Online monitoring and early warning method and system

Abstract

The application relates to the technical field of brain-computer interfaces, and discloses an online monitoring and early warning method and system. The method comprises the following steps: determining the size of a spherical electrode according to a preset image and placing the spherical electrode into a preset cavity; constructing an electrode impedance feature pair based on the spherical electrode to perform impedance identification; when the impedance identification result is abnormal, sending an early warning information and starting a neural signal acquisition instruction to construct a feature vector of a neural signal and perform neural signal identification; when the neural signal identification result is abnormal, sending an early warning information and starting a liquid extraction instruction to obtain a feature vector of a liquid; and obtaining a current state identification result based on the feature vectors of the electrode impedance, the neural signal and the liquid. The application comprehensively uses impedance, neural signals and tumor cavity liquid modal data to determine the postoperative condition to obtain the current state identification result, and automatically realizes the whole process of marker collection, data analysis and early warning, reduces the invasiveness of surgery and timely adjusts the treatment parameters.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of brain-computer interface, and particularly relates to an online monitoring and early warning method and system. BACKGROUND

[0002] The traditional method is to diagnose by CT or MRI, clinical indicators or in combination with brain tissue impedance, cerebrospinal fluid analysis and the like. Impedance testing needs a separate program by means of multiple puncture needles, and can easily bring additional pain to the user and increase the risk of complications. SUMMARY

[0003] The main purpose of the present application is to provide an online monitoring and early warning method and system, which aims to solve at least one of the above technical problems.

[0004] To achieve the above-mentioned purpose, the present application provides an online monitoring and early warning method, characterized in that it comprises:

[0005] determining the size of the spherical electrode according to a preset image, and placing the spherical electrode into a preset cavity;

[0006] constructing an electrode impedance feature pair based on the spherical electrode to collect electrode impedance, and performing impedance identification on the electrode impedance feature pair;

[0007] when the impedance identification result is abnormal, sending an early warning information and starting a neural signal acquisition instruction to construct a feature vector of the neural signal, and performing neural signal identification on the feature vector of the neural signal;

[0008] when the neural signal identification result is abnormal, sending an early warning information and starting a liquid extraction instruction of the preset cavity to obtain a feature vector of the liquid in the preset cavity;

[0009] performing state identification based on the electrode impedance feature pair, the feature vector of the neural signal and the feature vector of the liquid to obtain a current state identification result.

[0010] In some embodiments, the determination of the size of the spherical electrode according to the preset image and the placement of the spherical electrode into the preset cavity comprise:

[0011] obtaining a preset image, and fusing the preset image to obtain a tomographic fusion image;

[0012] training a preset segmentation model based on the tomographic fusion image to mark the target boundary of each of the tomographic fusion images;

[0013] performing three-dimensional reconstruction based on the tomographic fusion image with the marked target boundary to obtain a head three-dimensional image with a target three-dimensional boundary;

[0014] Determine a spherical electrode based on the head three-dimensional image, and place the spherical electrode into a preset cavity.

[0015] In some embodiments, the determining the spherical electrode based on the head three-dimensional image comprises:

[0016] Determine a volume proportion of a target volume in the head three-dimensional image;

[0017] Obtain a cavity physical space size according to an actual head size of a preset subject and the volume proportion;

[0018] Determine a size and a number of pieces of a lobed electrode according to the cavity physical space size;

[0019] Generate the spherical electrode according to the size and the number of pieces of the lobed electrode.

[0020] In some embodiments, the collecting an electrode impedance feature pair based on the spherical electrode and performing impedance identification on the electrode impedance feature pair comprises:

[0021] When the spherical electrode is placed into the preset cavity, test the electrode impedance of each electrode multiple times based on the spherical electrode;

[0022] Calculate an average value of the electrode impedance of each lobed electrode in one test according to the electrode impedance, and fit the average value of the electrode impedance of each lobed electrode to obtain a slope;

[0023] Form an electrode impedance feature pair according to the average value of the electrode impedance in the first test, the average value of the electrode impedance in the last test, and the slope;

[0024] Train a first classification model based on the electrode impedance feature pair, and perform impedance identification according to the trained first classification model.

[0025] In some embodiments, when the impedance identification result is abnormal, send a warning information and start a neural signal acquisition instruction to construct a feature vector of a neural signal, and perform neural signal identification on the feature vector of the neural signal, comprising:

[0026] When the impedance identification result is abnormal, send a warning information and start a neural signal acquisition instruction to acquire a preset time length signal of each electrode and process the preset time length signal to obtain a two-dimensional neural signal;

[0027] Convert the two-dimensional neural signal to a time domain and a frequency domain respectively to extract a time domain feature set and a frequency domain feature set of each lobed electrode;

[0028] Splice the electrode impedance feature pair, the time domain feature set, and the frequency domain feature set into a first feature vector;

[0029] training a second classification model based on the first feature vector, and performing neural signal recognition according to the trained second classification model.

[0030] In some embodiments, when the neural signal recognition result is abnormal, the method further comprises:

[0031] when the neural signal recognition result is abnormal, sending an early warning information and starting a liquid extraction instruction of a preset cavity to extract the liquid in the preset cavity;

[0032] imaging the liquid to obtain an image of the liquid;

[0033] color quantizing the image to obtain a color quantization value;

[0034] transparency quantizing the image to obtain a pixel value average;

[0035] determining protein quantification and white blood cell count of the liquid;

[0036] taking the color quantization value, the pixel value average, the protein quantification, and the white blood cell count as a feature vector of the liquid.

[0037] In some embodiments, the color quantizing the image to obtain a color quantization value comprises:

[0038] palette controlling the image to obtain a palette image;

[0039] color principal component analyzing the palette image to obtain color features of the palette image;

[0040] eliminating black color features in the color features to obtain updated color features;

[0041] obtaining a three-channel pixel mean value of each color feature in the updated color features, and determining a target color feature corresponding to a median value of the three-channel pixel mean value;

[0042] obtaining a color quantization value according to the target color feature.

[0043] In some embodiments, the transparency quantizing the image to obtain a pixel value average comprises:

[0044] converting the image to a Lab color mode, performing bilateral filtering on an L channel, and converting the L channel to a BGR color mode to obtain a processed image;

[0045] calculating a pixel value average of the processed image.

[0046] In some embodiments, the state recognition based on the electrode impedance feature pair, the feature vector of the neural signal and the feature vector of the liquid obtains a current state recognition result, comprising:

[0047] The electrode impedance feature pair, the feature vector of the neural signal and the feature vector of the liquid corresponding to each lobe-shaped electrode are spliced into a second feature vector;

[0048] A third classification model is trained based on the second feature vector and a preset label to obtain a target level corresponding to a signal collection area of each lobe-shaped electrode;

[0049] A second level target area is determined based on the target level, a preset amplification coefficient, a total area of a preset cavity, a normal tissue area, a first level target area;

[0050] A first level target variance and a second level target variance are determined according to the number of each lobe-shaped electrode;

[0051] A current state recognition result is obtained according to the first level target area, the second level target area, the first level target variance, the second level target variance and a preset weight.

[0052] In addition, to achieve the above-mentioned purpose, the application further provides an online monitoring and early warning system, comprising:

[0053] An electrode determination module is configured to determine the size of a spherical electrode according to a preset image and place the spherical electrode into a preset cavity;

[0054] An impedance recognition module is configured to construct an electrode impedance feature pair based on the electrode impedance collected by the spherical electrode and perform impedance recognition on the electrode impedance feature pair;

[0055] A neural signal recognition module is configured to, when the impedance recognition result is abnormal, send an early warning information and start a neural signal collection instruction to construct a feature vector of a neural signal and perform neural signal recognition on the feature vector of the neural signal;

[0056] A liquid extraction module is configured to, when the neural signal recognition result is abnormal, send an early warning information and start a liquid extraction instruction of a preset cavity to obtain a feature vector of a liquid in the preset cavity;

[0057] A state recognition module is configured to perform state recognition based on the electrode impedance feature pair, the feature vector of the neural signal and the feature vector of the liquid to obtain a current state recognition result. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 It is a flowchart of an embodiment of the online monitoring and early warning method of the application;

[0059] Figure 2 Structure diagram of electrode device involved in the embodiment of the present application;

[0060] Figure 3 Example diagram of size and number of petals of petal-shaped electrode involved in the embodiment of the present application;

[0061] Figure 4 Schematic diagram of petal number of petal-shaped electrode involved in the embodiment of the present application;

[0062] Figure 5 Structure block diagram of an embodiment of the online monitoring and early warning system of the present application;

[0063] Figure 6 Schematic diagram of technical process involved in the embodiment of the present application;

[0064] Figure 7 Schematic diagram of electrode style confirmation unit involved in the embodiment of the present application;

[0065] Figure 8 Schematic diagram of data acquisition unit involved in the embodiment of the present application;

[0066] Figure 9 Schematic diagram of condition monitoring unit involved in the embodiment of the present application;

[0067] Figure 10 Schematic diagram of early warning unit involved in the embodiment of the present application;

[0068] Figure 11 Structure schematic diagram of electronic device of hardware running environment involved in the embodiment of the present application.

[0069] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0070] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0071] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present application are only used to explain the relative positional relationship, movement condition, etc. between components in a certain posture (as shown in the drawings), and if the certain posture changes, the directional indications will also change accordingly.

[0072] In addition, the description related to "first", "second" and the like in the present application is only for the purpose of description, and cannot be understood as indicating or implying the relative importance of the same, or implying the number of the indicated technical features. Therefore, the features defined as "first", "second" can be explicitly or implicitly included at least one of the features. In addition, the technical solutions of various embodiments can be combined with each other, but it must be based on the realization of ordinary skilled in the art, when the combination of technical solutions appears contradictory or unachievable, it should be considered that the combination of technical solutions does not exist, nor in the protection scope required by the present application. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0073] The present application provides an online monitoring and early warning method and system.

[0074] The embodiment of the present application provides an online monitoring and early warning method, referring to Figure 1 , Figure 1 The flowchart of an embodiment of the online monitoring and early warning method of the present application.

[0075] As Figure 1 shown, the online monitoring and early warning method comprises:

[0076] Step S100: determining the size of the spherical electrode according to a preset image, and placing the spherical electrode into a preset cavity;

[0077] Step S200: constructing an electrode impedance feature pair based on the spherical electrode to collect electrode impedance, and performing impedance identification on the electrode impedance feature pair;

[0078] Step S300: when the impedance identification result is abnormal, sending early warning information and starting a neural signal acquisition instruction to construct a feature vector of neural signal, and performing neural signal identification on the feature vector of neural signal;

[0079] Step S400: when the neural signal identification result is abnormal, sending early warning information and starting a liquid extraction instruction of the preset cavity to obtain a feature vector of the liquid of the preset cavity;

[0080] Step S500: performing state identification based on the electrode impedance feature pair, the feature vector of neural signal and the feature vector of the liquid, to obtain a current state identification result.

[0081] It should be noted that the execution subject in the present embodiment can be an electronic device, which can be a computer device with data processing function, and can also be other devices that can realize the same or similar functions, and the present embodiment does not limit this. In the present embodiment, a computer device is taken as an example for description.

[0082] It can be understood that the present embodiment is illustrated by taking glioma postoperative as an example. Glioma postoperative needs to be regularly scanned by head CT or MRI, blood routine and tumor marker, etc. to evaluate the operation effect and monitor tumor recurrence. The method described in the present embodiment is the application of brain-computer interface technology combined with artificial intelligence technology in the field of neurosurgery.

[0083] In an embodiment, the size of the spherical electrode is determined according to a preset image, and the spherical electrode is placed in a preset cavity, comprising: obtaining a preset image, and fusing the preset image to obtain a tomographic fusion image; training a preset segmentation model based on the tomographic fusion image to mark the target boundary of each tomographic fusion image; performing three-dimensional reconstruction based on the tomographic fusion image with the marked target boundary to obtain a head three-dimensional image with a target three-dimensional boundary; determining the spherical electrode based on the head three-dimensional image, and placing the spherical electrode in the preset cavity.

[0084] It should be noted that, as shown in the electrode device, Figure 2 including: a flexible electrode shaft; a plurality of petal-shaped electrodes arranged in turn along the extension direction of the flexible electrode shaft, the petal-shaped electrodes are arranged to form a spherical electrode with a spherical hollow; a sampling tube, the sampling end of the sampling tube extends into the spherical center; a drug delivery tube, the drug delivery end of the drug delivery tube extends into the spherical center cavity. Specifically, as shown on the left side, Figure 2 the spherical electrode is unfolded like a petal, and each petal has a petal-shaped electrode, as shown on the right side, Figure 2 the petal-shaped electrodes are arranged to form a spherical electrode with a spherical hollow. The electrode device is also provided with a sampling tube, and the sampling end of the sampling tube extends into the spherical center. The electrode device is also provided with a drug delivery tube, which introduces drugs from the outside to the tumor cavity liquid in the spherical center.

[0085] For example, after the glioma is removed from the intracranial, the glioma removal leaves a tumor cavity, and the spherical electrode is placed in the tumor cavity. The tumor cavity fluid is obtained through the cerebrospinal fluid acquisition device, i.e. the spherical electrode. The current condition of the patient can be judged based on the electrode impedance and other clinical data. In the present embodiment, the spherical electrode is used to facilitate the deep penetration into the tumor, directly inhibit the tumor cells by electric field, and cooperate with drug delivery at the same time. It is convenient and fast, and can be treated online in time according to the patient's state evaluation.

[0086] Specifically, step S1, electrode style confirmation: step S1.1, tumor boundary (target boundary) confirmation: step S1.1.1: obtaining preoperative preset images such as MRI and CT tomographic images of a preset object (e.g. patient); step S1.1.2: fusing the MRI and CT corresponding single-layer scanning layer images one by one to obtain a tomographic fusion image img MCiHere, the image fusion manner includes but is not limited to a feature matching based method such as SIFT, a deep learning model based method such as IFCNN; step S1.1.3: fusing the images img MCi based on the tomography, and training a preset segmentation model (for example, a glioma segmentation model) to mark the target boundary (for example, a glioma boundary) on each tomography fused image img MCi ; step S1.1.4: performing three-dimensional reconstruction based on the tomography fused image img MCi with the glioma boundary to obtain a head 3D image 3D MC with a tumor three-dimensional boundary (a head three-dimensional image with a target three-dimensional boundary), wherein the three-dimensional reconstruction can be performed by using a commercial software such as VTK.

[0087] In an embodiment, determining a spherical electrode based on the head three-dimensional image includes: determining a volume proportion of the target volume in the head three-dimensional image; obtaining a cavity physical space size according to an actual head size of a preset subject and the volume proportion; determining a size and a number of petals of a petal-shaped electrode according to the cavity physical space size; and generating the spherical electrode according to the size and the number of petals of the petal-shaped electrode.

[0088] Specifically, step S1.2: electrode size selection; step S1.2.1: obtaining a real physical space size V LQ of the tumor cavity based on the volume proportion of the glioma volume to the head 3D image 3D MC with the tumor three-dimensional boundary in step S1.1.4 and an actual head size of the patient; step S1.2.2: selecting a petal-shaped electrode with a proper size and a number of petals according to the real physical size V LQ of the tumor cavity. As shown in FIG. 2, different sizes and different numbers of petals of the petal-shaped electrode, and the spherical electrode is generated according to the selected size and the number of petals of the petal-shaped electrode in the embodiment. Figure 3

[0089] In an embodiment, collecting an electrode impedance to construct an electrode impedance feature pair and performing impedance identification on the electrode impedance feature pair based on the spherical electrode includes: testing the electrode impedance of each electrode based on the spherical electrode multiple times when the spherical electrode is placed in a preset cavity; calculating an average value of the electrode impedance of each petal-shaped electrode in the same test according to the electrode impedance, and fitting the average value of the electrode impedance of each petal-shaped electrode to obtain a slope; constructing the electrode impedance feature pair according to the average value of the electrode impedance in the first test, the average value of the electrode impedance in the last test, and the slope; training a first classification model based on the electrode impedance feature pair, and performing impedance identification according to the trained first classification model. ​

[0090] For example, step S2, data acquisition, includes step S2.1 electrode acquisition and step S2.2 tumor cavity fluid acquisition. After the glioma is removed from the intracranial cavity, the glioma cavity left after removal is filled, and the spherical electrode selected according to step S1.2.2 above is placed into the tumor cavity.

[0091] Specifically, step S2.1, electrode acquisition: Step S2.1.1: Impedance value acquisition: ① After the spherical electrode is placed into the tumor cavity, the impedance of each electrode is tested simultaneously each time the tumor cavity fluid is aspirated; ② If... Figure 4 As shown, the petal-shaped electrode sheets are numbered. Since the spherical electrode is composed of N petal-shaped electrode sheets, the electrode impedance of each electrode is obtained and arranged in chronological order. The average value of the impedance of all electrodes contained in each petal-shaped electrode sheet during the same impedance acquisition is calculated (average electrode impedance). The average impedance of each petal-shaped electrode sheet during the first test is Ω. 1i ③ The slope k of the average value (average electrode impedance) obtained for each lobe electrode is obtained by fitting a straight line using the least squares method. ji ④ Based on the average electrode impedance Ω 1ji Slope k ji The average electrode impedance Ω of the last acquisition lastji Constituent features for list Ωji =(Ω) 1ji ,Ω lastji ,k ji (electrode impedance characteristic pairs); ⑤ Obtain a list of electrode impedance characteristic pairs for multiple patients. Ω The label is set to abnormal or normal. The classification model model1 (the first classification model) is trained. Here, the first classification model includes, but is not limited to, decision trees, random forests, and backpropagation neural networks, so as to perform impedance identification based on the trained first classification model.

[0092] In one embodiment, when the impedance identification result is abnormal, a warning message is sent and a neural signal acquisition command is initiated to construct a feature vector of the neural signal. The neural signal feature vector is then used for neural signal identification, including: when the impedance identification result is abnormal, sending a warning message and initiating a neural signal acquisition command to acquire a preset duration signal for each electrode and processing the preset duration signal to obtain a two-dimensional neural signal; converting the two-dimensional neural signal to the time domain and frequency domain respectively to extract the time domain feature set and frequency domain feature set of each valve electrode; concatenating the electrode impedance feature pairs, the time domain feature set, and the frequency domain feature set into a first feature vector; training a second classification model based on the first feature vector; and performing neural signal identification based on the trained second classification model.

[0093] Specifically, step S2.1.2: nerve signal acquisition: ① When the first classification model model1 identifies as abnormal, collect impedance synchronous start nerve signal acquisition instruction, measure and collect the signal of each electrode for a time length (preset time length signal), and store according to the petal, to constitute W ji = (w = Δt, h = m) two-dimensional signal (two-dimensional nerve signal); ② For each two-dimensional signal (two-dimensional nerve signal), the features in the time domain can be considered, such as kurtosis, skewness, maximum value, minimum value, mean value, median, mean absolute error, and root mean square, etc. Time domain features, get the time domain feature set list Tji of each petal electrode; ③ Consider converting the signal (two-dimensional nerve signal) to the frequency domain, extracting the frequency spectrum maximum power, power spectrum bandwidth, fundamental frequency, maximum peak value, frequency domain kurtosis, frequency domain skewness, etc. Frequency domain features, get the frequency domain feature set list Pji of each petal electrode; ④ Concatenate the electrode impedance feature pair list Ωji , the time domain feature set list Tji , and the frequency domain feature set list Pji into a one-dimensional feature vector (first feature vector), collect multiple patients, and train a classification model model2 (second classification model) with the region condition (normal / abnormal) corresponding to each petal electrode at different times as the label, to get the region condition of each petal electrode acquisition signal region. Here, the second classification model includes but is not limited to decision tree, random forest, and BP neural network, etc.

[0094] In an embodiment, when the nerve signal recognition result is abnormal, send a warning information and start a liquid extraction instruction of a preset cavity to obtain a feature vector of the liquid in the preset cavity, comprising: when the nerve signal recognition result is abnormal, send a warning information and start a liquid extraction instruction of a preset cavity to extract the liquid in the preset cavity; image the liquid to obtain an image of the liquid; color quantization is performed on the image to obtain a color quantization value; the transparency of the image is quantified to obtain the average value of the pixel value; determine the protein quantification and white blood cell count of the liquid; the color quantization value, the average value of the pixel value, the protein quantification and the white blood cell count are taken as the feature vector of the liquid.

[0095] It can be understood that, according to the impedance recognition result (first classification model model1 recognition result), it is determined whether to start nerve signal acquisition, according to the impedance recognition result (first classification model model1 recognition result) and the nerve signal recognition result (second classification model model2 recognition result), it is determined whether to extract the tumor cavity liquid.

[0096] Exemplarily, step S2.2: tumor cavity fluid acquisition: step S2.2.1, when the second classification model model2 identifies as abnormal, start the instruction of extracting tumor cavity fluid, and acquire cerebrospinal fluid through the sampling tube; step S2.2.2, take the first extraction of tumor cavity fluid as the benchmark, and extract a certain amount of tumor cavity fluid L1 (ensure that the first extraction of tumor cavity fluid is not given drugs); step S2.2.3, tumor cavity fluid routine examination, including the following contents: 1) color quantification; 2) transparency quantification; 3) regularly extract a certain amount of tumor cavity fluid through the sampling tube, and obtain color i and Color i ; 4) 5) protein quantification label 3i ; 6) white blood cell count label 4i .

[0097] In an embodiment, the image is color quantified to obtain a color quantification value, including: performing palette control on the image to obtain a palette image; performing color principal component analysis on the palette image to obtain color features of the palette image; removing black color features in the color features to obtain updated color features; obtaining a three-channel pixel mean value of each color feature in the updated color features, and determining a target color feature corresponding to a median value of the three-channel pixel mean value; and obtaining a color quantification value according to the target color feature.

[0098] Specifically, 1) color quantification: ① palette control, imaging and photographing a certain amount of tumor cavity fluid L1 to obtain an initial image img 11 , converting the initial image img 11 to a P color mode, increasing color dithering, and then controlling the number of palette colors to express the picture by a certain number of color features to obtain an image img P11 . In this embodiment, the number of palette color control can be set to 10. ② Color principal component analysis, all color features color_list = [(r1, g1, b1), (r2, g2, b2)…(r n ,g n ,b n )] in the image img P11 can be obtained by the getcolors() method of PIL, and new color features color_list_new = [(r1, g1, b1), (r2, g2, b2)…(r i ,g i ,b i) (updated color feature); ③ calculate the three-channel pixel mean value of each color feature in the updated color feature color_list_new, get the median value corresponding to the color feature in all pixel mean values (three-channel pixel mean value) (r m1 ,g m1 ,b m1 ) (target color feature), and get the color quantization value m1 ,g m1 ,b m1 ) through the target color feature (r L12 L12 12 12

[0099] In an embodiment, the transparency quantization of the image is performed to obtain the average value of the pixel value, comprising: converting the image to Lab color mode, and converting the L channel to BGR color mode after bilateral filtering to obtain a processed image; calculating the average value of the pixel value of the processed image.

[0100] Specifically, 2) transparency quantization: ① pour a certain amount of tumor cavity fluid L1 into a transparent glass beaker for the first time, irradiate one end with visible light with luminance a, and image the other end with a camera to obtain an image img L12 ; ② convert the image img L12 from BGR to Lab color mode, and convert the L channel to BGR color mode after bilateral filtering to obtain an image L 12 -img; ③ calculate the average pixel value Color1 of the image L 12 -img.

[0101] It should be noted that here, Lab color mode is a mode that simulates the recognition of colors by the human eye, which has three kinds of photoreceptor cells, one for recognizing brightness (L), one for distinguishing red and green, and one for distinguishing yellow and blue. In Lab color mode, there are three channels, namely brightness channel (L channel), a channel and b channel. Bilateral filtering is a non-linear filtering method that is a compromise between image spatial proximity and pixel value similarity. It simultaneously considers spatial information and gray similarity to achieve the purpose of edge preservation and denoising. The bilateral filter can achieve good edge preservation while smoothing and denoising, because the kernel of the filter is generated by two functions: one function determines the coefficients of the filter template by the pixel Euclidean distance, and the other function determines the coefficients of the filter by the pixel gray difference. The principle of bilateral filtering is as follows:

[0102]

[0103] wherein g(i,j) represents an output point; S(i,j) refers to a size range of (2N+1)(2N+1) centered at (i,j); f(k,l) represents an input point(s); w(i,j,k,l)=w s *w r , w s is a spatial proximity Gaussian function, w r is a pixel value similarity Gaussian function,

[0104] It can be understood that the bilateral filter is controlled by 3 parameters: filter half-width N, parameters δ s and δ r . The greater the filter half-width N, the stronger the smoothing effect; parameters δ s and δ r control the attenuation degree of spatial proximity factor w s and brightness similarity factor w r respectively. Here, in the embodiment, N=5, δ r =200, δ s =200 can be set to calculate the average pixel value Color1 of the image L 12 -img.

[0105] In an embodiment, the state recognition is performed based on the electrode impedance feature pair, the feature vector of the neural signal, and the feature vector of the liquid to obtain a current state recognition result, including: splicing the electrode impedance feature pair corresponding to each lobe-shaped electrode, the feature vector of the neural signal, and the feature vector of the liquid into a second feature vector; training a third classification model based on the second feature vector and a preset label to obtain a target level corresponding to a signal collection region of each lobe-shaped electrode; determining a second level target area based on the target level, a preset amplification coefficient, a total area of a preset cavity, a normal tissue area, and a first level target area; determining a first level target variance and a second level target variance according to the number of each lobe-shaped electrode; and obtaining the current state recognition result according to the first level target area, the second level target area, the first level target variance, the second level target variance, and a preset weight.

[0106] Exemplarily, the step S3, condition monitoring, includes: mode recognition, area determination, and level determination.

[0107] Specifically, step S3.1: mode recognition: step S3.1.1, the label 1i , label 2i , protein quantification label 3i , white blood cell count label 4i , electrode impedance feature pair listΩji temporal feature set list Tji frequency feature set list Pji concatenated into a one-dimensional feature vector (second feature vector) by concat(), wherein, color quantization value Color1 is the average pixel value. In step S3.1.2, a plurality of patients are collected and a classification model (third classification model) is trained with the label of the state of the glioma (normal brain tissue / low-grade glioma / high-grade glioma) of the region corresponding to each lobe-shaped electrode at different time points to obtain the glioma grade of the signal collection region of each lobe-shaped electrode. In this embodiment, the third classification model includes but is not limited to decision tree, random forest, BP neural network, etc.

[0108] Specifically, in step S3.2, area determination: if a certain lobe-shaped electrode piece is identified as high-grade glioma, only a few channel signals contained therein are extremely abnormal; but if a certain lobe-shaped electrode piece is identified as normal brain tissue, all the channel signals contained therein are normal. From this point of view, when a certain lobe-shaped electrode piece is identified as normal brain tissue, the corresponding tumor cavity avoidance area is appropriately enlarged; on the contrary, the tumor cavity avoidance area corresponding to high-grade glioma is appropriately reduced, and the lobe-shaped electrode piece corresponding to low-grade glioma corresponds to the remaining tumor cavity wall. Here, the tumor cavity avoidance area refers to a surgical resection area usually planned in brain tumor surgery, which includes the tumor itself and a small part of the surrounding normal tissue (referred to as safety margin or avoidance area) to prevent tumor recurrence. If a lobe-shaped electrode piece is identified as normal brain tissue, the corresponding tumor cavity avoidance area should be appropriately enlarged when planning surgery to ensure that enough normal tissue is removed as a safety margin. Conversely, if a lobe-shaped electrode piece is identified as high-grade glioma, the corresponding tumor cavity avoidance area can be appropriately reduced because this area is considered to have a high risk of tumor and needs to be removed as much as possible.

[0109] In this embodiment, the normal brain tissue amplification coefficient is α, the high-grade glioma amplification coefficient is β, the total area of the tumor cavity is S (the total area of the preset cavity), the area occupied by the normal brain tissue is s0 (the area of the normal tissue), the area occupied by the high-grade glioma is s2 (the first target area), and the area occupied by the low-grade glioma (the second target area) is s1=S-α·s0-β·s2; wherein α>1, for example, it can be set to 1.2 / 1.3, etc.; β<1, for example, it can be set to 0.7 / 0.65, etc.

[0110] Specifically, step S3.3: grade determination, step S3.3.1: number each high-grade glioma petal-shaped electrode piece and obtain the variance std2, and number each low-grade glioma petal-shaped electrode piece and obtain the variance std1 according to the result of step S2.1; step S3.3.2: obtain the current state recognition result according to the first grade target area s2, the second grade target area s1, the first grade target variance std2, the second grade target variance std1, the total area S of the preset cavity, and a preset weight, the disease condition (current state recognition result)

[0111]

[0112] In an example, step S4: early warning includes four functions: impedance value early warning, neural signal early warning, tumor cavity fluid early warning, and disease condition (state) early warning.

[0113] Specifically, step S4.1: impedance value early warning, when the identification result of the first classification model model1 indicates that the impedance value is abnormal, an early warning is issued, exemplarily, the early warning information "glioma may have recurrence, please turn on the spherical electrode neural signal acquisition function" is issued, and the neural signal is synchronously acquired while the impedance is acquired.

[0114] Specifically, step S4.2: neural signal early warning, step S4.2.1: when the second classification model model2 indicates that the neural signal is abnormal (for example, when the neural signal recognition result is abnormal), an early warning is issued, exemplarily, the early warning information "glioma may have recurrence, please turn on the collection tube collection function" is issued; step S4.2.2: according to the real physical space size (cavity physical space size) V LQ and the current disease condition (current state recognition result) le, the tumor cavity fluid extraction amount L i and the collection speed v i ; step S4.2.3: according to the collection speed v it ≤ 0.8 × v i , it is predicted that the tumor cavity fluid capacity is too small, and an early warning (tumor cavity fluid early warning) should be issued; the tumor cavity fluid extraction amount L it ≥ 0.9 × L i , it is predicted that the tumor cavity fluid collection task is about to be completed, and a reminder should be issued.

[0115] Specifically, step S4.3: disease condition early warning, step S4.3.1: when the ratio of the high-grade glioma occupied area s2 to the total tumor cavity area S is greater than or equal to a preset parameter, that is, , an early warning signal is issued, where ε is an empirical preset parameter, which can be set by the user, and the present embodiment does not limit it.

[0116] In the embodiment, the postoperative condition of glioma is comprehensively determined by integrating various modal data such as electrode impedance, neural signals and tumor cavity fluid, and the current state recognition result is obtained, and the whole process of marker collection, data analysis and early warning is automated. According to the impedance recognition result, it is determined whether to start the neural signal collection. According to the impedance recognition result combined with the neural signal recognition result, it is determined whether to extract the tumor cavity fluid. For online monitoring of postoperative conditions, real-time monitoring and early warning are realized by combining computer processing with various modal data. In addition, the size of the new spherical electrode proposed in the embodiment can be customized according to the size of the tumor cavity; the recognition of normal tissue, low-grade glioma and high-grade glioma is divided into units by each lobe-shaped electrode piece, and the final condition is considered based on all electrode pieces.

[0117] It should be noted that the online monitoring of postoperative conditions of glioma, combined with computer processing, realizes real-time monitoring and early warning through various modal data. Real-time monitoring helps to reduce unnecessary treatment and examination, thereby reducing the overall treatment cost. Compared with impedance testing through multiple puncture needle channels, using spherical electrodes combined with computer processing system for online treatment and real-time monitoring, since no additional puncture step is needed, the treatment process can be more rapid, reducing the treatment time of patients, and accordingly reducing the related complications such as infection and bleeding. Without multiple puncture of brain tissue like puncture needle channel, the invasiveness of surgery is reduced, which can reduce the damage to normal brain tissue, reduce the side effects of treatment, and improve the tolerance of patients. The computer processing system can monitor the electric field distribution and changes in tumor tissue in real time, providing higher monitoring accuracy and real-time feedback, which helps to adjust the treatment parameters in time. In addition, real-time monitoring can detect abnormal conditions that may occur in time, such as damage to normal brain tissue caused by improper electrode position or excessive electric field strength, so as to adjust the treatment plan in time and ensure the safety of treatment. Real-time monitoring by spherical electrodes combined with computer processing provides a more accurate, safe and efficient method for postoperative treatment of glioma.

[0118] The embodiment determines the size of the spherical electrode according to a preset image, and places the spherical electrode into a preset cavity; constructs an electrode impedance feature pair based on the electrode impedance collected by the spherical electrode, and performs impedance identification on the electrode impedance feature pair; when the impedance identification result is abnormal, sends a warning information and starts a neural signal collection instruction to construct a feature vector of the neural signal, and performs neural signal identification on the feature vector of the neural signal; when the neural signal identification result is abnormal, sends a warning information and starts a liquid extraction instruction of the preset cavity to obtain a feature vector of the liquid in the preset cavity; and performs state identification based on the electrode impedance feature pair, the feature vector of the neural signal and the feature vector of the liquid, to obtain a current state identification result. In the embodiment, the current state identification result is obtained by comprehensively determining the postoperative condition of the glioma through multiple modal data such as electrode impedance, neural signal and tumor cavity liquid, and the whole process of marker collection, data analysis and early warning is automated, which helps to reduce the invasiveness of surgery, facilitate timely adjustment of treatment parameters, reduce unnecessary treatment and examination, and thus reduce the overall treatment cost.

[0119] In addition, the embodiment of the present application also provides a storage medium, wherein the storage medium stores an online monitoring and early warning program, and the online monitoring and early warning program realizes the steps of the online monitoring and early warning method when executed by a processor.

[0120] Reference Figure 5 , Figure 5 is a structural block diagram of an embodiment of the online monitoring and early warning system of the present application.

[0121] As Figure 5 shown, the online monitoring and early warning system comprises:

[0122] The electrode determination module 10 is configured to determine the size of the spherical electrode according to a preset image, and place the spherical electrode into a preset cavity;

[0123] The impedance identification module 20 is configured to construct an electrode impedance feature pair based on the electrode impedance collected by the spherical electrode, and perform impedance identification on the electrode impedance feature pair;

[0124] The neural signal identification module 30 is configured to, when the impedance identification result is abnormal, send a warning information and start a neural signal collection instruction to construct a feature vector of the neural signal, and perform neural signal identification on the feature vector of the neural signal;

[0125] The liquid extraction module 40 is configured to, when the neural signal identification result is abnormal, send a warning information and start a liquid extraction instruction of the preset cavity to obtain a feature vector of the liquid in the preset cavity;

[0126] The state recognition module 50 is configured to perform state recognition based on the electrode impedance feature pair, the feature vector of the neural signal, and the feature vector of the liquid to obtain a current state recognition result.

[0127] For example, referring to the technical process shown in FIG. 1, the online monitoring and early warning system can be used for online monitoring of the postoperative condition of glioma, and the online monitoring and early warning system can include an electrode style confirmation unit, a data acquisition unit, a condition confirmation unit, and an early warning unit. Figure 6

[0128] For example, as shown in FIG. 2, the electrode style confirmation unit can include a tumor boundary confirmation module and an electrode size selection module. Figure 7

[0129] In an embodiment, the electrode determination module 10 can include a tumor boundary confirmation module and an electrode size selection module. The electrode determination module 10 is specifically configured to acquire a preset image and fuse the preset image to obtain a tomographic fusion image; train a preset segmentation model based on the tomographic fusion image to mark a target boundary of each of the tomographic fusion images; perform three-dimensional reconstruction based on the tomographic fusion image with the marked target boundary to obtain a head three-dimensional image with a target three-dimensional boundary; determine a spherical electrode based on the head three-dimensional image, and place the spherical electrode into a preset cavity.

[0130] In an embodiment, the electrode size selection module is configured to determine a volume proportion of a target volume in the head three-dimensional image; obtain a cavity physical space size according to an actual head size of a preset subject and the volume proportion; determine a size and a number of petals of a petal-shaped electrode according to the cavity physical space size; and generate a spherical electrode according to the size and the number of petals of the petal-shaped electrode.

[0131] For example, as shown in FIG. 3, the data acquisition unit can include an electrode acquisition module and a tumor cavity liquid acquisition module. Figure 8

[0132] In an embodiment, the electrode acquisition module can include an impedance recognition module 20, which is specifically configured to, when the spherical electrode is placed into the preset cavity, test the electrode impedance of each electrode based on the spherical electrode multiple times; calculate the average value of the electrode impedance of each petal-shaped electrode in the same test according to the electrode impedance, and fit the average value of the electrode impedance of each petal-shaped electrode to obtain a slope; construct an electrode impedance feature pair according to the average value of the electrode impedance in the first test, the average value of the electrode impedance in the last test, and the slope; train a first classification model based on the electrode impedance feature pair, and perform impedance recognition according to the trained first classification model.

[0133] ​​​In one embodiment, the electrode acquisition module may include a neural signal recognition module 30, which is specifically used for: sending a warning message and initiating a neural signal acquisition command when the impedance recognition result is abnormal, to acquire a preset duration signal for each electrode and process the preset duration signal to obtain a two-dimensional neural signal; converting the two-dimensional neural signal to the time domain and frequency domain respectively to extract the time domain feature set and frequency domain feature set of each lobe electrode; concatenating the electrode impedance feature pair, the time domain feature set, and the frequency domain feature set into a first feature vector; training a second classification model based on the first feature vector, and performing neural signal recognition according to the trained second classification model.

[0134] In one embodiment, the tumor cavity fluid acquisition module is specifically used for: sending an early warning message and initiating a fluid extraction command for a preset cavity to extract fluid from the preset cavity when the neural signal recognition result is abnormal; imaging the fluid to obtain an image of the fluid; performing color quantization on the image to obtain a color quantization value; performing transparency quantization on the image to obtain an average pixel value; determining the protein quantification and white blood cell count of the fluid; and using the color quantification value, average pixel value, protein quantification, and white blood cell count as the feature vector of the fluid.

[0135] In one embodiment, the tumor fluid acquisition module performs color quantization, specifically including: performing palette control on the image to obtain a palette image; performing principal component analysis on the palette image to obtain color features of the palette image; removing black color features from the color features to obtain updated color features; obtaining the three-channel pixel mean of each color feature in the updated color features, determining the target color feature corresponding to the median of the three-channel pixel mean; and obtaining a color quantization value based on the target color feature.

[0136] In one embodiment, the tumor fluid acquisition module performs transparency quantization, specifically including: converting the image to Lab color mode, performing bilateral filtering on the L channel, and then converting it to BGR color mode to obtain a processed image; calculating the average pixel value of the processed image.

[0137] For example, such as Figure 9 As shown, the condition confirmation / monitoring unit (status recognition module 50) may include a pattern recognition module, an area determination module, and a grade determination module.

[0138] In an embodiment, the state recognition module 50 is specifically configured to: (a mode recognition module) splice the electrode impedance feature pair corresponding to each petal-shaped electrode, the feature vector of the neural signal, and the feature vector of the liquid into a second feature vector; train a third classification model based on the second feature vector and a preset label to obtain a target level corresponding to a signal collection area of each petal-shaped electrode; (an area determination module) determine a second level target area based on the target level, a preset amplification coefficient, a total area of a preset cavity, a normal tissue area, and a first level target area; (a level determination module) determine a first level target variance and a second level target variance according to the number of each petal-shaped electrode; and obtain a current state recognition result according to the first level target area, the second level target area, the first level target variance, the second level target variance, and a preset weight.

[0139] As shown in Figure 10 , the early warning unit can include an impedance value early warning module, a neural signal early warning module, a tumor cavity liquid early warning module, and a condition early warning module.

[0140] Specifically, the impedance value early warning module is configured to: when the identification result of the first classification model model1 prompts an abnormal impedance value, issue a warning, for example, issue the early warning information "glioma may have recurrence, please turn on the spherical electrode neural signal collection function", and collect the impedance value while synchronously collecting the neural signal.

[0141] Specifically, the neural signal early warning module is configured to: when the second classification model model2 prompts an abnormality (for example, when the neural signal identification result is abnormal), issue a warning, for example, issue the early warning information "glioma may have recurrence, please turn on the collection tube collection function"; and step S4.2.2, according to the real physical space size of the tumor cavity (cavity physical space size) V LQ and the current condition (current state recognition result) le, determine the tumor cavity liquid extraction amount L i and the collection speed v i ; the tumor cavity liquid early warning module is configured to: according to the collection speed v it ≤ 0.8 x v i , it is predicted that the tumor cavity liquid capacity is too small, and a warning (tumor cavity liquid early warning) should be issued; and the tumor cavity liquid extraction amount L it ≥ 0.9 x L i , it is predicted that the tumor cavity liquid collection task will be completed soon, and a reminder should be issued.

[0142] Specifically, the condition early warning module is configured to: when the ratio of the high-grade glioma occupied area s2 to the total tumor cavity area S is greater than or equal to a preset parameter, that is, At this time, a high attention warning signal is sent, wherein ε is a preset parameter set empirically, which can be set by the user, and the embodiment does not limit this.

[0143] The embodiment provides an online monitoring and warning system. The system determines a current state recognition result of a postoperative condition of a glioma by comprehensively determining multiple modal data such as electrode impedance, nerve signals and tumor cavity fluid, and automatically realizes a whole process of marker collection, data analysis and warning, realizes online monitoring of the postoperative condition of the glioma, facilitates timely adjustment of treatment parameters, helps to reduce unnecessary treatment and examination, and thus reduces overall treatment cost.

[0144] It should be noted that technical details not described in detail in the embodiment of the online monitoring and warning system can refer to the application of the online monitoring and warning method as described above, provided by any embodiment of the application, which will not be described here.

[0145] Reference Figure 11 , Figure 11 The electronic device structure schematic diagram of the hardware running environment related to the embodiment scheme of the application is shown in FIG. 1.

[0146] As Figure 11 shown, the electronic device can include a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to realize the connection and communication between the components. The user interface 1003 can include a display screen (Display) and an input unit such as a keyboard (Keyboard). The optional user interface 1003 can also include a standard wired interface and a wireless interface. The network interface 1004 can optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 can be a high-speed random access memory (RAM memory) or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. The memory 1005 can also be a storage device independent of the aforementioned processor 1001.

[0147] Those skilled in the art can understand Figure 11 that the structure shown in the figure does not constitute a limitation on the electronic device, and can include more or fewer components than the figure, or combine certain components, or different component arrangements.

[0148] As Figure 11As shown, the memory 1005 as a storage medium can include an operating system, a network communication module, a user interface module, and an online monitoring and early warning program.

[0149] In Figure 11 In the electronic device shown, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the electronic device of the present application can be arranged in the electronic device, and the electronic device calls the online monitoring and early warning program stored in the memory 1005 through the processor 1001, and executes the online monitoring and early warning method provided by the embodiment of the present application.

[0150] It should be understood that the above is only for illustration, and does not constitute any limitation on the technical solutions of the present application. In specific applications, those skilled in the art can set it up according to the needs, and the present application does not limit it.

[0151] It should be noted that the above-described workflow is only illustrative and does not limit the scope of protection of the present application. In actual application, those skilled in the art can select part or all of them according to actual needs to achieve the purpose of the embodiment, which is not limited here.

[0152] In addition, it should be noted that in this paper, the term "includes", "contains" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or system including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or includes elements inherent to such process, method, article or system. Without more limitations, the element defined by the sentence "includes a" does not exclude the presence of other identical elements in the process, method, article or system including the element.

[0153] The above-mentioned serial numbers of the embodiments of the present application are only for description, not representing the advantages and disadvantages of the embodiments.

[0154] Those skilled in the art can clearly understand the above-mentioned embodiment method can be realized by means of software and the necessary general hardware platform, of course, can also be through hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application essentially or say the part of the prior art contribution can be embodied in the form of software products, the computer software product is stored in a storage medium (such as read only memory (Read Only Memory, ROM) / RAM, disk, optical disk), including a number of instructions to make a terminal device (may be a mobile phone, computer, server, or network equipment, etc.) executes the method described in various embodiments of the present application.

[0155] The above is only the preferred embodiment of the present application, not therefore limit the patent scope of the present application, any equivalent structure or equivalent process transformation using the content of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. An online monitoring and early warning system, characterized in that, The method comprises the following steps: An electrode determination module is used to determine the size of a spherical electrode according to a preset image and place the spherical electrode into a preset cavity; An impedance identification module is used to identify the impedance of the spherical electrode based on the spherical electrode to construct an electrode impedance feature pair and identify the impedance of the electrode impedance feature pair; A neural signal identification module is used to send a warning message and start a neural signal acquisition instruction to construct a feature vector of a neural signal when the impedance identification result is abnormal, and identify the neural signal based on the feature vector of the neural signal; A liquid extraction module is used to send a warning message and start a liquid extraction instruction of the preset cavity to obtain a feature vector of the liquid in the preset cavity when the neural signal identification result is abnormal; A state identification module is used to identify the state based on the electrode impedance feature pair, the feature vector of the neural signal, and the feature vector of the liquid to obtain a current state identification result; The method of determining the spherical electrode comprises the following steps: Determining the volume proportion of a target volume in a head three-dimensional image; 2. The system of claim 1, wherein, Obtaining the physical space size of the cavity based on the actual size of the head of the preset object and the volume proportion; Determining the size and the number of petals of the petal-shaped electrode based on the physical space size of the cavity; Generating the spherical electrode based on the size and the number of petals of the petal-shaped electrode. When the neural signal identification result is abnormal, the method of sending a warning message and starting a liquid extraction instruction of the preset cavity to obtain a feature vector of the liquid in the preset cavity comprises the following steps: When the neural signal identification result is abnormal, a warning message is sent and a liquid extraction instruction of the preset cavity is started to extract the liquid in the preset cavity; 3. The system of claim 1, wherein, An image of the liquid is obtained by imaging the liquid; A color quantization value is obtained by quantifying the color of the image; A pixel value average is obtained by quantifying the transparency of the image; Protein quantification and white blood cell count of the liquid are determined; The color quantization value, the pixel value average, the protein quantification, and the white blood cell count are used as the feature vector of the liquid. The method of determining the size of the spherical electrode according to the preset image and placing the spherical electrode into the preset cavity comprises the following steps: Obtaining a preset image and fusing the preset image to obtain a tomographic fusion image; Training a preset segmentation model based on the tomographic fusion image to mark the target boundary of each tomographic fusion image; Three-dimensional reconstruction is performed based on the tomographic fusion image with the marked target boundary to obtain a head three-dimensional image with a target three-dimensional boundary; The spherical electrode is determined based on the head three-dimensional image and placed into the preset cavity. The method of constructing an electrode impedance feature pair based on the spherical electrode and identifying the impedance of the electrode impedance feature pair comprises the following steps: When the spherical electrode is placed into the preset cavity, the electrode impedance of each electrode is tested multiple times based on the spherical electrode; The average electrode impedance of each petal-shaped electrode is calculated based on the electrode impedance, and the average electrode impedance of each petal-shaped electrode is fitted to obtain a slope; The electrode impedance feature pair is constructed based on the average electrode impedance of the first test, the average electrode impedance of the last test, and the slope. Train a first classification model based on the electrode impedance feature, and perform impedance recognition according to the trained first classification model.

4. The system of claim 1, wherein, When the impedance recognition result is abnormal, send a warning message and start a neural signal acquisition instruction to construct a feature vector of the neural signal, and perform neural signal recognition on the feature vector of the neural signal, including: When the impedance recognition result is abnormal, send a warning message and start a neural signal acquisition instruction to collect a preset time length signal of each electrode and process the preset time length signal to obtain a two-dimensional neural signal; Convert the two-dimensional neural signal into time domain and frequency domain, respectively, to extract a time domain feature set and a frequency domain feature set of each petal-shaped electrode; Concatenate the electrode impedance feature pair, the time domain feature set and the frequency domain feature set into a first feature vector; Train a second classification model based on the first feature vector, and perform neural signal recognition according to the trained second classification model.

5. The system of claim 1, wherein, The color quantization of the image includes: Performing palette control on the image to obtain a palette image; Performing color principal component analysis on the palette image to obtain color features of the palette image; Eliminate black color features in the color features to obtain updated color features; Obtain the three-channel pixel mean value of each color feature in the updated color features, and determine the target color feature corresponding to the median value of the three-channel pixel mean value; Obtain the color quantization value according to the target color feature.

6. The system of claim 1, wherein, The transparency quantization of the image includes: Convert the image to Lab color mode, and convert the L channel to BGR color mode after bilateral filtering to obtain a processed image; Calculate the pixel value average of the processed image.

7. The system of any one of claims 1 to 6, wherein, The state recognition based on the electrode impedance feature pair, the feature vector of the neural signal and the feature vector of the liquid includes: Concatenate the electrode impedance feature pair, the feature vector of the neural signal and the feature vector of the liquid corresponding to each petal-shaped electrode into a second feature vector; Train a third classification model based on the second feature vector and a preset label to obtain a target level corresponding to the signal acquisition region of each petal-shaped electrode; Determine a second level target area based on the target level, a preset amplification coefficient, a total area of the preset cavity, a normal tissue area, a first level target area; Determine a first level target variance and a second level target variance according to the number of each petal-shaped electrode; Obtain the current state recognition result according to the first level target area, the second level target area, the first level target variance, the second level target variance and a preset weight.

Images