Method and system for determining epilepsy focus

Abstract

The invention provides an epilepsy focus determination method and system. The epilepsy focus determination method comprises the following steps: performing multi-modal fusion in combination with PET-CT, clinical typical symptoms and EEG electroencephalogram signals to identify epilepsy focus types; through an adversarial generative network, establishing a mapping relation between a feature after the EEG neural signal and the clinical typical feature are spliced and a feature after the SEEG neural signal and the clinical typical feature are spliced; by changing the magnitude of electrical stimulation current, the contribution weight of each SEEG electrode to epileptic seizure and the epileptic signal coverage range of the position are determined, and the final precise range of epileptic focus is obtained through comprehensive evaluation; a focal epileptic focus boundary identification model in PET-CT is optimized according to the accurate epileptic focus range obtained according to the neural signals, so that the model has the ability of identifying lesions which cannot be identified by human eyes and has the ability of drawing unobvious epileptic focus boundaries.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medical assistance technology, and in particular to a method and system for determining an epileptic focus. Background Art

[0002] Usually, epilepsy can be diagnosed through consultation of medical history, clinical symptoms, cranial nerve signals, head MRI / CT, ​​etc. There are many factors that cause epilepsy: focal lesions with clear boundaries, such as intracranial tumors, cerebral hemorrhage, or vascular malformations, which can cause obvious lesions that compress cranial nerves, can cause epileptic seizures; epileptogenic lesions with unclear boundaries, such as those that cannot be confirmed with clear boundaries even with the help of imaging methods such as PET-CT / MRI; in addition, there are also epileptogenic lesions that cannot be distinguished with the help of imaging methods such as PET-CT / MRI, and can only be roughly focused on by electroencephalograms and clinical symptoms. Abnormal discharges in the entire functional area or brain structure cause epilepsy, such as abnormal discharges from visual stimulation that cause abnormal discharges of neurons in the entire occipital lobe, abnormal discharges of neurons in the temporal lobe caused by auditory stimulation, and abnormal discharges of neurons in the frontal lobe caused by abnormal emotions or movement stimulation; more serious ones, such as diffuse epilepsy, are also the most difficult and most difficult to cure.

[0003] At present, the preliminary diagnosis of epilepsy can be determined through the patient's clinical manifestations combined with the electroencephalogram (EEG). However, if you want to accurately locate the epileptogenic focus, it is best to use intracranial electrodes to determine the boundaries of the epileptogenic focus through electrical stimulation and neural signal analysis. Summary of the invention

[0004] The present invention provides an epileptic focus determination method and system, which are used to solve the defects in the prior art in accurately locating the epileptic focus of epilepsy.

[0005] In a first aspect, the present invention provides a method for determining an epileptogenic focus, comprising: Collect EEG, body movement features, voice features, and emotional features of epilepsy patients to extract epilepsy recognition features; The epilepsy recognition features are fused and spliced, PET-CT tomography images are extracted, an abnormal area segmentation model is trained, and the type of causative focus is determined; Based on the PET-CT tomographic image, a three-dimensional reconstruction image of the brain is obtained; Determining the implantation range of the stereoscopic electroencephalogram (SEEG) electrode according to the three-dimensional reconstructed brain image; Within the implantation range of the SEEG electrode, SEEG stimulation and signal acquisition are performed to identify typical epilepsy symptoms, and feature fusion and splicing of the typical epilepsy symptoms are performed to obtain mapped SEEG signal features; Determining the epileptic focus area boundary based on the mapped SEEG signal features; Based on the abnormal region segmentation model, the boundary of the epileptic focus region is optimized.

[0006] According to a method for determining an epileptic focus provided by the present invention, EEG, limb movement features, sound features, and emotional features of an epileptic patient are collected to extract epilepsy identification features, including: Collecting brain nerve signals of multiple channels of epilepsy patients through an electroencephalogram cap, decomposing the brain nerve signals of the multiple channels into multiple frequency band signals, and obtaining signal features; Decomposing the action video into continuous frames, inputting the trained first convolutional neural network, identifying a plurality of continuous images exceeding a preset confidence level, flattening the last feature layer of the first convolutional neural network, adding the continuous images, and randomly extracting features to obtain action features, wherein the length of the action feature is 1 greater than the length of the signal feature; The sound signal is window-cropped and input into the trained second convolutional neural network to identify a plurality of continuous window segments exceeding a preset confidence level, the last feature layer of the second convolutional neural network is flattened and then the plurality of continuous window segments are added, and features are randomly extracted to obtain sound features, wherein the length of the sound features is 1 greater than the length of the signal features; Decomposing the emotional video into continuous frames, inputting the frames into the trained third convolutional neural network, identifying a plurality of continuous images exceeding a preset confidence level, flattening the last feature layer of the third convolutional neural network, adding the continuous images, and randomly extracting features to obtain emotional features, wherein the length of the emotional features is 1 greater than the length of the signal features; The signal feature, the action feature, the sound feature and the emotion feature are averaged respectively to obtain the signal feature average, the action feature average, the sound feature average and the emotion feature average; and the signal feature average, the action feature average, the sound feature average and the emotion feature average are weightedly fitted by a machine learning algorithm to obtain the epilepsy recognition feature.

[0007] According to a method for determining an epileptic focus provided by the present invention, a brain nerve signal of a patient with epilepsy in multiple channels is collected by an electroencephalogram cap, and the brain nerve signal of the patient in the multiple channels is decomposed into multiple frequency band signals to obtain signal features, including: Based on the deep learning classification model, each frequency band is trained to identify the epilepsy model and obtain the frequency band with the highest accuracy; Based on the training of the highest accuracy frequency band, each channel is identified to identify the epilepsy model, the accuracy of each model is sorted, and several channels with the highest accuracy are selected; Randomly extracting the plane coordinates of 4 channels from the plurality of channels each time and the model recognition accuracy rates corresponding to the 4 channels; Performing equivalent coordinate processing on the combination with the highest accuracy among the combinations of the plane coordinates and the model recognition accuracy, and determining the attribution label attributes of the equivalent coordinates, wherein the attribution label attributes include frontal lobe, parietal lobe, occipital lobe and temporal lobe; The last feature layer of the model corresponding to the best combination is flattened and features are randomly extracted, and all features are concatenated to obtain the signal features.

[0008] According to a method for determining an epileptic focus provided by the present invention, the epilepsy recognition features are fused and spliced, a PET-CT tomography image is extracted, an abnormal region segmentation model is trained, and the type of the causative focus is determined, including: The signal feature, the action feature, the sound feature and the emotion feature are concatenated to form a feature square matrix, and each element in the feature square matrix is ​​scaled to a grayscale image in a preset grayscale range; Obtain PET-CT tomographic images of patients confirmed to be epileptic, train a normal / abnormal recognition deep learning model, and output abnormal PET-CT images and normal PET-CT images; The abnormal PET-CT image is used to train an abnormal region segmentation model. If it is determined that the image has a segmentation boundary, it means that the focal lesion has a clear boundary; otherwise, it means that the focal lesion has no clear boundary; The features of the last convolutional layer of the normal PET-CT image are flattened and concatenated with the concatenated features, and then input into a BP neural network for binary classification recognition to obtain the type of the lesion.

[0009] According to a method for determining an epileptogenic focus provided by the present invention, a three-dimensional reconstructed image of the brain is obtained based on the PET-CT tomographic image, comprising: The trained brain functional area and anatomical structure image segmentation model is obtained, and the brain functional area and anatomical structure are segmented and boundary lines are marked in combination with the PET-CT tomography image, and the three-dimensional reconstruction image of the brain is obtained using three-dimensional reconstruction software.

[0010] According to a method for determining an epileptogenic focus provided by the present invention, SEEG stimulation and signal acquisition are performed within the implantation range of the SEEG electrode to identify typical epilepsy symptoms, and feature fusion and splicing of the typical epilepsy symptoms are performed to obtain mapped SEEG signal features, including: One SEEG electrode is provided with current stimulation at any current value among several SEEG typical current values, and the other SEEG electrodes are responsible for receiving neural signals; The collected electrode numbers are stored in a two-dimensional matrix to obtain a number of two-dimensional matrices; The electrode signals in the two-dimensional matrices are respectively input into the trained epilepsy recognition model, the two-dimensional matrices that fail to recognize epilepsy signals are eliminated, the corresponding SEEG electrodes are silenced and no longer used, and the remaining SEEG electrode sets are taken; Input the remaining SEEG electrode set into the trained epilepsy recognition model, flatten the last feature layer of the model, randomly extract features, and concatenate them with the features corresponding to the remaining SEEG electrode set to obtain new features; The new features are processed in sequence according to the steps of signal features, action features, sound features and emotion features to obtain the mapped SEEG signal features.

[0011] According to a method for determining an epileptic focus provided by the present invention, the boundary of an epileptic focus area is determined by mapping SEEG signal features, comprising: Determining the centroids of electrode arrangement regions corresponding to different current magnitudes based on the similarity of the same current in the mapped SEEG signal features; According to the preset similarity ratio and centroid, the boundaries of the epileptic lesion area with different current sizes are determined.

[0012] In a second aspect, the present invention further provides an epileptogenic focus determination system, comprising: The acquisition module is used to collect EEG, body movement characteristics, sound characteristics and emotional characteristics of epilepsy patients and extract epilepsy recognition features; A type determination module is used to fuse and splice the epilepsy recognition features, extract positron emission tomography-computed tomography (PET-CT) tomography images, train an abnormal region segmentation model, and determine the type of the causative focus; A reconstruction module, used for obtaining a three-dimensional reconstructed image of the brain based on the PET-CT tomography image; A range determination module, used for determining the implantation range of the stereoscopic electroencephalogram signal SEEG electrode according to the three-dimensional reconstructed brain image; A mapping module is used to perform SEEG stimulation and signal acquisition within the implantation range of the SEEG electrode, identify typical epilepsy symptoms, perform feature fusion and splicing on the typical epilepsy symptoms, and obtain mapped SEEG signal features; A boundary determination module, used to determine the boundary of the epileptic focus area based on the mapped SEEG signal features; A boundary optimization module is used to optimize the boundary of the epileptic focus area based on the abnormal area segmentation model.

[0013] In a third aspect, the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for determining an epileptic focus as described above is implemented.

[0014] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described methods for determining an epileptic focus.

[0015] The epileptic focus determination method and system provided by the present invention identify the type of epileptic focus by combining PET-CT with typical clinical symptoms and EEG brain wave signals for multimodal fusion; establish a mapping relationship between the features after splicing the EEG neural signal with the typical clinical features and the features after splicing the SEEG neural signal with the typical clinical features through a generative adversarial network; determine the contribution weight of each SEEG electrode to epileptic seizures and the coverage range of the epileptic signal at the location by changing the magnitude of the electrical stimulation current, and obtain the final precise range of the epileptic focus through comprehensive evaluation; optimize the boundary recognition model of the focal epileptic focus in PET-CT according to the precise range of the epileptic focus obtained from the neural signal, so that the model has the ability to recognize lesions that cannot be distinguished by the human eye, and has the ability to outline the boundary of the epileptic focus with unclear boundaries. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 This is one of the flow charts of the method for determining the epileptogenic focus provided by the present invention; Figure 2 This is the second flow chart of the method for determining epileptogenic focus provided by the present invention; Figure 3 is a schematic diagram of an epilepsy identification unit provided by the present invention; Figure 4 is a schematic diagram of EEG electrode distribution provided by the present invention; Figure 5 It is a schematic diagram of a pathogenic focus type determination unit provided by the present invention; Figure 6 It is a schematic diagram of the implantation position of SEEG electrodes provided by the present invention; Figure 7 It is a schematic diagram of the SEEG stimulation and signal acquisition unit provided by the present invention; Figure 8This is a schematic diagram of the preliminary determination of epilepsy boundaries by the 0.5 mA electrode provided by the present invention; Fig. 9 is a schematic structural diagram of the epileptogenic focus determination system provided by the present invention; Fig.10 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Figure 1 FIG. 1 is one of the flow charts of the method for determining the epileptogenic focus provided by an embodiment of the present invention. Figure 1 As shown, including: Step 100: Collect EEG, body movement features, voice features and emotional features of epilepsy patients to extract epilepsy recognition features; Step 200: fusing and splicing the epilepsy recognition features, extracting PET-CT tomography images, training an abnormal region segmentation model, and determining the type of lesion focus; Step 300: Obtaining a three-dimensional reconstructed image of the brain based on the PET-CT tomography image; Step 400: Determine the implantation range of the stereoscopic electroencephalogram signal SEEG electrode according to the three-dimensional reconstructed brain image; Step 500: Perform SEEG stimulation and signal acquisition within the implantation range of the SEEG electrode, identify typical epilepsy symptoms, perform feature fusion and splicing on the typical epilepsy symptoms, and obtain mapped SEEG signal features; Step 600: determining the epileptic focus area boundary based on the mapped SEEG signal features; Step 700: Optimizing the epileptic lesion region boundary based on the abnormal region segmentation model.

[0020] The specific solution process of the embodiment of the present invention is as follows Figure 2 As shown, specifically including: Step S1, epilepsy identification unit, functional modules such as Figure 3 As shown, it includes four functional modules: electroencephalogram (EEG) signal feature extraction module, epileptic limb movement feature extraction module, typical sound feature extraction module and typical expression feature extraction module.

[0021] Step S1.1, EEG signal feature extraction module, when the patient has an epileptic seizure, the patient's brain nerve signals are collected through the EEG cap (a total of N channels, N can be 64 / 128 / 256, etc.), and the nerve signals are decomposed into multiple frequency bands delta (<4Hz), theta (4-7Hz), alpha (8-13Hz), beta (14-31Hz) and gamma (>32Hz), such as Figure 4 shown.

[0022] Step S1.1.1, building a model based on a deep learning classification model, such as 2DCNN, 2DCNN+RNN, etc., training a model for identifying epilepsy in each frequency band to obtain the frequency band with the highest accuracy; Step S1.1.2, using the optimal frequency band obtained in step S1.1.1 to perform epilepsy identification on N channels separately, the specific steps are as follows: 1) Build models based on deep learning classification models, such as 1DCNN, 1DCNN+RNN, etc., train models for identifying epilepsy in each channel (a total of N models), sort the accuracy of each model, and select the m channels with the highest accuracy, where m can be 10 / 20 or any other number; 2) Randomly select 4 channels from the m channels each time (record the plane coordinates of the 4 channels at the same time), and the model recognition accuracy in step 1), build a model through 2DCNN, 2DCNN+RNN and other methods, train each combination of epilepsy recognition models, and obtain the combination with the highest accuracy; 3) Equivalent coordinates are calculated for the combination with the highest accuracy in step 2), and one of the label attributes of the frontal lobe, parietal lobe, occipital lobe and temporal lobe is determined based on the coordinates; 4) Flatten the last feature layer of the model corresponding to the best combination obtained in step 2), and then randomly extract features to ensure that the feature length is L. In the embodiment of the present invention, L=35; 5) Concatenate the features obtained in step 4) with those obtained in step 3) to obtain new features.

[0023] Step S1.2: Typical limb movement feature extraction module, typical movements such as limb twitching, etc.

[0024] Step S1.2.1: Obtain the time-frequency of the patient's epileptic seizure. The video can be taken by mobile phone, video surveillance, etc., and the video is disassembled into frames (such as 25 frames / s). The trained convolutional neural network model1 (such as Resnet, VGG and other neural network models) is used to identify epileptic movements. Continuous frames (M1 consecutive images, M1 can be 10 / 20, etc.) are identified as epileptic movements, and the model confidence of the M1 consecutive images exceeds 0.95; Step S1.2.2: The model1 obtained in S1.2.1 performs feature extraction on M images, flattens the last feature layer of the model to obtain M1 image features, then performs the add() operation on the M1 image features, and finally randomly extracts features after add() to obtain and ensure that the feature length is L+1.

[0025] Step S1.3: Typical sound feature extraction module, typical sounds such as screaming, roaring, etc.

[0026] Step S1.3.1: Obtain the sound signal of the patient's epileptic seizure. The sound signal can come from a mobile phone recording, a voice recorder recording, etc., cut the sound signal into windows, and use the trained neural network model2 (such as 1DCNN, 1DCNN+RNN, RNN and other neural network models) to identify epileptic movements. Continuous windows (M2 consecutive windows, M2 can be 10 / 20, etc.) are identified as epileptic sounds, and the model confidence of the M2 consecutive windows exceeds 0.95; Step S1.3.2: The model2 obtained in S1.3.1 performs feature extraction on the M2 window segments, flattens the last feature layer of the model to obtain M2 window features, then performs the add() operation on the M2 window features, and finally randomly extracts features after add() to obtain and ensure that the feature length is L+1.

[0027] Step S1.4: Typical emotion feature extraction module, typical actions such as clenched teeth and blue face.

[0028] Step S1.4.1: Obtain the time-frequency of the patient's epileptic seizure. The video can be taken by mobile phone, video surveillance, etc., and the video is disassembled into frames (such as 25 frames / s). The epileptic movements are identified through the trained convolutional neural network model3 (such as Resnet, VGG and other neural network models). Continuous frames (M3 consecutive images, M3 can be 10 / 20, etc.) are identified as epileptic movements, and the model confidence of the M3 consecutive images is more than 0.95; Step S1.4.2: The model3 obtained in S1.4.1 performs feature extraction on M images, flattens the last feature layer of the model to obtain M3 image features, then performs the add() operation on the M3 image features, and finally randomly extracts features after add() to obtain and ensure that the feature length is L+1.

[0029] Step S1.5: Epilepsy identification, , , , Find the mean respectively, and get , , , , through machine learning algorithms (decision trees, random forests, etc.), , , , Perform weighted fitting for two categories: epilepsy or non-epileptic, and obtain , , , The corresponding weights , , , .

[0030] Step S2: Pathogenic focus type determination unit, functional modules such as Figure 5 As shown, it includes two functional modules: feature fusion and splicing module and PET-CT feature extraction module.

[0031] Step S2.1: Feature fusion and splicing module, which combines the features obtained from steps S1.1 to S1.4 , , , The four features are concatenated to obtain a feature concatenation A with a length of 144. Then, each 12 features are combined into a row to generate a matrix, and then each element in the matrix is ​​scaled to a grayscale image of [0,255]. .

[0032] Step S2.2: Obtain the PET-CT tomographic scan images of the patients confirmed as epileptic in step S1, and train the normal / abnormal recognition deep learning model model4 (Resnet, VGG, etc.), where normal includes diffuse lesions and no obvious lesions in the functional areas involved; abnormalities include focal lesions without clear boundaries and focal lesions with clear boundaries (these two situations cannot be distinguished here, but can be determined in the subsequent step S2.3).

[0033] Step S2.3: For the PET-CT images with abnormalities in step S2.2, train an abnormal region segmentation model (such as Unet, D-link, Mask-Rcnn). If the boundary can be segmented, it means that the lesion is focal with clear boundaries (such as tumors, blood clots, etc., which can be directly surgically removed without subsequent SEEG signal acquisition); if the boundary cannot be segmented, it means that the lesion is focal and has no clear boundaries.

[0034] Step S2.4: For the PET-CT images without abnormalities in step S2.2, when model4 identifies the PET-CT images without abnormalities, the features of the last convolutional layer are flattened and concatenated with the features obtained in step S2.1 A for concate() operation, and then input into the BP neural network for binary classification recognition, with the labels being diffuse and no obvious lesion areas in the involved functional areas (such as the temporal lobe, occipital lobe and other functional areas, but no obvious abnormal areas can be observed from the PET-CT images.) Step S3: The cranial skull 3D reconstruction unit uses the PET-CT tomography image obtained in step S2 to load the trained brain functional area and anatomical structure image segmentation model, such as Unet, D-link, Mask-Rcnn and other deep learning models, to segment the brain functional areas and anatomical structures and mark their boundaries, such as the temporal lobe, occipital lobe, frontal lobe, hippocampus and hypothalamus, and then uses VTK and other 3D reconstruction software to perform 3D reconstruction of the cranial skull.

[0035] Step S4: Electrode implantation range, combined with the location of the focal lesion without clear boundaries in step S2.3 or the target functional area, lock the stereo electroencephalography (SEEG) electrode implantation range. When the SEEG electrode locks the precise boundary of the epileptogenic focus, multiple SEEG electrodes are usually implanted around the target area. The effect is shown in the figure below. Figure 6 The embodiment of the present invention is described by taking an abnormality in a certain cerebral cortex functional area as an example. Around the boundary line of this functional area in the three-dimensional reconstruction image of the cranium obtained in step S3, K SEEG electrodes are implanted at equal intervals, and the value of K can be 6 / 8 / 10, etc., without limitation.

[0036] Step S5: SEEG stimulation and signal acquisition unit, functional modules such as Figure 7 As shown, it includes four functional modules: current stimulation module, neural signal acquisition module, neural signal pattern recognition module, and typical symptom recognition module.

[0037] Step S5.1: Current stimulation module, the current required for each electrode is no more than 5 mA. The present invention uses 0.5 mA, 3 mA, and 5 mA as examples, but the present invention is not limited to these three current values. When one SEEG electrode provides current stimulation, the other SEEG electrodes are only responsible for receiving nerve signals without providing stimulation function.

[0038] Step S5.2: Take a SEEG electrode as number 1, and number the other electrodes in sequence. According to the current stimulation method of step S5.1, store the collected electrode numbers in a two-dimensional matrix (each matrix has K rows), for a total of Two-dimensional matrices, .

[0039] Step S5.3: In a two-dimensional matrix , respectively input into the trained epilepsy recognition model, such as 2DCNN / 2DCNN+RNN convolutional neural network model, Step S5.3.1: Eliminate k two-dimensional matrices that fail to identify epilepsy, and the remaining A two-dimensional matrix is ​​created (it should be noted that if all signals collected by applying 5mA current to an electrode fail to identify epilepsy, then applying 3mA or 0.5mA current will not identify epilepsy either), and the corresponding SEEG electrode is silenced and no longer used. At this time, only SEEG electrodes to be analyzed, forming a set ; Step S5.3.2: The remaining electrodes are divided into three situations: only 0.5 mA current is needed to identify epilepsy, 3 mA current is needed to identify epilepsy, and 5 mA current is needed to identify epilepsy. The two-dimensional matrix gives the current label attributes One of them.

[0040] Step S5.3.3: The two-dimensional matrices are respectively input into the 2DCNN / 2DCNN+RNN SEEG neural signal epilepsy recognition model trained in step S5.3, and the last feature layer of the model is flattened, and then the features are randomly extracted. , ensure that the characteristic length is L, and and Perform feature splicing to obtain new features .

[0041] Step S5.4: Decompose the video of the patient when epilepsy can be identified in step S5.3 into frames (e.g., 25 frames / s) Then input the model1 model of step S1.2, select the m11 consecutive images with the highest confidence and the m12 consecutive images after, and ensure that m11+m12+1=M1, extract features from the M1 images at this time, flatten the last feature layer of the model to obtain M1 image features, then perform the add() operation on the M1 image features, and finally randomly extract features from the features after add() to obtain And ensure that the characteristic length is L.

[0042] Step S5.5: Input the patient's sound signal when epilepsy can be identified in step S5.3 into the model2 model of step S1.3, select the first m21 and the last m22 consecutive cropping windows with the highest confidence, and ensure that m21+m22+1=M2, extract features from the M2 windows at this time, flatten the last feature layer of the model to obtain M2 sound features, then perform add() operation on the M2 sound features, and finally randomly extract features from the features after add() to obtain And ensure that the characteristic length is L.

[0043] Step S5.6: When epilepsy can be identified in step S5.3, the patient's video is disassembled by frame (such as 25 frames / s) and input into the model3 model of step S1.4, and the continuous m31 and m32 images before and after the image frame with the highest confidence are selected, and m31+m32+1=M3 are ensured, and features are extracted from the M3 images at this time, and the last feature layer of the model is flattened to obtain M3 image features, and then the M3 image features are added () operation is performed, and finally the features after add () are randomly extracted to obtain features. And ensure that the characteristic length is L.

[0044] Step S5.7: Feature fusion and splicing module, which combines the features obtained from steps S5.3 to SS5.6 , , , The four features are concatenated to obtain feature concatenation B with a length of 144. . Each line is 12 long, generating The matrix, then Each element in the matrix is ​​scaled to a grayscale image of [0,255] .

[0045] Step S6: mapping relationship.

[0046] Step S6.1: is the input image, To create a training set for decoding images, train the generative adversarial network model5 (such as GAN / DCGAN, etc.), and establish a mapping relationship between feature A extracted based on EEG+clinical manifestations and feature B extracted based on SEEG+clinical manifestations.

[0047] Step S6.2: Concatenate the patient features to the feature image corresponding to A Generate model5 obtained by step S6.1 , and compare Feature image corresponding to feature concatenation B The similarity of the images is obtained by comparing the similarity of the images, such as the histogram method, SSIM (structural similarity) method, hash similarity method, etc. Similarity results , sort them from large to small according to the similarity of 0.5mA / 3mA / 5mA, and store the corresponding electrode numbers in the corresponding order.

[0048] Step S7: epilepsy boundary determination.

[0049] Step S7.1: 0.5 mA epilepsy boundary determination, 0.5 mA similarity in step S6.2 , It can be 60%, 80%, etc. without specific requirements; for SEEG electrodes that meet the similarity conditions in step S7.1, the single electrode pattern recognition accuracy is trained by 1DCNN / RNN / 1DCNN+RNN and other methods , retain such electrodes, It can be 88%, 92%, etc. There is no specific requirement. The electrode set at this time is The area enclosed by these electrodes is the initial epileptic boundary. Figure 8 ABC area in, calculate the combined centroid of ABC area .

[0050] Step S7.2: 3mA epilepsy boundary determination, determined in step S7.1 When the applied current is 3 mA, , the single electrode pattern recognition accuracy is trained by 1DCNN / RNN / 1DCNN+RNN and other methods for such electrode signals , It can be determined according to the actual situation, and no specific requirements are given here. All electrodes and points Maximum Euclidean distance At this time, the epilepsy boundary contour determined in step S7.1 is expanded proportionally ,in It can be 10 / 20 / 100, depending on the actual situation. No specific requirements are given here. Calculate the centroid of the area enclosed by the epileptic boundary at this time .

[0051] Step S7.3: 5mA epilepsy boundary determination, determined in step S7.1 When the applied current is 5 mA, , the single electrode pattern recognition accuracy is trained by 1DCNN / RNN / 1DCNN+RNN and other methods for such electrode signals , It can be determined according to the actual situation, and no specific requirements are given here. All electrodes and points Maximum Euclidean distance ,like , expand the epilepsy boundary contour determined in step S7.2 in equal proportion At this time, the expanded epilepsy boundary line is the final epilepsy boundary line.

[0052] Step S8: If the abnormal region segmentation model trained in step S2.3 cannot identify the abnormal region or cannot identify the abnormal region boundary, the epilepsy boundary determined in step S7 is recalibrated on the PET-CT tomography scan, the mask boundary of the segmentation model is re-trained, and the abnormal region boundary segmentation model is re-trained to achieve the purpose of model optimization.

[0053] The optimized model has the following functions: 1) The optimized segmentation model can automatically identify the epileptogenic lesions and their areas in PET-CT that even experienced doctors cannot easily identify with the naked eye; 2) The optimized model can locally segment the boundaries of the epileptogenic lesions in PET-CT (some lesions have edge infiltration).

[0054] The epileptic focus determination system provided by the present invention is described below. The epileptic focus determination system described below and the epileptic focus determination method described above can be referenced to each other.

[0055] Fig. 9 is a schematic diagram of the structure of the epileptogenic focus determination system provided by an embodiment of the present invention. Fig. 9 As shown, it includes: an acquisition module 91, a type determination module 92, a reconstruction module 93, a range determination module 94, a mapping module 95, a boundary determination module 96 and a boundary optimization module 97, wherein: The acquisition module 91 is used to collect the EEG, limb movement characteristics, sound characteristics and emotional characteristics of epilepsy patients, and extract epilepsy identification features; the type determination module 92 is used to fuse and splice the epilepsy identification features, extract the positron emission tomography PET-CT tomography image, train the abnormal area segmentation model, and determine the type of the causative focus; the reconstruction module 93 is used to obtain a three-dimensional reconstruction image of the brain based on the PET-CT tomography image; the range determination module 94 is used to determine the implantation range of the SEEG electrode of the stereo electroencephalogram signal according to the three-dimensional reconstruction image of the brain; the mapping module 95 is used to perform SEEG stimulation and signal acquisition within the implantation range of the SEEG electrode, identify typical epilepsy symptoms, and perform feature fusion and splicing on the typical epilepsy symptoms to obtain the mapping SEEG signal characteristics; the boundary determination module 96 is used to determine the boundary of the epilepsy focus area by the mapping SEEG signal characteristics; the boundary optimization module 97 is used to optimize the boundary of the epilepsy focus area based on the abnormal area segmentation model.

[0056] Fig.10 An example of a physical structure diagram of an electronic device is shown in FIG. Fig.10 As shown, the electronic device may include: a processor (processor) 1010, a communication interface (Communications Interface) 1020, a memory (memory) 1030 and a communication bus 1040, wherein the processor 1010, the communication interface 1020, and the memory 1030 communicate with each other through the communication bus 1040. The processor 1010 can call the logic instructions in the memory 1030 to execute the method for determining the epileptic focus, which includes: collecting EEG, limb movement characteristics, sound characteristics and emotional characteristics of epilepsy patients to extract epilepsy identification features; performing feature fusion and splicing on the epilepsy identification features, extracting PET-CT tomography images, training abnormal area segmentation models, and determining the type of causative focus; based on the PET-CT tomography images, obtaining a three-dimensional reconstruction image of the brain; according to the three-dimensional reconstruction image of the brain, determining the implantation range of the SEEG electrode for the stereoscopic brain wave signal; within the implantation range of the SEEG electrode, performing SEEG stimulation and signal acquisition to identify typical epileptic symptoms, performing feature fusion and splicing on the typical epileptic symptoms to obtain mapped SEEG signal features; determining the boundary of the epileptic focus area by the mapped SEEG signal features; and optimizing the boundary of the epileptic focus area based on the abnormal area segmentation model.

[0057] In addition, the logic instructions in the above-mentioned memory 1030 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0058] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the epileptic focus determination method provided by the above-mentioned methods, the method comprising: collecting EEG, limb movement characteristics, sound characteristics and emotional characteristics of epilepsy patients, and extracting epilepsy identification features; performing feature fusion and splicing on the epilepsy identification features, extracting PET-CT tomographic scan images, training abnormal area segmentation models, and determining the type of causative focus; obtaining a three-dimensional reconstructed image of the brain based on the PET-CT tomographic scan images; determining the implantation range of SEEG electrodes for stereoscopic electroencephalogram signals based on the three-dimensional reconstructed image of the brain; performing SEEG stimulation and signal acquisition within the SEEG electrode implantation range, identifying typical epileptic symptoms, performing feature fusion and splicing on the typical epileptic symptoms, and obtaining mapped SEEG signal features; determining the boundaries of the epileptic focus area by the mapped SEEG signal features; and optimizing the boundaries of the epileptic focus area based on the abnormal area segmentation model.

[0059] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0060] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for determining an epileptogenic focus, characterized in that: include: Collect EEG signals, body movement features, sound features and emotional features of epilepsy patients to extract epilepsy recognition features; The epilepsy recognition features are fused and spliced, positron emission tomography (PET-CT) tomography images are extracted, an abnormal region segmentation model is trained, and the type of causative focus is determined; Based on the PET-CT tomographic image, a three-dimensional reconstruction image of the brain is obtained; Determining the implantation range of the stereoscopic electroencephalogram (SEEG) electrode according to the three-dimensional reconstructed brain image; Within the implantation range of the SEEG electrode, SEEG stimulation and signal acquisition are performed to identify typical epilepsy symptoms, and feature fusion and splicing of the typical epilepsy symptoms are performed to obtain mapped SEEG signal features; Determining the epileptic focus area boundary based on the mapped SEEG signal features; Based on the abnormal region segmentation model, the boundary of the epileptic focus region is optimized.

2. The method for determining epileptogenic focus according to claim 1, characterized in that: Collect EEG, body movement features, sound features, and emotional features of epilepsy patients to extract epilepsy recognition features, including: Collecting brain nerve signals of multiple channels of epilepsy patients through an electroencephalogram cap, decomposing the brain nerve signals of the multiple channels into multiple frequency band signals, and obtaining signal features; Decomposing the action video into continuous frames, inputting the trained first convolutional neural network, identifying a plurality of continuous images exceeding a preset confidence level, flattening the last feature layer of the first convolutional neural network, adding the continuous images, and randomly extracting features to obtain action features, wherein the length of the action feature is 1 greater than the length of the signal feature; The sound signal is window-cropped and input into the trained second convolutional neural network to identify a plurality of continuous window segments exceeding a preset confidence level, the last feature layer of the second convolutional neural network is flattened and then the plurality of continuous window segments are added, and features are randomly extracted to obtain sound features, wherein the length of the sound features is 1 greater than the length of the signal features; Decomposing the emotional video into continuous frames, inputting the frames into the trained third convolutional neural network, identifying a plurality of continuous images exceeding a preset confidence level, flattening the last feature layer of the third convolutional neural network, adding the continuous images, and randomly extracting features to obtain emotional features, wherein the length of the emotional features is 1 greater than the length of the signal features; The signal feature, the action feature, the sound feature and the emotion feature are averaged respectively to obtain the signal feature average, the action feature average, the sound feature average and the emotion feature average; and the signal feature average, the action feature average, the sound feature average and the emotion feature average are weightedly fitted by a machine learning algorithm to obtain the epilepsy recognition feature.

3. The method for determining epileptogenic focus according to claim 2, characterized in that: The brain nerve signals of multiple channels of the epilepsy patient are collected by an electroencephalogram cap, and the brain nerve signals of the multiple channels of the patient are decomposed into multiple frequency band signals to obtain signal features, including: Based on the deep learning classification model, each frequency band is trained to identify the epilepsy model and obtain the frequency band with the highest accuracy; Based on the training of the highest accuracy frequency band, each channel is identified to identify the epilepsy model, the accuracy of each model is sorted, and several channels with the highest accuracy are selected; Randomly extracting the plane coordinates of 4 channels from the plurality of channels each time and the model recognition accuracy rates corresponding to the 4 channels; Performing equivalent coordinate processing on the combination with the highest accuracy among the combinations of the plane coordinates and the model recognition accuracy, and determining the attribution label attributes of the equivalent coordinates, wherein the attribution label attributes include frontal lobe, parietal lobe, occipital lobe and temporal lobe; The last feature layer of the model corresponding to the best combination is flattened and features are randomly extracted, and all features are concatenated to obtain the signal features.

4. The method for determining epileptogenic focus according to claim 2, characterized in that: The epilepsy recognition features are fused and spliced, PET-CT tomography images are extracted, an abnormal area segmentation model is trained, and the type of causative focus is determined, including: The signal feature, the action feature, the sound feature and the emotion feature are concatenated to form a feature square matrix, and each element in the feature square matrix is ​​scaled to a grayscale image in a preset grayscale range; Obtain PET-CT tomographic images of patients confirmed to be epileptic, train a normal / abnormal recognition deep learning model, and output abnormal PET-CT images and normal PET-CT images; The abnormal PET-CT image is used to train an abnormal region segmentation model. If it is determined that the image has a segmentation boundary, it means that the focal lesion has a clear boundary; otherwise, it means that the focal lesion has no clear boundary; The features of the last convolutional layer of the normal PET-CT image are flattened and concatenated with the concatenated features, and then input into a BP neural network for binary classification recognition to obtain the type of the lesion.

5. The method for determining epileptogenic focus according to claim 1, characterized in that: Based on the PET-CT tomography image, a three-dimensional reconstruction image of the brain is obtained, including: The trained brain functional area and anatomical structure image segmentation model is obtained, and the brain functional area and anatomical structure are segmented and boundary lines are marked in combination with the PET-CT tomography image, and the three-dimensional reconstruction image of the brain is obtained using three-dimensional reconstruction software.

6. The method for determining epileptogenic focus according to claim 1, characterized in that: Within the implantation range of the SEEG electrode, SEEG stimulation and signal acquisition are performed to identify typical epilepsy symptoms, and feature fusion and splicing of the typical epilepsy symptoms are performed to obtain mapped SEEG signal features, including: One SEEG electrode is provided with current stimulation at any current value among several SEEG typical current values, and the other SEEG electrodes are responsible for receiving neural signals; The collected electrode numbers are stored in a two-dimensional matrix to obtain a number of two-dimensional matrices; The electrode signals in the two-dimensional matrices are respectively input into the trained epilepsy recognition model, and the two-dimensional matrices that fail to recognize epilepsy signals are eliminated, and the corresponding SEEG electrodes are silenced and no longer used, and the remaining SEEG electrode sets are taken; Input the remaining SEEG electrode set into the trained epilepsy recognition model, flatten the last feature layer of the model, randomly extract features, and concatenate them with the features corresponding to the remaining SEEG electrode set to obtain new features; The new features are processed in sequence according to the steps of signal features, action features, sound features and emotion features to obtain the mapped SEEG signal features.

7. The method for determining epileptogenic focus according to claim 1, characterized in that: Determining the epileptic focus area boundary by mapping the SEEG signal features includes: Determining the centroids of electrode arrangement regions corresponding to different current magnitudes based on the similarity of the same current in the mapped SEEG signal features; According to the preset similarity ratio and centroid, the boundaries of the epileptic lesion area with different current sizes are determined.

8. A system for determining an epileptogenic focus, characterized in that: include: The acquisition module is used to collect EEG, body movement characteristics, sound characteristics and emotional characteristics of epilepsy patients and extract epilepsy recognition features; A type determination module is used to perform feature fusion and splicing on the epilepsy recognition features, extract positron emission tomography-computed tomography (PET-CT) tomography images, train an abnormal region segmentation model, and determine the type of the causative focus; A reconstruction module, used for obtaining a three-dimensional reconstructed image of the brain based on the PET-CT tomography image; A range determination module, used for determining the implantation range of the stereoscopic electroencephalogram signal SEEG electrode according to the three-dimensional reconstructed brain image; A mapping module is used to perform SEEG stimulation and signal acquisition within the implantation range of the SEEG electrode, identify typical epileptic symptoms, perform feature fusion and splicing on the typical epileptic symptoms, and obtain mapped SEEG signal features; A boundary determination module, used to determine the boundary of the epileptic focus area based on the mapped SEEG signal features; A boundary optimization module is used to optimize the boundary of the epileptic focus area based on the abnormal area segmentation model.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method for determining the epileptic focus according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for determining the epileptic focus according to any one of claims 1 to 7 is implemented.

Images