An EIT sensor design method for optimizing edge sensitivity

Abstract

This invention discloses an EIT sensor design method for optimizing edge sensitivity. First, n electrodes are placed on a horizontal plane 3 cm above the patient's eyebrows. One electrode is placed at the center of the forehead and the center of the back of the head as fixed electrodes. The remaining electrodes are placed at intervals of arc length #imgabs0#. This uniform electrode arrangement serves as the initial electrode position for initial imaging of the patient. The electrode positions are optimized based on the location of the lesion as shown in the imaging. All electrodes are divided into two groups, moved separately, and the RS and BR values ​​are recorded and saved. Based on the RS and BR values, the electrode group closer to the lesion is fixed to the optimal position. After deleting some data, the other electrode group farther from the lesion is fixed to the optimal position again based on the RS and BR values. This invention effectively improves the quality of edge imaging by changing the electrodes to allow more current to flow through the target area, thereby increasing sensitivity, resulting in a smaller imaging BR value and fewer artifacts.

Description

Technical Field

[0001] The invention belongs to the application field of electrical tomography, and in particular relates to an EIT sensor design method for optimizing edge sensitivity. Background Art

[0002] Electrical tomography (ET) is a process tomography technology that uses the sensitive mechanism of electrical properties. It uses boundary electrodes to measure voltage information and reversely infers the distribution of electrical property parameters inside the measured object through mathematical methods, thereby deriving the actual internal object distribution. Electrical tomography mainly includes three basic imaging forms, namely electrical capacitance tomography (ECT), electrical resistance tomography (ERT) and electromagnetic tomography (EMT). Among them, electrical resistance tomography and electrical capacitance tomography can be collectively referred to as electrical impedance tomography (EIT) technology. Electrical impedance tomography has a wide range of applications in the field of medical imaging, such as the Image Monitoring for Head Phantom of Intracranial Hemorrhage Using Electrical Impedance Tomography published by L Wang et al. at the Progress In Electromagnetic Research Symposium in Shanghai in 2016. The paper conducted bleeding experiments on head models with and without craniotomy, verifying the feasibility of EIT in head imaging. In 2019, B Yang et al. published an article titled "Comparison of electrical impedance tomography and intracranial pressure during dehydration treatment of cerebral edema" in NeuroImage: Clinical, Vol. 23, No. 101909. The article successfully detected changes in the process of dehydration treatment of cerebral edema using EIT, and compared it with the ICP curve to verify the accuracy of EIT in detecting changes in cerebral edema. These studies all used a sensor arrangement with 16 electrodes evenly placed around the patient's head. This uniform electrode arrangement is very effective for detecting unknown lesions because it distributes the sensitivity more evenly around the head, so that lesions at any location can be reflected intuitively and effectively.However, in actual applications, the head is not a regular model. Evenly distributed electrodes will cause the sensitivity of some areas to be too low, affecting the imaging quality. We need better imaging quality to help doctors judge the current condition of the patient. Therefore, it is necessary to optimize the edge areas, that is, the areas with uneven sensitivity distribution and poor imaging quality, to ensure that when the lesion is located in this area, there can be better imaging quality. Summary of the invention

[0003] The technical problem solved by the present invention is to propose an EIT sensor design method for optimizing edge sensitivity, the purpose of which is to increase the sensitivity of the head edge position, thereby improving the imaging quality of this area, which in turn helps doctors judge the condition and improves the patient's cure rate.

[0004] The present invention adopts the following technical solution to solve the above technical problems: a design method of an EIT sensor for optimizing edge sensitivity, characterized in that the specific steps are as follows:

[0005] Step 1: Place one electrode at the center of the front periphery of the skull and the center of the back periphery of the skull as a fixed electrode on the horizontal plane 3 cm above the patient's eyebrows. Place the remaining n-2 electrodes at an arc length of 1 / 4 of the center of the electrodes. Placement, where c is the circumference of the patient's brain at that level, and n is the number of electrodes;

[0006] Step 2: Using the relative excitation mode and adjacent measurement mode to collect boundary voltage, calculate the sensitivity matrix and perform initial image reconstruction;

[0007] Step 3: Determine the lesion location of the cerebral hemorrhage patient based on the initial image. If the lesion is in the θ1 region, i.e., the area from the forehead to the temples on both sides, or in the θ2 region, i.e., the area from the back of the ears to the back of the head, optimize the electrodes respectively, and calculate and retain the relative sensitivity RS and blur radius BR of the lesion area:

[0008]

[0009] In the formula, S θi represents the sensitivity value of the pixel in the θi area, and S represents the sensitivity value of the pixel in the entire brain area, where the value range of i is 1 or 2;

[0010]

[0011] In the formula, R t Indicates the radius of the lesion, A t It represents the area of ​​the lesion, R represents the radius of the entire area, and A represents the area of ​​the entire area. The smaller the BR value, the more accurate the imaging target and the fewer artifacts.

[0012] The specific steps for electrode optimization are as follows:

[0013] (3.1) If the lesion area is within the range of θ1, perform the following steps:

[0014] Step 3.11, the electrodes from the forehead to the ears are defined as electrode group 1, and the remaining electrodes are defined as electrode group 2;

[0015] Step 3.12: Keep the forehead electrodes fixed in place and change the arc lengths of the electrodes in electrode group 1 from Shorten them one by one, and the step length of each shortening is Where e is the length of a single electrode, and the relative sensitivity RS and blur radius BR are calculated and retained after each shortening. The constraint condition for shortening the arc length of the electrode center interval is that the spacing between any electrode edges cannot be less than e;

[0016] Step 3.13, filter out the maximum value of the RS data stored in all interval arc length positions, and determine whether the blur radius BR value at this time is smaller than the BR value of the initial image. If so, the position is determined as the optimal position of electrode group 1, and the electrode group optimization is terminated. If not, remove the position data, repeat this step, and filter the new RS maximum value position until the optimal electrode position is obtained;

[0017] Step 3.14: Fix electrode group 1 to the optimal position, keep the position of the back occipital fixed electrode unchanged, and change the arc length of the electrode center interval of the electrodes in electrode group 2 from Lengthen it gradually at the same time, and the lengthening step each time is At the same time, the relative sensitivity RS and the blur radius BR after each lengthening are calculated and retained. The constraint condition for lengthening the arc length of the electrode center interval is that the spacing between any electrode edges cannot be less than e;

[0018] Step 3.15: Remove all the electrodes from the ears to the back of the head with less than After obtaining the position data, the maximum value of the RS data stored in all interval arc length positions is screened out, and it is determined whether the blur radius BR value at this time is smaller than the initial image BR value. If so, the position is determined as the optimal position of the electrode group 2, and the electrode group optimization is terminated. If not, the position data is removed, and this step is repeated to screen a new RS maximum value position until the optimal electrode position is obtained;

[0019] Step 3.16, fix the electrode group 2 to the optimal position, and end the optimization of this area;

[0020] (3.2) If the lesion area is within the range of θ2, perform the following steps:

[0021] Step 3.21, the electrodes from the forehead to the ears are defined as electrode group 1, and the remaining electrodes are defined as electrode group 2;

[0022] Step 3.22: The position of the occipital fixed electrode remains unchanged, and the arc length of the interval between the centers of the electrodes in the second electrode group is increased from The synchronous shortening is done one by one, and the shortening step length is Where e is the length of a single electrode, and the relative sensitivity RS and blur radius BR are calculated and retained after each shortening. The constraint condition for shortening the arc length of the electrode center interval is that the spacing between any electrode edges cannot be less than e;

[0023] Step 3.23, filter out the maximum value of the RS data stored in all interval arc length positions, and determine whether the blur radius BR value at this time is smaller than the initial image BR value. If so, then the position is determined as the optimal position of the electrode group 2, and the electrode group optimization is terminated. If not, remove the position data, repeat this step, and filter the new RS maximum value position until the optimal electrode position is obtained;

[0024] Step 3.24, fix electrode set 2 to the optimal position, keep the forehead fixed electrode in place, and change the arc length of the electrode center of electrode set 1 from Lengthen it gradually at the same time, and the lengthening step each time is At the same time, the relative sensitivity RS and the blur radius BR after each lengthening are calculated and retained. The constraint condition for lengthening the arc length of the electrode center interval is that the spacing between any electrode edges cannot be less than e;

[0025] Step 3.25: Remove all the electrodes from the ears to the back of the head with less than After obtaining the position data, the maximum value of the RS data stored in all interval arc length positions is screened out, and it is determined whether the blur radius BR value at this time is smaller than the initial image BR value. If so, the position is determined as the optimal position of electrode group 1, and the electrode group optimization is terminated. If not, the position data is removed, and this step is repeated to screen a new RS maximum value position until the optimal electrode position is obtained;

[0026] Step 3.26: Fix the electrode group 1 to the optimal position, and end the optimization of this area.

[0027] It is further defined that the number of electrodes n is in the range of 8, 10, 12, 14, 16 or 18.

[0028] Further defined, the relative excitation mode is specifically: the forehead fixed electrode is electrode No. 1 and the n electrodes are arranged in reverse time, and the first excitation is to inject current into electrode No. 1, The No. 1 electrode, i.e. the fixed electrode at the back of the head, is grounded. The second stimulation is to inject current into the No. 2 electrode. Electrode No. is grounded; by analogy, when the number of grounded electrodes exceeds n, the actual electrode number is obtained by subtracting n from the electrode number, such as When the excitation electrode is The ground electrode is The actual electrode number is 1, until the current is injected into the nth electrode. The excitation is terminated when the No. 1 electrode is grounded, so that the length of the connection line between all injected current and the grounded electrode pairs is close to the longest diameter of the head, which is convenient for the current to penetrate the skull and obtain the best effect.

[0029] To further define, the sensitivity matrix is ​​specifically: the brain area is segmented and modeled using the finite element method, that is, it is divided into a finite number of small grids, each grid is a pixel point, and the sensitivity of each grid to the change in conductivity is calculated by measuring the voltage value. The higher the sensitivity, the greater the change in conductivity. The matrix composed of all pixel points is the sensitivity matrix.

[0030] Further defined, the calculation formula of the pixel sensitivity value is specifically: Where i and j represent the sensitivity pixel at row i and column j, They represent the excitation currents of the i-th electrode group and the j-th electrode group, respectively. i ,I j The ladder of the electric potential distribution in the field.

[0031] It is further defined that image reconstruction is specifically as follows: each time imaging is performed, n electrodes arranged around the patient's head are used, and a relative excitation mode is adopted to obtain measurement voltages counterclockwise from two adjacent electrodes except the excitation electrode and the ground electrode, and the sensitivity matrix obtained by the corresponding finite element model is used to invert the conductivity changes of the brain, that is, image reconstruction.

[0032] The present invention has the following beneficial effects: the present invention divides the electrodes into two groups, the forehead and the occipital, the electrode group close to the lesion area shortens the arc length of the electrode center interval, and the electrode group far from the lesion area lengthens the arc length of the electrode center interval, and finds the optimal electrode position through relative sensitivity and fuzzy radius. When the electrode optimization is completed, the sensitivity is concentrated on the edge position to improve the imaging quality. The present invention effectively solves the problems of inaccurate imaging at the edge position, affecting the judgment of the patient's condition, and effectively improves the patient's cure rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is the grouping and position of the electrodes at a certain moment in the process of optimizing the electrode arrangement in the θ1 region of the present invention.

[0034] Figure 2 A flowchart of the design of an EIT sensor for optimizing edge sensitivity provided by the present invention.

[0035] Figure 3 Graphs 1 and 2 are sensitivity images of uniform electrodes and sensitivity images of the final electrode distribution for the θ1 region according to the present invention.

[0036] Figure 4 It is a bar graph of the optimal relative sensitivity of the uniform electrode and the different areas of the present invention.

[0037] Figure 5 Image comparison of two electrode arrangements for six different lesions in the θ1 region.

[0038] Figure 6 BR values ​​of two electrode arrangements for six different lesion conditions in the θ1 region. DETAILED DESCRIPTION

[0039] The design of an EIT sensor for optimizing edge sensitivity of the present invention is described in detail with reference to the accompanying drawings and embodiments.

[0040] The EIT sensor design method for optimizing edge sensitivity described in the present invention aims to optimize edge sensitivity. By changing the electrode position, the path of the current is changed, so that the medium change of the voltage signal to the edge position is more significant. First, all electrodes are placed on a horizontal plane 3 cm above the eyebrows, and the forehead fixed electrode 3 and the back occipital fixed electrode 4 are set. Then, the electrodes are grouped into electrode group 1 and electrode group 2. The distance 5 between the electrode arrangement plane and the eyebrow plane is 3 cm. Figure 1 The specific process of performing different optimization operations on different regional locations is as follows Figure 2 As shown, the specific steps include:

[0041] Step 1: For patients with cerebral hemorrhage and cerebral hemorrhage accompanied by secondary ischemia, place one electrode at the center of the anterior periphery of the skull and one at the center of the occipital periphery on a horizontal plane 3 cm above the eyebrows as a fixed electrode, and place the remaining n-2 electrodes at an arc length of 1 / 2. Placement, where c is the circumference of the patient's brain at that horizontal plane, and the number of electrodes n ranges from 8, 10, 12, 14, 16 or 18.

[0042] Step 2: Use the relative excitation mode and adjacent measurement mode to collect boundary voltage, calculate the sensitivity matrix and perform initial image reconstruction.

[0043] The relative excitation mode is as follows: the forehead fixed electrode is electrode No. 1 and the n electrodes are arranged in reverse order. The first excitation is to inject current into electrode No. 1. The No. 1 electrode, i.e. the fixed electrode at the back of the head, is grounded. The second stimulation is to inject current into the No. 2 electrode. Electrode No. is grounded; by analogy, when the number of grounded electrodes exceeds n, the actual electrode number is obtained by subtracting n from the electrode number, such as When the excitation electrode is The ground electrode is The actual electrode number is 1, until the current is injected into the nth electrode. The excitation is terminated when the No. 1 electrode is grounded, so that the length of the connection line between all injected current and the grounded electrode pairs is close to the longest diameter of the head, which is convenient for the current to penetrate the skull and obtain the best effect.

[0044] The sensitivity matrix is ​​specifically as follows: the finite element method is used to segment and model the brain area, that is, it is divided into a finite number of small grids, each grid is a pixel point, and the sensitivity of each grid to the change in conductivity is calculated by measuring the voltage value. The higher the sensitivity, the greater the change in conductivity. The matrix composed of all pixel points is the sensitivity matrix.

[0045] The calculation formula of pixel sensitivity value is as follows: Where i and j represent the sensitivity pixel at row i and column j, They represent the excitation currents of the i-th electrode group and the j-th electrode group, respectively. i ,I j The gradient of the electric potential distribution in the field at time t.

[0046] Image reconstruction is as follows: Each time imaging is performed, n electrodes arranged around the patient's head are used, and a relative excitation mode is adopted to obtain measurement voltages counterclockwise from two adjacent electrodes except the excitation electrode and the ground electrode. The sensitivity matrix obtained by the corresponding finite element model is used to invert the conductivity changes of the brain, that is, image reconstruction.

[0047] Step 3: Determine the location of the patient's lesion based on the initial image. If the lesion is in the θ1 region, i.e., the area from the forehead to the temples on both sides, or in the θ2 region, i.e., the area from the back of the ears to the back of the head, optimize the electrodes respectively, and calculate and retain the relative sensitivity RS and blur radius BR of the lesion area:

[0048]

[0049] In the formula, S θi represents the sensitivity value of the pixel point in the θi area, and S represents the sensitivity value of the pixel point in the entire brain area, where the value range of i is 1 or 2.

[0050]

[0051] In the formula, R t Indicates the radius of the lesion, A t It represents the area of ​​the lesion, R represents the radius of the entire area, and A represents the area of ​​the entire area. The smaller the BR value, the more accurate the imaging target and the fewer artifacts.

[0052] The specific steps for electrode optimization are as follows:

[0053] (3.1) If the lesion area is within the range of θ1, perform the following steps:

[0054] Step 3.11: The electrodes from the forehead to the ears are defined as electrode group one, and the remaining electrodes are defined as electrode group two.

[0055] Step 3.12: Keep the forehead electrodes fixed in place and change the arc lengths of the electrodes in electrode group 1 from Shorten them one by one, and the step length of each shortening is Where e is the length of a single electrode. The relative sensitivity RS and blur radius BR are calculated and retained after each shortening. The constraint condition for shortening the arc length of the electrode center interval is that the spacing between any electrode edges cannot be less than e.

[0056] Step 3.13, filter out the maximum value in the RS data saved at all interval arc length positions, and determine whether the blur radius BR value at this time is smaller than the BR value of the initial image. If so, then define the position as the optimal position of electrode group one and end the optimization of the electrode group. Otherwise, remove the position data and repeat this step to filter the new RS maximum value position until the optimal electrode position is obtained.

[0057] Step 3.14: Fix electrode group 1 to the optimal position, keep the position of the back occipital fixed electrode unchanged, and change the arc length of the electrode center interval of the electrodes in electrode group 2 from Lengthen it gradually at the same time, and the lengthening step each time is At the same time, the relative sensitivity RS and the blur radius BR after each lengthening are calculated and retained. The constraint condition for lengthening the arc length of the electrode center interval is that the spacing between any electrode edges cannot be less than e.

[0058] Step 3.15: Remove all the electrodes from the ears to the back of the head with less than After obtaining the position data, the maximum value of the RS data saved in all interval arc length positions is screened out, and it is determined whether the blur radius BR value at this time is smaller than the initial image BR value. If so, the position is determined as the optimal position of electrode group 2, and the electrode group optimization is ended. Otherwise, the position data is removed, and this step is repeated to screen the new RS maximum value position until the optimal electrode position is obtained.

[0059] Step 3.16: Fix electrode group 2 to the optimal position and end the optimization of this area.

[0060] (3.2) If the lesion area is within the range of θ2, perform the following steps:

[0061] Step 3.21: The electrodes from the forehead to the ears are defined as electrode group one, and the remaining electrodes are defined as electrode group two.

[0062] Step 3.22: The position of the occipital fixed electrode remains unchanged, and the arc length of the interval between the centers of the electrodes in the second electrode group is increased from The synchronous shortening is done one by one, and the shortening step length is Where e is the length of a single electrode. The relative sensitivity RS and blur radius BR are calculated and retained after each shortening. The constraint condition for shortening the arc length of the electrode center interval is that the spacing between any electrode edges cannot be less than e.

[0063] Step 3.23, filter out the maximum value in the RS data saved at all interval arc length positions, and determine whether the blur radius BR value at this time is smaller than the initial image BR value. If so, then define the position as the optimal position of electrode group two and end the optimization of the electrode group. Otherwise, remove the position data and repeat this step to filter the new RS maximum value position until the optimal electrode position is obtained.

[0064] Step 3.24, fix electrode set 2 to the optimal position, keep the forehead fixed electrode in place, and change the arc length of the electrode center of electrode set 1 from Lengthen it gradually at the same time, and the lengthening step each time is At the same time, the relative sensitivity RS and the blur radius BR after each lengthening are calculated and retained. The constraint condition for lengthening the arc length of the electrode center interval is that the spacing between any electrode edges cannot be less than e.

[0065] Step 3.25: Remove all the electrodes from the ears to the back of the head with less than After obtaining the position data, the maximum value of the RS data saved in all interval arc length positions is screened out, and it is determined whether the blur radius BR value at this time is smaller than the initial image BR value. If so, the position is determined as the optimal position of electrode group one, and the electrode group optimization is ended. Otherwise, the position data is removed, and this step is repeated to screen the new RS maximum value position until the optimal electrode position is obtained.

[0066] Step 3.26: Fix the electrode group 1 to the optimal position, and end the optimization of this area.

[0067] (3.3) If the lesion area is located in other positions, no operation will be performed.

[0068] like Figure 3 The figure shows the sensitivity diagram of the uniform arrangement and the present invention. The dark area in the figure is the area with lower sensitivity, and the bright area is the area with higher sensitivity. It can be seen in the figure that the sensitivity distribution of the uniform arrangement is relatively uniform, and each part has a certain sensitivity. On the contrary, the sensitivity of the present invention is more concentrated in the θ1 area, so the voltage information collected by the electrode distribution of the present invention is almost all from the θ1 area, reducing irrelevant signals and improving the signal-to-noise ratio.

[0069] like Figure 4The figure shows the optimal relative sensitivity bar graph of uniform arrangement and the present invention. It can be seen that the relative sensitivity of the present invention in the θ1 region is 60% higher than that of the uniform arrangement, and the relative sensitivity of the present invention in the θ2 region is 50% higher than that of the uniform arrangement, which confirms the superiority of the present invention in sensitivity at the edge position.

[0070] like Figure 5 The figure shows the imaging comparison between the uniform arrangement and the present invention under different pathological conditions. Referring to the parameters of real human tissue, the brain background conductivity is 0.15S / m, which is the conductivity of human cerebrospinal fluid. In actual cerebral hemorrhage patients, brain hemorrhage is often accompanied by secondary ischemia. Therefore, the two lesion sites in this example are set as one for hemorrhage (both on the left side) and one for ischemia (both on the right side). The hemorrhage conductivity is set to 0.8S / m and the ischemia conductivity is set to 0.06S / m. The second row is the image of the uniform electrode; the third row is the electrode arrangement optimized by the present invention. It can be seen that the electrode imaging optimized by the present invention has fewer artifacts and is closer to the real model. The image has almost no deformation, which proves that the electrode arrangement optimized by the present invention has significantly improved the imaging accuracy of the θ1 area.

[0071] like Figure 6 Shown Figure 5 The BR values ​​of the six cases are compared, where the smaller the BR value, the more accurate the image and the fewer artifacts. It can be seen that in all six cases, the BR value of the imaging of the present invention is smaller than the BR value of the uniformly arranged imaging. Since the changes in the conductivity of hemorrhage and ischemia are opposite, the signals obtained from the boundary voltage will be offset, and the image quality will be reduced. Figure 5 As shown in the image of the uniform electrode arrangement in the last three columns, the present invention can still maintain a very low BR value and has very good imaging quality.

[0072] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A design method for an EIT sensor with optimized edge sensitivity, characterized in that The specific steps are as follows: Step 1: Place one electrode at the center of the front periphery of the skull and the center of the back periphery of the skull as a fixed electrode on the horizontal plane 3 cm above the patient's eyebrows. Place the remaining n-2 electrodes at an arc length of 1 / 4 of the center of the electrodes. Placement, where c is the circumference of the patient's brain at that level, and n is the number of electrodes; Step 2: Using the relative excitation mode and adjacent measurement mode to collect boundary voltage, calculate the sensitivity matrix and perform initial image reconstruction; Step 3: Determine the lesion location of the cerebral hemorrhage patient based on the initial image. If the lesion is in the θ1 region, i.e., the area from the forehead to the temples on both sides, or in the θ2 region, i.e., the area from the back of the ears to the back of the head, optimize the electrodes respectively, and calculate and retain the relative sensitivity RS and blur radius BR of the lesion area: In the formula, S θi represents the sensitivity value of the pixel in the θi area, and S represents the sensitivity value of the pixel in the entire brain area, where the value range of i is 1 or 2; In the formula, R t Indicates the radius of the lesion, A t It represents the area of ​​the lesion, R represents the radius of the entire area, and A represents the area of ​​the entire area. The smaller the BR value, the more accurate the imaging target and the fewer artifacts. The specific steps for electrode optimization are as follows: (3.1) If the lesion area is within the range of θ1, perform the following steps: Step 3.11, the electrodes from the forehead to the ears are defined as electrode group 1, and the remaining electrodes are defined as electrode group 2; Step 3.12: Keep the forehead electrodes fixed in place and change the arc lengths of the electrodes in electrode group 1 from Shorten them one by one, and the step length of each shortening is Where e is the length of a single electrode, and the relative sensitivity RS and blur radius BR are calculated and retained after each shortening. The constraint condition for shortening the arc length of the electrode center interval is that the spacing between any electrode edges cannot be less than e; Step 3.13, filter out the maximum value of the RS data stored in all interval arc length positions, and determine whether the blur radius BR value at this time is smaller than the BR value of the initial image. If so, the position is determined as the optimal position of electrode group 1, and the electrode group optimization is terminated. If not, remove the position data, repeat this step, and filter the new RS maximum value position until the optimal electrode position is obtained; Step 3.14, fix electrode group 1 to the optimal position, keep the position of the back occipital fixed electrode unchanged, and change the arc length of the electrode center interval of the electrodes in electrode group 2 from Lengthen it gradually at the same time, and the lengthening step each time is At the same time, the relative sensitivity RS and the blur radius BR after each lengthening are calculated and retained. The constraint condition for lengthening the arc length of the electrode center interval is that the spacing between any electrode edges cannot be less than e; Step 3.15: Remove all the electrodes from the ears to the back of the head with less than After obtaining the position data, the maximum value of the RS data stored in all interval arc length positions is screened out, and it is determined whether the blur radius BR value at this time is smaller than the initial image BR value. If so, the position is determined as the optimal position of the electrode group 2, and the electrode group optimization is terminated. If not, the position data is removed, and this step is repeated to screen a new RS maximum value position until the optimal electrode position is obtained; Step 3.16, fix the electrode group 2 to the optimal position, and end the optimization of this area; (3.2) If the lesion area is within the range of θ2, perform the following steps: Step 3.21, the electrodes from the forehead to the ears are defined as electrode group 1, and the remaining electrodes are defined as electrode group 2; Step 3.22: The position of the occipital fixed electrode remains unchanged, and the arc length of the interval between the centers of the electrodes in the second electrode group is changed from The synchronous shortening is done one by one, and the shortening step length is Where e is the length of a single electrode, and the relative sensitivity RS and blur radius BR are calculated and retained after each shortening. The constraint condition for shortening the arc length of the electrode center interval is that the spacing between any electrode edges cannot be less than e; Step 3.23, filter out the maximum value of the RS data stored in all interval arc length positions, and determine whether the blur radius BR value at this time is smaller than the initial image BR value. If so, then the position is determined as the optimal position of the electrode group 2, and the electrode group optimization is terminated. If not, remove the position data, repeat this step, and filter the new RS maximum value position until the optimal electrode position is obtained; Step 3.24, fix electrode set 2 to the optimal position, keep the forehead fixed electrode in place, and change the arc length of the electrode center of electrode set 1 from Lengthen it gradually at the same time, and the lengthening step each time is At the same time, the relative sensitivity RS and the blur radius BR after each lengthening are calculated and retained. The constraint condition for lengthening the arc length of the electrode center interval is that the spacing between any electrode edges cannot be less than e; Step 3.25: Remove all the electrodes from the ears to the back of the head with less than After obtaining the position data, the maximum value of the RS data stored in all interval arc length positions is screened out, and it is determined whether the blur radius BR value at this time is smaller than the initial image BR value. If so, the position is determined as the optimal position of electrode group 1, and the electrode group optimization is terminated. If not, the position data is removed, and this step is repeated to screen a new RS maximum value position until the optimal electrode position is obtained; Step 3.26, fix the electrode group to the optimal position, and end the optimization of this area.

2. The EIT sensor design method for optimizing edge sensitivity according to claim 1, characterized in that: The number of electrodes n ranges from 8, 10, 12, 14, 16 or 18.

3. The EIT sensor design method for optimizing edge sensitivity according to claim 1, characterized in that The relative excitation mode is as follows: the forehead fixed electrode is electrode No. 1 and the n electrodes are arranged in reverse order. The first excitation is to inject current into electrode No.

1. The No. 1 electrode, i.e. the fixed electrode at the back of the head, is grounded. The second stimulation is to inject current into the No. 2 electrode. Electrode No. is grounded; by analogy, when the number of grounded electrodes exceeds n, the actual electrode number is obtained by subtracting n from the electrode number, such as When the excitation electrode is The ground electrode is The actual electrode number is 1, until the current is injected into the nth electrode. The excitation is terminated when the No. 1 electrode is grounded, so that the length of the connection line between all injected current and the grounded electrode pairs is close to the longest diameter of the head, which is convenient for the current to penetrate the skull and obtain the best effect.

4. The EIT sensor design method for optimizing edge sensitivity according to claim 1, characterized in that The sensitivity matrix is ​​specifically as follows: the finite element method is used to segment and model the brain area, that is, it is divided into a finite number of small grids, each grid is a pixel point, and the sensitivity of each grid to the change in conductivity is calculated by measuring the voltage value. The higher the sensitivity, the greater the change in conductivity. The matrix composed of all pixel points is the sensitivity matrix.

5. The EIT sensor design method for optimizing edge sensitivity according to claim 1, characterized in that The calculation formula of pixel sensitivity value is as follows: Where i and j represent the sensitivity pixel at row i and column j, They represent the excitation currents of the i-th electrode group and the j-th electrode group, respectively. i ,I j The ladder of the electric potential distribution in the field.

6. The EIT sensor design method for optimizing edge sensitivity according to claim 1, characterized in that Image reconstruction is as follows: Each time imaging is performed, n electrodes arranged around the patient's head are used, and a relative excitation mode is adopted to obtain measurement voltages counterclockwise from two adjacent electrodes except the excitation electrode and the ground electrode. The sensitivity matrix obtained by the corresponding finite element model is used to invert the conductivity changes of the brain, that is, image reconstruction.

Images