Adaptive brain frequency difference electrical impedance tomography and lesion target recognition method

Abstract

The present invention discloses an adaptive brain frequency-difference electrical impedance imaging and lesion target identification method, comprising: first, acquiring multi-frequency EIT data of the brain; then calculating the average effective boundary voltage of each frequency point in the multi-frequency EIT data; then, adaptively filtering out background frames from all frequency points of the multi-frequency EIT data based on the average effective boundary voltage; then, performing differential imaging on the multi-frequency EIT data of all frequency points based on the background frames to obtain frequency-difference electrical impedance images of all frequency points. Finally, performing lesion feature image analysis on the frequency-difference electrical impedance images corresponding to all frequency points to identify the lesion target in the frequency-difference electrical impedance images, thereby achieving one-time rapid imaging of intracranial lesions.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical impedance imaging, and in particular to an adaptive cranial frequency difference electrical impedance imaging and lesion target recognition method. Background Art

[0002] During pre-hospital emergency rescue efforts for victims of war, earthquakes, and major accidents, traumatic brain injuries are easily identified, while closed brain injuries are often difficult to detect. If closed brain injuries are not addressed and appropriate treatment measures are not implemented, patients often miss the optimal treatment window. Therefore, a one-time, rapid imaging technology that can be used pre-hospital is needed.

[0003] Electrical impedance tomography (EIT) is a new, portable, noninvasive, and harmless imaging technology. It is highly sensitive to changes in electrical impedance caused by the pathophysiological state of biological tissue, and therefore has the potential to become a rapid, pre-hospital imaging tool for brain injury. Since its initial introduction, researchers have consistently pursued the goal of rapidly imaging brain lesions in one go. While the technology has progressed through several stages and achieved some progress, no significant breakthroughs have yet been achieved.

[0004] The first stage of one-time rapid electrical impedance tomography technology is the static EIT stage. This technology uses one frame of data to reconstruct the resistivity distribution inside the target being measured. Due to its own pathological nature, it is difficult to obtain high-quality images and has not been actually used in the human brain.

[0005] The second phase of one-shot rapid electrical impedance imaging technology is symmetrical electrical impedance imaging. Patent number ZL201210426035.2 discloses "A method for electrical impedance tomography with a self-constructed background frame," which allows for imaging of targets with abnormal resistivity distribution within structurally symmetrical objects. However, this symmetrical imaging method has very limited application in the human brain, being applicable only to brains with a certain degree of inherent symmetry. Currently, only images of stroke lesions within brains with relatively good symmetry have been obtained, and rapid one-shot imaging of less symmetrical brains is infeasible.

[0006] The Bi-Frequency Symmetry Difference EIT (BFSD-EIT) technology proposed abroad first performs symmetrical imaging on the EIT data of two frequency points separately, and then compares the imaging results of the two frequency points to highlight unilateral brain damage lesions. However, later studies found that this method is extremely sensitive to the asymmetry of the human brain itself, making it difficult to obtain images of lesions in the real human brain.

[0007] The third phase of one-shot rapid electrical impedance imaging technology is multi-frequency differential electrical impedance imaging. Patent number ZL201310006208.X discloses "An Image Reconstruction Method for Quasi-Static Electrical Impedance Imaging." This algorithm uses two-frequency electrical impedance imaging data for weighted differential imaging. However, the weighting factor is estimated using the least squares method applied to the overall boundary voltage, which introduces artificial errors. When the error exceeds the target signal strength, it is impossible to image brain lesions.

[0008] Patent number ZL202010152430 discloses "an electrical impedance imaging method based on the spatial distribution characteristics of tissues and the characteristics of impedance variation with frequency." This method discloses an electrical impedance imaging method that utilizes the impedance variation characteristics of the measured tissue with frequency. However, applying this technology to the human brain will face formidable challenges. Because the dielectric properties of brain tissue vary among healthy people of different races and ages, and because the acquisition of brain tissue dielectric properties requires ex vivo brain tissue, the dielectric properties of healthy human brains are difficult to obtain. The current knowledge of the dielectric properties of animal brain tissue cannot support the use of this technology to obtain information on intracranial lesions in the human brain.

[0009] Previous work on direct differential imaging of multi-frequency electrical impedance data from healthy individuals and stroke patients demonstrated that, within the 1kHz-100kHz range, images of unilateral lesions could be observed in the patient's multi-frequency imaging sequence if the background frame was appropriately selected. However, due to individual differences in the human brain, each individual has its own impedance spectrum. Using data from a fixed frequency point as the background frame during multi-frequency electrical impedance imaging, the differential results do not accurately reveal the lesion.

[0010] Therefore, there is an urgent need for a brain frequency difference electrical impedance imaging method that can adaptively select background frame frequency points to achieve one-time rapid imaging of intracranial lesions. Summary of the Invention

[0011] To address the shortcomings of existing technologies, the present invention proposes an adaptive brain frequency difference electrical impedance imaging and lesion target recognition method, which can adaptively filter out background frames to achieve one-time rapid imaging of intracranial lesions. The specific technical solution is as follows:

[0012] Provided is an adaptive brain frequency difference electrical impedance imaging and lesion target recognition method, including:

[0013] Collect multi-frequency EIT data of the brain;

[0014] Calculate the average effective boundary voltage of each frequency point in the brain multi-frequency EIT data;

[0015] From all the frequency points of the brain multi-frequency EIT data, the EIT data of the frequency point corresponding to the maximum effective boundary voltage average value is selected as the background frame;

[0016] Based on the background frame, differential imaging is performed on the EIT data of each frequency point in the multi-frequency EIT data to obtain the frequency difference electrical impedance images corresponding to all frequency points;

[0017] Perform lesion feature image analysis on the frequency difference electrical impedance images corresponding to all frequency points to determine the lesion targets in the frequency difference electrical impedance images.

[0018] Furthermore, the frequency difference electrical impedance image is constructed using a direct difference image reconstruction algorithm based on the EIT data corresponding to each frequency point in the multi-frequency EIT data of the brain and the EIT data of the background frame.

[0019] Furthermore, performing lesion characteristic image analysis on the frequency difference electrical impedance image includes:

[0020] The lesion target in the frequency difference electrical impedance image is determined by performing geometric symmetry analysis on the frequency difference electrical impedance image.

[0021] Furthermore, a geometric symmetry analysis is performed on the frequency difference electrical impedance image, including:

[0022] Determine the abnormal signal areas on the left and right sides of the frequency difference electrical impedance image;

[0023] Calculate the distances from the centroid of the abnormal signal areas on the left and right sides to the X-axis and Y-axis respectively;

[0024] Calculate the geometric symmetry index of the abnormal signal areas on the left and right sides according to the distances from the centroids of the abnormal signal areas on the left and right sides to the X-axis and the Y-axis;

[0025] The lesion target in the frequency difference electrical impedance image is determined by the geometric symmetry index.

[0026] Furthermore, performing lesion characteristic image analysis on the frequency difference electrical impedance image includes:

[0027] The lesion target in the frequency difference electrical impedance image is determined by performing intensity symmetry analysis on the frequency difference electrical impedance image.

[0028] Furthermore, performing intensity symmetry analysis on the frequency difference electrical impedance image includes:

[0029] Determine the abnormal signal areas on the left and right sides of the frequency difference electrical impedance image;

[0030] Calculate the average signal intensity of the abnormal signal areas on the left and right sides respectively;

[0031] Calculate the intensity symmetry index of the abnormal signal areas on the left and right sides according to the average signal intensity of the abnormal signal areas on the left and right sides;

[0032] The lesion target in the frequency difference electrical impedance image is determined by the intensity symmetry index.

[0033] Beneficial Effects: The adaptive brain frequency-difference electrical impedance imaging and lesion target identification method of the present invention can adaptively select background frames of frequency-difference imaging for all human brains under test through a set imaging method to perform differential imaging, thereby obtaining a frequency-difference electrical impedance image of the human brain under test. By performing lesion feature image analysis on the frequency-difference electrical impedance image obtained using the set imaging method, the lesion target in the frequency-difference electrical impedance image can be identified, thereby achieving one-time rapid imaging of intracranial lesions. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the specific embodiments of the present invention, the following briefly introduces the drawings required for use in the specific embodiments. In all the drawings, each element or part is not necessarily drawn according to the actual scale.

[0035] Figure 1 A flowchart of a method for adaptive cranial frequency difference electrical impedance imaging and lesion target identification provided by one embodiment of the present invention;

[0036] Figure 2 These are the brain frequency difference electrical impedance images of two healthy people at various frequencies obtained based on the imaging method provided by the embodiment of the present invention;

[0037] Figure 3 A comparison diagram of a cranial frequency difference impedance image and an MRI image of a patient's brain obtained using the imaging method provided by an embodiment of the present invention;

[0038] Figure 4 A comparison diagram of a cranial frequency difference electrical impedance image and a CT image of a patient's brain obtained based on the imaging method provided by an embodiment of the present invention;

[0039] Figure 5 The following are the lesion identification results obtained by testing 3 patients and 1 healthy person based on the adaptive brain frequency difference electrical impedance imaging and lesion target recognition method provided by the embodiment of the present invention. DETAILED DESCRIPTION

[0040] The following embodiments of the technical solution of the present invention will be described in detail with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention and are therefore only examples and are not intended to limit the scope of protection of the present invention.

[0041] like Figure 1 The flowchart of the adaptive brain frequency difference electrical impedance imaging and lesion target recognition method shown is performed by the brain impedance imaging system, including:

[0042] Step S1, collecting brain multi-frequency EIT data;

[0043] Step S2: calculating the average effective boundary voltage of each frequency point in the brain multi-frequency EIT data based on the corresponding brain EIT data;

[0044] Step S3, selecting the EIT data of the frequency point corresponding to the maximum effective boundary voltage average value from the brain multi-frequency EIT data as the background frame;

[0045] Step S4: performing differential imaging on the EIT data corresponding to all frequency points in the brain multi-frequency EIT data and the background frame to obtain frequency difference electrical impedance images corresponding to all frequency points;

[0046] Step S5: performing lesion characteristic image analysis on the frequency difference electrical impedance images corresponding to all frequency points to determine the lesion targets in the frequency difference electrical impedance images.

[0047] Specifically, the brain impedance imaging system can collect multi-frequency EIT data of the human brain at one time. Based on the EIT data of each frequency point in the collected multi-frequency EIT data of the brain, the effective boundary voltage average value of each frequency point can be calculated. By comparing the effective boundary voltage average values corresponding to the EIT data of each frequency point, the EIT data of the frequency point corresponding to the maximum value of the effective boundary average value can be selected as the background frame.

[0048] The EIT system can directly perform differential imaging of the EIT data corresponding to each frequency point in the multi-frequency EIT data of the brain with the background frame, thereby obtaining the frequency difference electrical impedance image corresponding to each frequency point. The EIT system can perform lesion feature image analysis on the frequency difference electrical impedance images of all frequencies, thereby identifying the lesion target in the frequency difference electrical impedance image and displaying the frequency difference electrical impedance image with the lesion target as the lesion image.

[0049] In this embodiment, the EEG system includes electrodes 1 through 16, for a total of 16 electrodes. These electrodes can be placed at designated locations on the human scalp. For example, electrode 1 is placed 1 cm above the right ear, electrode 5 is placed 1 cm above the brow, electrode 9 is placed 1 cm above the left ear, and electrode 13 is placed 1 cm above the occipital protuberance. The remaining electrodes are evenly distributed in a counterclockwise order between electrodes 1, 5, 9, and 13.

[0050] In step 1, the brain impedance imaging system can adopt the opposite excitation-proximity measurement mode, and apply an excitation current of 500 microamperes to the human brain through the excitation electrode pair. The frequency of the excitation current can include 18 frequency points, and the frequencies of these 18 frequency points can be 1kHz, 3kHz, 3.5kHz, 4kHz, 4.5kHz, 5kHz, 6kHz, 8kHz, 13kHz, 21kHz, 30kHz, 40kHz, 50kHz, 60kHz, 70kHz, 80kHz, 90kHz, and 100kHz respectively.

[0051] For example, the brain impedance imaging system can first apply the excitation current to the electrode pairs 1-9, and then measure the boundary voltage modulus U2_3, U3_4, U4_5, U5_6, U6_7, U7_8, U10_11, U11_12, U12_13, U13_14, U14_15, and U15-16 on the adjacent electrode pairs 2-3, 3-4, 4-5, 5-6, 6-7, 7-8, 10-11, 11-12, 12-13, 13-14, 14-15, and 15-16. _15, U15_16, and then apply the excitation current to the electrode pairs 2-10, and then measure the upper boundary voltage modulus values U3_4, U4_5, U5_6, U6_7, U7_8, U8_9, U11_12, U12_13, U13_14, U14_15, U15_16, U16_1 of the corresponding 12 adjacent electrode pairs, and repeat this cycle until the excitation current traverses all the opposing electrode pairs 1-9, 2-10, 3-11, ..., 16-8.

[0052] Thus, each frame of brain EIT data collected at each frequency point contains 192 effective boundary voltage moduli. The EIT system can calculate the effective boundary voltage average value for each frequency point based on the 192 boundary voltage moduli in the brain EIT data at each frequency point. Based on the effective boundary voltage average value, the EIT system can select the EIT data for the frequency point corresponding to the maximum effective boundary voltage average value from the multi-frequency EIT data of the brain as the background frame.

[0053] In this embodiment, the frequency difference electrical impedance image is constructed using a direct difference image reconstruction algorithm based on the EIT data corresponding to each frequency point in the multi-frequency EIT data of the brain and the background frame.

[0054] Specifically, the EI system can use a direct difference image reconstruction algorithm, such as a damped least squares image reconstruction algorithm, based on the filtered background frames to determine the resistivity change of each frequency point in the multi-frequency EIT data relative to the background frame. The EI system can then display the resistivity change region using different color scales, thereby generating a frequency-difference electrical impedance image for each frequency point.

[0055] The imaging method of this embodiment is used to perform frequency difference electrical impedance imaging on the brains of two healthy people. The imaging results are as follows: Figure 2 shown.

[0056] The image processing method of this embodiment was used to perform frequency difference electrical impedance imaging on the brains of two patients, and the imaging results obtained were as follows: Figure 3 、 Figure 4 As shown. After the selection of the adaptive background frame, Figure 3 The background frame of the corresponding patient's brain multi-frequency EIT imaging is 13 kHz EIT data; Figure 4 The background frame of the corresponding patient's brain multi-frequency EIT imaging is 13 kHz EIT data.

[0057] In this embodiment, in step S5, the brain impedance imaging system may determine the lesion target in the frequency difference electrical impedance image by performing a geometric symmetry analysis on the frequency difference electrical impedance image, including:

[0058] Step S5-1-1, determining the abnormal signal areas on the left and right sides of the frequency difference electrical impedance image;

[0059] Step S5-1-2, respectively calculating the distances from the centroids of the abnormal signal areas on the left and right sides to the X-axis and the Y-axis;

[0060] Step S5-1-3, calculating the geometric symmetry index of the abnormal signal areas on the left and right sides according to the distances from the centroids of the abnormal signal areas on the left and right sides to the X-axis and the Y-axis;

[0061] Step S5-1-4: determine the lesion target in the frequency difference electrical impedance image through the geometric symmetry index.

[0062] In step S5-1-1, the step of determining the abnormal signal area on the frequency difference EIT image first calculates the absolute value Amax of the maximum value of the reconstructed resistivity change value on the entire EIT image, and the pixel points on the image whose absolute value of the resistivity change value is in the range of 50%Amax-100%Amax form the abnormal signal area.

[0063] In step S5-1-2, the brain impedance imaging system constructs an XY coordinate system with the center of the frequency difference electrical impedance image as the origin. It can also calculate the distance from the center of mass of the abnormal signal areas on the left and right sides of the frequency difference electrical impedance image to the X axis, and the distance from the center of mass of the abnormal signal areas to the Y axis, thereby calculating the geometric symmetry index of the abnormal signal areas on the left and right sides of the frequency difference electrical impedance image. The specific calculation formula is as follows:

[0064] A1=0.5*{[(L x -R x ) / R x ] 2 +[(Ly -R y ) / R y ] 2} 0.5 *100%;

[0065] Among them, L x is the distance from the center of mass of the abnormal signal area on the left to the X axis, R x is the distance from the center of mass of the abnormal signal area on the right to the X axis, L y is the distance from the center of mass of the abnormal signal area on the left to the Y axis, R y is the distance from the centroid of the abnormal signal area on the right to the Y axis.

[0066] The maximum geometric symmetry index A1 of the frequency-difference impedance imaging of healthy individuals is 5.14%. If A1 is less than 5.14%, the abnormal signal areas on the left and right sides of the frequency-difference impedance imaging are geometrically symmetrical about the Y-axis (the midline of the brain), and EI imaging suggests that further intensity symmetry analysis is needed. Otherwise, the two abnormal signal areas on the left and right sides are asymmetrical. The larger abnormal signal area is the lesion target, and the smaller abnormal signal area is determined to be the artifact area. The EI imaging system marks the lesion target with a solid circle, and the artifact area with a dashed circle. The area of the abnormal signal area is measured by the number of pixels within the area.

[0067] In this embodiment, in step S5, the brain impedance imaging system can perform intensity symmetry analysis on the frequency difference electrical impedance image, based on the premise that the abnormal signal regions on the left and right sides of the frequency difference electrical impedance image are geometrically substantially symmetrical about the Y axis. The specific steps include:

[0068] Step S5-2-1, determining the abnormal signal areas on the left and right sides of the frequency difference electrical impedance image;

[0069] Step S5-2-2, respectively calculating the average signal strength of the abnormal signal areas on the left and right sides;

[0070] Step S5-2-3, calculating the intensity symmetry index of the abnormal signal areas on the left and right sides according to the average signal intensity of the abnormal signal areas on the left and right sides;

[0071] Step S5-2-4: determine the lesion target in the frequency difference electrical impedance image by using the intensity symmetry index.

[0072] In step S5-2-1, the step of determining the abnormal signal area on the frequency difference EIT image first calculates the absolute value Amax of the maximum value of the reconstructed resistivity change value on the entire EIT image, and the pixel points on the image whose absolute value of the resistivity change value is in the range of 50%Amax-100%Amax constitute the abnormal signal area.

[0073] In step S5-2-2, the brain impedance imaging system can calculate the average of the absolute values of the reconstructed resistivity change values of all pixel points in the abnormal signal area on the frequency difference EIT image as the average signal intensity of the abnormal signal area.

[0074] The EEG system can calculate the intensity symmetry index of the abnormal signal areas on the left and right sides based on the average signal intensity of the abnormal signal areas on the left and right sides. The specific technical formula is as follows:

[0075] A2=abs(I L -I R ) / I R *100%;

[0076] Among them, I L is the average signal intensity of the abnormal signal area on the left, I R is the average signal intensity of the abnormal signal area on the right. The maximum value of the intensity symmetry index A2 of the frequency difference impedance image for healthy people is 4.35%. If A2 is less than 4.35%, the signal intensity of the geometrically symmetrical abnormal signal areas on the left and right sides of the frequency difference impedance image is basically symmetrical about the brain midline. The EI system determines that the subject is healthy and the geometrically symmetrical abnormal signal intensity areas on the image are identified as artifacts, which are marked and displayed with dotted circles. Otherwise, the signal intensity of the two geometrically symmetrical abnormal signal areas on the left and right sides of the frequency difference impedance image is asymmetric. The area with higher signal intensity is the lesion target, while the area with lower signal intensity is the artifact. The EI system marks the artifact area with a dotted circle.

[0077] The processing method disclosed in the embodiment of the present invention was used to perform image analysis on the brains of three patients and one healthy person, and the analysis results were as follows: Figure 5 As shown. Figure 5 In the figure, the area circled by the solid line is the lesion target.

[0078] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and description of the present invention.

Claims

1. An adaptive brain frequency difference electrical impedance tomography and lesion target identification method, characterized in that: include: Collect multi-frequency EIT data of the brain; Calculate the average effective boundary voltage of each frequency point in the brain multi-frequency EIT data; From all the frequency points of the brain multi-frequency EIT data, the EIT data of the frequency point corresponding to the maximum effective boundary voltage average value is selected as the background frame; Based on the background frame, differential imaging is performed on the EIT data of each frequency point in the multi-frequency EIT data to obtain the frequency difference electrical impedance images corresponding to all frequency points; Perform lesion feature image analysis on the frequency difference electrical impedance images corresponding to all frequency points to determine the lesion targets in the frequency difference electrical impedance images, including: By performing geometric symmetry analysis on the frequency difference electrical impedance image, the geometric symmetry index of the abnormal signal areas on the left and right sides of the frequency difference electrical impedance image is calculated. The specific calculation formula is as follows: ； in, is the distance from the center of mass of the abnormal signal area on the left to the X axis, is the distance from the centroid of the abnormal signal area on the right to the X-axis, is the distance from the center of mass of the abnormal signal area on the left to the Y axis, is the distance from the centroid of the abnormal signal area on the right to the Y axis; like By performing intensity symmetry analysis on the frequency difference electrical impedance image, the intensity symmetry index of the abnormal signal areas on the left and right sides is calculated. The specific calculation formula is as follows: ； in, is the average signal intensity of the abnormal signal area on the left, is the average signal intensity of the abnormal signal area on the right; The lesion target in the frequency difference electrical impedance image is determined by the intensity symmetry index.

2. The method for adaptive cranial frequency difference electrical impedance imaging and lesion target identification according to claim 1, characterized in that: The frequency difference electrical impedance image is constructed using a direct difference image reconstruction algorithm according to the EIT data corresponding to each frequency point in the brain multi-frequency EIT data and the EIT data of the background frame.

3. The method for adaptive cranial frequency difference electrical impedance imaging and lesion target identification according to claim 1, characterized in that: Performing geometric symmetry analysis on the frequency difference electrical impedance image includes: Determine the abnormal signal areas on the left and right sides of the frequency difference electrical impedance image; Calculate the distances from the centroid of the abnormal signal areas on the left and right sides to the X-axis and Y-axis respectively; Calculate the geometric symmetry index of the abnormal signal areas on the left and right sides according to the distances from the centroids of the abnormal signal areas on the left and right sides to the X-axis and the Y-axis; The lesion target in the frequency difference electrical impedance image is determined by the geometric symmetry index.

4. The method for adaptive cranial frequency difference electrical impedance imaging and lesion target identification according to claim 1, characterized in that: Performing intensity symmetry analysis on the frequency difference electrical impedance image includes: Determine the abnormal signal areas on the left and right sides of the frequency difference electrical impedance image; Calculate the average signal intensity of the abnormal signal areas on the left and right sides respectively; Calculate the intensity symmetry index of the abnormal signal areas on the left and right sides according to the average signal intensity of the abnormal signal areas on the left and right sides; The lesion target in the frequency difference electrical impedance image is determined by the intensity symmetry index.

Images