CN106529126A - Processing method for inheriting monitoring image information after continuous monitoring interruption in brain dynamic electrical impedance imaging - Google Patents

Abstract

The invention discloses a processing method for inheriting monitoring image information after continuous monitoring interruption in brain dynamic electrical impedance imaging, and the method is used for recovering the monitoring image information lost after the clinical continuous monitoring interruption in the brain dynamic electrical impedance imaging. According to the method, measurement data before and after the continuous monitoring interruption is analyzed; first, the measurement data is processed by using a spline difference fitting method, and image artifacts caused by baseline change of the measurement data due to inconsistent electrode contact states before and after the continuous monitoring interruption are inhibited; and then, an image reconstruction matrix is subjected to augmentation processing, and prior information of electrode displacement is introduced in the reconstruction matrix, so that the influence on the monitoring image due to inconsistent electrode arrangement positions before and after the monitoring interruption is reduced. Through a test, the method can effectively inhibit the image artifacts caused by changes of electrode contact states and arrangement positions before and after the monitoring interruption and recover a normal monitoring target, so that the clinical practicality of electrical impedance imaging monitoring is improved.

Description

Monitoring image information of having no progeny in a kind of cranium brain dynamic electric impedance imaging on-line monitor is inherited Processing method

Technical field

The invention belongs to dynamic electric impedance technical field of imaging, and in particular to a kind of cranium brain dynamic electric impedance imaging is continuously supervised The processing method that monitoring image information of having no progeny in shield is inherited.

Background technology

, by being evenly distributed on surrounding's electrode of head, real-time continuous ground is to cranium for cranium brain dynamic electric impedance tomography technology Brain applies safe current excitation Measured Boundary voltage, using two data not in the same time, with the data at wherein more early moment For reference frame, the data at another moment are prospect frame, by two frame data difference after, with reference to certain imaging method, rebuild Go out inside the cranium in the two impedance variations not in the same time.Therefore, the gathered data of continuous-stable is that system is normally moved The most important condition of state impedance imaging.

During clinical practice use, often there is the situation for having to temporarily interrupt monitoring.Such as client need Carry out the imaging examinations such as CT, need to remove ring and be attached to the electrode of head, inspection finish it is follow-up it is continuous guarded if need Re-posted electrode.After re-posted electrode, compared with the state for initially starting to guard, the position of electrode and contact condition are likely to send out Raw to change, this change can directly affect gathered data.Restart monitoring when, if still using initially start guard when Reference frame, then as the change of electrode-scalp contact condition and electrode position may produce artifact in reconstruction image, fall into oblivion Normal impedance variations information.If selecting data when restarting monitoring as reference frame, although electrode-head can be eliminated Skin contact condition and electrode position change the impact for causing, but the image information before restarting to guard cannot be reflected directly in newly Monitoring image on, impedance variations information before causing is lost.The interruption of on-line monitor have a strong impact on the observation to the state of an illness and Diagnosis, is unfavorable for clinical expansion and the application of dynamic cranium brain electrical impedance imaging.

Therefore, in order to eliminate the impact of electrode position and contact condition change to image, and retain the image monitoring of early stage Information, needs a kind of method processed by data that can be to having no progeny in on-line monitor badly.

The content of the invention

It is an object of the invention to provide monitoring image information of having no progeny in a kind of cranium brain dynamic electric impedance imaging on-line monitor The processing method of succession, the method can effectively suppress the reconstruction figure that electrode position and electrode-scalp contact condition change are caused As artifact, and retain effective image monitoring information, improve the clinical applicability of dynamic electric impedance imaging.

The present invention is to be achieved through the following technical solutions：

The processing method that monitoring image information of having no progeny in a kind of cranium brain dynamic electric impedance imaging on-line monitor is inherited, the process Method is analyzed to the measurement data before and after on-line monitor interruption：First, measurement is processed using the method that batten difference is fitted Data, the image that the change of measurement data baseline causes caused by electrode contact state is inconsistent before and after suppressing on-line monitor to interrupt are pseudo- Shadow；Then, augmentation process is carried out to image reconstruction algorithm, the prior information of electrode displacement is introduced into reconstruction matrix, reduce monitoring Before and after interruption, electrode is because of the inconsistent impact to guarding image of riding position.

Above-mentioned processing method specifically includes following steps：

1) obtain the Baseline wander value of measurement data

Last n frame data and the front n frame data for restarting to guard before selected on-line monitor interruption, data cube computation is existed Together, it is data x (l), to i-th valid data channel data x_iL () uses Spline-Fitting, obtain fitting sequence Wherein, i ∈ N, N are Measurement channel sum；

It is then detected thatThe maximum of uplifted side at the baseline transitionWith the minimum for declining side Calculate Baseline wander value

2) Baseline wander is measured to restarting the data guarded

To restarting the ith measurement channel data sequence y after guarding_iL (), usesTo y_iL () carries out baseline and rectifys Just, compared with the data before monitoring interruption, if y_iL () is in the elevated side of baseline, then after correcting If y_iL side that () declines in baseline, then after correcting

3) augmentation process is carried out to Gauss-Newton image reconstruction formula

To Δ ρ=- [J^tJ+λR]^-1Jz, Δ ρ be not in the same time between electrical impedance change profile vector, z is not in the same time Boundary voltage difference vector；

By Jacobin matrixExpand toBy constraint matrixExtension ForAnd coefficient lambda and constraint matrix R is determined using prior information_augUnit item；

4) impedance variations of target field domain are rebuild using the reconstruction formula of data and augmentation after correction

I.e.WhereinFor the measurement voltage data at t1 moment,For the t2 moment Measurement voltage data.

Step 1) in, to i-th valid data channel data x_iL () uses Spline-Fitting, specifically carry out as the following formula：

Wherein, S (x) is data sequence x_iL the spline-fit function of (), asks S (x) to make the value of L get minimum；Wherein, p ∈ [0,1] reflect the degree of closeness of fitting function and measured data；S (x) is segmentation batten fitting function, is write as general type：

S_j(l)=a_j(l-l_j)³+b_j(l-l_j)²+c_j(l-l_j)+d_j；

Boundary condition is added to solve each piecewise fitting function coefficients of S (l) using least square method.The S (l) for obtaining is concrete After expression formula, fitting reconfiguration sequence is soughtDetectionThe maximum of uplifted side at the baseline transitionWith decline side MinimumCalculate Baseline wander value

Step 3) concrete operations are：

Gauss -- Newtonian image reconstruction formula such as following formula：

Δ ρ=- [J^tJ+λR]^-1Jz；

Wherein, J is Jacobin matrix, and λ R are regularization constraint item, Δ ρ be not in the same time between electrical impedance change profile to Amount, z is boundary voltage difference vector not in the same time；

For suppressing electrode position change influence on RT, augmentation process is carried out to the matrix in reconstruction formula：It is right In Jacobin matrixExpand ton_measFor the measurement voltage number that frame data are included, n_elemFor the FEM model unit number that imaging reconstruct used is used, n_dimFor being imaged dimension, two-dimensional imaging, institute is generally carried out To take n_dim=2, n_eFor number of electrodes, the expansion of Jacobin matrix is filled to due to the disturbance of measuring electrode change in location, I.e.：

Wherein, A is current excitation vector, and H is positive calculating matrix,Become for electrode position in an x or y direction The positive calculating matrix recalculated after change, the displacement generally unification are set to a priori constant；

Then, augmentation process is carried out to constraint matrix R：

WillExpand toExpansion is filled to the noise elder generation of reconstruct data The prior estimate estimated and reconstruct conductivity variations is tested, i.e.,：

Wherein, R_extraFor a Laplace filter, according to the prior information of imaging field domain determine regularization parameter λ and R_extraLaplce's template, then have：

Wherein, ave_noiseFor noise relative to measurement data average priori amplitude, ave_conductFor conductivity variations phase For the average priori amplitude of initial conductivity distribution, ave_moveIt is the average priori width of electrode displacement relative to field domain radius Value,For Laplace operator.

Compared with prior art, the present invention has following beneficial technique effect：

The process side that monitoring image information of having no progeny in cranium brain dynamic electric impedance imaging on-line monitor disclosed by the invention is inherited Method, for recovering the monitoring image information of loss of having no progeny in cranium brain dynamic electric impedance imaging clinic on-line monitor.The method is to even Measurement data before and after continuous monitoring is interrupted is analyzed, and processes measurement data first by the method for batten difference fitting, suppresses Before and after on-line monitor interrupts, the inconsistent caused measurement data baseline of electrode contact state changes the image artifacts for causing, then right Image Reconstruction matrix carries out augmentation process, and the prior information of electrode displacement is introduced reconstruction matrix, reduces electricity before and after monitoring is interrupted Pole is because of the inconsistent impact to guarding image of riding position.Jing is tested, and before and after the method can effectively suppress monitoring to interrupt, electrode connects Tactile state and riding position change caused image artifacts, recover normal monitoring target, improve electrical impedance imaging monitoring Clinical practicability.

Description of the drawings

Fig. 1 is method of the present invention schematic flow sheet；

Fig. 2 is the reconstructed image that on-line monitor interrupts previous moment.

Fig. 3 is that the inventive method is not used to process, and on-line monitor is interrupted data for the previous period and restarts prison One-dimensional data curve (a) after one piece of data is coupled together after shield, and reselect reference frame (c) and do not reselect reference Two-Dimensional Reconstruction image in the case of frame (b)；

Fig. 4 is, using one-dimensional data curve (a) after this method rectification step, innovatory algorithm (b) to be not used and uses This method improves Two-Dimensional Reconstruction image (c) of imaging algorithm.

Specific embodiment

With reference to specific embodiment, the present invention is described in further detail, it is described be explanation of the invention and It is not to limit.

The process side that monitoring image information of having no progeny in a kind of cranium brain dynamic electric impedance imaging on-line monitor of the present invention is inherited Method, comprises the following steps：

(1) obtain the Baseline wander value of measurement data.Selected on-line monitor interrupt before last n frame data and restart The front n frame data of monitoring, are data x (l) together by data cube computation, individual effectively to i-th (i ∈ N, N are Measurement channel sum) Data channel data x_iL () uses Spline-Fitting：

Wherein S (x) is data sequence x_iL the spline-fit function of (), S (x) make the value of L get minimum.Wherein p ∈ [0, 1] reflect the degree of closeness of fitting function and measured data.S (x) is segmentation batten fitting function, can be write as general type：

S_j(l)=a_j(l-l_j)³+b_j(l-l_j)²+c_j(l-l_j)+d_j

Boundary condition is added to solve each piecewise fitting function coefficients of S (l) using least square method.The S (l) for obtaining is concrete After expression formula, fitting reconfiguration sequence is soughtDetectionThe maximum of uplifted side at the baseline transitionWith decline side MinimumCalculate Baseline wander value

(2) Baseline wander is measured to restarting the data guarded.Lead to restarting the ith measurement after guarding Track data sequences y_iL (), usesTo y_iL () carries out Baseline wander, compared with the data before monitoring interruption, if y_i(l) place In the elevated side of baseline, then after correctingIf y_iL side that () declines in baseline, then correct Afterwards

Repeat step (1) (2), carries out above process to all of Measurement channel,.

(3) imaging algorithm is improved, augmentation process is carried out to Gauss-Newton image reconstruction formula.For Gauss-Newton is imaged Algorithmic formula：

Δ ρ=- [J^tJ+λR]^-1Jz

Wherein it is J restructuring matrixs, λ R are regularization constraint item, and z is boundary voltage difference vector not in the same time, and Δ ρ is not for Electrical impedance change profile vector between in the same time.

In order to suppress electrode position change influence on RT, then the matrix in reconstruction formula to be carried out at augmentation Reason.For Jacobin matrixExpand ton_measFor the measurement electricity that frame data are included Pressure number, n_elemFor the FEM model unit number that imaging reconstruct used is used, n_dimFor be imaged dimension, generally carry out two dimension into Picture, so take n_dim=2, n_eFor number of electrodes, the expansion of Jacobin matrix is filled to due to measuring electrode change in location Disturbance, i.e.,

Wherein

A is current excitation vector, and H is positive calculating matrix,After changing for electrode position in an x or y direction The positive calculating matrix for recalculating, the displacement generally unification are set to a priori constant.

Then augmentation process is carried out to constraint matrix R, willExpand toExpand For reconstructing the noise prior estimate of data and the prior estimate of reconstruct conductivity variations, i.e., exhibition is partially filled with

Wherein R_extraFor a Laplace filter.According to imaging field domain prior information determine regularization parameter λ and R_extraLaplce's template, have

Wherein ave_noiseFor noise relative to measurement data average priori amplitude, ave_conductIt is relative for conductivity variations In the average priori amplitude of initial conductivity distribution, ave_moveIt is the average priori amplitude of electrode displacement relative to field domain radius.

(4) impedance variations of target field domain are rebuild using the reconstruction formula of data and augmentation after correction.I.e.WhereinFor restarting the reference frame voltage data after guarding,For restarting Prospect frame voltage data after monitoring, can be abbreviated as Δ ρ=S again_augz。

Specific embodiment is as follows：

The impedance bioelectrical measurement electrode for all coating conductive paste is posted in subject's head, and winds head to fix with bandage Electrode, after all electrode contacts are normal, proceeds by data acquisition and image monitoring.Image such as Fig. 2 of normal on-line monitor It is shown, include obvious impedance variations target on image.In order to simulate the situation that monitoring is interrupted, by whole measuring electrodes from receiving Examination person's head is removed, and is pasted subject's head after the region wiped clean of electrode with gauze, re-posted whole impedance bioelectrical measurement electrode. Limited by operation actual conditions, the electrode of re-posted compared with original state, in distributing position and with the contact impedance of scalp all Can change, affect boundary voltage so that the measurement data baseline before and after restarting to guard is inconsistent, such as Fig. 3 (a) institutes Show.If not carrying out any process, the reference frame before still being interrupted using monitoring then can be shown on monitoring image strong Strong artifact, falls into oblivion original target, shown in such as Fig. 3 (b)；If the data after selection restarts to guard are background frames, scheme As the upper target information then not having before, shown in such as Fig. 3 (c).Therefore certain processing method is needed, is not reselecting imaging In the case of reference frame, suppress image artifacts and recover original target information on image.

Data and imaging algorithm, when monitoring is restarted, are processed by flow process according to the following steps according to Fig. 1：

Step one：Obtain the inconsistent caused measurement data baseline change of electrode tip skin contact condition.Read monitoring to interrupt The last n frame data of front monitoring process, and the front n frame data of the monitoring for restarting, are combined into measurement data sequence x L (), the composition of data sequence are expressed in matrix as：

x_measL () represents the measurement data of meas Measurement channel of l frame data, and have l=2n.Survey in EIT In amount system, Measurement channel number is relevant with the excitation-measurement pattern of number of electrodes and employing.It is per treatment to choose single measurement Data sequence x of passage_iL (), using spline function according to the following formula to x_iL () is fitted, seek fitting function S (l)：

Fitting function S (l) is segmental cubic polynomials, the general type that can be written as：

S_j(l)=a_j(l-l_j)³+b_j(l-l_j)²+c_j(l-l_j)+d_j

Add natural boundary conditions b₀=b_l=0, and each section boundaries condition and the derivative condition of continuity is substituted into, obtain matrix Equation：

h_i=x_i+1-x_i, q_i=2 (h_i-1+h_i),

And obtain coefficient a_j,c_jWith regard to b_j,d_jExpression formula：

L'=0 is made, corresponding cubic polynomial coefficient is tried to achieve.After obtaining the expression formula of spline-fit function S (l), calculate The fitting value sequence of correspondence positionDetectionThe maximum of uplifted side at the baseline transitionWith the pole for declining side Little valueCalculate Baseline wander value

Step 2：The measurement data baseline of correction change.Ith measurement channel data sequence y to monitoring process 2_i L (), usesTo y_iL () carries out Baseline wander.As shown in Fig. 3 (a), compared with the data before monitoring interruption, y_i(l) place In the elevated side of baseline, then after correctingShown in correction result such as Fig. 4 (a).Such as Fig. 4 (b) institutes Show, if be directly imaged, guard and occur on image because of the caused monitoring image artifacts of electrode position change, affect Identification to guarding target judges, needs to carry out suppression process to image artifacts.

Step 3：Make imaging algorithm into, the image artifacts for suppressing electrode position change to cause.Dynamic imaging is used Gauss-Newton image reconstruction formula carries out augmentation process.There is Gauss-Newton imaging algorithm formula：

Δ ρ=- [J^tJ+λR]^-1Jz

Wherein it is J restructuring matrixs, λ R are regularization constraint item, and z is boundary voltage difference vector not in the same time, and Δ ρ is not for Electrical impedance change profile vector between in the same time.

For Jacobin matrixExpand ton_measFor the survey that frame data are included Amount voltage number, n_elemFor the FEM model unit number that imaging reconstruct used is used, n_eFor measuring electrode quantity, Jacobi square The expansion of battle array is filled to the priori disturbance compensation value of measuring electrode change in location, i.e.,：

In order to try to achieve filling element therein, the positive computing formula of electrical impedance imaging to be utilized

Z=HA

Wherein A is that current excitation is vectorial, and H is positive calculating matrix, z relevant with the FEM model coordinate for imaging For boundary voltage vector.CalculateOne priori amount θ is first set, the x-axis direction coordinate u of No. 1 electrode is changed_xFor u_x =u_x+ θ, recalculates positive reconstruction matrix, obtains new boundary voltage vector：

z_move=H_moveA

Then have

Calculate other electrodes in the same manner and the J matrix augmentation elements in the case that y-axis coordinate is subjected to displacement, priori amount is all used θ。

Then augmentation process is carried out to constraint matrix R, willExpand toExtension Divide the prior estimate of the noise prior estimate and reconstruct conductivity variations for being filled to reconstruct data, i.e.,

Wherein R_extraFor discrete Laplace filter form, can be write as：

Specifically element is：R_i,j|(extra)=2.1 δ², R_i,j|(extra)=-1 δ²(i units and j units are adjacent), other Element is 0.Whereinave_conductChange the ratio system relative to initial conductivity distribution for target conductivity Number, ave_moveFor electrode average displacement relative to field domain radius proportionality coefficient.

Constraint factor λ is：

Wherein, ave_noiseFor noise signal relative to measurement data proportionality coefficient.

Shown in imaging results after correction such as Fig. 4 (c).With without algorithm process, the imaging of imaging reference frame is reselected As a result compare, the imaging results after process can effectively reflect the imageable target before monitoring interruption.With without algorithm process, do not weigh Newly it is chosen to compare as the imaging results of reference frame, the imaging results after process can effectively suppress the baseline of DATA REASONING to change Caused image artifacts.Through the algorithm compensation changed to electrode position, and the Baseline wander of measurement data, compared with Fig. 2, Jing Imaging results after this paper algorithm process effectively reduce original image information, the color range bar on the right side of reference picture, it is ensured that also Former image result is consistent with original image in numeric distribution.Therefore, the monitoring image information inherit the processing method that processes can be with Have no progeny in successfully managing cranium brain dynamic electric impedance imaging on-line monitor the situation that original monitoring-information is lost after restarting to guard.

Claims (4)

1. the processing method that monitoring image information of having no progeny in a kind of cranium brain dynamic electric impedance imaging on-line monitor is inherited, its feature exist In the processing method is analyzed to the measurement data before and after on-line monitor interruption：First, the method being fitted using batten difference Measurement data is processed, measurement data baseline change caused by electrode contact state is inconsistent before and after suppressing on-line monitor to interrupt causes Image artifacts；Then, augmentation process is carried out to image reconstruction algorithm, the prior information of electrode displacement is introduced into reconstruction matrix, Before and after reducing monitoring interruption, electrode is because of the inconsistent impact to guarding image of riding position.

2. the place that monitoring image information of having no progeny in cranium brain dynamic electric impedance imaging on-line monitor according to claim 1 is inherited Reason method, it is characterised in that comprise the following steps：

1) obtain the Baseline wander value of measurement data

Selected on-line monitor interrupt before last n frame data and restart the front n frame data guarded, by data cube computation one Rise, be data x (l), to i-th valid data channel data x_iL () uses Spline-Fitting, obtain fitting sequence Wherein, i ∈ N, N are Measurement channel sum；

It is then detected thatThe maximum of uplifted side at the baseline transitionWith the minimum for declining sideCalculate base Line correction value

2) Baseline wander is measured to restarting the data guarded

To restarting the ith measurement channel data sequence y after guarding_iL (), usesTo y_iL () carries out Baseline wander, Compared with the data before monitoring interruption, if y_iL () is in the elevated side of baseline, then after correcting If y_iL side that () declines in baseline, then after correcting

3) augmentation process is carried out to Gauss-Newton image reconstruction formula

To Δ ρ=- [J^tJ+λR]^-1Jz, Δ ρ be not in the same time between electrical impedance change profile vector, z is border not in the same time Voltage difference vector；

By Jacobin matrixExpand toBy constraint matrixExpand toAnd coefficient lambda and constraint matrix R is determined using prior information_augUnit item；

4) impedance variations of target field domain are rebuild using the reconstruction formula of data and augmentation after correction

I.e.WhereinFor the measurement voltage data at t1 moment,For the measurement at t2 moment Voltage data.

3. the place that monitoring image information of having no progeny in cranium brain dynamic electric impedance imaging on-line monitor according to claim 2 is inherited Reason method, it is characterised in that step 1) in, to i-th valid data channel data x_iL () uses Spline-Fitting, specifically Carry out as the following formula：

$L=p\underset{j=0}{\overset{n}{\Σ}}{(x_{i,j}-S(l_j\left)\right)}^2+(1-p){\∫}_m^n{\left(S^{\′\′}\right(l\left)\right)}^{2}dl;$

Wherein, S (x) is data sequence x_iL the spline-fit function of (), asks S (x) to make the value of L get minimum；Wherein, p ∈ [0,1] Reflect the degree of closeness of fitting function and measured data；S (x) is segmentation batten fitting function, is write as general type：

S_j(l)=a_j(l-l_j)³+b_j(l-l_j)²+c_j(l-l_j)+d_j；

Add boundary condition to solve each piecewise fitting function coefficients of S (l) using least square method, obtain S (l) expressions Afterwards, seek fitting reconfiguration sequenceDetectionThe maximum of uplifted side at the baseline transitionIt is minimum with decline side ValueCalculate Baseline wander value

4. the place that monitoring image information of having no progeny in cranium brain dynamic electric impedance imaging on-line monitor according to claim 2 is inherited Reason method, it is characterised in that step 3) concrete operations are：

Gauss -- Newtonian image reconstruction formula such as following formula：

Δ ρ=- [J^tJ+λR]^-1Jz；

Wherein, J is Jacobin matrix, and λ R are regularization constraint item, Δ ρ be not in the same time between electrical impedance change profile vector, z For boundary voltage difference vector not in the same time；

For suppressing electrode position change influence on RT, augmentation process is carried out to the matrix in reconstruction formula：For refined Respectively compare matrixExpand ton_measFor the measurement voltage number that frame data are included, n_elem For the FEM model unit number that imaging reconstruct used is used, n_dimFor being imaged dimension, two-dimensional imaging is generally carried out, so taking n_dim=2, n_eFor number of electrodes, the expansion of Jacobin matrix is filled to due to the disturbance of measuring electrode change in location, i.e.,：

${\mathrm{\Δv}}_{n_{electrode}|d_{move}}=AH-{\mathrm{AH}}_{n_{electrode}|d_{move}},d_{move}=xory;$

Wherein, A is current excitation vector, and H is positive calculating matrix,After changing for electrode position in an x or y direction The positive calculating matrix for recalculating, the displacement generally unification are set to a priori constant；

Then, augmentation process is carried out to constraint matrix R：

WillExpand toExpansion is filled to the noise prior estimate of reconstruct data With reconstruct conductivity variations prior estimate, i.e.,：

$R_{aug}=\left(\begin{array}{cc}R& \\ & R_{extra}\end{array}\right);$

Wherein, R_extraFor a Laplace filter, regularization parameter λ and R are determined according to the prior information of imaging field domain_extra Laplce's template, then have：

$\mathrm{\λ}=\frac{{\mathrm{ave}}_{noise}}{{\mathrm{ave}}_{conduct}},R_{extra}={\left(\frac{{\mathrm{ave}}_{conducl}}{{\mathrm{ave}}_{move}}\right)}^2{\▿}^2;$

Wherein, ave_noiseFor noise relative to measurement data average priori amplitude, ave_conductFor conductivity variations relative to The average priori amplitude of initial conductivity distribution, ave_moveIt is the average priori amplitude of electrode displacement relative to field domain radius, For Laplace operator.