Monitoring method and device and electronic equipment

Abstract

The invention discloses a monitoring method and device and electronic equipment. The method comprises the following steps: preparing a plasma sample; scanning the plasma sample by using a Raman spectrometer, and collecting a molecular vibration spectrum corresponding to the plasma sample and chemical component information of the plasma sample; measuring each target tissue in the plasma sample by using a Raman spectrometer to generate a Raman spectrum image; analyzing the Raman spectrum image in combination with circulating tumor DNA related parameter thresholds, and determining spectral characteristics and image characteristics; and determining the content and purity of the circulating tumor DNA in the plasma sample according to the spectral characteristics and the image characteristics so as to determine the chemotherapy effect. According to the monitoring scheme provided by the invention, the purpose of monitoring the chemotherapy effect in real time can be achieved, the operation is simple, the efficiency is high, and the monitoring result is accurate.

Description

Technical Field

[0001] The present invention relates to the field of biomedical detection technologies, and particularly to a method and device for real-time monitoring of chemotherapy effects using spectral imaging, and an electronic device. Background Art

[0002] The detection of chemotherapy effects is of great significance for the treatment of patients. To ensure the effectiveness and safety of chemotherapy, doctors need to comprehensively use a variety of detection means to accurately evaluate the tumor condition of patients and formulate personalized treatment plans according to the specific conditions of patients. Therefore, it is necessary to design a non-destructive, fast and efficient method for real-time monitoring of chemotherapy effects.

[0003] Nowadays, with the continuous progress of technologies such as imaging, pathology, and blood testing, although doctors can accurately evaluate the chemotherapy effects, there are still the following defects: First, most methods are too cumbersome to operate and will cause pain to patients to a certain extent; Second, the detection time is too long and real-time monitoring cannot be achieved; Third, the inspection methods are costly, inefficient and inaccurate; Fourth, they require a high professional level of technicians.

[0004] The characteristics of ctDNA (circulating tumor DNA) are that it is highly fragmented but relatively consistent in size; ctDNA circulates in the body with a short half-life and higher sensitivity, and can reflect the changes of the entire tumor tissue in real time, tracking the disappearance, spread and recurrence of tumors; and its concentration in tumor patients is extremely low, which is closely related to the clinical stage and degree of differentiation. Compared with tissue / cytology biopsy, the detection of ctDNA is based on peripheral blood, and its advantage lies in non-invasive detection and convenient sampling, solving the problem that some advanced patients do not have sufficient tissue samples. However, the most sensitive ctDNA detection nowadays (such as BEAMing technology) requires prior knowledge of the mutant genes to be detected, through processes such as biopsy, sequencing and molecular probes, or requires prior knowledge of tumor information, with a relatively high cost. Summary of the Invention

[0005] The purpose of the embodiments of the present invention is to provide a monitoring method, device and electronic device, which can solve at least one of the above problems existing in the prior art.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: The embodiments of the present invention provide a monitoring method, wherein the method includes: Prepare a plasma sample; Use a Raman spectrometer to scan the plasma sample, and collect the molecular vibration spectrum corresponding to the plasma sample and the chemical composition information of the plasma sample; Measure each target tissue in the plasma sample using a Raman spectrometer to generate a Raman spectroscopic image; Analyze the Raman spectroscopic image in combination with the circulating tumor DNA - related parameter thresholds to determine spectral features and image features; Based on the spectral features and image features, determine the content and purity of circulating tumor DNA in the plasma sample to determine the chemotherapy effect.

[0007] Optionally, the steps of preparing the plasma sample include: Obtain the blood collected through intravenous injection; After mixing the blood with a blood anticoagulant, collect the plasma through centrifugal filtration; Add a substrate for surface - enhanced Raman scattering to the blood to make a plasma sample.

[0008] Optionally, the steps of using a Raman spectrometer to scan the plasma sample and collect the molecular vibration spectrum corresponding to the serum sample and the chemical composition information of the plasma sample include: Place the plasma sample in a sample cell; Select a laser as the excitation light source for Raman spectroscopic analysis within a preset wavelength range; Set a filter or a physical barrier; Set scanning parameters, where the scanning parameters include: laser - to - sample energy, integration time, and the number of acquisitions for each spectrum; According to the scanning parameters, use the selected Raman spectrometer to perform spectral scanning on the plasma sample and generate a molecular vibration spectrum and the chemical composition information of the plasma sample based on the collected scattered light.

[0009] Optionally, the steps of combining the circulating tumor DNA - related parameter thresholds, analyzing the molecular vibration spectroscopic image, and determining spectral features and image features include: Determine multiple thresholds in combination with the characteristics of circulating tumor DNA; Use the multiple thresholds to segment the image into multiple regions, with each region corresponding to a different chemical composition or structure; Identify the characteristic peaks of the molecular vibration spectroscopic image among the respective thresholds; Based on the identified characteristic peaks, determine spectral features and image features.

[0010] Optionally, the steps of determining spectral features and image features based on the identified characteristic peaks include: For each identified characteristic peak, perform integral calculation on the characteristic peak to calculate the area of the characteristic peak; Measure the width of the characteristic peak; Calculate the ratio between different characteristic peaks; Determine spectral features and image features based on the area, the width, and the ratio.

[0011] Optionally, the step of determining the content and purity of circulating tumor DNA in the plasma sample based on the spectral features and image features to determine the chemotherapy effect includes: Input the spectral features and image features into a pre-trained convolutional neural network, where the convolutional neural network includes: multiple convolutional layers, pooling layers, normalization layers, and fully connected layers; Determine the content and purity of circulating tumor DNA in the plasma sample based on the output of the convolutional neural network to determine the chemotherapy effect.

[0012] Optionally, the step of determining multiple thresholds in combination with the characteristics of circulating tumor DNA includes: Judge whether the Raman spectrum image is uniform; If not, determine each threshold according to the adaptive threshold in the local threshold method and the local characteristics of the image.

[0013] An embodiment of the present invention provides a monitoring device, characterized in that the device includes: A preparation module for preparing a plasma sample; A collection module for scanning the plasma sample using a Raman spectrometer to collect the molecular vibration spectrum corresponding to the plasma sample and the chemical composition information of the plasma sample; A generation module for measuring each target tissue in the plasma sample using a Raman spectrometer to generate a Raman spectrum image; A first determination module for analyzing the Raman spectrum image in combination with the circulating tumor DNA related parameter thresholds to determine spectral features and image features; A second determination module for determining the content and purity of circulating tumor DNA in the plasma sample based on the spectral features and image features to determine the chemotherapy effect.

[0014] Optionally, the preparation module includes: A first sub-module for obtaining blood collected through a vein; A second sub-module for mixing the blood with a blood anticoagulant and collecting plasma through centrifugal filtration; A third sub-module for adding a substrate for surface-enhanced Raman scattering to the blood to make a plasma sample.

[0015] Optionally, the collection module includes: A fourth sub-module for placing the plasma sample in a sample cell; The fifth sub-module is used to select a laser as the excitation light source for Raman spectroscopy analysis within a preset wavelength range; The sixth sub-module is used to set a filter or a physical barrier; The seventh sub-module is used to set scanning parameters, where the scanning parameters include: laser-to-sample energy, integration time, and the number of spectral acquisitions per spectrum; The eighth sub-module is used to perform spectral scanning on the plasma sample using the selected Raman spectrometer according to the scanning parameters, and generate a molecular vibration spectrum and chemical composition information of the plasma sample based on the collected scattered light.

[0016] Optionally, the first determination module includes: The ninth sub-module is used to determine multiple thresholds by combining the characteristics of circulating tumor DNA; The tenth sub-module is used to segment the image into multiple regions using the multiple thresholds, and each region corresponds to a different chemical composition or structure; The eleventh sub-module is used to respectively identify the characteristic peaks of the molecular vibration spectrum image among the thresholds; The twelfth sub-module is used to determine spectral features and image features based on the identified characteristic peaks.

[0017] Optionally, the twelfth sub-module is specifically used for: For each identified characteristic peak, perform integral calculation on the characteristic peak to calculate the area of the characteristic peak; Measure the width of the characteristic peak; Calculate the ratio between different characteristic peaks; Determine spectral features and image features based on the area, the width, and the ratio.

[0018] Optionally, the second determination module is specifically used for: Input the spectral features and image features into a pre-trained convolutional neural network, where the convolutional neural network includes: multiple convolutional layers, pooling layers, normalization layers, and fully connected layers; Determine the content and purity of circulating tumor DNA in the plasma sample based on the output of the convolutional neural network to determine the chemotherapy effect.

[0019] Optionally, the ninth sub-module is specifically used for: Judge whether the Raman spectrum image is uniform; If not, determine each threshold according to the adaptive threshold in the local threshold method and the local characteristics of the image.

[0020] An embodiment of the present invention further provides an electronic device, where the electronic device includes a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete communication with each other through the communication bus; the memory is used to store a computer program; when the processor executes the program stored on the memory, the monitoring method described in any of the above embodiments is implemented.

[0021] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the monitoring method described in any of the above embodiments is implemented.

[0022] The monitoring solution provided by the embodiment of the present invention prepares a plasma sample; uses a Raman spectrometer to scan the plasma sample to collect the molecular vibration spectrum corresponding to the plasma sample and the chemical composition information of the plasma sample; uses a Raman spectrometer to measure each target tissue in the plasma sample to generate a Raman spectrum image; combines the relevant parameter thresholds of circulating tumor DNA to analyze the Raman spectrum image to determine the spectral characteristics and image characteristics; based on the spectral characteristics and image characteristics, determine the content and purity of circulating tumor DNA in the plasma sample to determine the chemotherapy effect. The monitoring method provided by the embodiment of the present invention detects tumor markers (ctDNA) in human blood, processes and processes data information, and then analyzes the detection results through an artificial intelligence algorithm, ultimately achieving the purpose of real-time monitoring of chemotherapy effects, with simple operation, high efficiency, and accurate monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flowchart showing the steps of a monitoring method according to an embodiment of the present application; Figure 2 It is a flowchart showing the steps of a method for real-time monitoring of chemotherapy effects using spectral imaging according to an embodiment of the present application; Figure 3 It is a block diagram showing the structure of a monitoring device according to an embodiment of the present application; Figure 4 It is a block diagram showing the structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] To make the technical problems, technical solutions, and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0025] The present invention proposes a Raman spectroscopy imaging analysis method for real-time monitoring of chemotherapy effects. By processing and analyzing data step by step and then using artificial intelligence algorithms to obtain the detection results, the purpose of real-time monitoring of chemotherapy effects can ultimately be achieved. Raman spectroscopy detection is a light scattering technique. Raman scattering refers to the inelastic collision between incident photons and sample molecules, which causes a change in frequency. Raman spectroscopy detection uses scattered light to obtain information related to molecular vibrations, and molecular vibrations can provide corresponding molecular information such as the molecular structure, symmetry, and chemical bonds of the molecule. Raman spectroscopy detection technology has now been widely applied in fields such as food safety, biomedicine, drug detection, materials science, environmental detection, and jewelry identification.

[0026] The following will, with reference to the accompanying drawings, elaborate in detail on the monitoring solution provided by the embodiments of the present application through specific embodiments and their application scenarios.

[0027] As shown Figure 1 in the accompanying drawings, the monitoring method of the embodiments of the present application includes the following steps: Step 101: Prepare a plasma sample.

[0028] In an optional embodiment, the method for preparing a plasma sample may be as follows: Obtain blood collected through venipuncture; after mixing the blood with a blood anticoagulant, collect plasma through centrifugation and filtration; add a substrate for surface-enhanced Raman scattering to the blood to make a plasma sample.

[0029] Venous blood is stratified into two parts: the upper light yellow plasma and the lower dark red blood cells. In the embodiments of the present application, the venous blood is further filtered to retain the upper plasma part.

[0030] Step 102: Use a Raman spectrometer to scan the plasma sample and collect the molecular vibration spectrum and chemical composition information of the plasma sample.

[0031] In an optional embodiment, the method for using a Raman spectrometer to scan the plasma sample and collect the molecular vibration spectrum and chemical composition information of the serum sample may include the following sub-steps: Sub-step 1: Place the plasma sample in a sample cell; Sub-step 2: Select a laser as the excitation light source for Raman spectroscopy analysis within a preset wavelength range; Sub-step 3: Set a filter or a physical barrier; Setting a filter or a physical barrier can filter the laser before it enters the detector to prevent the plasma sample from being irradiated by external radiation sources.

[0032] Sub-step 4: Set the scanning parameters; Among them, the scanning parameters include: laser energy to the sample, integration time, and the number of acquisitions for each spectrum; Sub-step 5: According to the scanning parameters, use the selected Raman spectrometer to perform spectral scanning on the plasma sample, and generate a molecular vibration spectrum and chemical composition information of the plasma sample based on the collected scattered light.

[0033] Since laser Raman spectroscopy has the same wavelength range as infrared spectroscopy and is relatively simple to operate, and has characteristics such as adjustable light source frequency, good resolution, high resolution, sharp spectral peaks, and small sample consumption, a laser is selected as the excitation light source for Raman spectroscopy analysis within the preset wavelength range in the embodiments of the present application.

[0034] When scanning the plasma sample according to the scanning parameters, the average of 5 acquisitions can be taken to collect the scattered light Raman spectral imaging.

[0035] Step 103: Use a Raman spectrometer to measure each target tissue in the plasma sample to generate a Raman spectral image.

[0036] Step 104: Combine the relevant parameter thresholds of circulating tumor DNA to analyze the Raman spectral image and determine the spectral characteristics and image characteristics.

[0037] ctDNA (circulating tumor DNA) refers to the somatic cell DNA of tumor cells that is shed or released into the circulatory system after cell apoptosis, and it is a characteristic tumor biomarker. Through ctDNA detection, the traces of tumors in the blood can be detected. ctDNA is usually DNA fragments actively secreted by tumor cells or released into the circulatory system during the apoptosis or necrosis of tumor cells, with a length of 132 - 145 bp and a short half-life. The characteristics of the ctDNA half-life are: it can reflect the changes of the entire tumor tissue in real time, track the disappearance, spread, and recurrence of tumors; and its concentration in tumor patients is extremely low and is closely related to the clinical stage and degree of differentiation.

[0038] In an optional embodiment, the manner of combining the relevant parameter thresholds of circulating tumor DNA to analyze the molecular vibration spectral image and determine the spectral characteristics and image characteristics can be as follows: Sub-step 1: Combine the characteristics of circulating tumor DNA to determine multiple thresholds; Optionally, when combining the characteristics of circulating tumor DNA to determine multiple thresholds, it can be judged whether the Raman spectral image is uniform; if not, according to the adaptive threshold in the local threshold method and the local characteristics of the image, each threshold is determined.

[0039] Sub-step 2: Use the multiple thresholds to segment the image into multiple regions, and each region corresponds to a different chemical composition or structure; Sub-step 3: Identify the characteristic peaks of the molecular vibration spectrum image among the respective thresholds; Sub-step 4: Determine the spectral features and image features based on the identified characteristic peaks.

[0040] More specifically, the method for determining the spectral features and image features based on the identified characteristic peaks can be as follows: First, for each of the identified characteristic peaks, perform integral calculation on the characteristic peak to calculate the area of the characteristic peak; Second, measure the width of the characteristic peak; Moreover, the height of the characteristic peak can also be calculated; Third, calculate the ratio between different characteristic peaks; Finally, determine the spectral features and image features based on the area, width, and ratio.

[0041] The ctDNA information can be determined by comparing the peak values of each characteristic peak, and the ctDNA content can be calculated based on the established calibration model.

[0042] Step 105: Determine the content and purity of circulating tumor DNA in the plasma sample based on the spectral features and image features to determine the chemotherapy effect.

[0043] In an optional embodiment, the method for determining the content and purity of circulating tumor DNA in the plasma sample based on the spectral features and image features to determine the chemotherapy effect can be as follows: Input the spectral features and image features into a pre-trained convolutional neural network; determine the content and purity of circulating tumor DNA in the plasma sample based on the output of the convolutional neural network to determine the chemotherapy effect.

[0044] Among them, the convolutional neural network includes: multiple convolutional layers, pooling layers, normalization layers, and fully connected layers.

[0045] The convolutional neural network model adopted in the embodiments of the present application is an artificial intelligence and auxiliary diagnosis model, and the training and application processes of this model are as follows: ① Input the collected image data into the convolutional neural network; ② Construct a CNN architecture, including multiple convolutional layers, pooling layers, normalization layers, and fully connected layers; ③ Divide the collected data into a training set, a validation set, and a test set for model training and evaluation; ④ Apply the model to actual clinical samples for further testing and verification to promote the clinical transformation of the model.

[0046] The main structural characteristics of the convolutional neural network are local area connection, weight sharing, downsampling, etc.

[0047] The monitoring method provided by the embodiments of the present invention includes: preparing a plasma sample; scanning the plasma sample using a Raman spectrometer to collect the molecular vibration spectrum corresponding to the plasma sample and the chemical composition information of the plasma sample; measuring each target tissue in the plasma sample using the Raman spectrometer to generate a Raman spectral image; analyzing the Raman spectral image in combination with the relevant parameter thresholds of circulating tumor DNA to determine the spectral features and image features; and determining the content and purity of circulating tumor DNA in the plasma sample based on the spectral features and image features to determine the chemotherapy effect. The monitoring method provided by the embodiments of the present invention detects tumor markers (ctDNA) in human blood, processes and processes data information, and then analyzes the detection results through an artificial intelligence algorithm, ultimately achieving the purpose of real-time monitoring of chemotherapy effects, with simple operation, high efficiency, and accurate monitoring results.

[0048] The following will illustrate the monitoring method provided by the embodiments of the present application with a specific example. Figure 2 Taking a specific example, the monitoring method provided by the embodiments of the present application will be described.

[0049] In this specific example, a method for detecting tumor markers in plasma by Raman spectral imaging to achieve real-time monitoring of chemotherapy effects is disclosed.

[0050] As Figure 2 shown, this method is executed by the following three modules: ① Data acquisition and processing module, which performs blood collection, plasma collection, surface-enhanced Raman scattering, etc. to prepare a plasma sample for Raman spectral imaging analysis; performs appropriate preprocessing to adapt to Raman spectral analysis, and collects its molecular vibration spectrum and the chemical composition information of the sample; ② Module for detecting the content and purity of ctDNA, which extracts the characteristic peaks related to ctDNA to obtain the content and purity of ctDNA; and establishes a calibration model to enhance the accuracy of quantitative analysis; ③ Artificial intelligence and auxiliary diagnosis module, which uses a convolutional neural network (CNN), namely partial least squares regression (PLSR), and combines relevant clinical information to perform feature learning and classification model training on Raman data. A smart diagnosis model is constructed. The present invention utilizes the characteristics of clear and highly accurate Raman spectroscopy detection to detect relevant data in real time and accurately evaluate the tumor condition of patients.

[0051] ① Data acquisition and processing module: The data acquisition and processing module is to obtain a plasma sample that can be used for Raman spectral imaging analysis; prepares a plasma sample through blood collection, plasma collection, surface-enhanced Raman scattering, etc., and performs appropriate preprocessing to adapt to Raman spectral analysis, and scans the sample using a Raman spectrometer to collect its molecular vibration spectrum and the chemical composition information of the sample.

[0052] The data acquisition and processing module is specifically used for: obtaining plasma samples that can be used for Raman spectroscopic imaging analysis; preparing plasma samples through blood collection, plasma collection, surface-enhanced Raman scattering, etc. Place the preprocessed sample in a sample cell, ensure the stability of the sample and appropriate lighting conditions, and perform appropriate pretreatment to adapt to Raman spectroscopic analysis; obtain human blood samples through intravenous blood collection, mix them with blood anticoagulants, put them into a centrifuge, centrifuge at a speed of 3000 revolutions per minute for about 30 minutes, and store them refrigerated at 2°C; retain the upper plasma part. Noble metal nanostructures (such as Au, Ag, Cu, etc.) can be used as substrates for surface-enhanced Raman scattering (SERS), and the Raman signal can be greatly enhanced through electromagnetic (EM) means. Add the collected plasma samples to the SERS-active nanostructures for molecular adsorption; eliminate the differences between blood samples from different individuals and the influence of the addition of different coagulants on the experimental results through preliminary experiments; Select a laser as the excitation light source for Raman spectroscopic analysis in the wavelength range from ultraviolet, visible to near-infrared. Set appropriate filters or physical barriers to filter the laser before it enters the detector to prevent the sample from being irradiated by external radiation sources; set the laser energy to the sample to be 100 - 200 mW, the integration time to be 10 s, and take the average value of 5 acquisitions for each spectrum; use a Raman spectrometer to scan the serum sample and collect its molecular vibration spectrum and the chemical composition information of the sample; use a Raman spectrometer to measure the target tissue, collect the scattered light, and record the wavelength intensity and position of the Raman scattered light to form a Raman spectroscopic image.

[0053] ② ctDNA content and purity detection module: The ctDNA content and purity detection module is to extract the spectral features and image features from the preprocessed Raman spectroscopic image; through image processing techniques, such as threshold processing, etc., extract the characteristic peaks related to ctDNA from the preprocessed Raman spectroscopic image to obtain the content and purity of ctDNA; and establish a calibration model to enhance the accuracy of quantitative analysis; The ctDNA content and purity detection module is specifically used for: determining one or more thresholds according to the characteristics of the spectrum. These thresholds can be fixed or dynamically calculated based on the data set, and are determined by analyzing the distribution of peak intensities; finding the thresholds related to ctDNA, calculating the spectral features and image features, including peak height, peak width, peak area, etc., to obtain the content and purity of ctDNA; and establishing a calibration model, namely partial least squares regression (PLSR), to enhance the accuracy of quantitative analysis; Design and use of thresholds; the threshold is equivalent to setting a threshold, and chromatographic peaks with a response intensity lower than the threshold will not be collected. Setting the threshold reasonably can effectively purify the chromatogram and reduce interference. Using multiple thresholds to segment the image into multiple regions, each region corresponding to different chemical components or structures; identifying characteristic peaks in the spectrum; ctDNA is usually DNA fragments actively secreted by tumor cells or released into the circulatory system during tumor cell apoptosis or necrosis, with a length of 132 - 145bp and a short half-life; integrating the detected peaks and calculating their areas as one of the characteristics; measuring the width of the peaks, achieved by full width at half maximum (FWHM), and performing peak width analysis to understand the uniformity or crystallinity of the sample; calculating the ratio between different peaks to provide information about the chemical composition of the sample; In particular, for non-uniform images, the adaptive threshold in the local threshold method is adopted to determine the threshold according to the local characteristics of the image.

[0054] By establishing a calibration model, namely partial least squares regression (PLSR), to associate spectral data with standard samples of known concentrations to achieve more accurate quantitative analysis; the specific steps are as follows: 1. Divide the preprocessed data set into a calibration set (for model establishment) and a validation set (for evaluating model performance); 2. Use feature selection techniques, principal component analysis (PCA) to select the features that contribute the most to the model; 3. Use the calibration set data to establish a model through the PLSR algorithm. Determine the optimal PLSR model parameters, such as the number of principal components (or latent variables) 4. Use the validation set data to evaluate the performance of the PLSR model. The performance indicators include the coefficient of determination (R²), root mean square error (RMSE), and correlation coefficient; 5. According to the validation results, adjust the PLSR model parameters, increasing or decreasing the number of principal components to optimize the model performance.

[0055] ③ Artificial intelligence and auxiliary diagnosis module: The artificial intelligence and auxiliary diagnosis module uses a convolutional neural network (CNN) and combines relevant clinical information to perform feature learning and training of a classification model on Raman data. Construct an intelligent diagnosis model to automatically identify the content of measured tumor markers, qualitatively and quantitatively determine the concentration of tumor markers, and monitor the chemotherapy effect in real time.

[0056] Using a convolutional neural network (CNN) to perform feature learning and training of a classification model on Raman spectral data, specifically the following key steps: Step 1: Input the collected image data into the convolutional neural network, convert it into a format suitable for CNN processing, reshape it into a three-dimensional image form, and the intensity of the spectrum can be represented as the pixel value of the image.

[0057] Step 2: The construction process of the CNN model can be as follows: Design a CNN architecture that includes multiple convolutional layers, pooling layers, normalization layers, and fully connected layers. The convolutional layers are responsible for extracting local features of the Raman spectroscopy images, identifying patterns and edges in the spectra; the pooling layers are used to reduce the spatial dimension of the features while increasing the invariance to image variations, which helps reduce the computational amount and improve the generalization ability of the model. And apply a non-linear activation function to introduce non-linearity, enabling the network to learn more complex feature representations; after multiple convolutional and pooling layers, the data is flattened and passed through one or more fully connected layers, which are responsible for mapping the learned features to the final classification labels. Finally, the output layer, i.e., the soft_max layer, outputs a probability distribution for each class, representing the probability that the model predicts the sample belongs to each class.

[0058] Step 3: Split the collected data into a training set, a validation set, and a test set for model training and evaluation. The splitting ratio is 7:1:2, that is, 70% is used for training, 10% is used for validation, and 20% is used for testing. Use the training set data to train the CNN model. During the training process, continuously update the weights of the network to reduce the prediction error. Use the validation set data to validate the model to monitor overfitting and underfitting, and adjust the network structure or parameters as needed to ensure the reliability of the model in practical applications.

[0059] Step 3 is the training and validation process of the CNN model. After the trained CNN model passes the validation, Step 4 can be executed to apply the trained model to actual clinical samples for further testing and validation to promote the clinical transformation of the model.

[0060] The monitoring method provided by the embodiments of the present application can make the Raman spectrum clearer and more accurate; the operation is simpler, more efficient than the existing related methods, and the labor cost is low. Moreover, it can also detect relevant data in real time to accurately evaluate the tumor status of patients.

[0061] Figure 3 It is a structural block diagram of a monitoring device for implementing the embodiments of the present application.

[0062] The monitoring device provided by the embodiments of the present application includes the following functional modules; The preparation module 301 is used to prepare plasma samples; The collection module 302 is used to scan the plasma samples using a Raman spectrometer to collect the molecular vibration spectra corresponding to the plasma samples and the chemical composition information of the plasma samples; The generation module 303 is used to measure each target tissue in the plasma samples using a Raman spectrometer to generate Raman spectroscopy images; The first determination module 304 is configured to analyze the Raman spectrum image in combination with the circulating tumor DNA related parameter thresholds to determine spectral features and image features; The second determination module 305 is configured to determine the content and purity of the circulating tumor DNA in the plasma sample according to the spectral features and image features, so as to determine the chemotherapy effect.

[0063] Optionally, the preparation module includes: The first sub-module is configured to obtain the blood collected through vein; The second sub-module is configured to mix the blood with a blood anticoagulant and then collect plasma through centrifugal filtration; The third sub-module is configured to add a substrate for surface enhanced Raman scattering to the blood to make a plasma sample.

[0064] Optionally, the collection module includes: The fourth sub-module is configured to place the plasma sample in a sample cell; The fifth sub-module is configured to select a laser as the excitation light source for Raman spectrum analysis within a preset wavelength range; The sixth sub-module is configured to set a filter or a physical barrier; The seventh sub-module is configured to set scanning parameters, where the scanning parameters include: laser energy to the sample, integration time, and number of acquisitions for each spectrum; The eighth sub-module is configured to perform spectral scanning on the plasma sample using the selected Raman spectrometer according to the scanning parameters, and generate a molecular vibration spectrum and chemical composition information of the plasma sample based on the collected scattered light.

[0065] Optionally, the first determination module includes: The ninth sub-module is configured to determine multiple thresholds in combination with the characteristics of circulating tumor DNA; The tenth sub-module is configured to use the multiple thresholds to segment the image into multiple regions, and each region corresponds to a different chemical composition or structure; The eleventh sub-module is configured to respectively identify the characteristic peaks of the molecular vibration spectrum image among the thresholds; The twelfth sub-module is configured to determine spectral features and image features according to the identified characteristic peaks.

[0066] Optionally, the twelfth sub-module is specifically configured to: For each identified characteristic peak, perform integral calculation on the characteristic peak to calculate the area of the characteristic peak; Measure the width of the characteristic peak; Calculate the ratio between different characteristic peaks; Determine spectral features and image features based on the area, the width, and the ratio.

[0067] Optionally, the second determination module is specifically configured to: Input the spectral features and image features into a pre-trained convolutional neural network, where the convolutional neural network includes: a plurality of convolutional layers, pooling layers, normalization layers, and fully connected layers; Determine the content and purity of circulating tumor DNA in the plasma sample according to the output of the convolutional neural network, so as to determine the chemotherapy effect.

[0068] Optionally, the ninth sub-module is specifically configured to: Determine whether the Raman spectrum image is uniform; If not, determine each threshold according to the adaptive threshold in the local threshold method and the local characteristics of the image.

[0069] The monitoring device provided by the embodiments of the present application Figure 3 shown can implement Figure 1 each process implemented by the method embodiments, and for the sake of brevity, details are not repeated here.

[0070] The embodiments of the present invention also provide an electronic device, as Figure 4 shown, including a processor 401, a communication interface 402, a memory 403, and a communication bus 404, where the processor 401, the communication interface 402, and the memory 403 communicate with each other through the communication bus 404.

[0071] The memory 403 is used to store a computer program; The processor 401 is configured to implement each method step in the above method embodiments when executing the program stored on the memory 403.

[0072] The communication bus mentioned in the above terminal may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity, only a thick line is shown in the figure, but it does not mean that there is only one bus or one type of bus.

[0073] The communication interface is used for communication between the above terminal and other devices.

[0074] The memory may include a Random Access Memory (RAM), or may also include non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located far from the aforementioned processor.

[0075] The aforementioned processor may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0076] In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that the computer can access, or a data storage device such as a server or data center that includes one or more integrated available media. The available medium may be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a Solid State Disk (SSD)).

[0077] It should be noted that in this text, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the presence of additional identical elements in the process, method, article or device comprising such element.

[0078] The above are the preferred embodiments of the present invention. It should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A monitoring method, characterized in that: The method comprises: Preparation of plasma samples; Scanning the plasma sample using a Raman spectrometer to collect molecular vibration spectra corresponding to the plasma sample and chemical composition information of the plasma sample; Using a Raman spectrometer to measure each target tissue in the plasma sample to generate a Raman spectrum image; Analyze the Raman spectral image in combination with the threshold of circulating tumor DNA related parameters to determine spectral features and image features; Based on the spectral characteristics and image characteristics, the content and purity of circulating tumor DNA in the plasma sample are determined to determine the chemotherapy effect.

2. The method according to claim 1, characterized in that The steps for preparing plasma samples include: Obtain blood collected through a vein; After mixing the blood with a blood anticoagulant, the blood plasma is collected by centrifugal filtration; A surface enhanced Raman scattering substrate is added to the blood to prepare a plasma sample.

3. The method according to claim 1, characterized in that: The step of scanning the plasma sample using a Raman spectrometer to collect the molecular vibration spectrum corresponding to the serum sample and the chemical composition information of the plasma sample comprises: placing the plasma sample in a sample cell; A laser is used as an excitation light source for Raman spectroscopy analysis within a preset wavelength range; Setting up filters or physical barriers; Set scanning parameters, including: laser to sample energy, integration time, and number of acquisitions per spectrum; According to the scanning parameters, a selected Raman spectrometer is used to perform a spectral scan on the plasma sample, and a molecular vibration spectrum and chemical composition information of the plasma sample are generated based on the collected scattered light.

4. The method according to claim 1, characterized in that The step of analyzing the molecular vibration spectrum image in combination with the circulating tumor DNA-related parameter threshold to determine the spectrum characteristics and image characteristics comprises: Combined with circulating tumor DNA characteristics, multiple thresholds were determined; Using the multiple thresholds, segment the image into multiple regions, each region corresponding to a different chemical component or structure; Respectively identifying characteristic peaks of the molecular vibration spectrum image in each of the thresholds; Based on the identified characteristic peaks, the spectral characteristics and image characteristics are determined.

5. The method according to claim 4, characterized in that The steps of determining the spectral features and the image features according to the identified characteristic peaks include: For each identified characteristic peak, integrating the characteristic peak to calculate the area of ​​the characteristic peak; measuring the width of the characteristic peak; Calculate the ratio between different characteristic peaks; A spectral feature and an image feature are determined according to the area, the width and the ratio.

6. The method according to claim 1, characterized in that The step of determining the content and purity of circulating tumor DNA in the plasma sample based on the spectral characteristics and image characteristics to determine the chemotherapy effect includes: Inputting the spectral features and image features into a pre-trained convolutional neural network, wherein the convolutional neural network includes: a plurality of convolutional layers, a pooling layer, a normalization layer and a fully connected layer; The content and purity of circulating tumor DNA in the plasma sample are determined based on the output of the convolutional neural network to determine the chemotherapy effect.

7. The method according to claim 4, characterized in that The steps to determine multiple thresholds based on circulating tumor DNA characteristics include: Determining whether the Raman spectrum image is uniform; If not, the thresholds are determined according to the adaptive threshold in the local threshold method and the local characteristics of the image.

8. A monitoring device, characterized in that: The device comprises: A preparation module, used for preparing plasma samples; A collection module, used to scan the plasma sample using a Raman spectrometer to collect the molecular vibration spectrum corresponding to the plasma sample and the chemical composition information of the plasma sample; A generating module, used to measure each target tissue in the plasma sample using a Raman spectrometer to generate a Raman spectrum image; A first determination module is used to analyze the Raman spectrum image in combination with a threshold of circulating tumor DNA-related parameters to determine spectrum features and image features; The second determination module is used to determine the content and purity of circulating tumor DNA in the plasma sample based on the spectral characteristics and image characteristics to determine the chemotherapy effect.

9. The device according to claim 8, characterized in that The preparation module comprises: The first submodule is used to obtain blood collected through a vein; The second submodule is used for mixing the blood with the blood anticoagulant and collecting the plasma by centrifugal filtration; The third submodule is used to add a surface enhanced Raman scattering substrate into the blood to prepare a plasma sample.

10. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor is used to implement the monitoring method according to any one of claims 1 to 7 when executing the program stored in the memory.

Images