CN115561690B - Magnetic resonance data processing method and device and computer equipment - Google Patents
Magnetic resonance data processing method and device and computer equipment Download PDFInfo
- Publication number
- CN115561690B CN115561690B CN202211162346.2A CN202211162346A CN115561690B CN 115561690 B CN115561690 B CN 115561690B CN 202211162346 A CN202211162346 A CN 202211162346A CN 115561690 B CN115561690 B CN 115561690B
- Authority
- CN
- China
- Prior art keywords
- magnetic resonance
- dictionary
- target
- hemodynamic response
- parameters
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R33/00—Arrangements or instruments for measuring magnetic variables
- G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
- G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
- G01R33/48—NMR imaging systems
- G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
Abstract
The application relates to a magnetic resonance data processing method, a magnetic resonance data processing device and computer equipment. The method comprises the following steps: determining a time sequence observation value corresponding to each voxel of a target part according to a magnetic resonance imaging sequence of the target part; searching target dictionary entries matched with each time series observation value from a preset magnetic resonance fingerprint dictionary; dictionary entries in the magnetic resonance fingerprint dictionary are used for recording the corresponding relation between the time sequence estimated value and the hemodynamic response function; and processing the magnetic resonance data of each voxel of the target part according to the hemodynamic response function corresponding to each target dictionary entry to obtain a magnetic resonance data processing result of the target part. By adopting the method, the accuracy of magnetic resonance data processing can be improved.
Description
Technical Field
The present application relates to the field of magnetic resonance technologies, and in particular, to a magnetic resonance data processing method, apparatus, and computer device.
Background
With the development of magnetic resonance technology, a BOLD-fMRI (Blood Oxygen Level Dependent-functional Magnetic Resonance Imaging, blood oxygen level dependent functional magnetic resonance imaging) technology has emerged, and a hemodynamic response is obtained by performing convolution operation on BOLD-fMRI data and HRF (Hemodynamic Response Function ), and then the hemodynamic response is input into a generalized linear model to perform fitting, so that activated voxels in the BOLD-fMRI data can be searched according to a fitting slope.
However, for different physiological regions of the human body, there may be differences in time delay, width and amplitude of HRF, in the prior art, it is difficult to perform differential design on HRF for voxels, and all voxels adopt the same HRF, so that deviation exists in estimation of hemodynamic response, and thus false positive or false negative easily occurs when activated voxels are searched in BOLD-fMRI data.
Therefore, the current data processing technology for BOLD-fMRI has the problem that differentiation processing is difficult to be performed on voxels, and the processing result is inaccurate.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a magnetic resonance data processing method, apparatus, computer device, and computer-readable storage medium capable of performing differentiation processing for voxels.
In a first aspect, the application provides a magnetic resonance data processing method. The method comprises the following steps:
determining a time sequence observation value corresponding to each voxel of a target part according to a magnetic resonance imaging sequence of the target part;
searching target dictionary entries matched with each time series observation value from a preset magnetic resonance fingerprint dictionary; dictionary entries in the magnetic resonance fingerprint dictionary are used for recording the corresponding relation between the time sequence estimated value and the hemodynamic response function;
And processing the magnetic resonance data of each voxel of the target part according to the hemodynamic response function corresponding to each target dictionary entry to obtain a magnetic resonance data processing result of the target part.
In one embodiment, before determining the time series observation value corresponding to each voxel of the target part according to the magnetic resonance imaging sequence of the target part, the method further comprises:
acquiring magnetic resonance fingerprint parameters; the magnetic resonance fingerprint parameters comprise hemodynamic response functions and tissue parameters;
according to a preset stimulation mode and a preset sequence code, performing simulation processing on the magnetic resonance fingerprint parameters to obtain a time sequence estimated value corresponding to the magnetic resonance fingerprint parameters;
and generating dictionary entries in the magnetic resonance fingerprint dictionary according to the corresponding relation among the hemodynamic response function, the tissue parameter and the time sequence estimated value.
In one embodiment, the acquiring magnetic resonance fingerprint parameters includes:
acquiring at least one of said hemodynamic response functions and at least one of said tissue parameters;
combining the at least one hemodynamic response function and the at least one tissue parameter to obtain at least one of the magnetic resonance fingerprint parameters.
In one embodiment, the searching for target dictionary entries matching each time series observation value from a preset magnetic resonance fingerprint dictionary includes:
finding out a time sequence estimated value matched with the time sequence observed value from the magnetic resonance fingerprint dictionary;
and determining dictionary entries corresponding to the time sequence estimation values as target dictionary entries.
In one embodiment, the finding a time series estimated value matched with the time series observed value from the magnetic resonance fingerprint dictionary includes:
determining a similarity between a time series estimated value and the time series observed value in the magnetic resonance fingerprint dictionary;
and determining a time sequence estimated value corresponding to the maximum value in the similarity as a time sequence estimated value matched with the time sequence observed value.
In one embodiment, after determining the dictionary entry corresponding to the time-series estimation value as the target dictionary entry, the method further includes:
and determining the tissue parameters corresponding to the target dictionary entries as the tissue parameters of each voxel of the target part.
In one embodiment, before determining the time series observation value corresponding to each voxel of the target part according to the magnetic resonance imaging sequence of the target part, the method further comprises:
acquiring a magnetic resonance imaging sequence of the target part; the magnetic resonance imaging sequence is obtained by carrying out magnetic resonance imaging on the target part according to the preset stimulation mode and the preset sequence code.
In one embodiment, the processing the magnetic resonance data of each voxel of the target location according to the hemodynamic response function corresponding to each target dictionary entry to obtain a magnetic resonance data processing result of the target location includes:
performing convolution operation on the magnetic resonance data of the voxels and the hemodynamic response function corresponding to the target dictionary entry to obtain hemodynamic response values corresponding to the voxels;
fitting the hemodynamic response values to obtain fitting results of the hemodynamic response values;
and determining the state of the voxel according to the slope corresponding to the fitting result.
In a second aspect, the application further provides a magnetic resonance data processing device. The device comprises:
The sequence determining module is used for determining a time sequence observation value corresponding to each voxel of the target part according to the magnetic resonance imaging sequence of the target part;
the dictionary searching module is used for searching target dictionary entries matched with the time sequence observation values from a preset magnetic resonance fingerprint dictionary; dictionary entries in the magnetic resonance fingerprint dictionary are used for recording the corresponding relation between the time sequence estimated value and the hemodynamic response function;
and the data processing module is used for processing the magnetic resonance data of each voxel of the target part according to the hemodynamic response function corresponding to each target dictionary entry to obtain a magnetic resonance data processing result of the target part.
In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor which when executing the computer program performs the steps of:
determining a time sequence observation value corresponding to each voxel of a target part according to a magnetic resonance imaging sequence of the target part;
searching target dictionary entries matched with each time series observation value from a preset magnetic resonance fingerprint dictionary; dictionary entries in the magnetic resonance fingerprint dictionary are used for recording the corresponding relation between the time sequence estimated value and the hemodynamic response function;
And processing the magnetic resonance data of each voxel of the target part according to the hemodynamic response function corresponding to each target dictionary entry to obtain a magnetic resonance data processing result of the target part.
In a fourth aspect, the present application also provides a computer-readable storage medium. The computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:
determining a time sequence observation value corresponding to each voxel of a target part according to a magnetic resonance imaging sequence of the target part;
searching target dictionary entries matched with each time series observation value from a preset magnetic resonance fingerprint dictionary; dictionary entries in the magnetic resonance fingerprint dictionary are used for recording the corresponding relation between the time sequence estimated value and the hemodynamic response function;
and processing the magnetic resonance data of each voxel of the target part according to the hemodynamic response function corresponding to each target dictionary entry to obtain a magnetic resonance data processing result of the target part.
The magnetic resonance data processing method, the device, the computer equipment and the storage medium comprise the steps of firstly determining a time sequence observation value corresponding to each voxel of a target part according to a magnetic resonance imaging sequence of the target part, then searching target dictionary entries matched with each time sequence observation value from a preset magnetic resonance fingerprint dictionary, and finally processing the magnetic resonance data of each voxel of the target part according to a hemodynamic response function corresponding to each target dictionary entry to obtain a magnetic resonance data processing result of the target part; the time sequence corresponding to each voxel of the target part can be extracted from the magnetic resonance imaging sequence, the hemodynamic response function corresponding to each time sequence is searched from the magnetic resonance fingerprint dictionary, the magnetic resonance data of each voxel is respectively processed by using the searched hemodynamic response function, and as different voxels correspond to different hemodynamic response functions, each voxel corresponds to a specific hemodynamic response function, the corresponding voxel is estimated by using the specific hemodynamic response function, so that the estimated hemodynamic response value is more accurate, and the accuracy of magnetic resonance data processing is improved.
Drawings
FIG. 1 is a flow chart of a method of magnetic resonance data processing in one embodiment;
figure 2 is a flow chart of a method of magnetic resonance data processing in another embodiment;
FIG. 3 is a flow chart of a functional magnetic resonance quantitative processing method based on magnetic resonance fingerprinting in one embodiment;
figure 4 is a block diagram of a magnetic resonance data processing apparatus in one embodiment;
fig. 5 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The magnetic resonance data processing method provided by the embodiment of the application can be applied to a terminal or a server. The terminal can be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things equipment and portable wearable equipment, and the internet of things equipment can be smart speakers, smart televisions, smart air conditioners, smart vehicle-mounted equipment and the like. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. The server may be implemented as a stand-alone server or as a server cluster composed of a plurality of servers.
In one embodiment, as shown in fig. 1, a magnetic resonance data processing method is provided, and the method is applied to a terminal for illustration, and includes the following steps:
step S110, determining a time sequence observation value corresponding to each voxel of the target part according to the magnetic resonance imaging sequence of the target part.
Wherein the magnetic resonance imaging sequence may be one or more magnetic resonance images arranged in time order.
The time series observation may be one or more magnetic resonance data arranged in time series, in particular a magnetic resonance data curve.
In specific implementation, a preset stimulation mode and a preset sequence code can be used for performing magnetic resonance imaging on a target part to generate a magnetic resonance imaging sequence, and corresponding magnetic resonance data is extracted from the magnetic resonance imaging sequence for each voxel of the target part, so that a time sequence observation value corresponding to each voxel can be obtained.
The stimulation pattern may be BOLD (Blood Oxygen Level Dependent ) stimulation pattern, among others. Specifically, the stimulation pattern may be a fixed period stimulation pattern, and may also be a random or pseudo-random stimulation pattern.
The sequence codes may be FA (Flip Angle) and TR (Repetition Time) of the radio frequency pulse sequence in the MRF (Magnetic Resonance Fingerprinting, magnetic resonance fingerprint) generation process.
For example, a radio frequency pulse sequence including K time points may be used to excite the brain, a BOLD stimulus with a fixed period is set, and (FA, TR) parameters of each time point are set to obtain a magnetic resonance imaging sequence including K magnetic resonance images, and for any voxel of the brain, magnetic resonance data corresponding to the voxel is extracted from each magnetic resonance image to obtain a set of time series observations including K magnetic resonance data, and assuming that a brain region includes L voxels, L sets of time series observations including K magnetic resonance data may be obtained.
Step S120, finding out target dictionary entries matched with each time sequence observation value from a preset magnetic resonance fingerprint dictionary; dictionary entries in the magnetic resonance fingerprint dictionary are used to record correspondence between time series estimates and hemodynamic response functions.
Wherein the hemodynamic response function may include an HRF expression and HRF parameters, wherein the HRF expression may be, but is not limited to, a dual gamma function expression or an exponential function expression.
The magnetic resonance fingerprint dictionary can be generated under a preset stimulation mode and a preset sequence code, and each dictionary entry in the magnetic resonance fingerprint dictionary is used for recording the corresponding relation between a group of time sequence estimated values and a hemodynamic response function.
The time series estimated value may be an estimated value of one or more magnetic resonance data arranged in time sequence, and may specifically be a magnetic resonance data curve obtained by simulation.
In a specific implementation, a correlation between the time series observation value and each dictionary entry in the magnetic resonance fingerprint dictionary may be calculated, the dictionary entry with the largest correlation is determined to be matched with the time series observation value, and the dictionary entry with the largest correlation is determined to be the target dictionary entry. The correlation between the time series observed value and the time series estimated value in each dictionary entry can also be calculated, the time series estimated value with the largest correlation is judged to be matched with the time series observed value, and the dictionary entry corresponding to the time series estimated value with the largest correlation is determined as the target dictionary entry. And performing the processing on the time sequence observed values corresponding to each voxel of the target part to obtain target dictionary entries corresponding to each time sequence observed value.
For example, for L groups of time-series observations corresponding to L voxels of the brain region, target dictionary entries corresponding to each group of time-series observations may be sequentially found from the magnetic resonance fingerprint dictionary, so as to obtain L target dictionary entries.
Step S130, processing the magnetic resonance data of each voxel of the target part according to the hemodynamic response function corresponding to each target dictionary entry to obtain a magnetic resonance data processing result of the target part.
In a specific implementation, because the dictionary entries record the correspondence between the time series estimated value and the hemodynamic response functions, the hemodynamic response functions recorded by each target dictionary entry can be obtained, where each hemodynamic response function corresponds to a voxel of the target site. And the corresponding hemodynamic response functions of all target dictionary entries can be obtained, and the corresponding voxel magnetic resonance data of the target part are respectively processed by using each hemodynamic response function to obtain the magnetic resonance data processing result of each voxel of the target part.
For example, L hemodynamic response functions corresponding to L target dictionary entries may be obtained, where each hemodynamic response function corresponds to a voxel of the brain region, and the magnetic resonance data of the L voxels of the brain region may be processed using the L hemodynamic response functions, respectively, to obtain a magnetic resonance data processing result of the brain region, where one hemodynamic response function processes data of a corresponding voxel.
According to the magnetic resonance data processing method, firstly, a time sequence observation value corresponding to each voxel of a target part is determined according to a magnetic resonance imaging sequence of the target part, then target dictionary entries matched with the time sequence observation values are found out from a preset magnetic resonance fingerprint dictionary, and finally, the magnetic resonance data of each voxel of the target part is processed according to a hemodynamic response function corresponding to each target dictionary entry, so that a magnetic resonance data processing result of the target part is obtained; the time sequence corresponding to each voxel of the target part can be extracted from the magnetic resonance imaging sequence, the hemodynamic response function corresponding to each time sequence is searched from the magnetic resonance fingerprint dictionary, the magnetic resonance data of each voxel is respectively processed by using the searched hemodynamic response function, and as different voxels correspond to different hemodynamic response functions, each voxel corresponds to a specific hemodynamic response function, the corresponding voxel is estimated by using the specific hemodynamic response function, so that the estimated hemodynamic response value is more accurate, and the accuracy of magnetic resonance data processing is improved.
In one embodiment, before step S110, the method specifically may further include: acquiring magnetic resonance fingerprint parameters; the magnetic resonance fingerprint parameters comprise hemodynamic response functions and tissue parameters; according to a preset stimulation mode and a preset sequence code, performing simulation processing on the magnetic resonance fingerprint parameters to obtain a time sequence estimated value corresponding to the magnetic resonance fingerprint parameters; dictionary entries in the magnetic resonance fingerprint dictionary are generated according to the correspondence between the hemodynamic response function, the tissue parameters, and the time series estimation values.
Wherein the hemodynamic response function may include an HRF expression and HRF parameters.
The tissue parameters may be physiological parameters of the human tissue, such as T1 (longitudinal relaxation time), T2 (transverse relaxation time) and T2 (transverse relaxation time based on free induction decay sequence) parameters of the human tissue.
The stimulation mode may be a BOLD stimulation mode.
The sequence codes can be FA and TR of the radio frequency pulse sequence in the MRF generation process.
In a specific implementation, a stimulus mode and (FA, TR) parameters may be preset at the terminal, a hemodynamic response function (HRF expression, HRF parameters) and tissue parameters (T1, T2) may be input to the terminal, the terminal may simulate the input hemodynamic response function (HRF expression, HRF parameters) and tissue parameters (T1, T2) under the preset stimulus mode and (FA, TR) parameters to obtain a set of time sequence estimation values, and the terminal may store the correspondence between the (HRF expression, HRF parameters), (T1, T2) and the obtained time sequence estimation values as a dictionary entry, change the hemodynamic response function and tissue parameters input to the terminal, and repeat the process to obtain other dictionary entries, and may use a set of multiple dictionary entries as a magnetic resonance fingerprint dictionary.
In this embodiment, by acquiring magnetic resonance fingerprint parameters, performing simulation processing on the magnetic resonance fingerprint parameters according to a preset stimulation mode and a preset sequence code to obtain a time sequence estimated value corresponding to the magnetic resonance fingerprint parameters, generating dictionary entries in a magnetic resonance fingerprint dictionary according to a corresponding relation among a hemodynamic response function, tissue parameters and the time sequence estimated value, simulating various conditions of the hemodynamic response function and the tissue parameters to obtain corresponding time sequence estimated values, and further forming the magnetic resonance fingerprint dictionary recorded with the corresponding relation among the hemodynamic response function, the tissue parameters and the time sequence estimated values, so as to facilitate searching the hemodynamic response function from the magnetic resonance fingerprint dictionary, and reduce complexity of magnetic resonance data processing.
In one embodiment, the step of acquiring the magnetic resonance fingerprint parameter may specifically include: acquiring at least one hemodynamic response function and at least one tissue parameter; at least one hemodynamic response function and at least one tissue parameter are combined to obtain at least one magnetic resonance fingerprint parameter.
In a specific implementation, at least one set of hemodynamic response functions (HRF expressions, HRF parameters) and at least one set of tissue parameters (T1, T2) may be input to the terminal, the terminal may perform various combinations of the at least one set of hemodynamic response functions (HRF expressions, HRF parameters) and the at least one set of tissue parameters (T1, T2) to obtain at least one set of magnetic resonance fingerprint parameters, and simulate each set of magnetic resonance fingerprint parameters under a preset stimulation mode and (FA, TR) parameters to obtain at least one set of time series estimated values, and the terminal may use the various combinations of the hemodynamic response functions (HRF expressions, HRF parameters) and the tissue parameters (T1, T2) and the corresponding time series estimated values as dictionary entries.
For example, 10 sets of (HRF expression, HRF parameters), 20 sets of (T1, T2) parameters may be input to the terminal, the terminal may simulate various combinations of (HRF expression, HRF parameters) and (T1, T2) according to the preset stimulation pattern and (FA, TR) parameters, to obtain 10×20=200 sets of time-series estimation values, and the terminal may use various combinations of (HRF expression, HRF parameters) and (T1, T2) and corresponding time-series estimation values as dictionary entries to obtain 200 dictionary entries.
In this embodiment, by acquiring at least one hemodynamic response function and at least one tissue parameter, and combining the at least one hemodynamic response function and the at least one tissue parameter to obtain at least one magnetic resonance fingerprint parameter, a plurality of dictionary entries may be generated in batch, and the generation efficiency of the magnetic resonance fingerprint dictionary may be improved.
In one embodiment, the step S120 may specifically include: finding out a time sequence estimated value matched with the time sequence observed value from the magnetic resonance fingerprint dictionary; and determining the dictionary entry corresponding to the time sequence estimation value as a target dictionary entry.
In a specific implementation, the correlation between the time series observation value and the time series estimation value in each dictionary entry can be calculated, the time series estimation value with the largest correlation is judged to be matched with the time series observation value, and the dictionary entry corresponding to the time series estimation value with the largest correlation is determined as the target dictionary entry.
For example, a time-series observation curve may be generated from the time-series observations, the time-series observation curve may be compared with a time-series estimation curve corresponding to the time-series estimation in each dictionary entry, and the correlation between the time-series observation curve and each time-series estimation curve may be calculated, and the dictionary entry corresponding to the time-series estimation curve having the greatest correlation may be determined as the target dictionary entry.
In practical application, the time series observed value D and the time series estimated value V in dictionary entry can be calculated i Correlation C between i The operation formula can be
C i ＝D·V i ，i＝1，2，……，N，
Where i is the dictionary entry index and N is the number of dictionary entries. Determination of { C i Dictionary entry index i corresponding to maximum value in } max Dictionary entry i max Is the target dictionary entry.
In this embodiment, the dictionary entry corresponding to the time-series estimated value is determined as the target dictionary entry by searching the time-series estimated value matched with the time-series observed value from the magnetic resonance fingerprint dictionary, so that the dictionary entry matched with the time-series observed value can be determined, and further the hemodynamic response function is determined according to the dictionary entry, so that different hemodynamic response functions corresponding to different voxels can be determined, and thus each voxel corresponds to a specific hemodynamic response function, and the hemodynamic response value of the corresponding voxel is estimated by using the specific hemodynamic response function, so that the obtained hemodynamic response value can be more accurate.
In one embodiment, the step of searching the time series estimated value matched with the time series observed value from the magnetic resonance fingerprint dictionary may specifically include: determining the similarity between the time series estimated value and the time series observed value in the magnetic resonance fingerprint dictionary; the time-series estimated value corresponding to the maximum value in the similarity is determined as the time-series estimated value that matches the time-series observed value.
In a specific implementation, the similarity between the time series observed value and each time series estimated value in the magnetic resonance fingerprint dictionary can be calculated, the maximum value is selected from the similarity, and the time series estimated value corresponding to the maximum value is determined as the time series estimated value matched with the time series observed value.
In practical application, the time series observed value D and each time series estimated value V in the magnetic resonance fingerprint dictionary can be calculated i Correlation C between i The operation formula can be
C i ＝D·V i ，i＝1，2，……，N，
Where i is the dictionary entry index and N is the number of dictionary entries. From { C i The maximum value C is selected max C is then max The corresponding time-series estimated value may be determined as a time-series estimated value that matches the time-series observed value D.
In this embodiment, by determining the similarity between the time-series estimated value and the time-series observed value in the magnetic resonance fingerprint dictionary and determining the time-series estimated value corresponding to the maximum value in the similarity as the time-series estimated value matched with the time-series observed value, the time-series estimated value matched with the time-series observed value can be quickly determined, and the efficiency of magnetic resonance data processing can be improved.
In one embodiment, after the step of determining the dictionary entry corresponding to the time-series estimation value as the target dictionary entry, the method specifically may further include: and determining the tissue parameters corresponding to the target dictionary entries as the tissue parameters of each voxel of the target part.
In a specific implementation, after determining the target dictionary entry according to the time sequence observation value of each voxel of the target part, the tissue parameters in the target dictionary entry may be determined as the tissue parameters of each voxel, so as to obtain the tissue parameters of each voxel of the target part.
For example, from the time-series observation value of the voxel d, the target dictionary entry P is determined, and the target dictionary entry P corresponds to the tissue parameter (T1, T2), and (T1, T2) may be determined as the tissue parameter of the voxel d.
The terminal may also process the magnetic resonance data of the voxels according to the tissue parameters. For example, the terminal may present tissue parameters for individual voxels on the magnetic resonance image.
In this embodiment, the tissue parameters corresponding to the target dictionary entries are determined as the tissue parameters of each voxel of the target portion, so that the tissue parameters of all voxels of the target portion can be determined at one time, and the efficiency of determining the tissue parameters of the voxels is improved.
In one embodiment, before step S110, the method specifically may further include: acquiring a magnetic resonance imaging sequence of a target part; the magnetic resonance imaging sequence is obtained by carrying out magnetic resonance imaging on the target part according to a preset stimulation mode and a preset sequence code.
In the specific implementation, the same stimulation mode and sequence code as those in the generation process of the magnetic resonance fingerprint dictionary can be used for carrying out magnetic resonance scanning on the target part to obtain a magnetic resonance imaging sequence of the target part.
For example, during a magnetic resonance fingerprint dictionary generation phase, a brain is excited using a radio frequency pulse sequence containing K time points, a BOLD stimulus of a fixed period is set, and (FA, TR) parameters for each time point, resulting in a magnetic resonance imaging sequence consisting of K magnetic resonance images. In the magnetic resonance data processing stage, the same fixed period BOLD stimulation can be set, and K magnetic resonance images of the brain part of the human body are acquired for the (FA, TR) parameters of K time points to form a magnetic resonance imaging sequence.
In this embodiment, the magnetic resonance imaging sequence is obtained by obtaining the magnetic resonance imaging of the target portion according to the preset stimulation mode and the preset sequence code, so that the obtained time sequence observation value can be compared with the time sequence estimation value, and further the hemodynamic response function of each voxel can be determined.
In one embodiment, the step S130 may specifically include: performing convolution operation on the magnetic resonance data of the voxels and the hemodynamic response function corresponding to the target dictionary entry to obtain hemodynamic response values corresponding to the voxels; fitting the hemodynamic response values to obtain fitting results of the hemodynamic response values; and determining the state of the voxel according to the slope corresponding to the fitting result.
Wherein the magnetic resonance data may be time series observations.
In specific implementation, a time sequence observed value corresponding to each voxel of the target part and a hemodynamic response function corresponding to a target dictionary entry can be subjected to convolution operation to obtain a hemodynamic response value of each voxel of the target part, the obtained hemodynamic response values are respectively input into a generalized linear model to obtain a fitting result of each hemodynamic response value, the state of the corresponding voxel can be determined to be an activated state under the condition that the slope of the fitting result is larger than a preset value, and otherwise, the state of the corresponding voxel can be determined to be an unactivated state under the condition that the slope of the fitting result is smaller than or equal to the preset value.
The generalized linear model can be an extension of a linear model, and a relation between a mathematical expected value of a response variable and a predicted variable of a linear combination is established through a connection function.
For example, in the data processing of functional magnetic resonance imaging based on BOLD signals, it is necessary to first convolve the data processing with a presumed hemodynamic response function according to a given stimulus pattern to obtain a signal level change due to the stimulus, then use the signal level change as an input for fitting a generalized linear model, find a voxel with a fitting slope significantly greater than 0, that is, an activated voxel, and a voxel with a fitting slope not significantly greater than 0, and deactivate the voxel.
In this embodiment, the hemodynamic response value corresponding to the voxel is obtained by performing convolution operation on the magnetic resonance data of the voxel and the hemodynamic response function corresponding to the target dictionary entry, the hemodynamic response value is fitted to obtain a fitting result of the hemodynamic response value, and the state of the voxel is determined according to the slope corresponding to the fitting result, so that the hemodynamic response value of the voxel can be estimated by using different hemodynamic response functions corresponding to different voxels, the accuracy of the hemodynamic response value estimation is increased, and the activation state of the voxel is determined according to the hemodynamic response value, so that the accuracy of the activation state identification can be improved.
In one embodiment, as shown in fig. 2, a magnetic resonance data processing method is provided, and the method is applied to a terminal for illustration, and includes the following steps:
Step S210, acquiring magnetic resonance fingerprint parameters; the magnetic resonance fingerprint parameters comprise hemodynamic response functions and tissue parameters;
step S220, performing simulation processing on the magnetic resonance fingerprint parameters according to a preset stimulation mode and a preset sequence code to obtain a time sequence estimated value corresponding to the magnetic resonance fingerprint parameters;
step S230, dictionary entries in a magnetic resonance fingerprint dictionary are generated according to the corresponding relation among the hemodynamic response function, the tissue parameters and the time sequence estimation value;
step S240, acquiring a magnetic resonance imaging sequence of a target part; the magnetic resonance imaging sequence is obtained by carrying out magnetic resonance imaging on the target part according to a preset stimulation mode and a preset sequence code;
step S250, determining a time sequence observation value corresponding to each voxel of the target part according to the magnetic resonance imaging sequence of the target part;
step S260, finding out the time sequence estimated value matched with each time sequence observed value from the magnetic resonance fingerprint dictionary;
step S270, determining dictionary entries corresponding to the time sequence estimation values as target dictionary entries;
step S281, processing the magnetic resonance data of each voxel of the target part according to the hemodynamic response function corresponding to each target dictionary entry to obtain the magnetic resonance data processing result of the target part;
In step S282, the tissue parameter corresponding to each target dictionary entry is determined as the tissue parameter of each voxel of the target region.
In this embodiment, by acquiring magnetic resonance fingerprint parameters, performing simulation processing on the magnetic resonance fingerprint parameters according to a preset stimulus mode and a preset sequence code, obtaining a time sequence estimation value corresponding to the magnetic resonance fingerprint parameters, generating dictionary entries in a magnetic resonance fingerprint dictionary according to a corresponding relation among a hemodynamic response function, tissue parameters and the time sequence estimation value, acquiring a magnetic resonance imaging sequence of a target part, determining a time sequence observation value corresponding to each voxel of the target part according to the magnetic resonance imaging sequence of the target part, finding out a time sequence estimation value matched with each time sequence observation value from the magnetic resonance fingerprint dictionary, determining the dictionary entries corresponding to each time sequence estimation value as target dictionary entries, processing magnetic resonance data of each voxel of the target part according to a hemodynamic response function corresponding to each target dictionary entry, obtaining a magnetic resonance data processing result of the target part, and determining the tissue parameters corresponding to each target dictionary entry as the tissue parameters of each voxel of the target part, different voxels can correspond to different hemodynamic response functions, thus the hemodynamic response function can be processed more accurately by the specific hemodynamic response function, and the hemodynamic response function can be processed more accurately.
In order to facilitate a thorough understanding of embodiments of the present application by those skilled in the art, the following description will be provided in connection with a specific example.
MRF is taken as a powerful quantitative imaging technology, is getting attention and getting good effect in the imaging of the T1, T2 and B0, B1 fields, and also shows better application prospect in the applications such as perfusion, diffusion, chemical saturation displacement imaging and the like. However, there is not enough attention paid to fMRI treatment. Based on MRF, the application provides a method for simultaneously estimating parameters and an activation area of HRF.
Fig. 3 is a flow chart of a functional magnetic resonance quantitative processing method based on magnetic resonance fingerprint. According to fig. 3, a functional magnetic resonance quantitative processing method based on magnetic resonance fingerprinting may comprise a dictionary generating phase, a scanning phase and a dictionary matching phase, wherein,
dictionary generation: a dictionary containing different tissue parameters and HRF parameters is obtained by simulation based on the possible HRF parameter combinations and given HRF expressions, given stimulation patterns, given flip angles and TR (repetition time) per firing, and tissue parameters. The HRF expression may take different forms such as a dual gamma function, an exponential function, and the like.
Scanning: performing BOLD scanning in a stimulation mode, a flip angle and TR consistent with the dictionary generation stage to obtain a time sequence of all voxels;
dictionary matching stage: and comparing the time sequence of each voxel with entries in a dictionary, finding out the most conforming entry, assigning the corresponding tissue parameters and HRF parameters to the voxels, and repeating the process for all the voxels to obtain the tissue parameters and the HRF parameters of all the voxels.
In this embodiment, a set of different HRFs is generated according to different parameters, then collected signal responses corresponding to different HRFs are calculated according to given stimulation modes and MRF sequence codes (changing flip angles and TRs), then a time sequence of the whole brain is obtained by scanning the time sequence of each voxel through preset stimulation modes and MRF sequence codes, finally the corresponding HRF parameters are found by matching the time sequence of each voxel with a pre-generated dictionary, and whether the voxel is an activated voxel is judged according to the HRF parameters. By applying the MRF quantitative method to the processing of BOLD data, the advantage of rapid quantification is fully exerted, and the parameters of T1, T2 and T2 of the tissue and the parameters of the HRF are estimated together, so that the accurate form of the HRF of each voxel can be obtained by one-time scanning, whether the voxel is activated or not can be judged according to the accurate form, and the T1 and T2 diagram of the tissue can be obtained.
It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.
Based on the same inventive concept, the embodiment of the application also provides a magnetic resonance data processing device for realizing the magnetic resonance data processing method. The implementation of the solution provided by the apparatus is similar to the implementation described in the above method, so the specific limitations in the embodiments of the magnetic resonance data processing apparatus provided below may be referred to the limitations of the magnetic resonance data processing method hereinabove, and will not be repeated here.
In one embodiment, as shown in fig. 4, there is provided a magnetic resonance data processing apparatus comprising: a sequence determination module 310, a dictionary lookup module 320, and a data processing module 330, wherein:
a sequence determining module 310, configured to determine a time-series observed value corresponding to each voxel of a target location according to a magnetic resonance imaging sequence of the target location;
the dictionary lookup module 320 is configured to find target dictionary entries that match each of the time-series observations from a preset magnetic resonance fingerprint dictionary; dictionary entries in the magnetic resonance fingerprint dictionary are used for recording the corresponding relation between the time sequence estimated value and the hemodynamic response function;
and the data processing module 330 is configured to process the magnetic resonance data of each voxel of the target location according to the hemodynamic response function corresponding to each target dictionary entry, so as to obtain a magnetic resonance data processing result of the target location.
In one embodiment, the magnetic resonance data processing apparatus further includes:
the parameter acquisition module is used for acquiring magnetic resonance fingerprint parameters; the magnetic resonance fingerprint parameters comprise hemodynamic response functions and tissue parameters;
The fingerprint simulation module is used for performing simulation processing on the magnetic resonance fingerprint parameters according to a preset stimulation mode and a preset sequence code to obtain a time sequence estimated value corresponding to the magnetic resonance fingerprint parameters;
and the dictionary generating module is used for generating dictionary entries in the magnetic resonance fingerprint dictionary according to the corresponding relation among the hemodynamic response function, the tissue parameter and the time sequence estimated value.
In one embodiment, the parameter obtaining module is further configured to obtain at least one of the hemodynamic response function and at least one of the tissue parameters; combining the at least one hemodynamic response function and the at least one tissue parameter to obtain at least one of the magnetic resonance fingerprint parameters.
In one embodiment, the dictionary lookup module 320 includes:
the sequence estimation value searching module is used for searching a time sequence estimation value matched with the time sequence observation value from the magnetic resonance fingerprint dictionary;
and the dictionary entry determining module is used for determining the dictionary entry corresponding to the time sequence estimated value as the target dictionary entry.
In one embodiment, the above sequence estimation value searching module is further configured to determine a similarity between a time sequence estimation value and the time sequence observation value in the magnetic resonance fingerprint dictionary; and determining a time sequence estimated value corresponding to the maximum value in the similarity as a time sequence estimated value matched with the time sequence observed value.
In one embodiment, the magnetic resonance data processing apparatus further includes:
and the tissue parameter determining module is used for determining the tissue parameters corresponding to the target dictionary entries as the tissue parameters of each voxel of the target part.
In one embodiment, the magnetic resonance data processing apparatus further includes:
the magnetic resonance imaging sequence acquisition module is used for acquiring a magnetic resonance imaging sequence of the target part; the magnetic resonance imaging sequence is obtained by carrying out magnetic resonance imaging on the target part according to the preset stimulation mode and the preset sequence code.
In one embodiment, the data processing module 330 is further configured to perform a convolution operation on the magnetic resonance data of the voxel and a hemodynamic response function corresponding to the target dictionary entry, so as to obtain a hemodynamic response value corresponding to the voxel; fitting the hemodynamic response values to obtain fitting results of the hemodynamic response values; and determining the state of the voxel according to the slope corresponding to the fitting result.
The modules in the magnetic resonance data processing apparatus described above may be implemented in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a terminal, and the internal structure of which may be as shown in fig. 5. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a magnetic resonance data processing method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the structure shown in FIG. 5 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In an embodiment, there is also provided a computer device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of the method embodiments described above when the computer program is executed.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when executed by a processor, carries out the steps of the method embodiments described above.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, or the like, but is not limited thereto.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.
Claims (10)
1. A method of magnetic resonance data processing, the method comprising:
acquiring magnetic resonance fingerprint parameters; the magnetic resonance fingerprint parameters comprise hemodynamic response functions and tissue parameters;
according to a preset stimulation mode and a preset sequence code, performing simulation processing on the magnetic resonance fingerprint parameters to obtain a time sequence estimated value corresponding to the magnetic resonance fingerprint parameters;
Generating dictionary entries in a magnetic resonance fingerprint dictionary according to the corresponding relation among the hemodynamic response function, the tissue parameter and the time sequence estimation value;
determining a time sequence observation value corresponding to each voxel of a target part according to a magnetic resonance imaging sequence of the target part;
searching target dictionary entries matched with each time series observation value from the magnetic resonance fingerprint dictionary; dictionary entries in the magnetic resonance fingerprint dictionary are used for recording the corresponding relation between the time sequence estimated value and the hemodynamic response function;
and processing the magnetic resonance data of each voxel of the target part according to the hemodynamic response function corresponding to each target dictionary entry to obtain a magnetic resonance data processing result of the target part.
2. The method of claim 1, wherein the acquiring magnetic resonance fingerprint parameters comprises:
acquiring at least one of said hemodynamic response functions and at least one of said tissue parameters;
combining the at least one hemodynamic response function and the at least one tissue parameter to obtain at least one of the magnetic resonance fingerprint parameters.
3. The method of claim 1, wherein finding target dictionary entries from the magnetic resonance fingerprint dictionary that match each of the time series observations comprises:
finding out a time sequence estimated value matched with the time sequence observed value from the magnetic resonance fingerprint dictionary;
and determining dictionary entries corresponding to the time sequence estimation values as target dictionary entries.
4. A method according to claim 3, wherein said finding a time series estimate from said magnetic resonance fingerprint dictionary that matches said time series observation comprises:
determining a similarity between a time series estimated value and the time series observed value in the magnetic resonance fingerprint dictionary;
and determining a time sequence estimated value corresponding to the maximum value in the similarity as a time sequence estimated value matched with the time sequence observed value.
5. A method according to claim 3, further comprising, after determining the dictionary entry corresponding to the time series estimate as the target dictionary entry:
and determining the tissue parameters corresponding to the target dictionary entries as the tissue parameters of each voxel of the target part.
6. The method of claim 1, further comprising, prior to determining a time series observation for each voxel of the target site from a magnetic resonance imaging sequence of the target site:
acquiring a magnetic resonance imaging sequence of the target part; the magnetic resonance imaging sequence is obtained by carrying out magnetic resonance imaging on the target part according to the preset stimulation mode and the preset sequence code.
7. The method according to claim 1, wherein the processing the magnetic resonance data of each voxel of the target region according to the hemodynamic response function corresponding to each target dictionary entry to obtain a magnetic resonance data processing result of the target region includes:
performing convolution operation on the magnetic resonance data of the voxels and the hemodynamic response function corresponding to the target dictionary entry to obtain hemodynamic response values corresponding to the voxels;
fitting the hemodynamic response values to obtain fitting results of the hemodynamic response values;
and determining the state of the voxel according to the slope corresponding to the fitting result.
8. A magnetic resonance data processing apparatus, the apparatus comprising:
The parameter acquisition module is used for acquiring magnetic resonance fingerprint parameters; the magnetic resonance fingerprint parameters comprise hemodynamic response functions and tissue parameters;
the fingerprint simulation module is used for performing simulation processing on the magnetic resonance fingerprint parameters according to a preset stimulation mode and a preset sequence code to obtain a time sequence estimated value corresponding to the magnetic resonance fingerprint parameters;
the dictionary generating module is used for generating dictionary entries in a magnetic resonance fingerprint dictionary according to the corresponding relation among the hemodynamic response function, the tissue parameters and the time sequence estimated value;
the sequence determining module is used for determining a time sequence observation value corresponding to each voxel of the target part according to the magnetic resonance imaging sequence of the target part;
the dictionary searching module is used for searching target dictionary entries matched with each time sequence observation value from the magnetic resonance fingerprint dictionary; dictionary entries in the magnetic resonance fingerprint dictionary are used for recording the corresponding relation between the time sequence estimated value and the hemodynamic response function;
and the data processing module is used for processing the magnetic resonance data of each voxel of the target part according to the hemodynamic response function corresponding to each target dictionary entry to obtain a magnetic resonance data processing result of the target part.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211162346.2A CN115561690B (en) | 2022-09-23 | 2022-09-23 | Magnetic resonance data processing method and device and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211162346.2A CN115561690B (en) | 2022-09-23 | 2022-09-23 | Magnetic resonance data processing method and device and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115561690A CN115561690A (en) | 2023-01-03 |
| CN115561690B true CN115561690B (en) | 2023-09-26 |
Family
ID=84740706
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211162346.2A Active CN115561690B (en) | 2022-09-23 | 2022-09-23 | Magnetic resonance data processing method and device and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115561690B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107194354A (en) * | 2017-05-23 | 2017-09-22 | 杭州师范大学 | A kind of quick dictionary search method for magnetic resonance fingerprint imaging |
| CN110133554A (en) * | 2018-02-08 | 2019-08-16 | 深圳先进技术研究院 | A kind of magnetic resonance fingerprint imaging method, apparatus and medium based on fractional model |
| CN110141237A (en) * | 2019-05-21 | 2019-08-20 | 上海海事大学 | A kind of blind deconvolution method of the functional MRI of data-driven |
| CN112617796A (en) * | 2020-12-30 | 2021-04-09 | 深圳市联影高端医疗装备创新研究院 | Magnetic resonance functional imaging method, device, equipment, storage medium and system |
| CN114972563A (en) * | 2022-05-24 | 2022-08-30 | 深圳市联影高端医疗装备创新研究院 | Organization parameter determination method, organization parameter determination device, computer equipment and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3382415A1 (en) * | 2017-03-30 | 2018-10-03 | Koninklijke Philips N.V. | Sub voxel resolution magnetic resonance fingerprinting imaging |
-
2022
- 2022-09-23 CN CN202211162346.2A patent/CN115561690B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107194354A (en) * | 2017-05-23 | 2017-09-22 | 杭州师范大学 | A kind of quick dictionary search method for magnetic resonance fingerprint imaging |
| CN110133554A (en) * | 2018-02-08 | 2019-08-16 | 深圳先进技术研究院 | A kind of magnetic resonance fingerprint imaging method, apparatus and medium based on fractional model |
| CN110141237A (en) * | 2019-05-21 | 2019-08-20 | 上海海事大学 | A kind of blind deconvolution method of the functional MRI of data-driven |
| CN112617796A (en) * | 2020-12-30 | 2021-04-09 | 深圳市联影高端医疗装备创新研究院 | Magnetic resonance functional imaging method, device, equipment, storage medium and system |
| CN114972563A (en) * | 2022-05-24 | 2022-08-30 | 深圳市联影高端医疗装备创新研究院 | Organization parameter determination method, organization parameter determination device, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115561690A (en) | 2023-01-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10698055B2 (en) | Method, neural network, and magnetic resonance apparatus for assigning magnetic resonance fingerprints | |
| US10627470B2 (en) | System and method for learning based magnetic resonance fingerprinting | |
| CN111798273A (en) | Training method of purchase probability prediction model of product and purchase probability prediction method | |
| CN109816630B (en) | fMRI visual coding model construction method based on transfer learning | |
| CN111383741B (en) | Method, device and equipment for establishing medical imaging model and storage medium | |
| US11961004B2 (en) | Predicting brain data using machine learning models | |
| CN112001856A (en) | Training method of denoising model, image noise removing method and related device | |
| CN111737922A (en) | Data processing method, device, equipment and medium based on recurrent neural network | |
| Plis et al. | High-order interactions observed in multi-task intrinsic networks are dominant indicators of aberrant brain function in schizophrenia | |
| CN107638179A (en) | Computational methods, system and the computer-readable recording medium of magnetic resonance flip angle | |
| CN114972563A (en) | Organization parameter determination method, organization parameter determination device, computer equipment and storage medium | |
| CN115561690B (en) | Magnetic resonance data processing method and device and computer equipment | |
| Silva et al. | A statistically motivated framework for simulation of stochastic data fusion models applied to multimodal neuroimaging | |
| CN108364067B (en) | Deep learning method based on data segmentation and robot system | |
| WO2019229472A1 (en) | Optimisation system and method | |
| Khalid et al. | A novel subject-wise dictionary learning approach using multi-subject fMRI spatial and temporal components | |
| Noè | Bayesian nonparametric inference in mechanistic models of complex biological systems | |
| Yao et al. | A Path Towards Clinical Adaptation of Accelerated MRI | |
| CN110648346A (en) | Tumor network construction method and device, computer equipment and storage medium | |
| CN115561691B (en) | Radio frequency emission field correction method, device, computer equipment and storage medium | |
| CN116256681A (en) | Magnetic resonance imaging method, apparatus, device and storage medium | |
| US20230200653A1 (en) | Diffusion kurtosis imaging method, computer device and storage medium | |
| CN116942133A (en) | Parameter determination method and device and medical equipment | |
| CN117519424A (en) | Signal down-conversion method, device, computer equipment and storage medium thereof | |
| CN117017277A (en) | Identity recognition method, device, equipment and medium based on photoelectric volume pulse wave |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |