Intracranial lesion positioning method, system, device and medium based on intracranial electroencephalogram features

Abstract

The application discloses a lesion positioning method, system, device and medium based on intracranial electroencephalogram characteristics, relates to the technical field of data transmission processing, and comprises the following steps: acquiring an electrode array, acquiring an initial electrode signal and a feedback electrode signal; acquiring real-time radians, and acquiring radian differences of each electrode group; if the radian differences of each electrode group exceed a preset radian threshold, acquiring a first distribution position of each electrode group corresponding to the radian differences exceeding the preset radian threshold in the electrode array, and acquiring a lesion positioning result according to the first distribution position; if the radian differences of each electrode group do not exceed the preset radian threshold, acquiring gradient values of each electrode group according to the radian differences corresponding to each electrode group, acquiring a second distribution position according to the gradient values of each electrode group, and acquiring a lesion positioning result according to the second distribution position. The application has the advantages of high positioning efficiency, high lesion recognition sensitivity and good positioning effect.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data transmission processing, and in particular to a lesion positioning method and system based on intracranial electroencephalogram features, a device and a medium. BACKGROUND

[0002] In the field of neurology, accurate positioning of lesions (such as epileptic foci, tumors or inflammatory areas) is a key prerequisite for developing effective treatment plans. In current clinical practice, intracranial electrode monitoring (such as stereoelectroencephalography, SEEG) can directly record deep brain electrical activity and provide high-resolution signals, but this method requires the implantation of electrodes through craniotomy surgery, which has significant drawbacks such as high trauma, high risk of infection, complex operation and high cost. While non-invasive scalp electroencephalography (EEG) avoids surgical risks, it suffers from spatial ambiguity due to the attenuation effect of the skull on electrical signals and the volume conduction effect, resulting in a severe lack of spatial resolution.

[0003] Specifically, the deep signal will be distorted by brain tissue impedance interference during transmission to the scalp, and the lesion will cause waveform phase shift (radial difference). At the same time, the conventional electrode layout has limited coverage and wide spacing, which cannot finely distinguish the electrical physiological differences between adjacent brain regions. In addition, existing algorithms usually rely on a single threshold value of signal amplitude or frequency spectrum, which can easily ignore the subtle distortion (such as phase lag between waveforms, frequency response drift) of the lesion area and the surrounding tissue on the electrical signal conduction path, lack a dynamic evaluation mechanism for the electroencephalogram conduction pattern (such as the phase consistency of waveforms), and are difficult to distinguish pathological features such as local conduction block, abnormal signal diffusion or cross-brain region synchronization activity. Especially when the lesion does not produce significant discharge, the traditional method is prone to missed diagnosis or positioning deviation. SUMMARY

[0004] In view of the fact that there is no non-invasive positioning method in the prior art that can overcome skull signal attenuation, quantitatively analyze electrical activity conduction characteristics and be suitable for micro-lesion detection, the present application provides a lesion positioning method and system based on intracranial electroencephalogram features, a device and a medium.

[0005] A lesion positioning method based on intracranial electroencephalogram features, comprising: acquiring an electrode array arranged at a plurality of extracranial body surface regions, acquiring two initial electrode signals continuously emitted by each electrode group in the electrode array at a preset interval and two feedback electrode signals collected after emission, wherein the waveform difference between the two initial electrode signals continuously emitted at the preset interval is a preset radian; acquiring the real-time radian difference between the waveforms of the two feedback electrode signals collected by each electrode group, and acquiring the radian difference of each electrode group according to the preset radian and the real-time radian corresponding to each electrode group; if there is a radian difference exceeding a preset radian threshold in the radian difference of each electrode group, acquiring a first distribution position of each electrode group in the electrode array corresponding to the radian difference exceeding the preset radian threshold, and acquiring a lesion positioning result according to the first distribution position; if there is no radian difference exceeding the preset radian threshold in the radian difference of each electrode group, acquiring a gradient value of each electrode group according to the radian difference corresponding to each electrode group, acquiring a second distribution position according to the gradient value of each electrode group, and acquiring a lesion positioning result according to the second distribution position.

[0006] Optionally, acquiring the gradient value of each electrode group according to the radian difference corresponding to each electrode group comprises: respectively acquiring the change amount of the radian difference between the ith electrode group and each adjacent electrode group; acquiring the absolute value of the change amount, and taking the change amount corresponding to the maximum absolute value as the gradient value of the ith electrode group.

[0007] Optionally, acquiring the second distribution position according to the gradient value of each electrode group comprises: identifying a plurality of continuous electrode groups as a continuous feature group, wherein a plurality of sequentially adjacent electrode groups with the same sign of the gradient value are identified as continuous; if the number of electrode groups in the continuous feature group exceeds a preset number threshold, acquiring a total gradient value of the continuous feature group; acquiring a continuous feature group with an absolute value of the total gradient value exceeding a preset gradient threshold as a target group, and acquiring a second distribution position of each electrode group in the electrode array in the target group.

[0008] Optionally, acquiring the radian difference of each electrode group according to the preset radian and the real-time radian corresponding to each electrode group comprises: acquiring the relative difference value between the preset radian and the real-time radian corresponding to the ith electrode group, taking the absolute value of the relative difference value, and taking the absolute value as the radian difference of the ith electrode group.

[0009] Optionally, acquiring the first distribution position of each electrode group in the electrode array corresponding to the radian difference exceeding the preset threshold comprises: acquiring the setting position of each electrode group corresponding to the radian difference exceeding the preset threshold in the extracranial body surface region according to the layout of the electrode array, and taking the setting position as the first distribution position, wherein the first distribution position comprises the extracranial body surface region to which the electrode group belongs and the relative position in the electrode array.

[0010] Optionally, the obtaining the lesion positioning result according to the first distribution position comprises: identifying a brain internal anatomical structure region corresponding to the extracranial body surface region in the first distribution position based on the extracranial body surface region in the first distribution position and the relative positions in the electrode array, and taking position information of the brain internal anatomical structure region as the lesion positioning result, wherein the lesion positioning result is used to indicate a position of a lesion in the brain.

[0011] Also provided is a lesion positioning system based on intracranial electroencephalogram features, comprising: a data acquisition module configured to acquire an electrode array arranged at a plurality of extracranial body surface regions, acquire two initial electrode signals continuously emitted by each electrode group in the electrode array at a preset interval, and acquire two feedback electrode signals collected after emission, wherein a waveform difference between the two initial electrode signals continuously emitted at the preset interval is a preset radian; a data processing module configured to acquire a real-time radian of a waveform difference between the two feedback electrode signals collected by each electrode group, and acquire a radian difference of each electrode group according to a preset radian and a real-time radian corresponding to each electrode group; a first data analysis and positioning module configured to, when there is a radian difference exceeding a preset radian threshold in the radian difference of each electrode group, acquire a first distribution position in the electrode array where each electrode group corresponding to the radian difference exceeding the preset radian threshold is located, and obtain a lesion positioning result according to the first distribution position; and a second data analysis and positioning module configured to, when there is no radian difference exceeding the preset radian threshold in the radian difference of each electrode group, acquire a gradient value of each electrode group according to the radian difference corresponding to each electrode group, acquire a second distribution position according to the gradient value of each electrode group, and obtain a lesion positioning result according to the second distribution position.

[0012] Optionally, the second data analysis and positioning module further comprises: acquiring a change amount of the radian difference between the i th electrode group and each adjacent electrode group, respectively; acquiring an absolute value of the change amount, and taking the change amount corresponding to the maximum absolute value as the gradient value of the i th electrode group.

[0013] Also provided is an electronic device, comprising: a memory having a computer program stored thereon; and a processor configured to execute the computer program in the memory to implement the above-mentioned lesion positioning method based on intracranial electroencephalogram features.

[0014] Also provided is a non-transitory computer readable storage medium having a computer program stored thereon, wherein the program is executed by a processor to implement the above-mentioned lesion positioning method based on intracranial electroencephalogram features.

[0015] The beneficial effects of the present application are embodied in:

[0016] In the whole method of lesion positioning based on intracranial EEG characteristics, the spatial ambiguity and insufficient sensitivity in traditional non-invasive positioning are solved simultaneously by a two-way decision model. The phase difference threshold detection path directly captures the significant phase distortion caused by the lesion conductivity mutation (such as the abnormality of three adjacent electrode groups in the insular region), and uses the anatomical projection model to reversely map the deep lesion from the skull surface high abnormal area (such as the insular prefrontal cortex), breaking through the physical limitation of skull signal attenuation. The gradient field analysis path specifically captures the lesion that does not cause significant phase shift but causes progressive deterioration of conduction efficiency (such as regional conduction block in white matter demyelination lesions) by identifying the same sign gradient continuous area (such as the five consecutive groups of negative gradient in the frontal lobe) and cumulative intensity judgment, which makes up for the blind area of threshold method to subtle pathological changes. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed in the specific embodiments or prior art description will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, each element or part is not necessarily drawn according to the actual proportion.

[0018] Figure 1 Part of the flowchart of the method of lesion positioning based on intracranial EEG characteristics of the present application;

[0019] Figure 2 Another part of the flowchart of the method of lesion positioning based on intracranial EEG characteristics of the present application;

[0020] Figure 3 Another part of the flowchart of the method of lesion positioning based on intracranial EEG characteristics of the present application;

[0021] Figure 4 The step diagram of the method of lesion positioning based on intracranial EEG characteristics of the present application;

[0022] Figure 5 Part of the step diagram of S4 in the method of lesion positioning based on intracranial EEG characteristics of the present application;

[0023] Figure 6 Another part of the step diagram of S4 in the method of lesion positioning based on intracranial EEG characteristics of the present application;

[0024] Figure 7 Part of the step diagram of S3 in the method of lesion positioning based on intracranial EEG characteristics of the present application;

[0025] Figure 8 The block diagram of an electronic device according to an embodiment of the present application is shown.

[0026] Reference numerals:

[0027] 700 - electronic device, 701 - processor, 702 - memory, 703 - multimedia component, 704 - I / O interface, 705 - communication component. DETAILED DESCRIPTION

[0028] To make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations.

[0029] Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without making creative efforts fall within the scope of protection of the present application.

[0030] It should be noted that: similar reference numerals and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. In addition, the terms "first", "second", etc. are only used to distinguish description, and cannot be understood as indicating or implying relative importance.

[0031] As shown in FIGS. 1-3, a method for lesion positioning based on intracranial electroencephalogram features is provided, in one embodiment, the method comprises: Figure 1 , Figure 2 , Figure 3 and Figure 4 As shown in FIGS. 1-3, a method for lesion positioning based on intracranial electroencephalogram features is provided, in one embodiment, the method comprises:

[0032] S1, an electrode array arranged at a plurality of extracranial body surface regions is obtained, two initial electrode signals continuously emitted by each electrode group in the electrode array at a preset interval and two feedback electrode signals collected after emission are obtained, wherein the waveform difference between the two initial electrode signals continuously emitted at the preset interval is a preset radian;

[0033] S2, the real-time radian difference between the waveforms of the two feedback electrode signals collected by each electrode group is obtained, and the radian difference of each electrode group is obtained according to the corresponding preset radian and real-time radian of each electrode group;

[0034] S3, if there is a radian difference exceeding the preset radian threshold in the radian difference of each electrode group, the first distribution position of each electrode group in the electrode array corresponding to the radian difference exceeding the preset radian threshold is obtained, and the lesion positioning result is obtained according to the first distribution position;

[0035] S4, if there is no radian difference exceeding the preset radian threshold in the radian differences of each electrode group, obtaining a gradient value of each electrode group according to the corresponding radian difference of each electrode group, obtaining a second distribution position according to the gradient values of each electrode group, and obtaining a lesion positioning result according to the second distribution position.

[0036] In the embodiment, it should be noted that in S1, the intracranial electrical signal conduction characteristics are actively detected by the electrode array with a specific layout. In a specific implementation, the electrode array needs to cover multiple extracranial surface areas corresponding to key brain regions (such as the temporal lobe and the hippocampus), and each area is provided with an electrode group (such as a temporal lobe electrode group and a left hippocampus electrode group) composed of multiple electrodes, forming a high spatial resolution detection network. Each electrode group continuously emits two initial electrode signals (i.e., actively applied stimulation signals) at a preset time interval, and the two signals have a preset waveform phase difference (for example, the first signal is a sine wave peak, and the second signal lags a certain radian to the wave trough). After emission, the same electrode group synchronously collects feedback signals conducted back through the skull and brain tissue. In theory, due to the consistency of the conduction path, the waveform phase difference of the two feedback signals should remain consistent with the preset radian of the initial signal. This design actively controls the phase relationship of the emitted signal, providing a reference for subsequent quantification of signal conduction distortion.

[0037] Further, taking a deep hippocampus lesion as an example: if the electrode array is deployed in the extracranial area covering the left hippocampus, and the preset phase difference of the initial signal emitted by the electrode group is π / 2 radian (i.e., one-fourth of the period). When the signal penetrates the skull and conducts to the deep part, the lesion (such as abnormal discharge tissue or tumor) in the hippocampus will distort the signal conduction path due to the change in electrical conductivity, causing the real-time phase difference of the feedback signal to deviate significantly from the preset value (such as only lagging π / 4 radian). The grouping and high-density layout strategy of the electrode array (such as 4 groups in the temporal lobe and 3 groups in the hippocampus) can form a local detection cluster group, thereby distinguishing the phase deviation caused by different anatomical structure conduction differences. By synchronously collecting the feedback signals of the electrode groups corresponding to multiple brain regions, a cross-regional phase distortion topology map can be established, laying a data foundation for subsequent identification of the accurate position and spatial gradient change of the lesion.

[0038] In S2, a non-invasive quantitative model of intracranial electrical signal conduction characteristics is established. In specific implementation, the waveform analysis is performed on the two feedback electrode signals collected by each electrode group, and the actual phase offset between the two is extracted through time-frequency analysis technology (such as waveform zero-crossing point or function peak value detection), that is, the real-time radian value; this value reflects the propagation delay characteristics of the electrical signal in the process of transmission through the skull, cerebrospinal fluid and brain parenchyma due to the difference in medium conductivity. Subsequently, the real-time radian corresponding to each electrode group is compared with its preset radian (determined by the transmission parameters of the initial signal), and the absolute value of the difference between the two is calculated. The physical meaning of this absolute difference is that when the brain tissue conduction path is uniform and free from lesion interference, the feedback signal should maintain the phase relationship set by the initial signal, and the difference tends to zero; while the lesion area will destroy the phase consistency of electromagnetic wave propagation due to the abnormal local conductivity (such as high conductivity of tumor cells, impedance mutation caused by ion channel disorder of epileptic focus), so that the measured radian deviates significantly from the preset value. This difference, as the radian difference, is essentially a scalar measure of the degree of interference of the lesion on the normal signal conduction path.

[0039] Further, taking a deep microglial tumor in the temporal lobe as an example, assuming that the initial signal of a certain temporal lobe electrode group is preset with a phase difference of 90 degrees (one quarter of the period difference under the ideal conduction model), when the feedback signal flows through the normal brain tissue after penetrating the skull, its phase shift should be close to this theoretical value; but if the tumor is located on the conduction path, its abnormal tissue structure (such as high water content or neovascular proliferation) will change the local dielectric properties, causing differences in speed and path bending of electromagnetic waves when passing through the lesion area, ultimately resulting in a significant reduction in the phase lag of the second feedback signal collected (for example, producing an excessive compressed phase difference with the first signal). By calculating the absolute deviation of the preset value and the real-time value of this electrode group (such as the 60-degree difference between the theoretical 90 degrees and the measured 30 degrees), the conduction abnormality at this site can be accurately captured. This mechanism is applicable to all electrode groups in the array - the radian difference of normal brain areas is distributed close to the zero baseline, while the lesion coverage area shows isolated convex peaks, and through spatial superposition, a two-dimensional thermal distribution reflecting the spatial boundary of brain conduction abnormalities can be constructed, providing high-resolution quantitative basis for the lesion location judgment in steps S3 / S4.

[0040] In S3, the preliminary spatial localization of the lesion is achieved by the arc difference threshold screening. When the arc difference (i.e., the absolute deviation of the preset arc and the measured arc) of the electrode group exceeds the preset threshold, it indicates that there is a significant abnormality in the corresponding brain region of the electrical signal conduction path - this abnormality is caused by the local destruction of the lesion to the electromagnetic wave propagation. Specifically, first, identify all electrode groups with excessive arc difference in the electrode array, and establish a first spatial distribution map according to their distribution positions on the extracranial surface (such as the middle temporal lobe, the posterior parietal lobe, etc.). The physical meaning of this map is that each excessive point corresponds to the end point of a conduction path that causes phase distortion due to impedance mutation of the lesion, and its position is defined by the anatomic coverage area of the electrode group and the relative coordinates in the array (for example, the three-dimensional index of the electrode group covering the left hippocampus in the array). It should be emphasized that the threshold itself reflects the minimum distinguishable boundary between normal tissue and pathological state in terms of phase consistency, and background interference caused by environmental noise or individual differences can be excluded through the screening mechanism.

[0041] Further, taking the deep insular lesion as an example to explain the positioning logic, it is assumed that the three adjacent electrode groups corresponding to the insular lobe on the cranial surface region all detect excessive arc difference (such as the deviation between the preset value and the measured value exceeding the upper limit of the threshold). According to the spatial layout model of the electrode array, these electrode groups are marked as "high abnormality region", and their first distribution position points to three closely arranged detection sites at the junction of the right anterior temporal bone and the sphenoid bone. According to the preset mapping rule between the cranial surface and the intracranial structure (for example, the conductive model constructed based on medical images), the distribution position is associated with the internal structure of the brain: the physical coordinates of the three electrode groups are corrected by spatial interpolation and anatomical markers, and converge to the anterior cortical region of the insular lobe, thereby generating the lesion positioning result (i.e., "right anterior insular abnormal conduction region"). It should be noted here that the essence of the result is the spatial projection of the conduction path abnormality region, and the actual positioning accuracy of the lesion depends on the electrode density, threshold sensitivity and the accuracy of the skull conduction model, but it does not require invasive means to lock the deep target.

[0042] It should also be noted that the determination of the preset arc threshold in S3 and S4 can be achieved by collecting a large amount of arc difference data of healthy subjects under standard detection conditions to construct a normal distribution model; taking the upper limit value of the 95% confidence interval of the arc difference distribution of the healthy population as the initial threshold benchmark (such as the mean value ± standard deviation of the arc difference of healthy people). Then test in known lesion patients (diagnosed by MRI / pathology), adjust the threshold to simultaneously cover more than 90% of true positive lesions (such as epilepsy focus arc difference > benchmark value) and exclude 95% of false positive interference (such as natural variation caused by skull thickness difference). Finally, test in an electromagnetic interference environment (simulate operating room equipment noise) to ensure that the threshold still maintains diagnostic specificity when the signal-to-noise ratio is below a certain level.

[0043] In S4, for the scenario that the arc difference does not exceed the threshold, the subtle conduction anomaly is captured by spatial gradient analysis. First, the gradient value of each electrode group is calculated: taking the i th electrode group as the center, the arc difference change amount (for example, the difference between the adjacent group arc difference and the group arc difference) between it and all adjacent electrode groups is calculated, and the absolute value of the largest one is selected, and the actual change amount (including sign) is defined as the gradient value. The physical meaning of this design is that the gradient value sign (positive / negative) reflects the mutation direction of local conduction characteristics (such as transition to high / low impedance area), and its amplitude represents the mutation strength. Subsequently, the electrode array is scanned, and multiple electrode groups with continuous and consistent gradient signs and spatial adjacency are aggregated into continuous feature groups (for example, 3 temporal lobe electrode groups with negative gradient), and this continuity reveals the boundary diffusion or regional block effect of the conduction anomaly between brain regions, and eliminates the interference of isolated noise points.

[0044] Further, taking the frontal lobe white matter small demyelination lesion as an example, assuming that the arc difference of this region does not exceed the threshold, but the gradient values of the 5 continuous electrode groups in the left frontal lobe are all negative and the intensity increases, indicating that the conduction characteristics continue to deteriorate in depth. When the number of continuous feature groups meets the preset requirement (such as not less than 4 groups), the total gradient value obtained by accumulating the gradient values of the continuous feature groups is a significant negative value; when this value exceeds the gradient threshold, it is determined as an abnormal diffusion area. According to the second distribution position of these electrode groups in the array (such as the midline of the frontal pole and both sides), combined with the spatial projection model of the skull surface coordinates to the white matter fiber bundle, the conduction attenuation core area of the dorsal white matter of the frontal lobe is located. This method has specific capture ability for deep lesions (such as early demyelination and microvascular ischemia) that do not cause significant phase distortion but cause regional conduction efficiency gradient, which makes up for the blind area of the threshold method.

[0045] In summary, the entire intracranial EEG feature-based lesion localization method synchronously solves the problems of spatial ambiguity and insufficient sensitivity in traditional non-invasive localization through a two-path decision model. The phase difference threshold detection path directly captures significant phase distortion caused by lesion conductivity mutation (e.g., three adjacent electrode groups in the insular region exceeding the abnormal threshold), and uses the anatomical projection model to reversely map the skull surface high abnormal area to the deep lesion (e.g., the insular prefrontal cortex), breaking through the physical limitation of skull signal attenuation. The gradient field analysis path specifically captures lesions that do not cause significant phase shift but lead to progressive deterioration of conduction efficiency (e.g., regional conduction block in white matter demyelination lesions) through the identification of continuous regions with the same sign gradient (e.g., five consecutive groups of negative gradient in the frontal lobe) and cumulative intensity judgment, making up for the blind spot of threshold method to subtle pathological changes. In summary, this phase-gradient dual-dimensional analysis mechanism, combined with high-density electrode layout, not only avoids the noise interference defects of traditional amplitude / spectrum methods, but also realizes accurate localization of deep small lesions under non-invasive conditions through the spatiotemporal correlation of conduction path distortion and regional gradual change mode. Its localization accuracy is close to invasive monitoring but completely eliminates the risk of surgery, providing safe and reliable technical support for precise delineation of epileptic foci, definition of tumor infiltration area, and early screening of inflammatory lesions.

[0046] As shown in Figure 5 In one embodiment, obtaining the gradient value of each electrode group according to the radian difference corresponding to each electrode group in S4 includes:

[0047] S41, respectively obtaining the change amount of the radian difference between the i-th electrode group and each adjacent electrode group;

[0048] S42, obtaining the absolute value of the change amount, and taking the change amount corresponding to the maximum absolute value as the gradient value of the i-th electrode group.

[0049] In this embodiment, it should be noted that the i-th electrode group is any electrode group, i is a positive integer, and is less than the number of electrode groups. S41, quantifying the spatial mutation trend of local conduction characteristics. For the i-th electrode group, the radian difference between it and all directly adjacent electrode groups (such as the upper, lower, left, right, and diagonal directions) is calculated, that is, the radian difference of the adjacent electrode group is subtracted from the radian difference of the i-th electrode group, to generate a set of change amounts reflecting the spatial gradient of conduction abnormality. For example, if the i-th group is located in the front of the temporal lobe, the radian difference of the adjacent upper electrode group is significantly higher than that of the group, and the change amount is positive, indicating that the conduction abnormality increases in the dorsal direction; if the radian difference of the left adjacent group is lower than that of the group, the change amount is negative, indicating that the abnormality decreases in the ventral direction. This change amount set essentially constructs a conduction characteristic change vector field centered on the i-th group.

[0050] In S42, the index most indicative of local conduction mutation is screened out from the change set generated in S41: first, the absolute values of all changes are calculated, and the one with the largest absolute value (i.e., the direction with the strongest spatial mutation) is selected, and then the corresponding original change (including the sign) is defined as the gradient value of the i-th electrode group. Taking the parietal calcification as an example: if the change between the i-th group and the adjacent group above is a negative extreme value (such as -5), it indicates that the conduction abnormality of this group is significantly lower than that of the brain area above, and the gradient value is defined as a negative value. The gradient value has a clear physical meaning: a positive gradient indicates outward diffusion of conduction abnormality, and a negative gradient indicates a conduction attenuation center pointing to the core of the lesion.

[0051] As shown in FIG. 1, in one embodiment, S4 includes obtaining the second distribution position according to the gradient value of each electrode group in S4. Figure 6

[0052] S43, a plurality of continuous electrode groups are identified as a continuous feature group, wherein the plurality of electrode groups are sequentially adjacent and have the same sign of the gradient value;

[0053] S44, if the number of electrode groups in the continuous feature group exceeds a preset number threshold, a total gradient value of the continuous feature group is obtained;

[0054] S45, a continuous feature group with an absolute value of the total gradient value exceeding a preset gradient threshold is obtained as a target group, and a second distribution position of each electrode group in the electrode array is obtained.

[0055] In the present embodiment, it should be noted that in S43, the potential lesion boundary is identified by scanning the spatial continuity of the gradient sign. The gradient values of all electrode groups are detected (positive / negative), and a plurality of electrode groups with the same gradient sign and continuously adjacent in physical position are aggregated as a continuous feature group. For example, in the scenario of damage to the cingulate gyrus fiber bundle, the gradient values of the four electrode groups distributed along the midline are all negative, indicating that the conduction efficiency in this region continuously decreases and is spatially continuous, and thus the four electrode groups are identified as a continuous feature group. This process needs to exclude isolated same-sign points (such as a single negative gradient point surrounded by positive gradients), so as to ensure that the continuity reflects the real regional lesion rather than noise interference.

[0056] In S44, the continuous feature group identified in S43 is verified for pathological significance: only when the number of electrode groups in the group exceeds a preset number threshold (for example, no less than 3 groups), the total gradient value (i.e., the algebraic sum of all gradient values in the group) is calculated. This design avoids false judgments caused by sporadic gradient changes - for example, a small ischemic lesion in the parahippocampal gyrus may form three continuous negative gradient groups, and the total gradient value is significantly negative; and two isolated negative gradient groups are excluded due to insufficient number. The physical meaning of the total gradient value is to reflect the cumulative strength of the conduction abnormality in the continuous region, and the larger the absolute value is, the wider the lesion range or the deeper the conduction degradation degree. ​

[0057] It should be noted that in S44, the determination of the preset number threshold can be based on the minimum spatial scale of brain anatomical functional partition (such as the width of the primary motor cortex hand area), combined with the electrode spacing (such as the standard electrode spacing on the scalp), to calculate the minimum number of electrode groups required to cover the lesion; and a random gradient sequence can be generated, the probability distribution of naturally formed continuous groups of the same sign is counted, and the number of groups with a probability of less than 5% is taken as the threshold (such as the probability of naturally forming 4 continuous negative gradient groups is only 0.3%); and the data of confirmed patients (such as multiple lacunar infarction lesions) can be analyzed to confirm that the threshold can cover the spatial extension range of more than 85% of the lesions.

[0058] In S45, the lesion projection area is locked according to the total gradient value intensity: when the absolute value of the total gradient value of the continuous characteristic group exceeds the gradient threshold (such as the total gradient value of the deep white matter lesion area reaches -9), it is marked as the target group, and the second distribution position of all electrode groups in the group is extracted (such as the 5 electrode coordinates of the corresponding area of the precentral gyrus on the scalp). According to the spatial density distribution of the electrode position (such as linear arrangement or ring aggregation), combined with the brain anatomical projection model, the position of the target group is mapped to a specific structure in the brain (such as the intersection area of the corona radiata fiber bundle), and finally the core area of the conduction abnormality is located. This step first realizes the accurate spatial delineation of regional progressive conduction lesions (such as subcortical arteriosclerosis) under non-invasive conditions.

[0059] It should be noted that in S45, the determination of the preset gradient threshold can collect the total gradient values of different types of lesions (tumor / ischemia / epilepsy). Then group according to lesion volume: early micro-lesions (<5mm) are concentrated in the 10th percentile of the distribution curve; progressive lesions are distributed in the 25th-75th percentile; and the 15th percentile value is taken as the preset gradient threshold.

[0060] In one embodiment, the obtaining of the radian difference of each electrode group according to the preset radian and the real-time radian corresponding to each electrode group in S2 comprises:

[0061] The relative difference value between the preset radian and the real-time radian corresponding to the i-th electrode group is obtained, and the absolute value of the relative difference value is taken, and the absolute value is taken as the radian difference of the i-th electrode group.

[0062] In the embodiment, it is to be noted that the distortion degree of the electroencephalogram on the specific conduction path is quantified by calculating the absolute difference between the preset radian and the real-time radian: for the i-th electrode group, first, the theoretical phase difference (preset radian) of the initial signal setting is obtained, and then the phase offset (real-time radian) actually generated by the feedback signal is analyzed, and the absolute value of the difference between the two is taken as the radian difference of the electrode group. Taking the occipital lobe micro-bleeding focus as an example, if the preset initial signal of the electrode group is a quarter cycle phase difference, and the actual feedback changes the local electrical conductivity due to lesion vascular bleeding, resulting in a phase delay of less than an eighth cycle, the absolute deviation of the preset value and the real-time value at this time is the radian difference, and the greater the value, the more significant the signal conduction path in the corresponding brain area is disturbed. This calculation method avoids the influence of positive and negative deviation directions, converts the phase mismatch in the electromagnetic wave conduction process into a single scalar, and objectively reflects the abnormal strength of the biological physical properties caused by the lesion.

[0063] As shown in Figure 7 In one embodiment, the first distribution position of each electrode group corresponding to the radian difference exceeding the preset threshold in S3 includes:

[0064] S31, according to the layout of the electrode array, obtaining the setting position of each electrode group corresponding to the radian difference exceeding the preset threshold in the extracranial body surface region, and taking the setting position as the first distribution position, wherein the first distribution position includes the extracranial body surface region to which the electrode group belongs and the relative position in the electrode array.

[0065] In the embodiment, it is to be noted that in S31, the electrode group with excessive radian difference is mapped to the cranial surface physical coordinates. In a specific implementation, according to the pre-defined electrode array layout model (including the three-dimensional space coordinates of each electrode group on the skull surface and the belonging anatomical partition), the identification numbers of all excessive electrode groups are extracted, and the accurate geometric positions thereof on the extracranial body surface (such as the temporal gyrus region A3 group, the top-down small leaf region B5 group, etc.) are queried. The position information of each electrode group is composed of two parts: the macroscopic anatomical attribution region (for example, "left hippocampal gyrus corresponding area") and the microscopic array topological positioning (such as "array 3rd row 4th column electrode cluster"), which together form the first distribution position data. This step constructs a spatial network of abnormal points exceeding the standard, providing position anchor points for intracranial projection.

[0066] Further, taking basal ganglia calcification as an example: assuming that in the cranial surface coverage area of the precentral gyrus, the curvature difference of 4 adjacent electrode groups exceeds the threshold (threshold detection trigger). S31 locates the 4 groups through the array layout library, respectively at 1 cm from the midline of the coronal suture (2 groups in area A) and at the upper edge of the lateral fissure (2 groups in area B), forming an "X" type spatial topology. This distribution not only marks the body surface projection of the abnormal highest point, but also reveals the trend of the lesion possibly spreading along the conduction path of the pyramidal tract through the relative position relationship (such as the B group being located below and outside the A group), assisting the doctor in judging whether it is a local calcification or a diffuse calcification.

[0067] As shown in the embodiment, in one embodiment, the step of obtaining the lesion positioning result according to the first distribution position in S3 comprises: Figure 7

[0068] S32, based on the extracranial body surface area in the first distribution position and the relative position in the electrode array, identifying the internal anatomical structure area of the brain corresponding to the extracranial body surface area, and taking the position information of the internal anatomical structure area of the brain as the lesion positioning result, wherein the lesion positioning result is used to indicate the position of the lesion in the brain.

[0069] In this embodiment, it should be noted that in S32, multi-level mapping is performed based on the first distribution position: first, the cranial surface area to which the electrode group belongs is mapped to the standard brain anatomical template (such as Brodmann partition), and then spatial weighted calculation is performed according to the relative position relationship in the array (such as the electrode group located in the midline of the cranial surface tends to be mapped to the tissue beside the longitudinal fissure of the brain). Combining individualized medical image data (such as three-dimensional occupation effect formed by MRI vascular flow void effect), combined with brain conduction topology rules (such as the attraction of high conductivity liquid area to signal), the lesion positioning result is finally generated. The result is essentially the equivalent projection area of the conduction abnormality in the intracranial structure, not a direct pathological diagnosis.

[0070] Further, when 3 electrode groups exceeding the threshold are distributed in a triangular manner at the junction of the temporal and occipital lobes (macroscopically belonging to the junction of the temporal and occipital lobes, and microscopically the sites are the high, middle and low layers of the Z-axis array), S32 first maps the high layer group to the posterior part of the inferior temporal gyrus and the low layer group to the lingual gyrus of the occipital lobe, and then converges according to the intermediate position coordinates of the middle layer group, finally locating the lesion core in the choroid plexus area at the posterior part of the inferior horn of the lateral ventricle. This mapping mechanism can solve the ambiguity problem of a single cranial surface area corresponding to multiple intracranial structures (such as the temporal bone covering the hippocampus and amygdala at the same time), and accurately distinguish the lesions between the adjacent visual radiation area and the white matter of the temporal lobe.

[0071] Also provided is a lesion positioning system based on intracranial electroencephalogram features, the system comprising:

[0072] ​The data acquisition module is configured to acquire an electrode array arranged on a plurality of extracranial body surface regions, acquire two initial electrode signals continuously emitted by each electrode group in the electrode array at a preset interval and two feedback electrode signals collected after the emission, and the waveform difference between the two initial electrode signals continuously emitted at the preset interval is a preset radian.

[0073] The data processing module is configured to acquire a real-time radian difference between the two feedback electrode signals collected by each electrode group, and acquire a radian difference of each electrode group according to the corresponding preset radian and real-time radian of each electrode group.

[0074] The first data analysis and positioning module is configured to acquire a first distribution position in the electrode array of each electrode group corresponding to the radian difference exceeding the preset radian threshold when the radian difference exceeding the preset radian threshold exists in the radian difference of each electrode group, and acquire a lesion positioning result according to the first distribution position.

[0075] The second data analysis and positioning module is configured to acquire a gradient value of each electrode group according to the radian difference of each electrode group when the radian difference exceeding the preset radian threshold does not exist in the radian difference of each electrode group, acquire a second distribution position according to the gradient value of each electrode group, and acquire a lesion positioning result according to the second distribution position.

[0076] In one embodiment, the second data analysis and positioning module further comprises: acquiring a change amount of the radian difference between the i-th electrode group and each adjacent electrode group, respectively; acquiring an absolute value of the change amount, and taking the change amount corresponding to the maximum absolute value as the gradient value of the i-th electrode group.

[0077] In the present embodiment, it should be noted that, as to the above-mentioned lesion positioning system based on intracranial EEG characteristics, the specific manner of performing operations has been described in detail in the embodiments of the lesion positioning method based on intracranial EEG characteristics, and will not be described in detail here.

[0078] Figure 8 is a block diagram of an electronic device for a lesion positioning method based on intracranial EEG characteristics according to an exemplary embodiment. As shown in Figure 8 the electronic device 700 can include a processor 701 and a memory 702. The electronic device 700 can also include one or more of a multimedia component 703, an I / O interface 704 (input / output interface), and a communication component 705.

[0079] The processor 701 is configured to control overall operations of the electronic device 700 to complete all or part of the steps of the above-described intracranial brain electrical feature-based lesion positioning method. The memory 702 is configured to store various types of data to support operations of the electronic device 700, which can include, for example, instructions for operating any application or method on the electronic device 700, and application-related data, such as contact data, sent and received messages, pictures, audio, video, and the like. The memory 702 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk. The multimedia component 703 can include a screen and an audio component. The screen can be, for example, a touch screen, and the audio component is configured to output and / or input audio signals. For example, the audio component can include a microphone configured to receive external audio signals. The received audio signals can be further stored in the memory 702 or transmitted through the communication component 705. The audio component also includes at least one speaker configured to output audio signals. The I / O interface 704 provides an interface between the processor 701 and other interface modules, which can be a keyboard, a mouse, a button, and the like. The buttons can be virtual buttons or physical buttons. The communication component 705 is configured to perform wired or wireless communication between the electronic device 700 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G, 4G, NB-IOT, eMTC, or other 5G, and the like, or a combination of one or more of them, is not limited herein. Therefore, the corresponding communication component 705 can include a Wi-Fi module, a Bluetooth module, an NFC module, and the like.

[0080] In an exemplary embodiment, the electronic device 700 can be implemented by one or more Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor or other electronic elements for executing the above-mentioned intracranial EEG feature-based lesion localization method.

[0081] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the above-mentioned intracranial EEG feature-based lesion localization method. For example, the computer-readable storage medium can be the above-mentioned memory 702 including program instructions, which can be executed by the processor 701 of the electronic device 700 to complete the above-mentioned intracranial EEG feature-based lesion localization method.

[0082] In another exemplary embodiment, a computer program product is also provided, which contains a computer program capable of being executed by a programmable device, and the computer program has code portions for executing the above-mentioned intracranial EEG feature-based lesion localization method when executed by the programmable device.

[0083] The preferred embodiments of the present disclosure are described in detail above with reference to the accompanying drawings, but the present disclosure is not limited to the specific details in the above-described embodiments. Within the technical concept range of the present disclosure, various simple modifications can be made to the technical solutions of the present disclosure, and these simple modifications all belong to the protection scope of the present disclosure.

[0084] In addition, it should be noted that each specific technical feature described in the above-described specific embodiments can be combined in any appropriate manner without contradiction. In order to avoid unnecessary repetition, various possible combination manners are not described again in the present disclosure.

[0085] In addition, any combination between various different embodiments of the present disclosure can also be made, as long as it does not deviate from the idea of the present disclosure, and it should also be considered as the disclosed content of the present disclosure.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can still be modified, or some or all of the technical features can be replaced by equivalent replacements; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should be covered in the scope of the claims and the description of the present application.

Claims

1. A method for lesion localization based on intracranial electroencephalographic features, characterized in that, The system comprises: obtaining an electrode array arranged on a plurality of extracranial body surface regions, obtaining two initial electrode signals continuously emitted by each electrode group in the electrode array at a preset interval and two feedback electrode signals collected after the emission, wherein the waveform difference between the two initial electrode signals continuously emitted at the preset interval is a preset radian; obtaining the real-time radian difference between the two feedback electrode signals collected by each electrode group, and obtaining the radian difference of each electrode group according to the corresponding preset radian and real-time radian of each electrode group; if there is a radian difference exceeding the preset radian threshold in the radian difference of each electrode group, obtaining a first distribution position of each electrode group in the electrode array corresponding to the radian difference exceeding the preset radian threshold, and obtaining a lesion positioning result according to the first distribution position; if there is no radian difference exceeding the preset radian threshold in the radian difference of each electrode group, obtaining a gradient value of each electrode group according to the corresponding radian difference of each electrode group, obtaining a second distribution position according to the gradient value of each electrode group, and obtaining a lesion positioning result according to the second distribution position.

2. The method of claim 1, wherein, The method comprises: respectively obtaining the change amount of the radian difference between the i-th electrode group and each adjacent electrode group; obtaining the absolute value of the change amount, and taking the change amount corresponding to the maximum absolute value as the gradient value of the i-th electrode group.

3. The lesion localization method based on intracranial electroencephalographic features according to claim 2, characterized in that, The method comprises: identifying a plurality of continuous electrode groups as a continuous feature group, wherein a plurality of sequentially adjacent electrode groups with the same sign of the gradient value are identified as continuous; if the number of electrode groups in the continuous feature group exceeds a preset number threshold, obtaining a total gradient value of the continuous feature group; obtaining a continuous feature group with an absolute value of the total gradient value exceeding a preset gradient threshold as a target group, and obtaining a second distribution position of each electrode group in the electrode array in the target group.

4. The method of claim 1, wherein the intracranial electroencephalography features are selected from the group consisting of: The method comprises: ​ obtaining the relative difference value between the preset radian and the real-time radian corresponding to the i-th electrode group, taking the absolute value of the relative difference value, and taking the absolute value as the radian difference of the i-th electrode group.

5. The method of claim 1, wherein the intracranial electroencephalography features are selected from the group consisting of: The method comprises: ​ obtaining the setting position of each electrode group in the extracranial body surface region according to the layout of the electrode array, and taking the setting position as the first distribution position, wherein the first distribution position comprises the extracranial body surface region to which the electrode group belongs and the relative position in the electrode array.

6. The method of claim 5, wherein the intracranial electroencephalography features are selected from the group consisting of: The method comprises: ​ based on the extracranial body surface region in the first distribution position and the relative position in the electrode array, identifying an internal anatomical structure region of the brain corresponding to the extracranial body surface region, and taking the position information of the internal anatomical structure region of the brain as the lesion positioning result, wherein the lesion positioning result is used to indicate the position of the lesion in the brain.

7. A lesion localization system based on intracranial electroencephalographic features, characterized by, The system comprises: The data acquisition module is configured to acquire an electrode array arranged on a plurality of extracranial body surface regions, acquire two initial electrode signals continuously emitted by each electrode group in the electrode array at a preset interval and two feedback electrode signals collected after the emission, and the waveform difference between the two initial electrode signals continuously emitted at the preset interval is a preset radian; The data processing module is configured to acquire a real-time radian difference between the two feedback electrode signals collected by each electrode group, and acquire a radian difference of each electrode group according to the preset radian and the real-time radian of each electrode group; The first data analysis and positioning module is configured to acquire a first distribution position in the electrode array of each electrode group corresponding to the radian difference exceeding the preset radian threshold when the radian difference exceeding the preset radian threshold exists in the radian difference of each electrode group, and acquire a lesion positioning result according to the first distribution position. The second data analysis and positioning module is configured to acquire a gradient value of each electrode group according to the radian difference of each electrode group when the radian difference exceeding the preset radian threshold does not exist in the radian difference of each electrode group, acquire a second distribution position according to the gradient value of each electrode group, and acquire a lesion positioning result according to the second distribution position.

8. The intracranial EEG feature based lesion localization system of claim 7, wherein, The second data analysis and positioning module further comprises: respectively acquiring a change amount of the radian difference between the ith electrode group and each adjacent electrode group; acquiring an absolute value of the change amount, and taking the change amount corresponding to the maximum absolute value as the gradient value of the ith electrode group.

9. An electronic device, comprising: comprise: a memory having a computer program stored thereon; a processor configured to execute the computer program in the memory to implement the lesion positioning method based on intracranial electroencephalogram features according to any one of claims 1 to 6.

10. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the lesion positioning method based on intracranial electroencephalogram features according to any one of claims 1 to 6.

Images