alignment can be achieved using a sliding window cross-correlation algorithm. This involves sorting the electrochemical and spectral signals by timestamp, then using a sliding window with the minimum sampling interval as the step size. The Pearson correlation coefficient within the window is calculated, and the window offset corresponding to the maximum correlation coefficient is used as the time calibration parameter. The electrochemical signal is then resampled or interpolated to achieve time synchronization between the heparin sodium detection spectral signal and the heparin sodium detection electrochemical signal. Filtering can be performed using a Savitzky-Golay filter combined with wavelet transform to filter the time-aligned heparin sodium detection spectral and electrochemical signal information, resulting in denoised information for both signals.

[0043] Step S202: Based on the preset heparin sodium detection signal feature extraction matrix, convolution calculation is performed on the denoised information of the heparin sodium detection spectral signal and the denoised information of the heparin sodium detection electrochemical signal to obtain multiple heparin sodium detection spectral signal sub-feature information and multiple heparin sodium detection electrochemical signal sub-feature information.

[0044] In this embodiment, the preset heparin sodium detection signal feature extraction matrix can be manually set. It can be achieved by arranging the denoised heparin sodium detection spectral signal information into a two-dimensional matrix in wavelength order, sliding the heparin sodium detection signal feature extraction matrix on the spectral matrix, and extracting local features such as absorption intensity changes and peak contours within a specific wavelength range through matrix multiplication and accumulation operations, thereby generating heparin sodium detection spectral signal sub-feature information. Alternatively, it can be achieved by constructing the potential-current information from the denoised heparin sodium detection electrochemical signal into a matrix form, and then capturing electrochemical sub-features such as the current response slope at potential abrupt changes and the peak spacing of the cyclic voltammetry curve through convolution operations of the heparin sodium detection signal feature extraction matrix on the matrix constructed from the potential-current information, thus obtaining multiple heparin sodium detection electrochemical signal sub-feature information.

[0045] Step S203: Based on multiple preset heparin sodium detection signal sub-feature enhancement matrices, interactive enhancement processing is performed on multiple heparin sodium detection spectral signal sub-feature information and multiple heparin sodium detection electrochemical signal sub-feature information to obtain heparin sodium detection spectral signal feature information and heparin sodium detection electrochemical signal feature information.

[0046] In this embodiment, the multiple preset heparin sodium detection signal sub-feature enhancement matrices can all be manually set. Each heparin sodium detection signal sub-feature enhancement matrix corresponds to the detection requirements of specific physicochemical properties of the heparin sodium molecule, such as highlighting the spectral absorption characteristics related to sulfate groups and enhancing the electrochemical current signal corresponding to redox reactions. This can be achieved by performing a dot product operation between the heparin sodium detection signal sub-feature enhancement matrix and the heparin sodium detection spectral and electrochemical signal sub-feature information, combined with an activation function to perform a nonlinear transformation on the result. Based on the intrinsic correlation between heparin sodium molecules in the spectral and electrochemical fields, higher weights are assigned to signal segments reflecting the same molecular structure or reaction process in the two types of sub-feature information. For example, when a specific wavelength absorption peak in the spectral signal and a current peak in a certain potential range in the electrochemical signal both indicate a certain functional group of heparin sodium, the weight of this part of the sub-feature is increased by the heparin sodium detection signal sub-feature enhancement matrix, suppressing background interference signals. This allows the heparin sodium detection spectral signal feature information and the heparin sodium detection electrochemical signal feature information to retain the unique advantages of their respective modes and enhance complementary characteristics during the fusion process, thus more accurately characterizing the essential features of the heparin sodium molecule.

[0047] The deep learning-based heparin sodium detection method provided in this application thoroughly eliminates the time-dimensional deviation and data noise interference of multimodal signals through time alignment and filtering. By using a preset heparin sodium detection signal feature extraction matrix and a preset heparin sodium detection signal sub-feature enhancement matrix, it achieves progressive processing of heparin sodium detection signal features from basic extraction to deep enhancement, significantly improving the accuracy of feature extraction and enhancing the ability of features to characterize the properties of heparin sodium in complex samples. At the same time, it deeply explores the complementary value of spectral detection signals and electrochemical detection signals, effectively addressing the challenges of complex sample composition and variable detection environment in heparin sodium detection scenarios, and greatly improving the accuracy, reliability and anti-interference performance of detection results.

[0048] Figure 3 The flowchart illustrating the implementation of the deep learning-based heparin sodium detection method provided in Embodiment 3 of this application is shown. Its difference from Embodiment 2 described above lies in:

[0049] Multiple preset heparin sodium detection signal feature enhancement matrices include a preset first heparin sodium detection signal feature enhancement matrix, a preset second heparin sodium detection signal feature enhancement matrix, and a preset third heparin sodium detection signal feature enhancement matrix.

[0050] Step S203 specifically includes:

[0051] Step S301: Perform feature interaction fusion processing on multiple heparin sodium detection spectral signal sub-feature information and multiple heparin sodium detection electrochemical signal sub-feature information to obtain heparin sodium detection signal sub-feature fusion information.

[0052] In this embodiment, the feature interaction fusion processing can employ a combination of tensor splicing and fully connected layer mapping. This involves expanding multiple heparin sodium detection spectral signal sub-features and multiple heparin sodium detection electrochemical signal sub-features along the channel dimension, forming a high-dimensional feature tensor containing both spectral and electrochemical features through tensor splicing. Subsequently, a fully connected layer is used to perform linear transformations and nonlinear activations on this tensor, enabling cross-transfer of information between spectral and electrochemical sub-features and capturing the correlation between different modal features. This results in fused heparin sodium detection signal sub-feature information that integrates the advantages of both signal types.

[0053] Step S302: Based on the heparin sodium detection signal feature fusion information and the preset first heparin sodium detection signal feature enhancement matrix, the first heparin sodium detection signal feature enhancement information is obtained.

[0054] In this embodiment, the preset first heparin sodium detection signal feature enhancement matrix, the preset second heparin sodium detection signal feature enhancement matrix, and the preset third heparin sodium detection signal feature enhancement matrix can all be manually set. The preset first heparin sodium detection signal feature enhancement matrix can focus on the detection of physicochemical properties related to the sulfate group in the heparin sodium molecule. Alternatively, the fused information of the heparin sodium detection signal features can be multiplied with the first enhancement matrix. By amplifying the signal components related to the sulfate group features, such as changes in spectral absorption intensity and electrochemical current response, through the weight parameters of the first heparin sodium detection signal feature enhancement matrix, irrelevant background features are suppressed. Then, a nonlinear transformation is performed using an activation function to highlight the signal features closely related to the sulfate group structure, ultimately obtaining the first heparin sodium detection signal feature enhancement information with enhanced sulfate group detection specificity.

[0055] Step S303: Based on the heparin sodium detection signal feature fusion information and the preset second heparin sodium detection signal feature enhancement matrix, the second heparin sodium detection signal feature enhancement information is obtained.

[0056] In this embodiment, a preset second heparin sodium detection signal sub-feature enhancement matrix can be used to target the redox reaction characteristics of heparin sodium molecules. This can be achieved by performing a dot product operation between the second heparin sodium detection signal sub-feature enhancement matrix and the heparin sodium detection signal sub-feature fusion information. Based on the characteristic performance of the redox reaction in spectral and electrochemical signals, the second heparin sodium detection signal sub-feature enhancement matrix assigns higher weights to current change characteristics in corresponding potential ranges and spectral absorption change characteristics at specific wavelengths, while weakening other interfering features. After processing with an activation function, second heparin sodium detection signal sub-feature enhancement information that highlights the redox reaction characteristics is generated.

[0057] Step S304: Based on the heparin sodium detection signal feature fusion information and the preset third heparin sodium detection signal feature enhancement matrix, the third heparin sodium detection signal feature enhancement information is obtained.

[0058] In this embodiment, a preset third heparin sodium detection signal sub-feature enhancement matrix can be used to focus on characterizing the overall structural integrity of the heparin sodium molecule. This can be achieved by weighting the heparin sodium detection signal sub-feature fusion information with the third heparin sodium detection signal sub-feature enhancement matrix. The third heparin sodium detection signal sub-feature enhancement matrix, through pre-trained parameters, comprehensively considers the overall distribution of functional group vibrational absorption in the spectrum and the dynamic characteristics of the molecular-electrode interface interaction in electrochemistry. This enhances the signal features reflecting the complete molecular structure and suppresses abnormal features caused by sample impurities or detection noise, thereby obtaining the third heparin sodium detection signal sub-feature enhancement information that reflects the structural characteristics of the heparin sodium molecule.

[0059] Step S305: Based on the first heparin sodium detection signal feature enhancement information and the second heparin sodium detection signal feature enhancement information, the heparin sodium detection signal feature interaction enhancement information is obtained.

[0060] In this embodiment, a bidirectional interactive operation can be performed on the first heparin sodium detection signal feature enhancement information and the second heparin sodium detection signal feature enhancement information. Alternatively, the first and second heparin sodium detection signal feature enhancement information can be added element-wise, fusing sulfate group characteristics and redox reaction characteristics to obtain interactive enhancement information for the heparin sodium detection signal feature that integrates both key characteristics.

[0061] Step S306: Based on the heparin sodium detection signal feature interaction enhancement information and the third heparin sodium detection signal feature enhancement information, the heparin sodium detection signal feature information is obtained.

[0062] In this embodiment, the interaction enhancement information of the heparin sodium detection signal sub-features and the enhancement information of the third heparin sodium detection signal sub-features can be deeply fused. This can be achieved by first performing a nonlinear mapping on the interaction enhancement information of the heparin sodium detection signal sub-features and the enhancement information of the third heparin sodium detection signal sub-features using a multilayer perceptron. This can be done by utilizing the multilayer network structure of the multilayer perceptron to mine higher-order correlations between features, while incorporating a residual connection network to avoid the gradient vanishing problem caused by increasing network depth. Then, the interaction enhancement information of the heparin sodium detection signal sub-features and the enhancement information of the third heparin sodium detection signal sub-features, after nonlinear mapping by the residual connection network, are multiplied together, and the result is used as the heparin sodium detection signal feature information.

[0063] Step S307: The heparin sodium detection signal feature information is separated and processed to obtain the heparin sodium detection spectral signal feature information and the heparin sodium detection electrochemical signal feature information.

[0064] In this embodiment, two branch convolutional neural networks can be used to perform convolution operations on the heparin sodium detection signal feature information. One branch network performs feature filtering based on the wavelength distribution characteristics of the spectral signal features, while the other branch network focuses on the potential-current change characteristics of the electrochemical signal features for feature extraction. Each branch network gradually filters out features of non-corresponding modes through convolutional layers and pooling layers, and outputs the heparin sodium detection spectral signal feature information and the heparin sodium detection electrochemical signal feature information, respectively.

[0065] The deep learning-based heparin sodium detection method provided in this application progressively enhances the spectral and electrochemical signal characteristics of heparin sodium detection using multiple preset heparin sodium detection signal feature enhancement matrices. Employing a feature interaction fusion and separation mechanism, it achieves deep synergy between spectral and electrochemical signals while preserving the independence of each modal feature. This effectively balances the complementarity and specificity of multimodal information, thereby improving the anti-interference capability of complex heparin sodium sample detection, significantly enhancing the specificity and sensitivity of heparin sodium detection, and providing more reliable technical support for heparin sodium quality control.

[0066] Figure 4 The flowchart illustrating the implementation of the deep learning-based heparin sodium detection method provided in Embodiment 4 of this application is shown. The difference between this method and Embodiment 1 is that step S103 specifically includes:

[0067] Step S401: The spectral signal feature information and electrochemical signal feature information of heparin sodium detection are fused and encoded to generate multiple individual information of heparin sodium detection signal features.

[0068] In this embodiment, the spectral and electrochemical signal features of heparin sodium detection can be first expanded into a sequence of feature vectors according to their dimensions. A multi-level feature interaction module can be designed to enable correlation calculations between features of different modalities at multiple abstract levels. For example, in the first layer of interaction, the correlation weight between the spectral absorption peak position features and the electrochemical redox potential features is calculated to generate preliminary cross-features. In subsequent layers, nonlinear transformations and secondary combinations are performed on the preliminary cross-features to gradually extract more abstract cross-modal feature correlations. Each individual feature contains multi-level feature expressions from the underlying signal to the high-level semantics, thus fully exploring the complementary information between the two signal modalities.

[0069] Step S402: Based on the preset heparin sodium detection signal feature optimization boundary, generate multiple initial heparin sodium detection signal feature individuals according to the information of multiple heparin sodium detection signal feature individuals.

[0070] In this embodiment, the preset optimization boundary for heparin sodium detection signal features can be artificially set to define the physical feasible region of the feature parameters. For example, the spectral absorption intensity feature is restricted to the [0,1] interval, and the electrochemical current response feature is constrained within the linear range of the detection device. For multiple individual heparin sodium detection signal feature information, a boundary constraint mapping algorithm is used to map them to the optimization boundary: for feature values ​​exceeding the upper limit, they are compressed proportionally to the boundary value; for feature values ​​below the lower limit, they are adjusted to the feasible region through interpolation. At the same time, a feature smoothing mechanism is introduced to perform gradient correction on feature values ​​near the boundary, avoiding optimization oscillations caused by abrupt boundary changes, thereby generating initial feature individuals that conform to physical laws.

[0071] Step S403: Calculate the fitness of multiple heparin sodium detection signal features based on the multiple initial heparin sodium detection signal feature individuals.

[0072] In this embodiment, a comprehensive evaluation function for the fitness of heparin sodium detection signal features can be designed to calculate the fitness of heparin sodium detection signal features. This function can include three core metrics: first, inter-class separation, which measures the ability of a feature individual to distinguish between heparin sodium samples and interfering samples, evaluated by calculating the distance between the two classes in the feature space; second, information gain ratio, which evaluates the predictive ability of a feature individual for target attributes such as heparin sodium concentration and purity; and third, robustness, which calculates the change in feature classification accuracy before and after adding noise simulating detection environment disturbances to the feature individual. The three metrics can be weighted and fused to obtain the final fitness value. A higher fitness value indicates a stronger ability of the feature individual to recognize heparin sodium in complex environments, similar to performing a multi-dimensional capability test on each feature individual, with higher scores providing an advantage in subsequent optimization.

[0073] Step S404: The initial heparin sodium detection signal feature individual corresponding to the maximum value of the fitness of the plurality of heparin sodium detection signal features is used as the reference heparin sodium detection signal feature individual.

[0074] In this embodiment, the fitness of all heparin sodium detection signal features can be compared, and the initial heparin sodium detection signal feature individual corresponding to the maximum value can be selected as the reference heparin sodium detection signal feature individual. In the subsequent iterative optimization process, the reference heparin sodium detection signal feature individual will serve as a guiding direction to ensure that the optimization direction iterates in the direction of improving the accuracy of heparin sodium detection.

[0075] Step S405: Generate multiple target heparin sodium detection signal feature individuals based on the multiple initial heparin sodium detection signal feature individuals, reference heparin sodium detection signal feature individuals, preset heparin sodium detection signal feature optimization iteration number, and preset heparin sodium detection signal feature optimization amplitude information.

[0076] In this embodiment, an iterative optimization method is used to generate target heparin sodium detection signal feature individuals. The process can terminate with a preset number of iterations for optimizing the heparin sodium detection signal features. Each iteration includes the following operations: First, the parameters of the initial feature individual are randomly adjusted, with the adjustment range controlled by a preset optimization range information. Then, the adjusted individual is combined with a reference feature individual to generate a new experimental individual. Finally, the fitness of the experimental individual is compared with the original individual, and the individual with higher fitness is retained for the next generation. During the iteration process, the parameter adjustment strategy is dynamically adjusted according to the optimization progress. In the early stage, different feature combinations are widely explored, while in the later stage, high-quality feature combinations are finely optimized. Through this continuous iterative evolution mechanism, multiple target feature individuals with stronger distinguishing abilities for heparin sodium characteristics are gradually generated. Through continuous mutation, combination, and screening, the feature individuals gradually become more suitable for the needs of heparin sodium detection.

[0077] Step S406: Generate multiple optimized feature information for heparin sodium detection signals based on the multiple target heparin sodium detection signal feature individuals.

[0078] In this embodiment, the individual characteristics of the target heparin sodium detection signal can be classified and analyzed, and features can be screened. This can be done by first calculating the similarity between all target individuals, clustering similar feature individuals into one class, with the center of each class representing a set of feature combinations with similar recognition patterns, and then evaluating the importance of each feature. The contribution of each feature dimension to the heparin sodium detection result is calculated by an algorithm, and feature dimensions with high contribution are screened out. Then, the screened features are standardized, and combined with professional knowledge of heparin sodium detection, the abstract feature vectors are transformed into optimized feature information with clear physical meaning, such as "sulfate content feature" and "molecular chain length feature", making the feature information easier to understand and apply to actual heparin sodium detection and analysis.

[0079] The deep learning-based heparin sodium detection method provided in this application achieves complementary advantages of spectral and electrochemical signals through deep fusion of multi-dimensional features, effectively extracting features that better reflect the essential properties of heparin sodium. By using a preset heparin sodium detection signal feature optimization boundary, the physical rationality and detection reliability of the heparin sodium detection signal features are ensured. An iterative optimization strategy, combined with the guidance of reference heparin sodium detection signal feature individuals, improves the efficiency and accuracy of heparin sodium detection signal feature optimization, thereby enhancing the accuracy and robustness of heparin sodium detection.

[0080] Figure 5 The flowchart of the deep learning-based heparin sodium detection method provided in Embodiment 5 of this application is shown. The difference between this method and Embodiment 4 is that step S403 specifically includes:

[0081] Step S501: Dimensionality reduction processing is performed on multiple initial heparin sodium detection signal feature individuals to obtain multiple initial heparin sodium detection signal feature dimensionality reduction information.

[0082] In this embodiment, the dimensionality reduction process can employ a multi-layer feature mapping approach. This involves constructing a nonlinear transformation function to project individual initial heparin sodium detection signal features from a high-dimensional feature space to a low-dimensional space. Each initial feature can be initially treated as a high-dimensional vector, and a nonlinear mapping function can be used to capture the complex relationships between features. For example, for initial heparin sodium detection signal features containing multi-dimensional features such as spectral absorption peak position, peak intensity, and electrochemical current response, a nonlinear transformation can be used to map them to a low-dimensional space. This preserves key information such as the correlation between specific wavelength absorption peaks and corresponding potential current peaks, avoiding the loss of feature relationships caused by traditional linear dimensionality reduction. This results in dimensionality-reduced initial heparin sodium detection signal features that retain core discriminative capabilities.

[0083] Step S502: Extract information from the dimensionality reduction information of the multiple initial heparin sodium detection signal features to obtain multiple dimensionality reduction information of heparin sodium detection spectral signal features and dimensionality reduction information of heparin sodium detection electrochemical signal features.

[0084] In this embodiment, feature-based modal decoupling can be used to extract information. For the dimensionality-reduced features of the initial heparin sodium detection signal, two independent feature selection modules are designed: one module focuses on identifying wavelength-related feature components, such as the low-dimensional coordinates corresponding to the UV absorption peak; the other module focuses on potential-current related feature components, such as the low-dimensional coordinates corresponding to the peak spacing of the voltammetric curve. By calculating the correlation coefficient between each low-dimensional feature component and the original spectral / electrochemical signal, components with correlations higher than a threshold are selected to form the dimensionality-reduced features of the heparin sodium detection spectral signal and the heparin sodium detection electrochemical signal, respectively. For example, if a low-dimensional component is highly correlated with the intensity change of the 230nm UV absorption peak, it is included in the spectral dimensionality-reduced information; if it is correlated with the current response at a potential of 0.5V, it is included in the electrochemical dimensionality-reduced information, thereby achieving decoupling and targeted retention of multimodal features.

[0085] Step S503: Based on the multiple heparin sodium detection spectral signal feature dimensionality reduction information, the heparin sodium detection electrochemical signal feature dimensionality reduction information, and the preset heparin sodium detection signal attenuation calculation function, multiple heparin sodium detection signal attenuation information are calculated.

[0086] In this embodiment, the preset heparin sodium detection signal attenuation calculation function can be set manually. It can be used to quantify the degree of information loss of features after dimensionality reduction. It can take spectral and electrochemical dimensionality reduction information as input and calculate the attenuation value through the following steps: (1) For spectral dimensionality reduction information, calculate the root mean square error (RMSE) between the original spectral features and the features after dimensionality reduction to reflect the information loss of spectral features in dimensionality reduction; similarly, calculate the RMSE for electrochemical dimensionality reduction information; (2) By calculating the change in mutual information between spectral-electrochemical feature pairs before and after dimensionality reduction, such as the change in mutual information between a certain wavelength absorption peak and the corresponding potential current peak, evaluate the degree of retention of cross-modal correlation information; (3) Weight the intra-modal attenuation and inter-modal correlation attenuation according to preset weights (such as 0.6:0.4) to obtain the heparin sodium detection signal attenuation information. It can be understood that the lower the heparin sodium detection signal attenuation information, the more effective information and cross-modal correlation of the features after dimensionality reduction are retained in the original signal.

[0087] Step S504: Based on the dimensionality reduction information of the multiple heparin sodium detection spectral signals, the dimensionality reduction information of the heparin sodium detection electrochemical signals, and the preset fitness weight information of the heparin sodium detection signals, the fitness of the multiple heparin sodium detection signals is calculated.

[0088] In this embodiment, the preset fitness weight information for heparin sodium detection signal features can be manually set and may include a spectral feature weight matrix and an electrochemical feature weight matrix, which correspond to the importance coefficients of each feature component in the spectral / electrochemical dimensionality reduction information, respectively. The spectral dimensionality reduction information can be performed first, multiplying each feature component by its corresponding weight coefficient and then summing the results to obtain the spectral modal fitness. Similarly, the electrochemical modal fitness is calculated, with the weight coefficients preset according to the molecular characteristics of heparin sodium; for example, spectral absorption features related to sulfate groups have higher weights. Then, the covariance of the spectral and electrochemical dimensionality reduction information is calculated to reflect the collaborative distinguishing ability of the two types of features in the dimensionality reduction space; a higher covariance indicates stronger cross-modal complementarity. Finally, the intra-modal fitness (spectral × 0.5 + electrochemical × 0.5) and the inter-modal collaborative fitness (× 0.5) are weighted and summed to obtain the fitness of the heparin sodium detection signal features. Understandably, a higher fitness of heparin sodium detection signal features indicates that the initial heparin sodium detection signal features perform better in preserving single-modal effectiveness and cross-modal synergy, and are more suitable for heparin sodium detection.

[0089] The deep learning-based heparin sodium detection method provided in this application reduces feature redundancy during dimensionality reduction while retaining key information on the molecular structure and electrochemical reaction of heparin sodium. This significantly improves the optimization efficiency of heparin sodium detection signal features and the generalization ability of the detection model. It is suitable for accurately distinguishing heparin sodium from interfering substances in complex samples, effectively improving the detection accuracy of heparin sodium and enhancing the efficiency and reliability of heparin sodium quality control.

[0090] Figure 6 The flowchart illustrating the implementation of the deep learning-based heparin sodium detection method provided in Embodiment Six of this application is shown. The difference between this method and Embodiment Four is that step S405 specifically includes:

[0091] Step S601: Based on the reference heparin sodium detection signal feature individuals and the preset heparin sodium detection signal feature optimization amplitude information, iterative optimization processing is performed on multiple initial heparin sodium detection signal feature individuals to obtain multiple intermediate heparin sodium detection signal feature individuals and the number of iterations for the heparin sodium detection signal feature individuals.

[0092] In this embodiment, the preset optimization amplitude information for heparin sodium detection signal features can be manually set. Using a reference heparin sodium detection signal feature as the optimization guide, the following operations are performed on each initial heparin sodium detection signal feature: The parameters of the initial heparin sodium detection signal feature are randomly perturbed according to the preset optimization amplitude information, such as by a weight adjustment of ±10%. The perturbation amplitude dynamically decays with the iteration process, ensuring global exploration in the early stage and local fine optimization in the later stage. Then, the perturbed initial heparin sodium detection signal feature is fused with the reference heparin sodium detection signal feature in a dimension-wise weighted manner. For example, for spectral features, the weight of the initial feature is added to the weight of the reference heparin sodium detection signal feature; for electrochemical features, the weight of the initial feature is added to the weight of the reference heparin sodium detection signal feature, resulting in multiple intermediate heparin sodium detection signal feature individuals. After each round of optimization to generate intermediate heparin sodium detection signal feature individuals, the iteration optimization count of the heparin sodium detection signal feature individual is automatically incremented by 1 to record the current optimization progress.

[0093] Step S602: Determine whether the number of individual iterations for optimizing the heparin sodium detection signal features is equal to the preset number of iterations for optimizing the heparin sodium detection signal features; if yes, proceed to step S603; if no, proceed to step S604.

[0094] In this embodiment, a preset number of iterations for optimizing the heparin sodium detection signal features can be used as a fixed threshold. When the number of iterations reaches this threshold, it is considered that the feature space search has covered the main optimization direction, and the iteration stops; if it has not reached this threshold, subsequent optimization steps continue to ensure the sufficiency of feature optimization. The preset number of iterations for optimizing the heparin sodium detection signal features can be 100.

[0095] Step S603: The multiple intermediate heparin sodium detection signal feature individuals are used as multiple target heparin sodium detection signal feature individuals.

[0096] In this embodiment, intermediate heparin sodium detection signal feature individuals can be used as multiple target heparin sodium detection signal feature individuals. This not only retains the basic distinguishing ability of the initial heparin sodium detection signal feature individuals, but also integrates the optimal feature combination mode of the reference heparin sodium detection signal feature individuals, which can more accurately characterize the spectral and electrochemical properties of heparin sodium molecules.

[0097] Step S604: Based on the multiple intermediate heparin sodium detection signal feature individuals, calculate the fitness of multiple heparin sodium detection signal feature iteration optimization.

[0098] In this embodiment, the inter-class separation degree (e.g., Mahalanobis distance), information gain ratio (e.g., Gini index), and robustness index (the difference in classification accuracy before and after noise perturbation) of intermediate feature individuals can be calculated. Then, the fitness improvement rate of the current intermediate feature individual relative to the individual in the previous iteration can be calculated. For example, the improvement rate = (current fitness - previous fitness) / previous fitness × 100%. The basic evaluation results are then weighted and fused with the evolutionary efficiency to obtain the iterative optimization fitness of the heparin sodium detection signal feature. A high fitness value indicates that the feature individual has achieved effective evolution during the iteration process.

[0099] Step S605: Select the multiple intermediate heparin sodium detection signal feature individuals as initial heparin sodium detection signal feature individuals, select the intermediate heparin sodium detection signal feature individual corresponding to the maximum value of the fitness of the multiple heparin sodium detection signal feature individuals as reference heparin sodium detection signal feature individuals, and return to step S601.

[0100] In this embodiment, the currently generated intermediate feature individuals are used as the initial input for the next iteration to ensure the continuity and cumulative nature of the optimization process; the intermediate feature individuals corresponding to the maximum fitness value of the iteration are selected as new reference standards to guide subsequent iterations to search for higher fitness regions; thus, each iteration is based on the current optimal feature pattern, accelerating convergence to the global optimal feature combination.

[0101] The deep learning-based heparin sodium detection method provided in this application balances the optimization accuracy of different modal features, enhances feature complementarity through cross-modal weighted fusion, and dynamically adjusts the optimization direction to avoid getting trapped in local optima. This enables the accurate capture of heparin sodium molecule features, effectively improves the sensitivity and anti-interference ability of heparin sodium detection, and enhances the efficiency and effectiveness of heparin sodium quality control.

[0102] Figure 7 The flowchart of the deep learning-based heparin sodium detection method provided in Embodiment 7 of this application is shown. The difference between this method and Embodiment 1 is that step S104 specifically includes:

[0103] Step S701: Based on the preset number of optimized features to be extracted for heparin sodium detection signals, multiple optimized features for heparin sodium detection signals are randomly extracted to obtain multiple optimized features for central heparin sodium detection signals.

[0104] In this embodiment, the preset number of optimized features for heparin sodium detection signals can be manually set, such as 5 or 10, to determine the number of initial classification centers. A random sampling algorithm can be used to extract a corresponding number of features as center samples from all optimized features for heparin sodium detection signals. For example, 5 features can be randomly selected from 100 optimized features for heparin sodium detection signals as initial centers. The sampling process employs a sampling strategy without replacement to ensure that each center feature has a certain degree of randomness in its distribution within the feature space, avoiding classification bias caused by an overly concentrated initial center.

[0105] Step S702: Calculate the logical distance between the multiple optimized feature information of the heparin sodium detection signal and the multiple optimized feature information of the central heparin sodium detection signal to obtain the multiple optimized feature distance information of the heparin sodium detection signal.

[0106] In this embodiment, the logical distance can be calculated as the Euclidean distance. The Euclidean distance between each optimized feature information of the heparin sodium detection signal and the optimized feature information of the central heparin sodium detection signal is calculated as the optimized feature distance information of the heparin sodium detection signal.

[0107] Step S703: Based on the multiple optimized feature distance information and the optimized feature information of the central heparin sodium detection signal, multiple optimized feature class information of heparin sodium detection signal is obtained; the optimized feature class information of heparin sodium detection signal corresponds one-to-one with the optimized feature information of the central heparin sodium detection signal; one optimized feature class information of heparin sodium detection signal includes multiple optimized feature information of heparin sodium detection signal.

[0108] In this embodiment, clustering can be performed using the nearest neighbor assignment principle: for each optimized feature, it is assigned to the class of the central feature with the smallest distance. For example, if feature A has the smallest distance to center 3, then feature A is assigned to the class corresponding to center 3. Through this assignment process, all optimized feature information is divided into classes with an equal number of centers. Each class contains several feature information, forming optimized feature class information for heparin sodium detection signals, thereby achieving a preliminary division of the feature space and grouping optimized feature distance information for heparin sodium detection signals with similar spectral-electrochemical feature combinations into one class.

[0109] Step S704: Calculate the average value of each of the optimized feature class information of the heparin sodium detection signal to obtain the average value information of multiple optimized features of the heparin sodium detection signal.

[0110] In this embodiment, mean calculation is performed for each feature class: for all optimized feature information within a class, the average value is calculated for each feature dimension. For example, if a class contains 10 feature information, and each feature information contains 5 dimensions such as the position of the ultraviolet absorption peak and the intensity of infrared vibration, then the mean value of the class in each dimension is calculated to generate a new feature vector as the average feature information of the class. The average value information of the optimized features of the heparin sodium detection signal can be used to represent the overall feature distribution center of the corresponding class, which can eliminate the random fluctuation of features within the class and highlight common features.

[0111] Step S705: Calculate the average value of the optimized features of multiple heparin sodium detection signals and the convergence distance value of the optimized features of the central heparin sodium detection signal to obtain the classification convergence value information of the optimized features of the heparin sodium detection signal.

[0112] In this embodiment, the convergence distance value is calculated by the Euclidean distance between the average feature information of each class and the original center feature information. The convergence distances of all classes are summed to obtain the convergence value information of the optimized feature classification of the heparin sodium detection signal, which is used to reflect the update magnitude of the classification center. The smaller the convergence value information of the optimized feature classification of the heparin sodium detection signal, the more stable the classification result.

[0113] Step S706: Determine whether the convergence value of the optimized feature classification of the heparin sodium detection signal is less than the preset convergence threshold of the feature classification of the heparin sodium detection signal; if yes, proceed to step S707; if no, proceed to step S708.

[0114] In this embodiment, the preset heparin sodium detection signal feature classification convergence threshold can be set manually. When the optimized feature classification convergence value of the heparin sodium detection signal is less than the preset heparin sodium detection signal feature classification convergence threshold, it indicates that after the current iteration, the change in the classification center is small enough and the classification result is stable. If the optimized feature classification convergence value of the heparin sodium detection signal is not less than the preset heparin sodium detection signal feature classification convergence threshold, it is necessary to continue iterating and optimizing the classification center to ensure the accuracy of the classification.

[0115] Step S707: Optimize feature class information based on multiple heparin sodium detection signals to obtain heparin sodium detection information.

[0116] In this embodiment, after classification convergence, each feature class corresponds to a specific attribute of heparin sodium: the average spectral absorption intensity of the feature information within the class corresponds to the standard concentration curve of the electrochemical current value, which determines the sample concentration range; the dispersion of the feature within the class reflects the sample purity, and the lower the dispersion, the higher the purity; if the distance between a certain feature and the standard heparin sodium feature exceeds a preset threshold, it is determined to be an abnormal sample; then the attribute information corresponding to each class is integrated to output heparin sodium detection information, such as "sample concentration is 2.5 IU / mL, purity is 98.3%, no abnormal interference".

[0117] Step S708: The average value of the multiple optimized features of the heparin sodium detection signal is used as the central optimized feature information of the heparin sodium detection signal, and the process returns to step S702.

[0118] In this embodiment, the currently calculated average feature information of the class is used as the new center feature information to replace the original center feature. Then, the distance between all optimized features and the new center is recalculated and the categories are redistributed. By continuously updating the center position, the classification result gradually approaches the true distribution of the feature space, avoiding classification bias caused by the initial random center, until the convergence distance meets the threshold requirement.

[0119] The deep learning-based heparin sodium detection method provided in this application utilizes an iterative update strategy for classification centers to dynamically adapt to the distribution characteristics of the feature space. In complex heparin sodium sample detection, it can accurately infer key indicators such as the concentration and purity of heparin sodium based on the average attributes of feature classes. At the same time, it ensures the stability of classification results through convergence judgment and exhibits stronger robustness when there are large differences between batches of heparin sodium samples or when the detection environment changes, thereby improving the accuracy and stability of heparin sodium detection.

[0120] Corresponding to the method in the above embodiments, Figure 8 The diagram shows a structural block diagram of a deep learning-based heparin sodium detection system provided in an embodiment of this application. For ease of explanation, only the parts relevant to the embodiments of this application are shown. Figure 8 The example deep learning-based heparin sodium detection system can be the execution entity of the deep learning-based heparin sodium detection method provided in Embodiment 1 above.

[0121] Reference Figure 8 The deep learning-based heparin sodium detection system includes:

[0122] The heparin sodium detection signal information acquisition module 810 is used to acquire heparin sodium detection spectral signal information and heparin sodium detection electrochemical signal information.

[0123] The heparin sodium detection signal feature information generation module 820 is used to extract features from the heparin sodium detection spectral signal information and the heparin sodium detection electrochemical signal information to obtain heparin sodium detection spectral signal feature information and heparin sodium detection electrochemical signal feature information.

[0124] The heparin sodium detection signal optimization feature information generation module 830 is used to generate multiple heparin sodium detection signal optimization feature information based on the heparin sodium detection spectral signal feature information, the heparin sodium detection electrochemical signal feature information, the preset heparin sodium detection signal feature optimization iteration number, and the preset heparin sodium detection signal feature optimization amplitude information.

[0125] The heparin sodium detection information generation module 840 is used to classify and process multiple optimized feature information of heparin sodium detection signals based on preset heparin sodium detection signal feature classification convergence threshold information to obtain heparin sodium detection information.

[0126] The process by which each module in the deep learning-based heparin sodium detection system provided in this application implements its respective function can be found in the foregoing. Figure 1 The description of Embodiment 1 shown will not be repeated here.

[0127] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0128] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0129] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0130] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0131] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance. It should also be understood that although the terms "first," "second," etc., are used in the text to describe various elements in some embodiments of this application, these elements should not be limited by these terms. These terms are merely used to distinguish one element from another. For example, a first table may be named a second table, and similarly, a second table may be named a first table, without departing from the scope of the various described embodiments. Both the first table and the second table are tables, but they are not the same table.

[0132] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0133] The deep learning-based heparin sodium detection method provided in this application can be applied to terminal devices such as mobile phones, tablets, wearable devices, in-vehicle devices, augmented reality (AR) / virtual reality (VR) devices, laptops, ultra-mobile personal computers (UMPCs), netbooks, and personal digital assistants (PDAs). This application does not impose any restrictions on the specific type of terminal device.

[0134] For example, the terminal device may be a station (STAION, ST) in a WLAN, a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA) device, a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, an in-vehicle device, a vehicle networking terminal, a computer, a laptop computer, a handheld communication device, a handheld computing device, a satellite wireless device, a wireless modem card, a set-top box (STB), customer premises equipment (CPE), and / or other devices used for communication over a wireless system, as well as next-generation communication systems, such as mobile terminals in 5G networks or mobile terminals in future evolved Public Land Mobile Network (PLMN) networks.

[0135] As an example and not a limitation, when the terminal device is a wearable device, the term "wearable device" can also refer to any device that utilizes wearable technology to intelligently design and develop everyday wearables, such as glasses, gloves, watches, clothing, and shoes. Wearable devices are portable devices worn directly on the body or integrated into a user's clothing or accessories. Wearable devices are not merely hardware devices; they achieve powerful functions through software support, data interaction, and cloud interaction. Broadly defined, wearable smart devices include those with comprehensive functions, large sizes, and the ability to perform complete or partial functions without relying on a smartphone, such as smartwatches or smart glasses, as well as those focused on a specific application function that require interaction with other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0136] Figure 9 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. For example... Figure 9 As shown, the terminal device 9 of this embodiment includes: at least one processor 90 ( Figure 9 Only one is shown in the image), and a memory 91 is stored in which a computer program 92 that can run on the processor 90 is stored. When the processor 90 executes the computer program 92, it implements the steps in the various embodiments of the deep learning-based heparin sodium detection method described above, for example... Figure 1 Steps S101 to S104 are shown. Alternatively, when the processor 90 executes the computer program 92, it implements the functions of each module / unit in the above system embodiments, for example... Figure 8The functions of modules 810 to 840 are shown.

[0137] The terminal device 9 can be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor 90 and a memory 91. Those skilled in the art will understand that... Figure 9 This is merely an example of terminal device 9 and does not constitute a limitation on terminal device 9. It may include more or fewer components than shown, or combine certain components, or different components. For example, the terminal device may also include input transmission devices, network access devices, buses, etc.

[0138] The processor 90 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0139] In some embodiments, the memory 91 may be an internal storage unit of the terminal device 9, such as a hard disk or memory of the terminal device 9. The memory 91 may also be an external storage device of the terminal device 9, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the terminal device 9. Furthermore, the memory 91 may include both internal and external storage units of the terminal device 9. The memory 91 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 91 can also be used to temporarily store data that has been sent or will be sent.

[0140] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0141] This application also provides a terminal device, which includes at least one memory, at least one processor, and a computer program stored in the at least one memory and executable on the at least one processor. When the processor executes the computer program, it causes the terminal device to implement the steps in any of the above method embodiments.

[0142] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0143] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0144] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or system capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0145] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0146] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0147] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0148] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A deep learning-based method for detecting heparin sodium, characterized in that, include: Acquire spectral signal information and electrochemical signal information for heparin sodium detection; Feature extraction is performed on the heparin sodium detection spectral signal information and the heparin sodium detection electrochemical signal information to obtain heparin sodium detection spectral signal feature information and heparin sodium detection electrochemical signal feature information; Based on the heparin sodium detection spectral signal feature information, the heparin sodium detection electrochemical signal feature information, the preset number of heparin sodium detection signal feature optimization iterations, and the preset heparin sodium detection signal feature optimization amplitude information, multiple heparin sodium detection signal optimization feature information are generated. Based on the preset heparin sodium detection signal feature classification convergence threshold information, the optimized feature information of multiple heparin sodium detection signals is classified and processed to obtain heparin sodium detection information. The step of generating multiple optimized heparin sodium detection signal feature information based on the heparin sodium detection spectral signal feature information, the heparin sodium detection electrochemical signal feature information, the preset number of heparin sodium detection signal feature optimization iterations, and the preset heparin sodium detection signal feature optimization amplitude information specifically includes: The spectral signal feature information and electrochemical signal feature information of heparin sodium detection are fused and encoded to generate multiple individual information of heparin sodium detection signal features; Based on the preset heparin sodium detection signal feature optimization boundary, multiple initial heparin sodium detection signal feature individuals are generated according to the information of multiple heparin sodium detection signal feature individuals. Based on the individual initial heparin sodium detection signal features, multiple heparin sodium detection signal feature fitnesss are calculated. The initial heparin sodium detection signal feature individual corresponding to the maximum value of the fitness of the multiple heparin sodium detection signal features is used as the reference heparin sodium detection signal feature individual; Based on the multiple initial heparin sodium detection signal feature individuals, the reference heparin sodium detection signal feature individuals, the preset number of heparin sodium detection signal feature optimization iterations, and the preset heparin sodium detection signal feature optimization amplitude information, multiple target heparin sodium detection signal feature individuals are generated. Based on the individual characteristics of the target heparin sodium detection signal, multiple optimized feature information of heparin sodium detection signal is generated; The step of calculating the fitness of multiple heparin sodium detection signal features based on multiple initial heparin sodium detection signal feature individuals specifically includes: Dimensionality reduction processing is performed on multiple initial heparin sodium detection signal feature individuals to obtain multiple initial heparin sodium detection signal feature dimensionality reduction information; Information is extracted from the dimensionality reduction information of multiple initial heparin sodium detection signal features to obtain multiple dimensionality reduction information of heparin sodium detection spectral signal features and dimensionality reduction information of heparin sodium detection electrochemical signal features. Based on the dimensionality reduction information of the multiple heparin sodium detection spectral signals, the dimensionality reduction information of the heparin sodium detection electrochemical signals, and the preset heparin sodium detection signal attenuation calculation function, multiple heparin sodium detection signal attenuation information is calculated. Based on the dimensionality reduction information of the multiple heparin sodium detection spectral signal features, the dimensionality reduction information of the heparin sodium detection electrochemical signal features, and the preset heparin sodium detection signal feature fitness weight information, the fitness of multiple heparin sodium detection signal features is calculated.

2. The deep learning-based heparin sodium detection method as described in claim 1, characterized in that, The step of extracting features from the heparin sodium detection spectral signal information and the heparin sodium detection electrochemical signal information to obtain heparin sodium detection spectral signal feature information and heparin sodium detection electrochemical signal feature information specifically includes: Time alignment and filtering are performed on the heparin sodium detection spectral signal information and the heparin sodium detection electrochemical signal information to obtain the heparin sodium detection spectral signal denoising information and the heparin sodium detection electrochemical signal denoising information. Based on the preset heparin sodium detection signal feature extraction matrix, the heparin sodium detection spectral signal denoising information and the heparin sodium detection electrochemical signal denoising information are convolved to obtain multiple heparin sodium detection spectral signal sub-feature information and multiple heparin sodium detection electrochemical signal sub-feature information. Based on multiple preset heparin sodium detection signal sub-feature enhancement matrices, interactive enhancement processing is performed on multiple heparin sodium detection spectral signal sub-feature information and multiple heparin sodium detection electrochemical signal sub-feature information to obtain heparin sodium detection spectral signal feature information and heparin sodium detection electrochemical signal feature information.

3. The deep learning-based heparin sodium detection method as described in claim 2, characterized in that, Multiple preset heparin sodium detection signal feature enhancement matrices include a preset first heparin sodium detection signal feature enhancement matrix, a preset second heparin sodium detection signal feature enhancement matrix, and a preset third heparin sodium detection signal feature enhancement matrix. The step of interactively enhancing multiple heparin sodium detection spectral signal feature information and multiple heparin sodium detection electrochemical signal feature information based on multiple preset heparin sodium detection signal feature enhancement matrices to obtain heparin sodium detection spectral signal feature information and heparin sodium detection electrochemical signal feature information specifically includes: The heparin sodium detection spectral signal sub-feature information and the heparin sodium detection electrochemical signal sub-feature information are subjected to feature interaction fusion processing to obtain heparin sodium detection signal sub-feature fusion information. Based on the heparin sodium detection signal sub-feature fusion information and the preset first heparin sodium detection signal sub-feature enhancement matrix, the first heparin sodium detection signal sub-feature enhancement information is obtained. Based on the heparin sodium detection signal sub-feature fusion information and the preset second heparin sodium detection signal sub-feature enhancement matrix, the second heparin sodium detection signal sub-feature enhancement information is obtained; Based on the heparin sodium detection signal feature fusion information and the preset third heparin sodium detection signal feature enhancement matrix, the third heparin sodium detection signal feature enhancement information is obtained. Based on the first heparin sodium detection signal feature enhancement information and the second heparin sodium detection signal feature enhancement information, the interactive enhancement information of the heparin sodium detection signal feature is obtained. Based on the heparin sodium detection signal feature interaction enhancement information and the third heparin sodium detection signal feature enhancement information, the heparin sodium detection signal feature information is obtained. The heparin sodium detection signal feature information is separated and processed to obtain the heparin sodium detection spectral signal feature information and the heparin sodium detection electrochemical signal feature information.

4. The deep learning-based heparin sodium detection method as described in claim 1, characterized in that, The step of generating multiple target heparin sodium detection signal feature individuals based on multiple initial heparin sodium detection signal feature individuals, reference heparin sodium detection signal feature individuals, a preset number of heparin sodium detection signal feature optimization iterations, and preset heparin sodium detection signal feature optimization amplitude information specifically includes: Based on the reference heparin sodium detection signal feature individuals and the preset heparin sodium detection signal feature optimization amplitude information, the initial heparin sodium detection signal feature individuals are iteratively optimized to obtain multiple intermediate heparin sodium detection signal feature individuals and the number of iterations for the heparin sodium detection signal feature individuals. Determine whether the number of individual iterations for optimizing the heparin sodium detection signal features is equal to the preset number of iterations for optimizing the heparin sodium detection signal features; If so, then the multiple intermediate heparin sodium detection signal feature individuals are taken as multiple target heparin sodium detection signal feature individuals; If not, then based on the multiple intermediate heparin sodium detection signal feature individuals, the fitness of multiple heparin sodium detection signal features is calculated through iterative optimization. Multiple intermediate heparin sodium detection signal feature individuals are used as initial heparin sodium detection signal feature individuals. The intermediate heparin sodium detection signal feature individual corresponding to the maximum value of the fitness of the multiple heparin sodium detection signal feature iterative optimization is used as a reference heparin sodium detection signal feature individual. Then, the process returns to the step of iteratively optimizing the multiple initial heparin sodium detection signal feature individuals based on the reference heparin sodium detection signal feature individual and the preset heparin sodium detection signal feature optimization amplitude information to obtain multiple intermediate heparin sodium detection signal feature individuals and the number of iterations of the heparin sodium detection signal feature individuals.

5. The deep learning-based heparin sodium detection method as described in claim 1, characterized in that, The step of classifying multiple optimized features of heparin sodium detection signals based on a preset heparin sodium detection signal feature classification convergence threshold to obtain heparin sodium detection information specifically includes: Based on the preset number of optimized features to be extracted from the heparin sodium detection signal, multiple optimized features of the heparin sodium detection signal are randomly extracted to obtain multiple optimized features of the central heparin sodium detection signal. Calculate the logical distance between the multiple optimized feature information of the heparin sodium detection signal and the multiple optimized feature information of the central heparin sodium detection signal to obtain the multiple optimized feature distance information of the heparin sodium detection signal. Based on the distance information of multiple optimized features of heparin sodium detection signals and the optimized feature information of the central heparin sodium detection signal, multiple optimized feature class information of heparin sodium detection signals is obtained; the optimized feature class information of heparin sodium detection signals corresponds one-to-one with the optimized feature information of the central heparin sodium detection signal; one optimized feature class information of heparin sodium detection signals includes multiple optimized feature information of heparin sodium detection signals. Calculate the average value of each of the optimized feature class information of the heparin sodium detection signal to obtain the average value information of multiple optimized features of the heparin sodium detection signal. Calculate the average value of multiple optimized features of heparin sodium detection signals and the convergence distance value of the central optimized feature of heparin sodium detection signals to obtain the classification convergence value information of optimized features of heparin sodium detection signals. Determine whether the convergence value of the optimized feature classification of the heparin sodium detection signal is less than the preset convergence threshold value of the heparin sodium detection signal feature classification. If so, then the heparin sodium detection information is obtained by optimizing the feature class information based on the multiple heparin sodium detection signals. If not, the average value of the multiple optimized features of the heparin sodium detection signal is used as the optimized feature information of the central heparin sodium detection signal, and the process is returned to the step of calculating the logical distance between the multiple optimized features of the heparin sodium detection signal and the multiple optimized features of the central heparin sodium detection signal to obtain the distance information of the multiple optimized features of the heparin sodium detection signal.

6. The detection system for the heparin sodium detection method based on deep learning as described in claim 1, characterized in that, include: The heparin sodium detection signal information acquisition module is used to acquire the heparin sodium detection spectral signal information and the heparin sodium detection electrochemical signal information; The heparin sodium detection signal feature information generation module is used to extract features from the heparin sodium detection spectral signal information and the heparin sodium detection electrochemical signal information to obtain heparin sodium detection spectral signal feature information and heparin sodium detection electrochemical signal feature information. The heparin sodium detection signal optimization feature information generation module is used to generate multiple heparin sodium detection signal optimization feature information based on the heparin sodium detection spectral signal feature information, the heparin sodium detection electrochemical signal feature information, the preset heparin sodium detection signal feature optimization iteration number, and the preset heparin sodium detection signal feature optimization amplitude information. The heparin sodium detection information generation module is used to classify and process multiple optimized feature information of heparin sodium detection signals based on preset heparin sodium detection signal feature classification convergence threshold information to obtain heparin sodium detection information.

7. A terminal device, characterized in that, The terminal device includes a memory and a processor. The memory stores a computer program that can run on the processor. When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 5.

Images