An intracranial pressure monitoring method, system and medium based on electrical impedance imaging

Abstract

The application provides an intracranial pressure monitoring method and system based on electrical impedance imaging and a medium, and relates to the technical field of intracranial pressure monitoring. The method comprises the following steps: collecting an extracranial original boundary voltage signal of a monitoring object by an electrical impedance tomography system, and extracting a voltage signal reflecting a dynamic blood perfusion frequency band; using a circular finite element model conforming to an actual electrode distribution position to solve a positive problem of the voltage signal to obtain an electric potential on each electrode node; using inverse problem calculation to obtain a reconstructed image; based on the reconstructed image, extracting an average perfusion velocity, a contraction wave amplitude, an inflow volume velocity and an angle between a descending branch and a baseline as characteristic parameters; using a random forest algorithm to construct an intracranial pressure increase change detection model, and monitoring an intracranial pressure state by using the trained intracranial pressure increase change detection model. The method obtains a cerebral blood perfusion parameter by electrical impedance imaging, and monitors the intracranial pressure in real time according to the change law of the cerebral blood perfusion parameter with the intracranial pressure.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intracranial pressure monitoring, in particular to an intracranial pressure monitoring method and system based on electrical impedance tomography and a medium. BACKGROUND

[0002] Neurological critical illness diseases, such as craniocerebral trauma, stroke, brain tumor, etc., are often accompanied by the condition of elevated intracranial pressure (ICP). The continuously elevated ICP can reduce cerebral blood perfusion, leading to cerebral ischemia and hypoxia, and if not treated in time, it will eventually lead to severe neurological damage and brain death. Monitoring and warning the elevated ICP are crucial for the correct treatment of patients in neurological intensive care and for improving the prognosis. Therefore, real-time monitoring of ICP to prevent insufficient cerebral blood perfusion has very important clinical value in the diagnosis and treatment of craniocerebral diseases.

[0003] The existing ICP monitoring methods are mainly divided into two kinds of invasive ICP monitoring and non-invasive ICP monitoring. The invasive ICP monitoring is widely considered as the gold standard for ICP monitoring, however, this method can cause intracranial hemorrhage, infection and other complications during the placement of the pressure sensor. The existing non-invasive ICP monitoring technologies, such as transcranial Doppler ultrasound, optic nerve sheath diameter measurement method, magnetic resonance imaging, etc., mostly obtain ICP information through indirect methods. However, these methods mostly need professional personnel to operate and interpret, lack of ease of use, and are not suitable for long-term dynamic non-invasive monitoring. Based on this, there is an urgent need for a bedside ICP monitoring method which is non-invasive, harmless and easy to interpret the results, in order to realize the monitoring and warning of ICP changes. SUMMARY

[0004] To solve the above problems, the present application provides an intracranial pressure monitoring method based on electrical impedance tomography, which obtains cerebral blood perfusion parameters through electrical impedance tomography, and monitors intracranial pressure in real time according to the change rule of cerebral blood perfusion parameters with intracranial pressure.

[0005] To achieve the above purpose, the present application provides the following technical solutions.

[0006] The present application provides an intracranial pressure monitoring method based on electrical impedance tomography, comprising the following steps:

[0007] The extracranial original boundary voltage signal of the monitoring object is collected by an electrical impedance tomography system, and the voltage signal reflecting the dynamic blood perfusion condition frequency band is extracted;

[0008] A finite element model conforming to the electrode node distribution position of the actual electrical impedance tomography system is used to solve the forward problem of the voltage signal, and the potential on each electrode node is obtained;

[0009] The damping least square method is used to perform inverse problem calculation on the potential on each electrode node to obtain a reconstructed image reflecting the dynamic change of blood flow; wherein the reconstructed image is a dynamic change image of resistivity caused by the pulsatility change of the brain tissue congestion state;

[0010] Based on the reconstructed image, the average perfusion velocity, the systolic wave amplitude, the inflow volume velocity and the angle between the descending branch and the baseline are extracted as characteristic parameters representing the change of intracranial pressure;

[0011] A random forest algorithm is used to construct an intracranial pressure increase change detection model, and the characteristic parameters representing the change of intracranial pressure are constructed into a training set to train the intracranial pressure increase change detection model; the characteristic parameters of the monitoring object are obtained, and the intracranial pressure state monitoring is performed through the trained intracranial pressure increase change detection model.

[0012] Preferably, the voltage signal reflecting the frequency band of the dynamic blood flow perfusion condition is extracted, including the following steps:

[0013] The voltage signal reflecting the dynamic blood flow perfusion condition is extracted through a band-pass filter with a passband frequency of 1Hz-5Hz.

[0014] Preferably, the damping least square method is used to perform inverse problem calculation on the potential on each electrode node to obtain a reconstructed image reflecting the dynamic change of blood flow, including the following steps:

[0015] According to the potential on each electrode node, the resistivity distribution between the current frame and the reference frame is obtained:

[0016] Δρ=(J T J+λR) -1 J T ΔV

[0017] Δρ is the change of resistivity distribution between the current frame and the reference frame, ΔV is the normalized boundary voltage change between the current frame and the reference frame, J is the Jacobian matrix, λ is the regularization parameter, and R is the regularization matrix;

[0018] The average reconstructed resistivity related to the whole brain in the electrical impedance tomography image sequence is obtained by the following formula:

[0019]

[0020] In the formula, Δρ i represents the reconstructed resistivity of the i th pixel in the electrical impedance tomography image; N represents the total number of pixels in the region of interest.

[0021] Preferably, the average perfusion velocity is obtained as shown in the following formula:

[0022]

[0023] where ARV i is the average reconstructed resistivity of the i-th frame of electrical impedance tomography images; ARV p is the peak value of the resistivity; ARV v is the valley value of the resistivity; ARV p drops from the peak value ARV v is the descending branch, T d is the time interval of the descending branch; ARV v goes back to the next peak value ARV p is the ascending branch, T a is the time interval of the ascending branch; N represents the total number of pixels of the region of interest.

[0024] Preferably, the acquisition of the contraction wave amplitude is as follows:

[0025] H s = ARV p - ARV v

[0026] where ARV p is the peak value of the resistivity; ARV v is the valley value of the resistivity.

[0027] Preferably, the acquisition of the inflow volume velocity is as follows:

[0028]

[0029] where H s is the contraction wave amplitude, T d is the time interval of the descending branch.

[0030] Preferably, the acquisition of the angle between the descending branch and the baseline is as follows:

[0031]

[0032] where the baseline is defined as a vector starting from the peak value ARV p of the resistivity and its direction is parallel to the x-axis; X p , Y p are the x and y coordinates of the peak value ARV p respectively; X v , Y v are the x and y coordinates of the valley value ARV v respectively.

[0033] The present application provides an intracranial pressure monitoring system based on electrical impedance imaging, the system comprising:

[0034] a processor;

[0035] a memory having stored thereon a computer program executable on the processor;

[0036] Wherein, the computer program is executed by the processor to realize the steps of the intracranial pressure monitoring method based on electrical impedance imaging.

[0037] The application provides a computer readable storage medium, the computer readable storage medium has stored thereon a data processing program, the data processing program is executed by a processor to realize the steps of the intracranial pressure monitoring method based on electrical impedance imaging.

[0038] The application has the following beneficial effects:

[0039] The application provides an intracranial pressure monitoring method, system and medium based on electrical impedance imaging. The method collects and analyzes the change rule of the EIT image parameter of cerebral blood perfusion with ICP through an EIT system, develops four EIT parameters related to cerebral blood perfusion, i.e., average perfusion rate, systolic wave height, inflow volume velocity and angle between descending branch and baseline, and analyzes and determines the change of ICP according to the above parameters. Through the above conclusion, the EIT image of cerebral blood perfusion is monitored, and the monitoring result of ICP can be non-destructively judged, which effectively assists medical staff in monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 is the intracranial pressure monitoring flowchart based on electrical impedance imaging of the embodiment of the application;

[0041] Figure 2 is the experimental method and process schematic diagram of the embodiment of the application; wherein, Figure 2 (A) of is a schematic diagram of 16 copper electrode positions, indwelling needle positions and ICP probe positions; Figure 2 (B) of is an experimental scene diagram; Figure 2 (C) of is experimental time, EIT data collection starts after about 10 minutes of baseline, blood injection interval is about 10 minutes, and there are 5 times of blood injection; Figure 2 (D) of is an ARV curve and perfusion parameter extraction schematic diagram;

[0042] Figure 3 is the EIT signal change with ICP of the embodiment of the application; wherein, Figure 3 (A) of is the original measurement data of EIT and ICP; Figure 3 (B) of is a comparison diagram of ARV and ICP during blood injection; Figure 3 (C) of is the average transmission impedance and perfusion cycle EIT image; Figure 3 (D) of is a reconstructed EIT basic impedance image with t1 as reference data;

[0043] Figure 4 are the average perfusion parameters and standard deviations of all samples of the embodiments of the present application at different ICP levels; wherein, Figure 4 (A) of is the average perfusion velocity; Figure 4 (B) of is the systolic wave height; Figure 4 (C) of is the inflow volume velocity; Figure 4 (D) of is the included angle;

[0044] Figure 5 are the four independent perfusion parameters and the ROC curves of the random forest model for detecting the ICP increase relative to the baseline of the embodiments of the present application; wherein, Figure 5 (A) of is the perfusion parameter for detecting the ICP increase greater than 5 mmHg; Figure 5 (B) of is the perfusion parameter for detecting the ICP increase greater than 10 mmHg; Figure 5 (C) of is the ROC curve of the random forest model for detecting the ICP increase greater than 5 mmHg; Figure 5 (D) of is the ROC curve of the random forest model for detecting the ICP increase greater than 10 mmHg;

[0045] Figure 6 are the detection performances of the random forest model for detecting the ICP increase when the reference value is changed; wherein, Figure 6 (A) of is the detection of the ICP increase greater than 5 mmHg; Figure 6 (B) of is the detection of the ICP increase greater than 10 mmHg. DETAILED DESCRIPTION

[0046] In order to make the objectives, technical solutions and advantages of the present application clearer, further detailed description will be made to the present application in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0047] Embodiment 1

[0048] The intracranial pressure monitoring method based on electrical impedance imaging of the present application has the specific process as shown in Figure 1 , which comprises the following steps:

[0049] S1: Collecting the extracranial original boundary voltage signal of the monitoring object through the electrical impedance tomography system, and extracting the voltage signal reflecting the dynamic blood flow perfusion condition frequency band.

[0050] Further, the voltage signal reflecting the dynamic blood flow perfusion condition is extracted through the band-pass filter with a passband frequency of 1 Hz-5 Hz.

[0051] S2: Using the circular finite element model conforming to the actual electrode distribution position to perform the forward problem calculation on the voltage signal to obtain the potential on each electrode node.

[0052] S3: The inverse problem calculation of the potential at each electrode node is performed using the damped least squares method to obtain a reconstructed image reflecting the dynamic changes in blood flow; wherein, the reconstructed image is a dynamic image of resistivity changes caused by the pulsatile changes in the state of cerebral congestion.

[0053] Specifically, based on the potential at each electrode node, the resistivity distribution between the current frame and the reference frame is obtained:

[0054] Δρ=(J T J+λR) -1 J T ΔV

[0055] Δρ is the change in resistivity distribution between the current frame and the reference frame, ΔV is the change in normalized boundary voltage between the current frame and the reference frame, J is the Jacobian matrix, λ is the regularization parameter, and R is the regularization matrix.

[0056] The average reconstructed resistivity of the whole brain in electrical impedance tomography sequences is obtained by the following formula:

[0057]

[0058] In the formula, Δρ i represents the reconstructed resistivity of the i-th pixel in the electrical impedance tomography image; N represents the total number of pixels in the region of interest.

[0059] S4: Based on the reconstructed image, the average perfusion velocity, systolic wave amplitude, inflow volume velocity, and the angle between the descending limb and the baseline are extracted as characteristic parameters to characterize changes in intracranial pressure.

[0060] Specifically, the average infusion rate is obtained as shown in the following formula:

[0061]

[0062] In the formula, ARV i The average reconstructed resistivity of the electrical impedance tomography image in the i-th frame; ARV p Peak resistivity; ARV v The valley of resistivity; derived from the peak ARV p Dropped to the lowest ARV v For the descending branch, T d The time interval for the descending branch; from the valley ARV v Return to the next peak ARV p For the ascending branch, T a denoted as the time interval of the rising branch; N represents the total number of pixels in the region of interest. A complete brain perfusion cycle is defined as the period from peak ARV. p Initially, after the valley value ARVv , to the next peak ARV p .

[0063] The contraction wave amplitude is obtained as follows:

[0064] H s = ARV p - ARV v

[0065] In the formula, ARV p is the peak value of the resistivity; and ARV v is the valley value of the resistivity.

[0066] The inflow volume velocity is obtained as follows:

[0067]

[0068] In the formula, H s is the contraction wave amplitude, T d is the time interval of the descending branch.

[0069] Preferably, the descending branch and the baseline included angle are obtained as follows:

[0070]

[0071] wherein the baseline is defined as a vector starting from the peak value ARV p of the resistivity and parallel to the x-axis in direction; X p , Y p are the x and y coordinates of the peak value ARV p respectively; X v , Y v are the x and y coordinates of the valley value ARV v respectively.

[0072] S5: An intracranial pressure increase change detection model is constructed by using a random forest algorithm, feature parameters representing intracranial pressure changes are constructed into a training set to train the intracranial pressure increase change detection model; feature parameters of a monitoring object are acquired, and intracranial pressure state monitoring is performed through the trained intracranial pressure increase change detection model.

[0073] In order to verify the feasibility of the above method, the following experiment is given in the present application:

[0074] A domestic pig intracranial hypertension model is constructed, and invasive ICP monitoring is used as a control method to analyze the change rule of EIT cerebral blood perfusion parameters with ICP, aiming to verify the ability of EIT to indirectly reflect the ICP level by monitoring the cerebral blood perfusion state by observing the correlation between ICP and the cerebral blood perfusion state reflected by EIT.

[0075] In particular,

[0076] Six healthy pigs with a body weight of 17.2 ± 1.5 kg were used in the experiment. The experimental protocol and procedures were approved by the Animal Ethics Committee of the Fourth Military Medical University (Ethical Permit Number: IACUC-20241299) and followed the "Guide for the Care and Use of Laboratory Animals" published by the National Institutes of Health (National Academy Press, Washington, DC, revised in 1996). All pigs were induced with anesthetic using 5 mg / kg of Xylazine 50, intubated, and maintained with 1-3% isoflurane. The femoral artery was punctured with an indwelling needle to establish a blood sampling channel. After monitoring the arterial blood pressure, oxygen saturation, and end-tidal carbon dioxide, the scalp of each pig was cut longitudinally, the skull was separated and exposed, the periosteum was scraped, and a window with a sagittal length of about 60 mm and a coronal length of about 60 mm was formed. A skull drill was used to drill holes at about 15 mm right of the sagittal suture and about 15 mm anterior of the coronal suture for placing the ICP probe. Holes were drilled at about 15 mm right of the sagittal suture and about 10 mm posterior of the coronal suture, and then a 22G indwelling needle was inserted into the skull hole with a depth of about 18 mm. In addition, 16 sterilized 1.2 mm dental root canals were drilled into the edge of the skull at equal intervals, as shown in Figure 2 (A), none of which penetrated the dura mater. The above-mentioned skull holes were sealed with bone wax to reduce the possibility of cerebrospinal fluid leakage.

[0077] ICP data were collected by an ICP monitor (SOPHYSA Inc., Auxerre, France) with a frame rate of 2 frames per second, as shown in Figure 2 (B).

[0078] The EIT signal was collected using a newly developed EIT system (UTRON Technology Co., Ltd., Hangzhou, China). The working frequency range of the system is 10 kHz-250 kHz, the output current range is 10 μΑ-1250 μΑ, and the signal-to-noise ratio is greater than 90 dB. During the experiment, the frame rate of data collection was set to 40 frames per second, and the excitation frequency was set to 50 kHz.

[0079] Experimental protocol:

[0080] The intracranial hypertension model was prepared by injecting autologous non-anticoagulated blood into the porcine brain parenchyma in batches. After connecting the electrodes to the EIT system, baseline EIT data was collected, and ICP data acquisition was started. About 10 min later, 2 ml of blood was extracted through the indwelling needle pre-positioned in the femoral artery, and then the blood was injected into the brain parenchyma region at a constant speed within 1 min through the indwelling needle pre-positioned in the intracranial region. This process was repeated about 10 min later, for a total of 5 times, and the time sequence was as shown in Figure 2 The norepinephrine was used to maintain the blood pressure basically unchanged during the whole operation. After the operation was completed, the animals were deeply anesthetized and sacrificed, and the brain was removed to observe the location of the hematoma.

[0081] The EIT image reconstruction and perfusion parameter extraction were obtained through the above steps S1-S4. Specifically, as shown in Figure 2 (D) of the present application, Figure 2 (D) of the present application is an ARV curve and a perfusion parameter extraction schematic diagram.

[0082] Data analysis:

[0083] Statistical analysis was performed using SPSS 27.0 statistical software (SPSS Inc, Chicago, IL, USA). The significance level of statistical analysis was 0.05. The Shapiro-Wilk test was used to test the normality of the data. According to the ICP classification standard, the ICP values were divided into the following four groups: 0-15 mmHg, 15-20 mmHg, 20-40 mmHg, and greater than 40 mmHg. Then, for all samples, the paired t-test was used to compare the average perfusion parameter differences between adjacent groups and the differences between 0-15 mmHg and other groups, respectively. The pixel value of functional EIT (fEIT) is not just impedance change, but valuable functional information that affects clinical decision-making. The EIT image sequence of 5 perfusion cycles was calculated for each group, the waveform of each pixel was extracted to calculate the average perfusion parameter, and these perfusion parameters were presented in the form of an image to generate the fEIT image.

[0084] According to the International Brain Trauma Foundation intracranial hypertension treatment intervention standard, the data set was divided into two intervals with 22 mmHg as the boundary, and the Spearman correlation coefficient between the perfusion parameters and ICP in the two intervals was calculated, and then the paired t-test was used to evaluate the difference between the correlation coefficients of the two intervals.

[0085] In order to evaluate the detection performance of EIT parameters in detecting elevated ICP, the average MV, H s , IV and A db of each sample at baseline were determined as reference values, and the normalized change rates (ΔMV, ΔH s , ΔIV and ΔA dbThe receiver operating characteristic (ROC) curves were analyzed for 3000 samples from each animal, including 500 samples from each animal. The efficacy of each independent EIT parameter in detecting ICP increases greater than 5 mmHg and 10 mmHg relative to baseline was evaluated. Finally, the area under the ROC curve, sensitivity, and specificity were calculated.

[0086] Finally, a random forest algorithm was used to establish a model for detecting ICP elevation changes. The research dataset consists of ΔMV and ΔH values ​​from 3000 samples. s ΔIV, ΔA db Composed of Y, ΔMV, ΔH s ΔIV and ΔA db MV and H respectively s , IV and A db The normalized rate of change relative to the reference value, Y = {0, 1}, is the label (where 0 and 1 represent ΔICP less than and greater than a certain threshold, respectively). The dataset was then randomly divided into training (70%) and test (30%) datasets, and the random forest model was trained using the training set. The detection performance of the model was comprehensively evaluated by calculating AUC, sensitivity, and specificity using the test set. The selected reference values ​​were normal ICP values ​​and intracranial hypertension values ​​(between 15-40 mmHg). This reference range was chosen primarily because patients often already have intracranial hypertension upon initial monitoring, making it crucial to explore the random forest algorithm's ability to detect elevated ICP across different reference values.

[0087] in conclusion:

[0088] (1) EIT signal variation with ICP

[0089] like Figure 3 As shown, Figure 3 (A) represents the raw measurement data for EIT and ICP. The top shows the average transmission impedance from 256 channels, with red arrows indicating the injection time points; the middle shows the raw and filtered data for ICP; and the bottom shows the mean arterial pressure. Figure 3 (B) is a comparison chart of ARV and ICP during blood injection. Figure 3 (C) shows the average transmission impedance and EIT images for the injection period. The upper part shows the average transmission impedance for the injection periods corresponding to t1, t2, and t3, and the lower part shows the corresponding EIT images for the injection periods. The image sequence interval is 25ms. The ICP corresponding to t1, t2, and t3 is approximately 10mmHg, 25mmHg, and 40mmHg, respectively. Figure 3 (D) is the reconstructed EIT basic impedance image with t1 as the reference data.

[0090] existFigure 3 In (A) of FIG. 9, exemplary changes of EIT transmission impedance, ICP and mean arterial pressure during the whole experiment are presented. In this experiment, five blood injections were performed, each of which caused a rapid increase in transmission impedance. Meanwhile, ICP increased rapidly and then decreased slowly. During the whole experiment, mean arterial pressure remained stable and no obvious fluctuation was observed. Figure 3 (B) of FIG. 9 shows the changes of ICP and ARV before and after blood injection. It can be observed that, after blood injection, ICP increased rapidly and then decreased slowly, while the amplitude of ARV decreased rapidly and then increased slowly. In order to study the effect of ICP change on blood perfusion EIT images, three time points corresponding to three different ICP levels were selected, i.e., t1, t2 and t3 in (B) of FIG. 9, each of which represents a cardiac cycle. The difference imaging of cardiac cycles at these three time points was performed using the start of systole as the reference frame, and the results are shown in (C) of FIG. 9. Figure 3 Figure 3 (C) of FIG. 9. By observing the dynamic cerebral blood perfusion EIT images at the selected time points, it can be seen that, as ICP gradually increases, the cerebral perfusion in the edge region of the image gradually decreases. After removing the pulsatile signal in EIT using a low-pass filter with an upper cutoff frequency of 0.1 Hz, the baseline impedance is determined. Figure 3 (D) of FIG. 9 shows the results of EIT baseline impedance imaging at t1, t2 and t3 time points with t1 as the reference time. The continuous EIT images show that the blue region in the lower left corner represents the blood injection area, and as the number of blood injections increases, the impedance in this region gradually increases.

[0091] (2) Effect of ICP change on EIT perfusion parameters

[0092] Figure 4 (A) of FIG. 10 shows the mean perfusion parameters and standard deviations of all samples under different ICP levels. Among them, Figure 4 (A) is the mean perfusion velocity; Figure 4 (B) is the systolic wave height; Figure 4 (C) is the inflow volume velocity; Figure 4 (D) is the angle.

[0093] ​The Shapiro-Wilk test showed that the data were normally distributed (p>0.05). In the study of the effect of different ICP levels on EIT parameters, paired t-test was used to compare the mean perfusion parameters between adjacent ICP levels. The results showed that the significance of the differences in the parameters was not consistent when comparing the perfusion parameters at ICP less than 15 mmHg versus 15-20 mmHg, and 15-20 mmHg versus 20-40 mmHg. However, when ICP increased from 20-40 mmHg to 40 mmHg, all perfusion parameters showed significant differences (P<0.01). In addition, there were also significant differences between the group with ICP less than 15 mmHg and the groups with ICP at 15-20 mmHg and higher than 40 mmHg (P<0.05). The fEIT images below the graphs also revealed the trend of decreasing perfusion parameters in the peripheral regions of the images as ICP increased.

[0094] Table 1 shows the mean Spearman correlation coefficients and standard deviations between ICP and each perfusion parameter. It is clear from the table that the correlation coefficients for the groups with ICP above 22 mmHg were significantly higher than those for the groups with ICP below 22 mmHg (P<0.05). This indicates that at higher ICP levels, there is a stronger association between ICP and cerebral blood flow perfusion, i.e. the effect of ICP on cerebral blood flow is more significant: as ICP increases further, the decrease in cerebral blood flow is more pronounced.

[0095] Table 1 Mean Spearman correlation coefficients and standard deviations between ICP and each perfusion parameter

[0096]

[0097] (3) Evaluation of the diagnostic performance of EIT perfusion parameters in detecting ICP elevation

[0098] Figure 5 Figures (A) and (B) show the ROC curves for the ability of ΔMV, ΔH s , ΔIV and ΔA db to detect an increase in ICP greater than 5 mmHg and 10 mmHg from baseline. The AUC for ΔMV, ΔH s , ΔIV and ΔA db to detect an increase in ICP greater than 5 mmHg were 0.82, 0.71, 0.70 and 0.69, respectively, and for an increase greater than 10 mmHg were 0.85, 0.72, 0.72 and 0.71, respectively. Table 2 and Table 3 list the specific values of AUC, sensitivity and specificity for each parameter. This shows that the use of these EIT perfusion parameters alone has some accuracy in detecting ICP elevation.

[0099] Table 2 Evaluation of different parameters in ICP>5 mmHg detection

[0100]

[0101] Table 3 Evaluation index of different parameters in ICP > 5mmHg detection

[0102]

[0103] To further improve the accuracy of detection, the embodiment uses a random forest method to establish a comprehensive model to detect the ICP increase. Figure 5 Fig. 6 (C) and (D) are ROC curves of the random forest model detecting ICP increase relative to the baseline, and the ROC analysis results show that in the detection of ICP increase of 5mmHg, the AUC, sensitivity and specificity of the model can reach 0.90, 92% and 90% respectively, and the variable importance of AMV, AH s , AIV and AA db is 6.03, 2.81, 2.34 and 3.23 respectively. In the detection of ICP increase of 10mmHg, the AUC, sensitivity and specificity can reach 0.92, 93% and 90% respectively, and the variable importance is 7.4, 2.50, 2.96 and 1.76 respectively. Figure 6 Fig. 6 (A) and (B) show the performance of the random forest model in detecting ICP increase of more than 5mmHg and 10mmHg at different reference values. Specifically, the AUC, sensitivity and specificity of the model are all excellent, and the values are all more than 0.9. More notably, as the reference value increases, these performance indicators also show an upward trend.

[0104] The above is an intracranial pressure monitoring method based on electrical impedance imaging provided by one embodiment of the embodiment. Based on the same idea, the embodiment also provides a corresponding intracranial pressure monitoring system based on electrical impedance imaging. For specific limitations of the intracranial pressure monitoring system based on electrical impedance imaging, refer to the limitations of the intracranial pressure monitoring method based on electrical impedance imaging described above, which will not be repeated here. Each module in the above intracranial pressure monitoring system based on electrical impedance imaging can be realized by software, hardware and combinations thereof, in whole or in part. The above modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to each of the above modules.

[0105] The embodiment also provides a computer readable storage medium, which stores a computer program, and the computer program can be used to execute the intracranial pressure monitoring method based on electrical impedance imaging provided by the above embodiment. Figure 1 The embodiment also provides a computer readable storage medium, which stores a computer program, and the computer program can be used to execute the intracranial pressure monitoring method based on electrical impedance imaging provided by the above embodiment.

[0106] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiments of the methods. Any reference to memory, storage, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. The volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0107] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An intracranial pressure monitoring method based on electrical impedance tomography, characterized in that, The method comprises the following steps: An extracranial original boundary voltage signal of a monitoring object is collected by an electrical impedance tomography system, and a voltage signal reflecting a dynamic blood perfusion condition frequency band is extracted; A finite element model conforming to the electrode node distribution position of an actual electrical impedance tomography system is used to solve a forward problem of the voltage signal, and an electric potential on each electrode node is obtained; A damping least square method is used to calculate an inverse problem of the electric potential on each electrode node, and a reconstructed image reflecting dynamic changes of blood flow is obtained; wherein the reconstructed image is a dynamic change image of electrical resistivity caused by pulsatility of a brain tissue congestion state; Based on the reconstructed image, an average perfusion velocity, a contraction wave amplitude, an inflow volume velocity, and an angle between a descending branch and a baseline are extracted as characteristic parameters representing changes in intracranial pressure; A random forest algorithm is used to construct an intracranial pressure increase change detection model, the characteristic parameters representing changes in intracranial pressure are constructed into a training set to train the intracranial pressure increase change detection model; the characteristic parameters of the monitoring object are obtained, and intracranial pressure state monitoring is performed through the trained intracranial pressure increase change detection model.

2. The intracranial pressure monitoring method based on electrical impedance tomography according to claim 1, characterized in that, The extraction of the voltage signal reflecting the dynamic blood perfusion condition frequency band comprises the following steps: A band-pass filter with a passband frequency of 1 Hz-5 Hz is used to extract the voltage signal reflecting the dynamic blood perfusion condition.

3. The intracranial pressure monitoring method based on electrical impedance tomography according to claim 1, characterized in that, The inverse problem calculation of the electric potential on each electrode node by the damping least square method to obtain the reconstructed image reflecting the dynamic changes of blood flow comprises the following steps: According to the electric potential on each electrode node, the electrical resistivity distribution between a current frame and a reference frame is obtained: Δρ = (J T J + λR) -1 J T ΔV Δρ is the change in the electrical resistivity distribution between the current frame and the reference frame, ΔV is the normalized boundary voltage change between the current frame and the reference frame, J is the Jacobian matrix, λ is the regularization parameter, and R is the regularization matrix: The average reconstructed electrical resistivity related to the whole brain in the electrical impedance tomography image sequence is obtained by the following formula: where Δρ i represents the reconstructed resistivity of the i-th pixel in the electrical impedance tomography image; N represents the total number of pixels in the region of interest.

4. The intracranial pressure monitoring method based on electrical impedance tomography according to claim 1, characterized in that, The average perfusion velocity is obtained as shown in the following formula: In the formula, ARV i The average reconstructed resistivity of the electrical impedance tomography image in the i-th frame; ARV p Peak resistivity; ARV v The valley of resistivity; derived from the peak ARV p Dropped to the lowest ARV v For the descending branch, T d The time interval for the descending branch; from the valley ARV v Return to the next peak ARV p For the ascending branch, T a The interval for the rising branch; N represents the total number of pixels in the region of interest.

5. The intracranial pressure monitoring method based on electrical impedance tomography according to claim 4, characterized in that, The contraction wave amplitude is obtained as shown in the following formula: H s = ARV p - ARV v where ARV p is the peak value of the resistivity; and ARV v is the valley value of the resistivity.

6. The intracranial pressure monitoring method based on electrical impedance tomography according to claim 5, characterized in that, The inflow volume velocity is obtained as shown in the following formula: where H s is the amplitude of the compression wave, T d is the time interval of the descending branch.

7. The intracranial pressure monitoring method based on electrical impedance tomography as claimed in claim 5, wherein, The angle between the descending branch and the baseline is obtained as shown in the following formula: wherein the baseline is defined as the peak value of the resistivity ARV p , and the direction of which is parallel to the x-axis; X p , Y p are the x and y coordinates of the peak value of the resistivity ARV p , respectively. v , Y v are the x and y coordinates of the valley value of the resistivity ARV v , respectively.

8. An intracranial pressure monitoring system based on electrical impedance imaging, characterized in that, The system comprises: a processor; a memory having a computer program stored thereon and executable on the processor; wherein the computer program, when executed by the processor, implements the steps of the intracranial pressure monitoring method based on electrical impedance imaging according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a data processing program, and the data processing program, when executed by the processor, implements the steps of the intracranial pressure monitoring method based on electrical impedance imaging according to any one of claims 1 to 7.

Images