Cerebral stroke monitoring device and method based on non-contact electrical impedance imaging and current focusing

Abstract

The present invention discloses a device and method for monitoring cerebral stroke based on non-contact electrical impedance imaging and current focusing. The device is composed of a non-contact current focusing sensor, an electrode switching module, a data acquisition module and an image reconstruction host computer. The non-contact current focusing sensor includes a focusing electrode layer, a detection electrode layer and two insulating layers. Among them, the focusing electrode is used to achieve a current focusing effect, thereby improving the detection depth and sensitivity of the signal; the detection electrode is responsible for obtaining the detection current containing brain electrical impedance information. The image reconstruction host computer is used to achieve image reconstruction and obtain real-time images and quantitative parameters of cerebral stroke. The present invention has the advantages of being non-contact, wearable, safe and friendly, highly sensitive, low cost, and having a fast imaging speed. It is suitable for scenarios such as homes, clinics, ambulances, etc. that require rapid diagnosis or real-time monitoring and are not covered by equipment such as CT / MRI, and avoids the measurement requirements of existing electrical impedance imaging equipment where patients need to shave their heads or implant electrodes.

Description

Technical Field

[0001] The present invention relates to electrical impedance tomography technology, and in particular to a non-contact electrical impedance tomography and current focusing cerebral stroke monitoring device and method. Background Art

[0002] Stroke is a serious cerebrovascular disease characterized by its rapid onset, high mortality and disability rates, and is a major concern worldwide. Stroke can be caused by either hemorrhagic or ischemic lesions. Interventions for different types of stroke are distinct and require absolute differentiation. Patients with hemorrhagic stroke require prompt emergency surgery, while patients with ischemic stroke require thrombolytic therapy with tissue-type plasminogen activator within hours of onset. Because information such as stroke type, location, and size plays a crucial role in selecting a treatment plan for a patient, rapid diagnosis of stroke and timely intervention and treatment are crucial to improving the prognosis of stroke patients.

[0003] Computed tomography (CT) and magnetic resonance imaging (MRI) are the gold-standard imaging methods currently used in clinical stroke diagnosis. Physicians use the high-resolution intracranial images obtained by CT or MRI to reliably assess the condition. However, the high cost and bulk of CT and MRI equipment make it difficult to schedule CT or MRI scans at the patient's home, in an ambulance, or at a primary care facility, a major limitation in rapidly diagnosing the condition. It has been reported that the process of transporting patients to the hospital, waiting for CT / MRI scans, and awaiting scan reports and diagnostic results often takes several hours, leaving only approximately one-quarter of patients receiving intervention within this prime time window. Furthermore, CT / MRI scans pose radiation hazards, making them unsuitable for long-term monitoring during treatment and recovery. Therefore, research into novel stroke monitoring technologies remains of great scientific significance and medical value. Clinically, a more convenient and safe technical means is needed that can be used for rapid diagnosis and real-time monitoring of stroke, improve the intervention rate of stroke patients during the golden period, and monitor the development of the lesion area during treatment, so as to improve the prognosis of stroke patients to the greatest extent and protect their life, health and quality of life.

[0004] Electrical impedance tomography (EIT), an electrical imaging technique proposed in the 1980s, offers advantages such as simple structure, rapid response, low cost, high safety, and good reliability, making it a hot topic in the field of biomedical imaging. However, existing EIT technology still fails to meet the requirements of practical applications. Its technical bottlenecks primarily lie in contact measurement and the complex structure of the brain. Firstly, the EIT system requires close contact between the electrodes and the scalp, making it a patient-unfriendly measurement method that not only causes additional physical and psychological harm to patients but also presents contact impedance issues. To address this issue, researchers have proposed capacitively coupled electrical impedance tomography (CCEIT). By incorporating the principle of capacitive coupling, CCEIT achieves contactless electrical impedance measurement and has demonstrated its feasibility in brain imaging, demonstrating its potential as a patient-friendly stroke monitoring tool. However, as a new technology, relevant research remains insufficient. On the other hand, the presence of the scalp and skull in the brain structure will significantly hinder the measurement path of the electrical impedance signal, affecting the detection depth of the effective signal. In EIT stroke monitoring, the presence of the scalp and skull in the brain structure will significantly hinder the measurement path of the electrical impedance signal, affecting the detection depth of the effective signal. This problem is also an important factor restricting the further development of non-contact electrical impedance imaging technology in stroke monitoring. To address this problem, in existing EIT stroke monitoring, researchers use methods such as implanted electrodes, but this method is not suitable for non-contact electrical impedance imaging. Therefore, there is an urgent need to seek more effective non-contact stroke monitoring devices and methods that can simultaneously overcome the problems of non-contact measurement and reduce the adverse effects of the measurement path of the brain structure. Summary of the Invention

[0005] In response to the above problems, the present invention discloses a cerebral stroke monitoring device and method based on non-contact electrical impedance imaging and current focusing. By introducing the focusing excitation principle and combining it with CCEIT technology, a non-contact wearable cerebral stroke monitoring device is designed, and a corresponding new method for cerebral stroke monitoring is proposed, which can reduce the influence of brain structure on the electrical impedance measurement path while performing non-contact measurement, and obtain cerebral stroke monitoring results with higher sensitivity and reliability. Compared with the cerebral stroke monitoring device and method based on CT / MRI, the present invention has the advantages of being wearable, safe and friendly, low cost, and fast imaging speed. It is suitable for scenes that require rapid diagnosis or real-time monitoring, such as homes, clinics, and ambulances that cannot be covered by equipment such as CT / MRI. Compared with the existing cerebral stroke monitoring device and method based on electrical impedance imaging, the present invention can achieve non-contact measurement, avoiding the measurement requirements of the patient to shave or implant electrodes in the scalp in the existing electrical impedance imaging device and method, while improving the electrical impedance measurement and imaging sensitivity of the stroke area.

[0006] The technical solutions of the present invention are as follows:

[0007] In a first aspect, the present invention provides a non-contact current focusing sensor for cerebral stroke monitoring, wherein the sensor is in the shape of a hemispherical shell and comprises, from the outside to the inside, a focusing electrode layer, a first insulating layer, a detection electrode layer, and a second insulating layer;

[0008] The focusing electrode layer includes n spherical flexible metal focusing electrodes of the same shape and size. Adjacent focusing electrodes have gaps between them and do not touch each other, forming a hemispherical shell shape to cover the human brain area. The focusing electrodes are used to achieve a current focusing effect after an excitation signal is applied, and n ≥ 2.

[0009] The detection electrode layer includes multiple detection electrodes, which are used to obtain detection current containing brain electrical impedance information. All detection electrodes are divided into n groups, and each of the n groups of detection electrodes corresponds one-to-one to the n focusing electrodes. Each group of detection electrodes is arranged in the coverage area of ​​its corresponding focusing electrode, and each detection electrode in the group is evenly distributed within the coverage area.

[0010] The focusing electrode layer and the detection electrode layer are separated by a first insulating layer to achieve isolation of the excitation and detection signals; the second insulating layer serves as the innermost layer of the non-contact current focusing sensor.

[0011] In a second aspect, the present invention provides a cerebral stroke monitoring device based on non-contact electrical impedance imaging and current focusing, which includes the non-contact current focusing sensor described above, an electrode switching module, a data acquisition module, and an image reconstruction host computer;

[0012] The electrode switching module is used to control and switch the states of the focusing electrode and the detection electrode;

[0013] The data acquisition module is used to control the electrode switching module and generate an excitation signal to the focusing electrode. The data acquisition module collects the detection signal of the detection electrode and communicates with the image reconstruction host computer;

[0014] The image reconstruction host computer is used to realize image reconstruction and obtain real-time images and quantitative parameters for cerebral stroke monitoring.

[0015] In a third aspect, the present invention further provides a method for monitoring cerebral stroke based on the device, which comprises the following steps:

[0016] 1) The image reconstruction host computer sends data acquisition instructions through the serial port, and the data acquisition module enters the acquisition mode and initializes the electrode state. All focusing electrodes are initialized to the ground state, and all detection electrodes are initialized to the suspended state;

[0017] 2) Selection of focusing electrodes: When n = 2, one focusing electrode is selected and switched to the energized state, and the focusing electrodes that are not in the energized state are switched to the grounded state; when n ≥ 3, n-2 or n-1 adjacent focusing electrodes are selected and switched to the energized state, and the focusing electrodes that are not in the energized state are switched to the grounded state;

[0018] 3) After the selection of the focusing electrode is completed, one detection electrode is selected to be in a detection state to realize the collection of the detection signal, and the detection electrodes that are not in the detection state are switched to a suspended state; wherein, the detection electrodes in the detection state are selected from the detection electrode group corresponding to the area covered by the focusing electrode in the grounded state, and one detection electrode is selected from the group and switched to the detection state each time, while the remaining electrodes are in the suspended state; after completing one detection electrode selection, the image reconstruction host computer controls the data acquisition module to perform a data collection and sends the data to the image reconstruction host computer; when the data collection is completed, the detection electrodes are switched until the detection electrodes in the detection electrode group corresponding to the area covered by the focusing electrode in the grounded state are switched to the detection state in sequence and the signal collection is completed, thereby completing the signal measurement under a set of focusing electrode selection schemes;

[0019] 4) Return to step 2) and change the focusing electrode selection scheme. When n=2, select another focusing electrode and switch it to the energized state. When n≥3, select n-1 or n-2 adjacent focusing electrodes that have not been previously combined and switch them to the energized state. Repeat step 3) to perform a set of signal measurements under the current focusing electrode selection scheme. This continues until all optional focusing electrode combination selection schemes have been selected and measured.

[0020] 5) Image reconstruction The host computer calculates all independent electrical impedance measurement values ​​based on the collected data, and reconstructs the image using the image reconstruction algorithm to obtain a visual image for cerebral stroke monitoring and calculate the quantitative parameters for cerebral stroke monitoring.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] 1) Compared with existing stroke monitoring devices and methods, the present invention has the advantages of being non-contact, wearable, safe and friendly, highly sensitive, low cost, and fast imaging speed. It is suitable for scenarios requiring rapid diagnosis or real-time monitoring, such as homes, clinics, and ambulances, which are not covered by CT / MRI and other equipment.

[0023] 2) Compared with existing stroke monitoring devices and methods based on electrical impedance imaging, the present invention does not require patients to shave their heads or implant electrodes, and can achieve stroke imaging in a non-contact, friendly manner, providing a new optimization solution for real-time monitoring of stroke.

[0024] 3) Compared with existing stroke monitoring devices and methods based on electrical impedance imaging, the present invention introduces a current focusing effect combined with non-contact electrical impedance imaging, improves the detection depth and sensitivity of the signal through focused excitation, effectively reduces the adverse effects of the insulating layer and the scalp and skull in the brain structure on the electrical impedance measurement, obtains more effective electrical impedance signals carrying stroke information, and thus obtains higher quality visual images and quantitative parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The overall framework of the brain stroke monitoring device using non-contact electrical impedance imaging and current focusing;

[0026] Figure 2 This is a schematic diagram of the structure of a contactless current focusing sensor;

[0027] Figure 3 This is a front view diagram of a contactless current focusing sensor;

[0028] Figure 4 This is a schematic diagram of the electrode switching module;

[0029] Figure 5 Schematic diagram of the PCB layout of the electrode switching module;

[0030] Figure 6 This is a schematic diagram of the data acquisition module;

[0031] Figure 7 The present invention is a flowchart of a method for monitoring brain stroke based on non-contact electrical impedance tomography and current focusing;

[0032] Figure 8 Schematic diagram of the focusing electrode state in the combined excitation scheme;

[0033] Figure 9 is the sensitivity index S under focusing excitation io Compared with the traditional non-focused excitation sensitivity index Sio The ratio of

[0034] Figure 10 These are the image reconstruction results under two schemes: unfocused excitation (276 projections) and focused excitation (72 projections). DETAILED DESCRIPTION

[0035] The present invention will be further described and illustrated below in conjunction with specific embodiments. The embodiments are merely illustrative of the present disclosure and do not limit its scope. The technical features of the various embodiments of the present invention may be combined accordingly, provided that there is no conflict between them.

[0036] like Figure 1 The figure shows the overall framework of the present invention. The stroke monitoring device based on non-contact electrical impedance imaging and current focusing primarily consists of a non-contact current focusing sensor, an electrode switching module, a data acquisition module, and an image reconstruction host computer. The image reconstruction host computer controls the data acquisition module to generate electrode switching control signals, which in turn controls the data acquisition module to collect waveform data, calculate electrical impedance based on the waveform, and image the electrical impedance using an image reconstruction algorithm.

[0037] like Figure 2 As shown, the non-contact current focusing sensor of the present invention is composed of a four-layer structure made of wearable material, which includes, from the outside to the inside, a focusing electrode layer, a first insulating layer, a detection electrode layer, and a second insulating layer. Among them, the focusing electrode layer includes n (n ≥ 2) spherical flexible metal focusing electrodes of the same shape and size, with gaps between adjacent focusing electrodes, which do not touch each other and together form a hemispherical shell shape for covering the human brain area; the focusing electrode is used to achieve a current focusing effect after an excitation signal is applied; the detection electrode layer includes multiple detection electrodes, which are used to obtain a detection current containing brain electrical impedance information. All detection electrodes are divided into n groups, and the n groups of detection electrodes correspond one-to-one to the n focusing electrodes. Each group of detection electrodes is arranged in the coverage area of ​​its corresponding focusing electrode, and the detection electrodes in the group are evenly distributed within the coverage area.

[0038] The non-contact current focusing sensor of the present invention has two states: energized and grounded. The detection electrode has two states: detection and suspended. When using the sensor, the focusing electrode is first selected, followed by a single detection electrode for signal acquisition. By switching detection electrodes and changing the selection scheme for the focusing electrode, a comprehensive and effective electrical impedance signal carrying stroke information can be obtained.

[0039] Specifically, the signal measurement process of the present invention includes:

[0040] (1) When n=2, one focusing electrode is selected and switched to the excited state, and the focusing electrodes that are not in the excited state are switched to the grounded state; when n≥3, n-2 or n-1 adjacent focusing electrodes are selected and switched to the excited state, and the focusing electrodes that are not in the excited state are switched to the grounded state;

[0041] (2) After the selection of the focusing electrode is completed, one detection electrode is selected to be in the detection state to realize the collection of the detection signal, and the detection electrodes that are not in the detection state are switched to the suspended state; wherein, the detection electrodes in the detection state are selected from the detection electrode group corresponding to the area covered by the focusing electrode in the grounded state, and one detection electrode is selected from the group and switched to the detection state each time, and the remaining electrodes are in the suspended state; after completing one detection electrode selection, data collection is performed once; when one data collection is completed, the detection electrodes are switched until the detection electrodes in the detection electrode group are switched to the detection state in sequence and the signal collection is completed, thereby completing the signal measurement under a set of focusing electrode selection schemes;

[0042] (3) Return to step (1) and change the selection scheme of the focusing electrode. When n=2, select another focusing electrode to switch to the excitation state; when n≥3, select n-1 blocks or n-2 adjacent focusing electrodes that have not been previously combined to switch to the excitation state, and repeat step (2) to perform a set of signal measurements under the current focusing electrode selection scheme; until all optional combination selection schemes of the focusing electrodes are selected and measured.

[0043] Figure 3A side view schematic diagram of a non-contact current focusing sensor in a specific embodiment is shown. In this embodiment, the focusing electrode layer comprises four spherical flexible metal electrodes of identical shape and size. Each spherical flexible metal electrode serves as a focusing electrode, with an area of ​​approximately 1 / 8 of a sphere. Adjacent focusing electrodes are separated by gaps, preventing contact, and together form a hemispherical shell that essentially covers the area of ​​the human brain. The gap width is selected to be as small as possible while ensuring that adjacent focusing electrodes do not contact each other. The focusing electrodes are used to achieve a current focusing effect. When a high-frequency excitation signal is applied, they effectively reduce the adverse effects of the scalp and skull on electrical impedance measurement, thereby obtaining more effective signals carrying stroke information, and thus obtaining higher-quality visual images and quantitative parameters. In this embodiment, the detection electrode layer comprises 24 circular flexible metal electrodes of identical shape and size, divided into four groups. Each group of six detection electrodes is evenly distributed below the area covered by its corresponding focusing electrode. The detection electrodes are responsible for acquiring detection currents containing brain electrical impedance information. The focusing electrode layer and the detection electrode layer are separated by a first insulating layer to isolate the excitation and detection signals. The second insulating layer serves as the innermost layer of the contactless current focusing sensor. To improve detection accuracy, the first insulating layer should be as thin as possible while ensuring electrical isolation between the focusing and detection electrode layers. For example, an insulating film can be used. The second insulating layer should also be as thin as possible while ensuring wearer comfort, and should be made of a wearable insulating material.

[0044] In this embodiment, the four 1 / 8 spherical flexible metal electrodes of identical shape and size in the focusing electrode layer are numbered 1 to 4. The 24 circular flexible metal electrodes of identical shape and size in the detection electrode layer are divided into four equal groups based on their specific locations. Each group contains six detection electrodes, and their areas are covered by corresponding focusing electrodes. The corresponding relationships are as follows: the detection electrodes corresponding to the covered area below focusing electrode No. 1 are numbered 1.1 to 1.6; the detection electrodes corresponding to the covered area below focusing electrode No. 2 are numbered 2.1 to 2.6; the detection electrodes corresponding to the covered area below focusing electrode No. 3 are numbered 3.1 to 3.6; and the detection electrodes corresponding to the covered area below focusing electrode No. 4 are numbered 4.1 to 4.6. The focusing electrodes have two states: excitation and grounding, and the detection electrodes have two states: detection and suspension. The electrode states are controlled by the electrode switching module controlled by the data acquisition module. The non-contact current focusing sensor uses a combined excitation method.

[0045] During each set of measurements, two or three adjacent focusing electrodes are switched to an energized state, while the remaining focusing electrodes are switched to a grounded state. One detection electrode is selected to acquire detection signals, while the remaining detection electrodes are switched to a suspended state. The detection electrodes in the detection state are selected from the detection electrode group (each group has six detection electrodes) corresponding to the area covered by the grounded focusing electrodes. Each time, one detection electrode is selected and switched to the detection state, while the remaining electrodes remain suspended until all the detection electrodes in the group are sequentially switched to the detection state and signal acquisition is complete, thus completing a set of measurements.

[0046] For the next set of measurements, the focusing electrode selection scheme is changed, selecting two or three adjacent focusing electrodes that have not been previously combined and switching them to an energized state. The non-energized focusing electrodes are then switched to a grounded state, and the aforementioned single-group measurement process is repeated to obtain the corresponding signal. Based on this combination scheme, the four focusing electrodes and 24 detection electrodes in the non-contact current focusing sensor can complete a total of 72 independent impedance measurements (wherein, the scheme of selecting three adjacent focusing electrodes and switching them to an energized state includes 4 groups x 6 times / group = 24 impedance measurements, and the scheme of selecting two adjacent focusing electrodes and switching them to an energized state includes 4 groups x 12 times / group = 48 impedance measurements), thereby improving the current focusing effect while obtaining as many independent measurement values ​​as possible.

[0047] The electrode switching module mainly includes 28 relays for controlling the focusing electrode state and the detection electrode state. Figure 4 In the electrode switching module, an NMOS transistor and a high-frequency monostable relay HF3_03S are used to control / switch the states of the focusing electrode and the detection electrode. The NMOS transistor is connected to the GPIO port of the data acquisition module to receive control signals from the data acquisition module.

[0048] Figure 5 Figure 2 shows the PCB layout of the electrode switching module. To better enable wearable sensors and optimize electrical impedance measurements, the electrode switching module's PCB is designed in a circular ring shape, allowing it to be nested within the maximum radius of the hemispherical shell-shaped contactless current focusing sensor.

[0049] like Figure 6The data acquisition module is shown as a schematic diagram. It includes an I / V conversion unit for current-to-voltage conversion of detection signals, a control unit, a signal generator, an oscilloscope, a communication unit, and a power supply unit. The I / V conversion unit connects to the detection electrodes to acquire detection signals and perform current-to-voltage conversion. The control unit controls the electrode switching state of the electrode switching module. The signal generator generates a high-frequency sinusoidal excitation signal and transmits it to the currently excited focusing electrode via the electrode switching module. The oscilloscope, connected to the I / V conversion unit, collects the detection signal output by the I / V conversion unit. The communication unit communicates between the data acquisition module and the image reconstruction host computer. The power supply unit provides a stable DC voltage for the entire system. The control unit's main chip is the STM32F103ZET6, which has a rich set of GPIO peripherals to meet the independent control requirements of the four focusing electrodes and 24 detection electrodes in the focus sensor. In the present invention, the FreeRTOS operating system is transplanted to the main control chip, and two tasks are created and managed by FreeRTOS: the task of controlling GPIO state switching and the task of USART communication. The two tasks are synchronized by semaphores, thereby realizing the collaborative work between the data acquisition module and the image reconstruction host computer. The communication unit is composed of a USB-to-serial port chip (CH340N). The data acquisition module and the image reconstruction host computer realize the synchronization between waveform acquisition and electrode state switching through the serial port and the USB-to-serial port module. The oscilloscope and the image reconstruction host computer transmit waveform data via the VISA protocol. The specific steps are as follows:

[0050] Step 1: The image reconstruction host computer continuously reads serial data from the data acquisition module and starts collecting data from the oscilloscope when a specific trigger signal is detected;

[0051] Step 2: Connect the image reconstruction host computer to the oscilloscope via the VISA address and set the parameters required for communication with the oscilloscope, including the timeout and termination character.

[0052] Step 3: Send commands to the oscilloscope to configure acquisition parameters, including setting AC coupling, data format, and acquisition channels;

[0053] Step 4: Start oscilloscope data acquisition, query binary waveform data from the oscilloscope, and use the queried vertical scale factor to convert the original binary data into voltage values;

[0054] Step 5: Send an acquisition completion signal to the data acquisition module, indicating that the waveform data acquisition is completed.

[0055] Figure 7 Shown is a flow chart of a method for monitoring brain stroke based on non-contact electrical impedance tomography and current focusing, comprising the following steps:

[0056] Step 1: Power on the device and initialize the serial port and GPIO port in the data acquisition module. The image reconstruction host computer runs and establishes a connection with the data acquisition module. The image reconstruction host computer sends data acquisition instructions via the serial port. The data acquisition module enters acquisition mode and begins setting the electrode states. All focusing electrodes are initialized to grounded, and all detection electrodes are initialized to suspended.

[0057] Step 2: Set the electrode switching logic. The focusing electrode and the detection electrode are switched to their respective corresponding states according to the electrode switching logic. Then, the state switching of the focusing electrode and the detection electrode is realized by steps 3 and 4, and the acquisition of the electrical impedance data is realized by step 5. When two adjacent focusing electrodes are in the energized state, the 12 detection electrodes in the two detection electrode groups corresponding to the areas covered by the two focusing electrodes in the grounded state are switched to the detection state in sequence; when three adjacent focusing electrodes are in the energized state, the 6 detection electrodes in the one detection electrode group corresponding to the area covered by the one focusing electrode in the grounded state are switched to the detection state in sequence; until 72 independent electrical impedance measurements are completed.

[0058] For the 4 focusing electrodes and 24 detection electrodes of the non-contact current focusing sensor in the present invention, the specific electrode switching logic is: switch the focusing electrodes No. 1 and No. 2 to the excitation state (the focusing electrodes No. 3 and No. 4 are in the ground state), and switch the detection electrodes 3.1 to 3.6 and 4.1 to 4.6 to the detection state in turn to achieve a set of electrical impedance measurements, that is, 12 electrical impedance measurements; switch the focusing electrodes No. 2 and No. 3 to the excitation state (the focusing electrodes No. 1 and No. 4 are in the ground state), and switch the detection electrodes 1.1 to 1.6 and 4.1 to 4.6 to the detection state in turn to achieve a set of electrical impedance measurements, that is, 12 electrical impedance measurements; switch the focusing electrodes No. 3 and No. 4 to the excitation state (the focusing electrodes No. 1 and No. 2 are in the ground state), and switch the detection electrodes 1.1 to 1.6 and 2.1 to 2.6 to the detection state in turn to achieve a set of electrical impedance measurements, that is, 12 electrical impedance measurements; switch the focusing electrodes No. 1 and No. 4 to the excitation state (the focusing electrodes No. 2 and No. 3 are in the ground state), and switch the detection electrodes 2.1 to 1.6 and 4.1 to 4.6 to the detection state in turn to achieve a set of electrical impedance measurements, that is, 12 electrical impedance measurements. ~2.6 and 3.1~3.6 are switched to the detection state to realize a group of electrical impedance measurements, that is, 12 electrical impedance measurements; the focusing electrodes No. 1, 2, and 3 are switched to the excitation state (the focusing electrode No. 4 is in the ground state), and the detection electrodes 4.1~4.6 are switched to the detection state in turn to realize a group of electrical impedance measurements, that is, 6 electrical impedance measurements; the focusing electrodes No. 2, 3, and 4 are switched to the excitation state (the focusing electrode No. 1 is in the ground state), and the detection electrodes 1.1~1.6 are switched to the detection state in turn to realize a group of electrical impedance measurements, that is, 6 electrical impedance measurements; the focusing electrodes No. 1, 3, and 4 are switched to the excitation state (the focusing electrode No. 2 is in the ground state), and the detection electrodes 2.1~2.6 are switched to the detection state in turn to realize a group of electrical impedance measurements, that is, 6 electrical impedance measurements; the focusing electrodes No. 1, 2, and 4 are switched to the excitation state (the focusing electrode No. 3 is in the ground state), and the detection electrodes 3.1~3.6 are switched to the detection state in turn to realize a group of electrical impedance measurements, that is, 6 electrical impedance measurements.

[0059] Step 3: Set the focusing electrode state. According to the order of the electrode switching logic, switch the corresponding 2 or 3 focusing electrodes to the energized state, and the remaining focusing electrodes to the grounded state.

[0060] Step 4: Set the detection electrode state. According to the order of the electrode switching logic, switch the detection electrode with the corresponding sequence number to the detection state, and the remaining detection electrodes to the floating state.

[0061] Step 5: Data acquisition and transmission: After completing the state setting of the focusing electrode and the detection electrode, the image reconstruction host computer sends a command to control the data acquisition module to collect data and send it to the image reconstruction host computer.

[0062] Step 6: Repeat steps 3 to 5 according to the electrode switching logic in step 2 until 72 independent data collections are completed.

[0063] Step 7: Image reconstruction The host computer calculates 72 independent electrical impedance measurement values ​​based on the collected data, and uses the image reconstruction algorithm to reconstruct the image to obtain a visual image for stroke monitoring and calculate the quantitative parameters for stroke monitoring.

[0064] like Figure 8 Shown is a schematic diagram of the focusing electrode states in the combined excitation scheme adopted in the cerebral stroke monitoring method based on non-contact electrical impedance imaging and current focusing, including combined excitation of focusing electrodes No. 1 and 2, combined excitation of electrodes No. 2 and 3, combined excitation of focusing electrodes No. 3 and 4, combined excitation of focusing electrodes No. 1 and 4, combined excitation of focusing electrodes No. 1, 2 and 3, combined excitation of focusing electrodes No. 2, 3 and 4, combined excitation of focusing electrodes No. 1, 3 and 4, and combined excitation of focusing electrodes No. 1, 2 and 4.

[0065] The effectiveness of the present invention's stroke monitoring method based on non-contact electrical impedance tomography and current focusing was experimentally verified. Image reconstruction using CCEIT is the process of inverting the distribution of the medium within the measured field based on the sensor's measurement data (projection data). This can be summarized as solving the following inverse problem:

[0066] P=SG (1)

[0067] Where P = [p1, p2, ..., p N ] is the projection value vector, which is calculated by the independent electrical impedance measurement values ​​under the current distribution and the independent electrical impedance measurement values ​​under the background distribution. N represents the number of independent electrical impedance measurement values, S = [s nm ], n=1,2,...,N, m=1,2,...,M is the sensitivity matrix, M represents the number of pixels, and the sensitivity matrix is ​​related to the geometric size of the sensor and the arrangement of the electrode plates. Based on the non-contact current focusing sensor structure disclosed in the present invention, after determining the size parameters, the sensitivity matrix can usually be obtained by finite element method in this field. G=[g1,g2,...,g M ] is the reconstructed image vector. Therefore, the advantages of the present invention over the traditional non-focused single-electrode excitation scheme are verified from three dimensions: projection value, sensitive field, and imaging result.

[0068] 1) Comparison of projection values. Under a 3.3V, 10MHz sinusoidal excitation signal, the projection value vectors calculated from 72 independent electrical impedance measurements under focused excitation and the projection value vectors calculated from 276 independent electrical impedance measurements of 24 detection electrodes under the traditional non-focused single-electrode excitation scheme were obtained, and the average projection values ​​were calculated. The higher the average projection value, the stronger the response amplitude of the method to the stroke area. The results show that under focused excitation, the average projection value is 1.314e-03, and under traditional non-focused single-electrode excitation, the average projection value is 2.840e-04. The average projection value under focused excitation is an order of magnitude higher than the average projection value under single-electrode excitation, indicating that the response amplitude of this method to the stroke area is stronger.

[0069] 2) Sensitive field comparison results.

[0070] Finite element simulation is used to calculate the sensitivity matrices under focused excitation and traditional non-focused single electrode excitation schemes, and the average sensitivity S is introduced to quantify the sensitive field. a and uniformity coefficient S u , and compare the two sensitive field matrices. Their definitions are:

[0071]

[0072] Where v is the standard deviation of all sensitivities in the sensitive field.

[0073] Under the focus incentive scheme, S a is 1.1576e-05, S u is 0.7916. Under the traditional non-focused single-electrode excitation scheme, S a is 1.6387e-06, S u The results show that compared with the traditional non-focused single-electrode excitation scheme, the focused excitation scheme has higher average sensitivity and more uniform sensitive field distribution.

[0074] According to the characteristics of the object being measured, a new indicator is introduced to measure the difference between the center sensitivity and the boundary sensitivity in the sensing area, that is, the sensitivity ratio of the inner and outer circles S io , defined as follows:

[0075]

[0076] Among them, S i is the sum of the inner ring sensitivities, S o The sum of the outer circle sensitivities. The sum of the inner circle sensitivities refers to the sum of the sensitivities of all pixels within the specified radius in the sensing area, and the sum of the outer circle sensitivities refers to the sum of the sensitivities of all pixels outside the specified radius in the sensing area.

[0077] Sio The higher the value, the higher the sensitivity of the inner circle, and the smaller the difference in sensitivity between the inner and outer circles. This shows that the excitation scheme can effectively improve the central sensitivity of the sensing area and better obtain the electrical impedance information inside the measured area. Figure 9 The ratio of this indicator under the focused excitation scheme and the traditional non-focused single-electrode excitation detection scheme is shown. It can be concluded that the focused excitation scheme can significantly improve the center sensitivity, increase the detection depth, and reduce the sensitivity difference between the inner and outer circles in the sensing area.

[0078] 3) Comparison of imaging results. The imaging results under the two excitation detection schemes are compared, and the imaging performance is measured by relative image error (RIE). It is defined as:

[0079]

[0080] in, and They represent the grayscale of the m-th pixel in the reconstructed image vector and the grayscale of the m-th pixel in the actual image vector, respectively.

[0081] Figure 10 The image reconstruction results using traditional non-focused single-electrode excitation (276 projection values) and focused excitation (72 projection values) when the stroke area is at the center of the lower sensitivity are shown. The white circular marked area is the actual stroke area. In the reconstructed image, white represents the background, and black represents the reconstructed stroke area. The RIE of image reconstruction using traditional non-focused single-electrode excitation (276 projection values) and focused excitation (72 projection values) are 0.28 and 0.18, respectively. From the results, it can be concluded that the reconstructed image using the focused excitation scheme has fewer artifacts, is more focused on the reconstruction and positioning of the stroke area, and has a lower relative image error. In other words, the brain stroke monitoring device and method based on non-contact electrical impedance imaging and current focusing of the present invention have the best image reconstruction effect and can significantly improve the imaging capability of the central area.

[0082] The above description of the present invention and its embodiments is not limited thereto, and the accompanying drawings are only one embodiment of the present invention. Without departing from the purpose of the present invention, any structure or embodiment similar to the technical solution without creative design shall fall within the scope of protection of the present invention.

Claims

1. A non-contact current focusing sensor for cerebral stroke monitoring, characterized in that: The sensor is in the shape of a hemispherical shell and comprises, from the outside to the inside, a focusing electrode layer, a first insulating layer, a detection electrode layer and a second insulating layer; The focusing electrode layer includes n spherical flexible metal focusing electrodes of the same shape and size. There are gaps between adjacent focusing electrodes, so they do not touch each other, and together form a hemispherical shell shape to cover the human brain area; The focusing electrode is used to achieve a current focusing effect after an excitation signal is applied, n≥2; The detection electrode layer includes multiple detection electrodes, which are used to obtain detection current containing brain electrical impedance information. All detection electrodes are divided into n groups, and each of the n groups of detection electrodes corresponds one-to-one to the n focusing electrodes. Each group of detection electrodes is arranged in the coverage area of ​​its corresponding focusing electrode, and each detection electrode in the group is evenly distributed within the coverage area. The focusing electrode layer and the detection electrode layer are separated by a first insulating layer to achieve isolation of the excitation and detection signals; The second insulating layer serves as the innermost layer of the contactless current focusing sensor.

2. The non-contact current focusing sensor for cerebral stroke monitoring according to claim 1, characterized in that: The shapes and sizes of the detection electrodes are exactly the same, and the detection electrodes are circular flexible metal electrodes; the diameter of the detection electrodes in each group of detection electrodes is ≥10 mm.

3. A cerebral stroke monitoring device based on non-contact electrical impedance tomography and current focusing, characterized in that: It includes the non-contact current focusing sensor according to claim 1, and also includes an electrode switching module, a data acquisition module and an image reconstruction host computer; The electrode switching module is used to control and switch the states of the focusing electrode and the detection electrode; The data acquisition module is used to control the electrode switching module and generate an excitation signal to the focusing electrode. The data acquisition module collects the detection signal of the detection electrode and communicates with the image reconstruction host computer; The image reconstruction host computer is used to realize image reconstruction and obtain real-time images and quantitative parameters for cerebral stroke monitoring.

4. The stroke monitoring device according to claim 3, characterized in that: The focusing electrode in the non-contact current focusing sensor has two states: excitation and grounding; the detection electrode has two states: detection and suspension; the data acquisition module generates a control signal and sends it to the electrode switching module, which controls and switches the states of the focusing electrode and the detection electrode.

5. The stroke monitoring device according to claim 3, characterized in that: The electrode switching module includes relays and corresponding NMOS tubes. The number of relays is the same as the total number of focusing electrodes and detection electrodes. The NMOS tubes and relays realize independent control and switching of the states of the focusing electrodes and detection electrodes. The NMOS tubes are connected to the data acquisition module to receive the control signals sent by the data acquisition module.

6. The stroke monitoring device according to claim 3, characterized in that: The data acquisition module includes an I / V conversion unit for detecting signal current-voltage conversion, a control unit, a signal generator, an oscilloscope, a communication unit and a power supply unit; the I / V conversion unit is connected to the detection electrode to obtain the detection signal, and the control unit is responsible for controlling the state of the electrode switching module; the signal generator is responsible for generating a high-frequency sinusoidal excitation signal, which is transmitted to the focusing electrode after passing through the electrode switching module; the oscilloscope is connected to the I / V conversion unit and is responsible for collecting the detection signal output by the I / V conversion unit; the communication unit is responsible for communication between the data acquisition module and the image reconstruction host computer; and the power supply unit is responsible for providing a stable DC voltage for the entire system.

7. The stroke monitoring device according to claim 3, characterized in that: The PCB of the electrode switching module is designed to be in a circular shape and is nested in the outer periphery of the maximum radius of the hemispherical shell-shaped non-contact current focusing sensor.

8. A method for monitoring cerebral stroke based on the device according to claim 3, characterized in that: The following steps are involved: 1) The image reconstruction host computer sends data acquisition instructions through the serial port, and the data acquisition module enters the acquisition mode and initializes the electrode state. All focusing electrodes are initialized to the ground state, and all detection electrodes are initialized to the suspended state; 2) Selection of focusing electrodes: When n = 2, one focusing electrode is selected and switched to the energized state, and the focusing electrodes that are not in the energized state are switched to the grounded state; when n ≥ 3, n-2 or n-1 adjacent focusing electrodes are selected and switched to the energized state, and the focusing electrodes that are not in the energized state are switched to the grounded state; 3) After the selection of the focusing electrode is completed, one detection electrode is selected to be in a detection state to realize the collection of the detection signal, and the detection electrodes that are not in the detection state are switched to a suspended state; wherein, the detection electrodes in the detection state are selected from the detection electrode group corresponding to the area covered by the focusing electrode in the grounded state, and one detection electrode is selected from the group and switched to the detection state each time, while the remaining electrodes are in the suspended state; after completing one detection electrode selection, the image reconstruction host computer controls the data acquisition module to perform a data collection and sends the data to the image reconstruction host computer; when the data collection is completed, the detection electrodes are switched until the detection electrodes in the detection electrode group corresponding to the area covered by the focusing electrode in the grounded state are switched to the detection state in sequence and the signal collection is completed, thereby completing the signal measurement under a set of focusing electrode selection schemes; 4) Return to step 2) and change the focusing electrode selection scheme. When n=2, select another focusing electrode and switch it to the energized state. When n≥3, select n-1 or n-2 adjacent focusing electrodes that have not been previously combined and switch them to the energized state. Repeat step 3) to perform a set of signal measurements under the current focusing electrode selection scheme. This continues until all optional focusing electrode combination selection schemes have been selected and measured. 5) Image reconstruction The host computer calculates all independent electrical impedance measurement values ​​based on the collected data, and reconstructs the image using the image reconstruction algorithm to obtain a visual image for cerebral stroke monitoring and calculate the quantitative parameters for cerebral stroke monitoring.

9. The method for monitoring cerebral stroke according to claim 8, characterized in that: The number of the focusing electrodes is 4, and each focusing electrode covers an area containing 6 detection electrodes; in step 2), when measuring a single group, two or three adjacent focusing electrodes are selected and switched to an excitation state; When two adjacent focusing electrodes are in the excited state, the 12 detection electrodes in the two detection electrode groups corresponding to the areas covered by the two focusing electrodes in the grounded state are switched to the detection state in sequence; When three adjacent focusing electrodes are in the excited state, six detection electrodes of a detection electrode group corresponding to the area covered by one focusing electrode in the grounded state are switched to the detection state in sequence.

Images