Magnetic resonance temperature measurement imaging method and application thereof

Abstract

The invention provides a magnetic resonance temperature measurement imaging method. According to the method, based on proton resonance frequency temperature dependence, phase self-calibration echoes are added in a temperature measurement imaging sequence; the phase unwrapping of the spatial dimension is carried out on the phase image by using the prior art, and the phase unwrapping of the time dimension is carried out on the temperature measurement phase image by using the self-calibration phase image. According to the method provided by the invention, the problem of reference basis failure of magnetic resonance temperature measurement imaging can be solved, the temperature measurement precision is not lost, and additional temperature measurement tools and intraoperative waiting time are not increased. According to the method provided by the invention, the temperature measurement precision can be improved by prolonging the echo time TE. The method provided by the invention can be used for thermal ablation treatment guided by magnetic resonance.

Description

Technical Field

[0001] The present invention belongs to the technical field of magnetic resonance imaging, and in particular relates to a magnetic resonance temperature measurement imaging method and application thereof. Background Art

[0002] Compared with other temperature measurement technologies, magnetic resonance temperature imaging technology can perform multi-slice and multi-angle imaging of internal human tissues at any angle. It has the advantages of high temperature measurement accuracy, high spatial resolution and coverage, high soft tissue contrast, no bone artifacts, non-invasiveness, and no ionizing radiation. During continuous scanning, neither patients nor doctors will be harmed by radiation. It is particularly suitable for continuous quantitative monitoring of the distribution and changes of internal human tissue temperature during thermal ablation therapy. With the advancement of related technologies, magnetic resonance-guided thermal ablation therapy is being used more and more widely in tumor treatment and other disease treatments.

[0003] Thermal ablation is a method of treating tumors. Its principle is to control the temperature of the target tissue to above 47°C for a period of time to kill tumor cells. Thermal ablation for tumor treatment has the advantages of minimally invasive / non-invasive and less blood loss. It can also improve the body's immune function and enhance anti-tumor ability by affecting the tumor microenvironment. Therefore, thermal ablation plays an irreplaceable role in tumor treatment. Currently, common thermal ablation methods include high-intensity focused ultrasound ablation (HIFU), radiofrequency ablation (RFA) and microwave ablation (MWA).

[0004] In thermal ablation therapy, real-time and non-invasive acquisition of the temperature distribution of the treatment target and its surrounding tissues is crucial to ensure the safety and effectiveness of the treatment. Magnetic resonance temperature measurement is mainly based on the temperature dependence of magnetic resonance parameters such as relaxation time, diffusion coefficient, proton resonance frequency, and magnetization transfer. In MRI-guided thermal ablation therapy, the most widely used method is magnetic resonance temperature imaging based on the temperature dependence of proton resonance frequency. The principle is that the proton resonance frequency has a linear relationship with temperature (-0.01ppm / ºC) in a large temperature range (-15ºC~100ºC). The gradient echo imaging sequence can be used to obtain the magnetic resonance phase map, and then the phase difference between the magnetic resonance phase maps at different times is calculated. The change in temperature distribution is calculated based on the echo time TE and the main magnetic field strength of the magnetic resonance imaging system (hereinafter referred to as the main magnetic field strength) B0.

[0005] In practical applications, before thermal ablation is performed, a frame of 3D magnetic resonance phase image or multi-layer 2D magnetic resonance phase image (hereinafter referred to as magnetic resonance phase image) is collected as a reference phase image. The reference phase image corresponds to the distribution of the basic temperature of the treatment target area and its surrounding tissues (hereinafter referred to as basic temperature). During thermal ablation treatment, the magnetic resonance phase image obtained at a certain moment is used as a temperature measurement phase image, which is subtracted from the reference phase image to obtain a phase difference image, and then the change of temperature distribution relative to the basic temperature distribution is calculated based on the echo time TE and the main magnetic field strength B0 of the imaging sequence, and finally the temperature distribution at that moment is obtained based on the basic temperature distribution.

[0006] According to the properties of inverse trigonometric functions, the phase value range of each pixel in the magnetic resonance phase image is between -π and π. Therefore, before displaying the phase image or the phase difference image between two frames, it is necessary to perform phase unwrapping in the spatial dimension to eliminate the phase jump between adjacent pixels in the image.

[0007] During thermal ablation therapy, a frame of temperature measurement phase map is usually collected every few seconds to tens of seconds according to the needs of treatment and temperature monitoring. When the temperature changes slowly, the phase difference between two adjacent frames will not exceed half a phase cycle (one phase cycle is 2π). During continuous monitoring, the temperature measurement phase map is continuously collected. By recording the phase difference between two adjacent frames, it can be determined how many phase cycles are spanned between the current temperature measurement phase map and the reference phase map, so as to correctly calculate the current temperature distribution. That is, in magnetic resonance-guided thermal ablation therapy, in addition to the phase unwrapping in the spatial dimension, the magnetic resonance phase maps obtained at different times also need to be phase unwrapped in the time dimension.

[0008] In actual applications, due to the influence of various factors during thermal ablation treatment and temperature monitoring, it is often necessary to suspend temperature measurement imaging. During the period of suspending temperature measurement imaging, if the temperature change of the tissue exceeds a certain range (for example, the temperature of the tissue rises sharply after the application of the thermal dose, or there are large blood vessels around the tissue, and the temperature of the tissue drops rapidly after the heating is stopped), the phase difference between the two frames of temperature measurement phase images before and after the temperature measurement is suspended will exceed half or one or even more phase cycles. Since it is impossible to know the specific number of phase cycles, the reference phase image cannot be used as a reference when the temperature measurement imaging is started again, resulting in the inability to correctly calculate the temperature distribution corresponding to the temperature measurement phase image, that is, the reference reference is invalid.

[0009] There are currently three solutions to solve the above-mentioned reference failure problem. The first solution is to wait for the tissue temperature to return to the basal temperature and re-collect the reference phase map. The disadvantage of this solution is that it prolongs the intraoperative waiting time, which may affect the implementation of the treatment plan and the treatment effect. The second solution is to use other temperature measurement tools as an auxiliary (such as fiber optic thermometers, temperature measurement probes, etc.) for minimally invasive temperature measurement. The disadvantage of this solution is that it adds additional tools and operations, which in turn increases the intraoperative workload and operation complexity. In addition, since auxiliary temperature measurement tools are usually invasive, the second solution is difficult to use for non-invasive treatment. The third solution is to shorten the echo time TE of the imaging sequence. For the same temperature change, the shorter the TE, the smaller the phase change, and the less likely it is to have phase wrapping. That is, shortening TE can avoid reference failure. However, in magnetic resonance temperature imaging based on the temperature dependence of the proton resonance frequency, the echo time TE is proportional to the temperature measurement accuracy, and shortening TE will reduce the temperature measurement accuracy. Therefore, the third solution is only suitable for situations where the temperature measurement accuracy is not required or the temperature change range is not large.

[0010] Chinese patent announcement number CN 101507603 B discloses a high-temporal-resolution magnetic resonance temperature imaging method and device, which belongs to the field of magnetic resonance temperature imaging technology. The high-temporal-resolution magnetic resonance temperature imaging method includes: using a radial sequence to scan the area to be diagnosed based on the magnetic resonance coil to obtain k-space data corresponding to the area to be diagnosed; grouping the k-space data, the spokes of the radial sequence, and the weight of the radial sequence based on the time frame to obtain at least two groups of grouped data; reconstructing at least two groups of grouped data based on the sensitivity map of the magnetic resonance coil to obtain multiple frames of target time frame images; there are no under-sampling artifacts in the multiple frames of target time frame images; based on the temperature difference between any two adjacent frames of target time frame images, the target temperature change image is obtained. The high-temporal-resolution magnetic resonance temperature imaging method can obtain the temperature change process of the area to be diagnosed that changes continuously, and improves the temporal resolution of the temperature change process.

[0011] International Publication No. WO2019126934A1 A local shimming system and shimming method for magnetic resonance imaging, comprising the following steps: using two-dimensional gradient echo to collect B0 field map information (301); calculating and evaluating the uniformity of B0 (302); optimizing the current of each channel shimming coil (303); determining whether the minimum standard deviation value of Δf is obtained (304); outputting the optimal current combination value, and setting the optimal current value corresponding to each channel shimming coil on the current control software (305); and testing and evaluating the uniformity of B0 to achieve the shimming target (306).

[0012] Chinese patent application number CN202410086540.X discloses a high temporal resolution real-time magnetic resonance temperature imaging method, which relates to the field of temperature imaging technology; the method includes: constructing a temperature prediction equation and determining various parameters; predicting the temperature of each subsequent frame image and obtaining predicted K-space data; executing a variable density sampling strategy to obtain actual under-sampled K-space data; performing data filling on the actual under-sampled K-space data to obtain updated K-space data; performing phase difference calculation based on the PRF method to obtain several temperature difference images and output them as temperature animations. The present invention uses the predicted K-space data to perform data filling on the K-space data obtained by under-sampled scanning to obtain complete K-space data; it not only greatly saves time at the data acquisition level, but also eliminates artifacts caused by under-sampled scanning. At the same time, the data processing level does not require as much time as the compressed sensing technology, which ensures the real-time nature of the data and greatly improves the temporal resolution of the treatment temperature change process.

[0013] Chinese patent announcement number CN 111568390B discloses a method for reducing system temperature errors in real-time magnetic resonance temperature imaging: placing a number of reference water films around the imaging object, and measuring the temperature of the reference water films in real time through magnetic resonance compatible temperature optical fibers; collecting magnetic resonance phase images before and during heating, calculating the phase difference of each reference water film temperature measurement point, and the actual temperature change of each reference water film temperature measurement point, and calculating the main magnetic field deviation of each reference water film temperature measurement point; estimating the overall plane distribution of the main magnetic field deviation according to the real-time main magnetic field deviation of the reference water film temperature measurement point; correcting the temperature directly calculated according to the phase image for each pixel of the magnetic resonance imaging according to the overall plane distribution of the main magnetic field deviation, and obtaining the actual temperature. The method provided by the present invention reduces the temperature measurement error in magnetic resonance temperature imaging by accurately estimating the main magnetic field offset value and spatial distribution during magnetic resonance temperature imaging.

[0014] In addition, Chinese patent publication number CN 113842210 A discloses a method and device for simulating microwave ablation surgery for vertebral tumors, which relates to the field of medical devices and simulation technology, wherein the method includes: obtaining a magnetic resonance image of the patient in the current state during surgery, performing image segmentation on the magnetic resonance image, and establishing an individual model of the patient based on the segmented image, and registering the individual model according to the preoperative simulation model to obtain an accurate model of the patient in the current state; inserting a microwave probe according to the target pin position and heating it, collecting actual temperature data during heating, and using the actual temperature data to correct the microwave probe model; performing temperature simulation on the target pin position of the corrected microwave probe model at the accurate model, and performing thermal damage assessment based on the temperature simulation results to correct the target pin position and target heating time, and generating a surgical simulation plan for the area to be ablated. This method can improve the accuracy and credibility of surgical simulation and increase the success rate of microwave thermal ablation surgery for vertebral tumors. Summary of the invention

[0015] The purpose of the present invention is to provide a magnetic resonance temperature measurement imaging method in view of the deficiencies in the prior art.

[0016] Another object of the present invention is to provide an application of the above-mentioned magnetic resonance temperature imaging method.

[0017] To achieve the above object, the present invention adopts the following technical scheme: a magnetic resonance temperature measurement imaging method, based on the temperature dependence of the proton resonance frequency, adding a phase self-calibration echo in the temperature measurement imaging sequence; performing phase unwrapping in the spatial dimension on the phase image, and performing phase unwrapping in the time dimension on the temperature measurement phase image using the self-calibration phase image; calculating the temperature distribution using the temperature measurement phase image that has been unwrapped in the time dimension and the spatial dimension; comprising the following steps: (1) Add phase self-calibration echo to the temperature imaging sequence: In the temperature measurement imaging sequence, a phase self-calibration echo is added. The phase self-calibration echo has the same composition and operation mode as the temperature measurement imaging echo, and the positive and negative polarities of the convergence gradients corresponding to the two echoes are opposite; the temperature measurement imaging echo time is TE, and the phase self-calibration echo time is TE', and the difference between the two is ; The phase self-calibration echo precedes the temperature imaging echo Or after the temperature imaging echo ; The absolute value of is less than the time threshold t max And it is greater than the minimum time t allowed by the magnetic resonance imaging instrument min , Determined by the manufacturing technology of magnetic resonance imaging equipment, the time threshold t max for Where γ represents the nuclear gyromagnetic ratio, B0 represents the main magnetic field strength, Indicates the maximum value of temperature change between two adjacent temperature measurements; especially It can be used to indicate the maximum value of temperature change that may occur between two adjacent temperature measurements in thermal ablation technology; (2) Execute the temperature measurement imaging sequence to obtain the temperature measurement phase map and self-calibration phase map: Before implementing the thermal ablation technique, a temperature measurement imaging sequence is executed to collect a frame of temperature measurement phase map and self-calibration phase map as a reference temperature measurement phase map and a reference self-calibration phase map, corresponding to the basic temperature distribution; during the thermal ablation technique, each time the temperature measurement imaging sequence is executed, a frame of temperature measurement phase map and self-calibration phase map is obtained, corresponding to the temperature distribution at that time; (3) Existing technology can be used to perform spatial phase unwrapping on the reference temperature measurement phase map, the reference self-calibration phase map, the temperature measurement phase map, and the self-calibration phase map, so as to eliminate phase jumps between adjacent pixels in the image; (4) Use the self-calibration phase diagram to perform phase unwrapping in the time dimension of the temperature measurement phase diagram: If the phase value of a pixel in the reference temperature measurement phase image is , the phase value in the reference self-calibration phase diagram is , then the phase difference between the reference self-calibration phase diagram and the reference temperature measurement phase diagram is ; If the phase value of a certain pixel in the temperature measurement phase map is , the phase value in the self-calibration phase diagram is , then the phase difference between the self-calibration phase diagram and the temperature measurement phase diagram is ; Use the following formula to calculate the number of phase cycles n that the pixel spans between the temperature measurement phase map and the reference temperature measurement phase map, is the phase change accumulation. According to the relationship between phase, frequency and time, The ratio of TE is equal to ( - ) to ΔTE. That is, using The frequency calculated by TE is the same as that calculated by ( - ) and the frequency calculated by ΔTE are equal; then n is used to calculate the phase value of the pixel in the temperature measurement phase image after unwrapping ; Among them, fix() means rounding the value in the brackets to zero; Perform the above operation on each pixel in the above phase image; (5) Calculate the temperature distribution using the temperature measurement phase image unwrapped in the time and space dimensions: After deconvolution in time and space, if the phase value of a pixel in the temperature measurement phase map is , the phase value in the reference temperature measurement phase diagram is , the base temperature is , then use the following formula to calculate the temperature T corresponding to the pixel m : in, ; The above operation is performed on each pixel in the above phase image to obtain the temperature distribution.

[0018] The principle of the present invention is: based on the temperature dependence of the proton resonance frequency, a phase self-calibration echo is added to the temperature measurement imaging sequence; the existing technology is used to perform phase unwrapping on the phase image in the spatial dimension, and the self-calibration phase map is used to perform phase unwrapping on the temperature measurement phase map in the time dimension; the temperature distribution is calculated using the temperature measurement phase map that has been unwrapped in the time dimension and the space dimension.

[0019] The method of the present invention can solve the problem of reference base failure in magnetic resonance temperature measurement imaging without losing temperature measurement accuracy, adding additional temperature measurement tools and intraoperative waiting time; the method proposed by the present invention can also be combined with extending the echo time TE to improve temperature measurement accuracy.

[0020] Based on the above scheme, the order of step (3) and step (4) is exchanged.

[0021] Preferably, based on the fact that the proton resonance frequency has a linear relationship with the temperature of -0.01 ppm / °C in the temperature range of -15°C to 100°C, a magnetic resonance phase image is obtained using a gradient echo imaging sequence, and then the phase difference between the magnetic resonance phase images at different times is calculated, and then the change in temperature distribution is calculated according to the echo time TE and the main magnetic field strength B0 of the imaging sequence; that is, in the process of temperature measurement imaging, the phase is first obtained, and then the phase difference is calculated according to the phase difference. , the proton resonance change Δf and the temperature change ΔT are used to calculate ΔT, and then the current temperature Tm is calculated based on the basic temperature Tb, where: The temperature measurement accuracy is proportional to the temperature imaging echo time TE. If TE is long, a small change in temperature can cause a significant change in the actual phase measurement value, that is, the phase difference Increase, ≤ half phase period.

[0022] Furthermore, the phase self-calibration echo time TE' differs from the temperature imaging echo time TE ΔTE ; Phase difference Changes and temperature changes ΔT Proportional, when ΔT≤ 65°C, the phase difference between two adjacent frames The change between them does not exceed half a phase cycle, combined with ( ) and relationship, yes Unwrap the time dimension.

[0023] The present invention provides an application of the magnetic resonance temperature measurement imaging method in a magnetic resonance guided thermal ablation device, which is suitable for magnetic resonance guided thermal ablation technology.

[0024] The beneficial effects of the present invention are as follows: the method proposed by the present invention can solve the problem of reference failure of magnetic resonance temperature imaging without losing temperature measurement accuracy, adding additional temperature measurement tools and intraoperative waiting time. The method proposed by the present invention can also be combined with extending the echo time TE to improve temperature measurement accuracy. The method proposed by the present invention is particularly suitable for magnetic resonance-guided thermal ablation therapy. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 In the temperature measurement imaging sequence of the present invention, the phase self-calibration echo is after the temperature measurement imaging echo; Figure 2 In the temperature measurement imaging sequence of the present invention, the phase self-calibration echo precedes the temperature measurement imaging echo; Figure 3 is a flow chart of the present invention; Description of reference numerals: TR—pulse sequence repetition time; TE—temperature measurement imaging echo time; TE'—phase self-calibration echo time; 101, 102—RF excitation pulse; 201, 202—slice selection gradient pulse; 301—phase encoding gradient pulse; 400—dephasing gradient pulse; 401, 402—converging gradient pulses; 501—temperature measurement imaging echo; 502—phase self-calibration echo. DETAILED DESCRIPTION

[0026] The principle of magnetic resonance thermometry based on the temperature dependence of proton resonance frequency is that the proton resonance frequency has a linear relationship with temperature (-0.01ppm / ℃) in a large temperature range (-15ºC~100ºC). The gradient echo imaging sequence can be used to obtain the magnetic resonance phase map, and then the phase difference between the magnetic resonance phase maps at different times can be calculated. Then, the change in temperature distribution can be calculated based on the echo time TE and the main magnetic field strength B0 of the imaging sequence; that is, in the process of thermometry imaging, the phase is first obtained, and then the phase difference is calculated based on the phase difference. , the proton resonance change Δf and the temperature change ΔT are used to calculate ΔT, and then the current temperature Tm is calculated based on the basic temperature Tb, where the phase difference for: The proton resonance change Δf is: The current temperature Tm is: According to the above principle, the temperature measurement accuracy is proportional to the echo time TE of the temperature measurement imaging. That is, if TE is long, a small change in temperature can cause a significant change in the actual measurement value (phase). Taking the temperature measurement imaging using a 3T magnetic resonance imaging system as an example, the proton resonance frequency ( ) is about 128MHz. If the temperature measurement accuracy needs to reach 1ºC, considering the influence of image noise, TE should be no less than 50ms, and the phase difference corresponding to a temperature change of 1ºC should be no less than 0.128π. Under this condition, if the temperature difference corresponding to the temperature measurement phase image of two adjacent frames is If the temperature exceeds 10 ºC, the phase difference between the corresponding pixels of the two phase images will be More than half a phase period; if If the temperature exceeds 20°C, More than one phase period. If the phase is more than half a period or more, the phase unwrapping in the time dimension will fail. That is, the phase continuity cannot be used to correctly unwrap the phase image of the next frame, so the temperature distribution cannot be calculated using the reference phase, that is, the reference benchmark fails. For example, based on the incorrectly unwrapped phase, the actual temperature difference of 10 ºC between two adjacent frames will be calculated as -5 ºC; the actual temperature difference of 20 ºC will be calculated as 5 ºC.

[0027] Under the above conditions, if TE is shortened to 15ms, even if Over 20°C, Still, it will not exceed half a phase cycle, so reference failure will not occur, but the temperature measurement accuracy drops to 3°C. Further shortening TE, the interval between two adjacent frames Larger values ​​can be allowed without causing phase wrapping and reference failure, but the temperature measurement accuracy will be further reduced. Therefore, the reference failure problem cannot be solved directly by shortening TE.

[0028] See also Figure 1 , Figure 2 The present invention adds a phase self-calibration echo 502 to the imaging sequence. Phase self-calibration echo time The difference with the temperature imaging echo time TE . In temperature changes Under the same conditions, The smaller the value, the greater the phase difference between the corresponding pixels in the self-calibration phase image and the temperature measurement phase image ( ) is smaller. The phase difference ( ) changes with temperature changes Generally, the basic temperature of human tissue is above 35 ºC, and the temperature of thermal ablation treatment is below 100 ºC. Therefore, even if the pause time during the temperature measurement imaging is long, the temperature difference between two adjacent frames will not be greater than 65 ºC. According to the law of heat conduction, the closer to the basic temperature, the slower the temperature change. Therefore, the temperature difference before and after the temperature measurement is paused during the treatment is generally 20 ºC~40 ºC. It can be seen that as long as The value of is reasonable, which can ensure that when the temperature changes to the maximum ( ) between two adjacent frames does not exceed half a phase cycle. Then use ( ) does not change more than half a period as a constraint, combined with ( ) and relationship, yes Unwrap the time dimension.

[0029] The present invention is described in detail below with reference to specific embodiments.

[0030] Example 1 See also Figure 1~Figure 3 The magnetic resonance temperature imaging method provided by the present invention comprises the following specific steps: (1) Add phase self-calibration echo to the temperature imaging sequence: In the thermometry imaging sequence, a phase self-calibration echo 502 is added, which has the same composition and operation mode as the thermometry imaging echo 501; the convergence gradients 402 and 401 corresponding to the two echoes have opposite positive and negative polarities. The thermometry imaging echo time is TE, and the phase self-calibration echo time is TE'. The difference between the two is ; The phase self-calibration echo 502 may precede the temperature imaging echo 501 ( ) or after ( ); The absolute value of the echo time threshold t max And greater than the minimum value allowed by the magnetic resonance imaging device , echo time threshold t max for: In the formula, if t min 1ms, which is determined by the manufacturing technology of magnetic resonance imaging equipment; represents the nuclear gyromagnetic ratio, B0 represents the main magnetic field strength, and taking the 3T magnetic resonance imaging system for temperature measurement as an example, the proton resonance frequency ≈128MHz; It indicates the maximum value of the temperature change between two adjacent temperature measurements. It is used in thermal ablation technology and has a value range of 15°C to 65°C. Preferably, in step (1), .

[0031] If the actual application is complicated and the temperature changes dramatically between two adjacent frames, it is difficult to accurately estimate ΔTmax, then it is set to 65°C and the absolute value of ΔTE is 5ms or 1ms. If the main magnetic field strength B0 is larger, such as 5T or 7T, the same Corresponding On the contrary, if the main magnetic field strength B0 is smaller, such as 1T or 1.5T, the same Corresponding The preferred absolute value of is larger.

[0032] For a magnetic resonance imaging system with a main magnetic field strength B0 of 7T, the proton resonance frequency ≈298MHz; t min is 1ms; when The absolute value of the phase difference between two adjacent frames is 1ms. The phase period should not exceed half.

[0033] (2) Execute the temperature measurement imaging sequence to obtain the temperature measurement phase map and self-calibration phase map: Before implementing thermal ablation technology, a thermometric imaging sequence is performed to acquire a frame of thermometric phase map and self-calibration phase map as a reference thermometric phase map and a reference self-calibration phase map, corresponding to the basic temperature distribution. During thermal ablation treatment, each time a thermometric imaging sequence is performed, a frame of thermometric phase map and self-calibration phase map is obtained, corresponding to the temperature distribution at that time.

[0034] (3) Using existing technology, the reference temperature measurement phase map, the reference self-calibration phase map, the temperature measurement phase map and the self-calibration phase map are phase unwrapped in the spatial dimension to eliminate the phase jump between adjacent pixels in the image.

[0035] (4) Use the self-calibration phase diagram to perform phase unwrapping in the time dimension of the temperature measurement phase diagram: If the phase value of a pixel in the reference temperature measurement phase image is , the phase value in the reference self-calibration phase diagram is , the phase value in the temperature measurement phase diagram is , the phase value in the self-calibration phase diagram is , the following formula is used to calculate the number of phase cycles n that the pixel spans between the temperature measurement phase map and the reference temperature measurement phase map, and then n is used to calculate the phase value of the pixel in the temperature measurement phase map after unwrapping .

[0036] fix() means rounding the value in the brackets toward zero.

[0037] The above operation is performed for each pixel in the above phase map.

[0038] (5) Calculate the temperature distribution using the temperature measurement phase image unwrapped in the time and space dimensions: After deconvolution in time and space, if the phase value of a pixel in the temperature measurement phase map is , the phase value in the reference temperature measurement phase diagram is , the base temperature is , then use the following formula to calculate the temperature corresponding to the pixel for The above operation is performed on each pixel in the above phase image to obtain the temperature distribution.

[0039] The order of the above steps (3) and (4) can be swapped.

[0040] Example 2 The method proposed in the present invention can also be combined with extending the echo time TE to improve the temperature measurement accuracy.

[0041] Taking the temperature measurement imaging of 1.5T magnetic resonance imaging system as an example, the proton resonance frequency ( ) is about 64MHz. If the temperature measurement accuracy is required to be 1ºC, considering the influence of image noise, TE needs to be set to no less than 100ms, and the phase difference corresponding to a temperature change of 1ºC is approximately no less than 0.128π. If the existing technology is used, when the temperature difference corresponding to two adjacent frames of temperature measurement images exceeds 10 ºC, the phase difference corresponding to the temperature change is approximately 1.28π (more than half the phase cycle π). However, the situation where the temperature difference between two adjacent frames is 10 ºC is relatively common in thermal ablation therapy. At this time, it is necessary to shorten TE to meet the temperature difference requirement between two adjacent frames, but at the same time the temperature measurement accuracy decreases.

[0042] If the temperature difference requirement between two adjacent frames is met by shortening TE without losing the temperature measurement accuracy, the phase difference corresponding to a temperature change of 1°C will become smaller, which will make the phase measurement or calculation of the phase difference more sensitive to noise, that is, the signal-to-noise ratio of the phase diagram will decrease, thus affecting the accuracy of the temperature measurement. Therefore, using the existing technology, it is necessary to make a trade-off between meeting the temperature difference requirement between two adjacent frames and maintaining the temperature measurement accuracy.

[0043] By adopting the method proposed in the present invention, TE can be extended to meet the requirements of temperature measurement accuracy. At the same time, the temperature measurement imaging echo and the phase self-calibration echo are used. If the time difference between the two echoes is If TE and TE are small enough, the difference between the temperature measurement phase diagrams of two adjacent frames can be avoided to exceed half a phase period, so that the temperature difference requirement of two adjacent frames can be met in a larger range. , which can simultaneously meet the requirements of temperature difference between two adjacent frames, temperature measurement accuracy and temperature measurement precision.

[0044] For specific steps, please refer to Example 1.

Claims

1. A magnetic resonance temperature imaging method, characterized in that: Based on the temperature dependence of the proton resonance frequency, a phase self-calibration echo is added to the temperature measurement imaging sequence; the phase image is phase unwrapped in the spatial dimension, and the temperature measurement phase image is phase unwrapped in the time dimension using the self-calibration phase image; the temperature distribution is calculated using the temperature measurement phase image that has been unwrapped in the time dimension and the spatial dimension; the steps include: (1) Add phase self-calibration echo to the temperature imaging sequence: In the temperature measurement imaging sequence, a phase self-calibration echo is added. The phase self-calibration echo has the same composition and operation mode as the temperature measurement imaging echo, and the positive and negative polarities of the convergence gradients corresponding to the two echoes are opposite; the temperature measurement imaging echo time is TE, and the phase self-calibration echo time is TE', and the difference between the two is ; The phase self-calibration echo precedes the temperature imaging echo Or after the temperature imaging echo ; The absolute value of is less than the time threshold t max And it is greater than the minimum time t allowed by the magnetic resonance imaging instrument min , Where γ represents the nuclear gyromagnetic ratio, B0 represents the main magnetic field strength, Indicates the maximum value of temperature change between two adjacent temperature measurements; (2) Execute the temperature measurement imaging sequence to obtain the temperature measurement phase map and self-calibration phase map: A frame of temperature measurement phase map and self-calibration phase map is collected as a reference temperature measurement phase map and a reference self-calibration phase map, corresponding to the basic temperature distribution; each time the temperature measurement imaging sequence is executed, a frame of temperature measurement phase map and self-calibration phase map is obtained, corresponding to the temperature distribution at that time; (3) performing spatial phase unwrapping on the reference temperature measurement phase image, the reference self-calibration phase image, the temperature measurement phase image, and the self-calibration phase image, respectively, to eliminate phase jumps between adjacent pixels in the image; (4) Use the self-calibration phase diagram to perform phase unwrapping in the time dimension of the temperature measurement phase diagram: If the phase value of a pixel in the reference temperature measurement phase image is , the phase value in the reference self-calibration phase diagram is , then the phase difference between the reference self-calibration phase diagram and the reference temperature measurement phase diagram is ; If the phase value of a certain pixel in the temperature measurement phase map is , the phase value in the self-calibration phase diagram is , then the phase difference between the self-calibration phase diagram and the temperature measurement phase diagram is ; Use the following formula to calculate the number of phase cycles n that the pixel spans between the temperature measurement phase map and the reference temperature measurement phase map, is the phase change accumulation. According to the relationship among phase, frequency and time, The ratio of TE is equal to ( - ) and ΔTE; that is, using The frequency calculated by TE is the same as that calculated by ( - ) and the frequency calculated by ΔTE are equal; then n is used to calculate the phase value of the pixel in the temperature measurement phase image after unwrapping ; Among them, fix() means rounding the value in the brackets to zero; Perform the above operation on each pixel in the above phase image; (5) Calculate the temperature distribution using the temperature measurement phase image unwrapped in the time and space dimensions: After deconvolution in time and space, if the phase value of a pixel in the temperature measurement phase map is , the phase value in the reference temperature measurement phase diagram is , the base temperature is , then use the following formula to calculate the temperature T corresponding to the pixel m : in, , perform the above operation on each pixel in the above phase image to obtain the temperature distribution.

2. The magnetic resonance temperature imaging method according to claim 1, characterized in that: The order of step (3) and step (4) is exchanged.

3. The magnetic resonance temperature imaging method according to claim 1, characterized in that: Based on the fact that the proton resonance frequency has a linear relationship with temperature of -0.01ppm / ℃ in the temperature range of -15℃~100℃, the magnetic resonance phase image is obtained by using the gradient echo imaging sequence, and then the phase difference between the magnetic resonance phase images at different times is calculated, and then the change of temperature distribution is calculated according to the echo time TE and the main magnetic field strength B0 of the imaging sequence; that is, in the process of temperature measurement imaging, the phase is first obtained, and then the phase difference is calculated according to the phase difference. , the proton resonance change Δf and the temperature change ΔT are used to calculate ΔT, and then the current temperature Tm is calculated based on the basic temperature Tb, where: The temperature measurement accuracy is proportional to the temperature imaging echo time TE. If TE is long, a small change in temperature can cause a significant change in the actual phase measurement value, that is, the phase difference Increase, ≤ half phase period.

4. The magnetic resonance temperature imaging method according to claim 1, characterized in that: The difference between the phase self-calibration echo time TE' and the temperature imaging echo time TE ΔTE ; Phase difference Changes and temperature changes ΔT Proportional, when ΔT≤ 65°C, the phase difference between two adjacent frames The change between them does not exceed half a phase cycle, combined with ( ) and relationship, yes Unwrap the time dimension.

5. The magnetic resonance temperature imaging method according to any one of claims 1 to 4, characterized in that: Use a 3T magnetic resonance imaging system to perform temperature measurement imaging, according to the following steps: (1) Add phase self-calibration echo to the temperature imaging sequence: In the thermometry imaging sequence, a phase self-calibration echo 502 is added, which has the same composition and operation mode as the thermometry imaging echo 501; the positive and negative polarities of the convergence gradients 402 and 401 corresponding to the two echoes are opposite; the thermometry imaging echo time is TE, and the phase self-calibration echo time is , the difference between the two ; Wherein, the phase self-calibration echo 502 is before the temperature measurement imaging echo 501 ( ) or after ( ); The absolute value of the echo time threshold t max And greater than the minimum value allowed by the magnetic resonance imaging device , echo time threshold t max for: In the formula, t min 1ms, determined by the manufacturing technology of magnetic resonance imaging equipment; represents the nuclear gyromagnetic ratio, B0 represents the main magnetic field strength, and the proton resonance frequency ≈128MHz; It indicates the maximum value of the temperature change between two adjacent temperature measurements, ranging from 15℃ to 65℃; (2) Execute the temperature measurement imaging sequence to obtain the temperature measurement phase map and self-calibration phase map: A frame of temperature measurement phase map and self-calibration phase map is collected as a reference temperature measurement phase map and a reference self-calibration phase map, corresponding to the basic temperature distribution; each time the temperature measurement imaging sequence is executed, a frame of temperature measurement phase map and self-calibration phase map is obtained, corresponding to the temperature distribution at that time; (3) Using existing technology, phase unwrapping in the spatial dimension is performed on the reference temperature measurement phase image, the reference self-calibration phase image, the temperature measurement phase image, and the self-calibration phase image to eliminate phase jumps between adjacent pixels in the image; (4) Use the self-calibration phase diagram to perform phase unwrapping in the time dimension of the temperature measurement phase diagram: If the phase value of a pixel in the reference temperature measurement phase image is , the phase value in the reference self-calibration phase diagram is , the phase value in the temperature measurement phase diagram is , the phase value in the self-calibration phase diagram is , then use formulas 2 to 6 to calculate the number of phase cycles n that the pixel spans between the temperature measurement phase map and the reference temperature measurement phase map, and then use n to calculate the phase value of the pixel in the temperature measurement phase map after unwrapping ; Perform the above operation on each pixel in the above phase image; (5) Calculate the temperature distribution using the temperature measurement phase image unwrapped in the time and space dimensions: After deconvolution in time and space, if the phase value of a pixel in the temperature measurement phase map is , the phase value in the reference temperature measurement phase diagram is , the base temperature is , then use formula 7 to calculate the temperature corresponding to the pixel ; The above operation is performed on each pixel in the above phase image to obtain the temperature distribution.

6. The magnetic resonance temperature imaging method according to claim 5, characterized in that: In step (1), If the actual situation is complicated and the temperature changes dramatically between two adjacent frames, it is difficult to accurately estimate ΔTmax, then it is set to 65ºC. The absolute value of ΔTE is 5ms or 1ms, which is related to the main magnetic field strength B0.

7. The magnetic resonance temperature imaging method according to claim 5, characterized in that: If the main magnetic field strength B0 is 5T or 7T, the same Corresponding On the contrary, if the main magnetic field strength B0 is 1T or 1.5T, the same Corresponding The preferred absolute value of is larger.

8. The magnetic resonance temperature imaging method according to claim 6, characterized in that: For a magnetic resonance imaging system with a main magnetic field strength B0 of 7T, the proton resonance frequency ≈298MHz; t min is 1ms; when The absolute value of the phase difference between two adjacent frames is 1ms. The phase period should not exceed half.

9. The magnetic resonance temperature imaging method according to any one of claims 1 to 4, characterized in that: Use a 1.5T magnetic resonance imaging system to perform temperature measurement imaging, according to the following steps: Proton resonance frequency ≈64MHz; if the temperature measurement accuracy is required to be 1°C, considering the influence of image noise, TE is set to be no less than 100ms, and the phase difference corresponding to a temperature change of 1°C is no less than 0.128π; Set TE and , in order to simultaneously meet the requirements of temperature difference between two adjacent frames, temperature measurement precision and temperature measurement accuracy.

10. Use of the magnetic resonance temperature measurement imaging method according to any one of claims 1 to 9 in a magnetic resonance-guided thermal ablation device.

Images