Fetal spinal cord neural tube malformation prenatal screening method based on MRI (Magnetic Resonance Imaging)

Abstract

The invention discloses a fetal spinal cord neural tube malformation prenatal screening method based on MRI, and relates to the technical field of medical detection. The fetal spinal cord neural tube malformation antenatal screening method based on MRI comprises the following steps: obtaining MRI image data; obtaining an artifact identification correlation coefficient; acquiring a signal quality evaluation value; and obtaining a coding quality evaluation value. According to the invention, the artifact identification correlation coefficient is obtained by combining the artifact identification data and the reference artifact identification data, and then the signal quality evaluation value is obtained according to the signal correlation data and the reference signal correlation data so as to judge whether to carry out signal quality adjustment. And finally, according to the coding direction related data and the reference coding direction data, obtaining a coding quality evaluation value to judge whether to carry out coding quality adjustment and prenatal screening of the fetal spinal cord neural tube deformity, thereby achieving the effect of improving the accuracy of the prenatal screening of the fetal spinal cord neural tube deformity. The problem that prenatal screening accuracy of fetal spinal cord neural tube malformation is not high in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the field of medical detection technology, and in particular to a prenatal screening method for fetal spinal neural tube defects based on MRI. Background Art

[0002] Congenital neural tube defects (NTDs), also known as congenital neural tube defects, are a type of serious congenital malformation caused by incomplete closure of the neural tube due to various factors in the early stages of embryonic development. The neural tube is the central nervous system of the fetus, including cranial neural tube defects and spinal neural tube defects. Among them, spinal neural tube defects include open spina bifida and closed spina bifida. Open spina bifida includes meningocele, myelomeningocele, and exposed spinal cord; closed spina bifida includes simple tethered cord, lipoma-type tethered cord, lipoma myelomeningocele, diastematomyelia, intraspinal cysts and other diseases.

[0003] Research on MRI (Magnetic Resonance Imaging) began in 1983. Initially, due to technical limitations, the image acquisition time of MRI was long, resulting in few indications for MRI. In recent years, MRI scanners have improved scanning time and signal-to-noise ratio, significantly shortening scanning time and reducing the impact of artifacts caused by fetal movement. This has led to more and more widespread research and use of MRI.

[0004] Existing methods are based on comprehensive assessment by detecting the concentration of specific substances in maternal serum (such as human chorionic gonadotropin, free estriol, and alpha-fetoprotein) combined with factors such as gestational weight, expected date of delivery, and age.

[0005] However, in the process of implementing the technical solution of the present invention, the present invention finds that the above technology has at least the following technical problems:

[0006] In the prior art, the movement of the fetus in the uterus may cause the quality of image detail information to deteriorate, resulting in a problem of low accuracy in prenatal screening for fetal spinal neural tube defects. Summary of the invention

[0007] The present invention provides a method for prenatal screening of fetal spinal neural tube malformations based on MRI, which obtains MRI image data, then obtains an artifact recognition correlation coefficient and an artifact image according to artifact recognition data and reference artifact recognition data, then obtains a signal quality evaluation value according to signal-related data corresponding to the artifact image and reference signal-related data to determine whether to perform signal quality adjustment, and finally obtains a coding quality evaluation value according to coding direction-related data corresponding to the artifact image and reference coding direction data to determine whether to perform coding quality adjustment and prenatal screening of fetal spinal neural tube malformations, thereby achieving evaluation and adjustment of image signals and coding quality, and further achieving improvement in the accuracy of prenatal screening of fetal spinal neural tube malformations.

[0008] To solve the above-mentioned invention object, the technical solution provided by the present invention is as follows: a prenatal screening method for fetal spinal neural tube defects based on MRI, comprising the following steps: S1, obtaining corresponding MRI image data according to the acquired MRI image, wherein the MRI image is used to display fetal information at different angles and different levels, and the MRI image data includes artifact recognition data, signal-related data, and encoding direction-related data; S2, obtaining an artifact recognition correlation coefficient according to the artifact recognition data and reference artifact recognition data obtained from a preset database, and classifying the MRI image based on the artifact recognition correlation coefficient and a preset artifact recognition threshold obtained from the preset database to obtain an artifact image and a non-artifact image, wherein the artifact recognition correlation coefficient is used to evaluate the generation of motion artifacts. The possibility of an artifact; S3, obtaining a signal quality evaluation value according to signal-related data corresponding to the artifact image and reference signal-related data obtained from a preset database, judging whether to perform signal quality adjustment based on the signal quality evaluation value and a preset signal quality threshold obtained from the preset database, wherein the signal quality evaluation value is used to evaluate the signal quality corresponding to the artifact image; S4, obtaining a coding quality evaluation value according to coding direction-related data corresponding to the artifact image and reference coding direction data obtained from a preset database, judging whether to perform coding quality adjustment and prenatal screening for fetal spinal neural tube defects based on the coding quality evaluation value and a preset coding quality threshold obtained from the preset database, wherein the coding quality evaluation value is used to evaluate the quality of the coding method corresponding to the artifact image.

[0009] Optionally, the artifact identification data includes echo frequency, magnetic field strength, acquisition time, imaging field of view and number of sampling points in the horizontal sampling direction; the signal-related data includes MRI signal strength, MRI noise strength and number of MRI signal acquisition repetitions; the encoding direction-related data includes MRI bandwidth, MRI frequency resolution and MRI phase resolution.

[0010] Optionally, the specific process of obtaining the artifact identification correlation coefficient based on the artifact identification data and the reference artifact identification data obtained from the preset database is as follows: the reference artifact identification data includes a preset phase change threshold and a preset spatial resolution threshold; the phase change influence factor is obtained in combination with the reference artifact identification data, echo frequency, magnetic field strength and acquisition time, and the phase change influence factor is used to reflect the degree of influence of the phase change in the MRI image on the artifact identification correlation coefficient; the spatial resolution influence factor is obtained in combination with the preset spatial resolution threshold, imaging field of view and number of sampling points in the horizontal sampling direction, and the spatial resolution influence factor is used to reflect the degree of influence of the spatial resolution in the MRI image on the artifact identification correlation coefficient; the artifact identification correlation coefficient is obtained in combination with the obtained phase change influence factor and spatial resolution influence factor.

[0011] Optionally, the restriction expression of the artifact identification correlation coefficient is as follows:

[0012]

[0013] In the formula, represents the artifact recognition correlation coefficient of the mth MRI image, m=1,2,...,n, m represents the number of the MRI image, n represents the total number of MRI images, xw m represents the phase change influencing factor of the mth MRI image, KJ m represents the spatial resolution influencing factor of the mth MRI image, γ m Echo frequency, CB m represents the magnetic field strength of the mth MRI image, t m represents the acquisition time of the mth MRI image, FOV m represents the imaging field of view of the mth MRI image, SN m represents the number of sampling points in the horizontal sampling direction of the mth MRI image, R0 represents the preset phase change threshold, J0 represents the preset spatial resolution threshold, and e represents a natural constant.

[0014] Optionally, the specific process of performing image classification on MRI images based on the artifact recognition correlation coefficient and a preset artifact recognition threshold obtained from a preset database to obtain artifact images and non-artifact images is as follows: judging whether the artifact recognition correlation coefficient is higher than the preset artifact recognition threshold; when the artifact recognition correlation coefficient is higher than the preset artifact recognition threshold, classifying the corresponding MRI image as an artifact image, and numbering the artifact image; otherwise, classifying the corresponding MRI image as a non-artifact image, and numbering the non-artifact image.

[0015] Optionally, the specific process of obtaining the signal quality evaluation value based on the signal-related data corresponding to the artifact image and the reference signal-related data obtained from a preset database is as follows: the reference signal-related data includes a preset MRI signal-to-noise ratio threshold and a preset average signal quantity threshold; the MRI signal-to-noise ratio influence factor is obtained by combining the preset MRI signal-to-noise ratio threshold, MRI signal intensity and MRI noise intensity; the MRI average signal quantity influence factor is obtained by combining the preset average signal quantity threshold, MRI signal intensity, MRI noise intensity and the number of MRI signal acquisition repetitions; the signal quality evaluation value is obtained by combining the artifact recognition correlation coefficient corresponding to the acquired artifact image, the MRI signal-to-noise ratio influence factor and the MRI average signal quantity influence factor; the specific process of obtaining the MRI signal-to-noise ratio influence factor The whole process is as follows: W1, performing a ratio operation on the MRI signal intensity and the MRI noise intensity of the artifact image; W2, performing a sum operation on the preset MRI signal-to-noise ratio threshold and the result of the ratio operation in W1; W3, performing a ratio operation on the result of the sum operation in W2 and the preset MRI signal-to-noise ratio threshold to obtain the MRI signal-to-noise ratio influence factor; the specific process of obtaining the MRI average signal quantity influence factor is as follows: A1, performing a product operation on the MRI signal intensity, the MRI noise intensity and the number of MRI signal acquisition repetitions of the artifact image; A2, performing a sum operation on the preset average signal quantity threshold and the result of the product operation in A1; A3, performing a ratio operation on the result of the sum operation in A2 and the preset average signal quantity threshold to obtain the MRI average signal quantity influence factor.

[0016] Optionally, the specific process of determining whether to perform signal quality adjustment based on the signal quality evaluation value and the preset signal quality threshold obtained from the preset database is as follows: SS1, determining whether the signal quality evaluation value is not lower than the preset signal quality threshold. When the signal quality evaluation value is not lower than the preset signal quality threshold, no signal quality adjustment is performed, otherwise SS2 is executed; SS2, performing median filtering processing, stopping execution when the signal quality evaluation value corresponding to the monitored artifact image is not lower than the preset signal quality threshold and the artifact recognition correlation coefficient is not higher than the preset artifact recognition threshold, otherwise SS3 is executed; SS3; performing wavelet transform processing, stopping execution when the signal quality evaluation value corresponding to the monitored artifact image is not lower than the preset signal quality threshold and the artifact recognition correlation coefficient is not higher than the preset artifact recognition threshold, otherwise feedback is given to the preset personnel to obtain the coding quality evaluation value.

[0017] Optionally, the specific process of obtaining the coding quality evaluation value according to the coding direction related data corresponding to the artifact image and the reference coding direction data obtained from a preset database is as follows: the reference coding direction data includes a preset frequency coding direction field of view threshold and a preset phase coding direction field of view threshold; the frequency coding influence factor is obtained by combining the preset frequency coding direction field of view threshold, MRI bandwidth and MRI frequency resolution; the phase coding influence factor is obtained by combining the preset phase coding direction field of view threshold, MRI bandwidth and MRI phase resolution; the coding quality evaluation value is obtained by combining the artifact recognition correlation coefficient, frequency coding influence factor and phase coding influence factor corresponding to the acquired artifact image; the specific process of obtaining the frequency coding influence factor The process is as follows: Q1, performing a ratio operation on the MRI bandwidth and MRI frequency resolution of the artifact image; Q2, summing the result of the ratio operation in Q1 and the preset frequency encoding direction field of view threshold; Q3, performing a ratio operation on the result of the summation operation in Q2 and the preset frequency encoding direction field of view threshold to obtain the frequency encoding influence factor; the specific process of obtaining the phase encoding influence factor is as follows: E1, performing a ratio operation on the MRI bandwidth and MRI phase resolution of the artifact image; E2, summing the result of the ratio operation in E1 and the preset phase encoding direction field of view threshold; E3, performing a ratio operation on the result of the summation operation in E2 and the preset phase encoding direction field of view threshold to obtain the phase encoding influence factor.

[0018] Optionally, the specific process of judging whether to adjust the coding quality based on the coding quality evaluation value and the preset coding quality threshold obtained from the preset database is as follows: WW1, judging whether the coding quality evaluation value is not lower than the preset coding quality threshold, and not adjusting the coding quality when the coding quality evaluation value is not lower than the preset coding quality threshold, otherwise executing WW2; WW2, performing frequency coding direction resolution improvement processing, stopping execution when the coding quality evaluation value corresponding to the monitored artifact image is not lower than the preset coding quality threshold and the artifact recognition correlation coefficient is not higher than the preset artifact recognition threshold, otherwise executing WW3, the frequency coding direction resolution improvement processing means improving the resolution in the frequency coding direction through a non-Cartesian sampling trajectory; WW3, performing phase coding direction resolution improvement processing, stopping execution when the coding quality evaluation value corresponding to the monitored artifact image is not lower than the preset coding quality threshold and the artifact recognition correlation coefficient is not higher than the preset artifact recognition threshold, otherwise feeding back to the preset personnel to re-acquire the MRI image, the phase coding direction resolution improvement processing means increasing the number of phase coding steps through an oversampling method.

[0019] Optionally, the specific process of prenatal screening for fetal spinal neural tube defects is as follows: determine whether the coding quality evaluation value is not lower than a preset coding quality threshold; when the coding quality evaluation value is not lower than the preset coding quality threshold, obtain an artifact image corresponding to when the coding quality evaluation value is not lower than the preset coding quality threshold; combine the acquired artifact image corresponding to when the coding quality evaluation value is not lower than the preset coding quality threshold and the non-artifact image to perform image fusion; and feed back the acquired fused image to the preset personnel for prenatal screening for fetal spinal neural tube defects.

[0020] One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages:

[0021] 1. The artifact recognition correlation coefficient and the artifact image are obtained by acquiring the artifact recognition data and the reference artifact recognition data, and then the signal quality evaluation value is obtained according to the signal-related data corresponding to the artifact image and the reference signal-related data to determine whether to adjust the signal quality, and finally the coding quality evaluation value is obtained according to the coding direction-related data corresponding to the artifact image and the reference coding direction data to determine whether to adjust the coding quality and perform prenatal screening for fetal spinal neural tube defects, thereby achieving accurate adjustment of the image signal and coding quality, and further achieving improved accuracy in prenatal screening for fetal spinal neural tube defects, and solving the problem of low accuracy in prenatal screening for fetal spinal neural tube defects.

[0022] 2. The corresponding MRI images are classified into artifact images and non-artifact images through the artifact recognition correlation coefficient and the preset artifact recognition threshold, thereby realizing the distinction between artifact images and non-artifact images, and further providing a guarantee for improving the quality of artifact images.

[0023] 3. The MRI signal-to-noise ratio influence factor is obtained by combining the preset MRI signal-to-noise ratio threshold, MRI signal intensity and MRI noise intensity, and then the MRI average signal quantity influence factor is obtained by combining the preset average signal quantity threshold, MRI signal intensity, MRI noise intensity and the number of MRI signal acquisition repetitions. Finally, the signal quality assessment value is obtained by combining the artifact recognition correlation coefficient corresponding to the acquired artifact image, the MRI signal-to-noise ratio influence factor and the MRI average signal quantity influence factor, thereby realizing the quantification of image signal quality and then achieving the improvement of image quality reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 A flow chart of a method for prenatal screening of fetal spinal neural tube defects based on MRI provided in an embodiment of the present invention;

[0026] Figure 2 A statistical diagram of changes in artifact recognition correlation coefficients provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] Unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one", "one" or "the" do not indicate a quantitative limitation, but indicate the existence of at least one. Words such as "include" or "comprise" mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Words such as "connect" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.

[0029] It should be noted that the terms "up", "down", "left", "right", "front", "back", etc. used in the present invention are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0030] The present invention provides a method for prenatal screening of fetal spinal neural tube malformations based on MRI, which obtains corresponding MRI image data through an acquired MRI image, then obtains an artifact recognition correlation coefficient and an artifact image according to artifact recognition data and reference artifact recognition data, then obtains a signal quality assessment value and performs signal quality adjustment according to signal-related data corresponding to the artifact image and reference signal-related data, and finally obtains a coding quality assessment value and performs coding quality adjustment and prenatal screening of fetal spinal neural tube malformations according to coding direction-related data corresponding to the artifact image and reference coding direction data, thereby achieving evaluation and adjustment of image signals and coding quality, thereby achieving improvement in the accuracy of prenatal screening of fetal spinal neural tube malformations, and solving the problem of low accuracy of prenatal screening of fetal spinal neural tube malformations.

[0031] like Figure 1 As shown, it is a flow chart of a method for prenatal screening of fetal spinal neural tube defects based on MRI provided by an embodiment of the present invention. The method for prenatal screening of fetal spinal neural tube defects based on MRI includes the following steps: S1, acquiring corresponding MRI image data according to the acquired MRI image, the MRI image is used to display fetal information at different angles and different levels, the MRI image data includes artifact recognition data, signal-related data and encoding direction-related data, the artifact recognition data is used to describe the artifact information of the MRI image, the signal-related data is used to describe the signal information of the MRI image, and the encoding direction-related data is used to describe the encoding information of the MRI image; S2, acquiring an artifact recognition correlation coefficient according to the artifact recognition data and reference artifact recognition data obtained from a preset database, performing image classification on the MRI image based on the artifact recognition correlation coefficient and a preset artifact recognition threshold obtained from the preset database to obtain an artifact image and a non-artifact image, and the artifact recognition correlation coefficient Used to evaluate the possibility of generating motion artifacts; S3, obtaining a signal quality evaluation value according to the signal-related data corresponding to the artifact image and the reference signal-related data obtained from the preset database, judging whether to perform signal quality adjustment based on the signal quality evaluation value and the preset signal quality threshold obtained from the preset database, the signal quality adjustment is used to adjust the signal quality evaluation value to no less than the preset signal quality threshold, and the signal quality evaluation value is used to evaluate the signal quality corresponding to the artifact image; S4, obtaining a coding quality evaluation value according to the coding direction-related data corresponding to the artifact image and the reference coding direction data obtained from the preset database, judging whether to perform coding quality adjustment and prenatal screening for fetal spinal neural tube defects based on the coding quality evaluation value and the preset coding quality threshold obtained from the preset database, the coding quality adjustment is used to adjust the coding quality evaluation value to no less than the preset coding quality threshold, and the coding quality evaluation value is used to evaluate the quality of the coding method corresponding to the artifact image.

[0032] In this embodiment, the artifact recognition correlation coefficient is an indicator for evaluating the possibility of motion artifacts, and the preset artifact recognition threshold is a standard for classifying whether an image has artifacts. The preset artifact recognition threshold is represented by the standard deviation of the resolution data in the MRI (Magnetic Resonance Imaging) image of the historical period in the preset database, the preset signal quality threshold is represented by the standard deviation of the signal intensity data in the MRI image of the historical period in the preset database, and the preset coding quality threshold is represented by the standard deviation of the bandwidth data in the MRI image of the historical period in the preset database. The signal quality evaluation value and the coding quality evaluation value can affect each other. For example, when performing high-resolution imaging, it is necessary to balance the signal strength and the coding direction to avoid blurring artifacts. If the artifact recognition correlation coefficient increases, it indicates that the possibility of artifacts in the image increases, which may interfere with the stability and consistency of the signal, thereby reducing the signal quality evaluation value. If the signal quality evaluation value decreases, it may affect the coding process, resulting in a decrease in the coding quality evaluation value; prenatal ultrasound examination and fetal MRI provide clinicians with effective tools for detecting spinal neural tube malformations in early pregnancy. Although prenatal ultrasound examination is still the preferred method for fetal screening, fetal MRI imaging has high value and applicability for the prenatal diagnosis and prognosis assessment of most spinal neural tube defects, and is an important supplementary method to prenatal ultrasound; it has improved the accuracy of prenatal screening for fetal spinal neural tube defects.

[0033] Among them, artifact identification data include echo frequency, magnetic field strength, acquisition time, imaging field of view and number of sampling points in the horizontal sampling direction; signal-related data include MRI signal intensity, MRI noise intensity and number of MRI signal acquisition repetitions; encoding direction-related data include MRI bandwidth, MRI frequency resolution and MRI phase resolution.

[0034] It should be understood that image extraction is performed on MRI images to obtain MRI image data. Image extraction means extracting data from MRI images through SPM (Statistical Parametric Mapping). Artifact recognition data, signal-related data, and encoding direction-related data interact and influence each other. Artifacts may affect signal strength and quality, causing image distortion, and may cause errors in the encoding direction. Signal quality and the choice of encoding direction can influence each other. For example, when performing high-resolution imaging, it is necessary to balance signal strength and encoding direction to avoid blurring artifacts. This improves the accuracy of prenatal screening for fetal spinal neural tube defects.

[0035] Among them, the specific process of obtaining the artifact identification correlation coefficient according to the artifact identification data and the reference artifact identification data obtained from the preset database is as follows: the reference artifact identification data includes a preset phase change threshold and a preset spatial resolution threshold; the phase change influence factor is obtained in combination with the reference artifact identification data, the echo frequency, the magnetic field strength and the acquisition time, and the phase change influence factor is used to reflect the influence of the phase change in the MRI image on the artifact identification correlation coefficient; the spatial resolution influence factor is obtained in combination with the preset spatial resolution threshold, the imaging field of view and the number of sampling points in the horizontal sampling direction, and the spatial resolution influence factor is used to reflect the influence of the spatial resolution in the MRI image on the artifact identification correlation coefficient; the artifact identification correlation coefficient is obtained by combining the obtained phase change influence factor and the spatial resolution influence factor.

[0036] In this embodiment, the preset phase change threshold is represented by the average value of the phase change data in the MRI images of the historical time period in the preset database, and the preset spatial resolution threshold is represented by the average value of the spatial resolution data in the MRI images of the historical time period in the preset database. In the screening of fetal spinal neural tube malformations, artifacts may mask or imitate the true pathological features, leading to misdiagnosis. Signal quality is crucial to the clarity of MRI images and the accuracy of diagnosis. Screening for fetal spinal neural tube malformations requires high signal quality to ensure that the image can clearly show the spinal cord and neural tube of the fetus. The encoding quality is directly related to the resolution and detail of the MRI image. In the screening of fetal spinal neural tube malformations, high encoding quality is the key to ensuring accurate diagnosis; the accuracy of prenatal screening for fetal spinal neural tube malformations is improved.

[0037] Specifically, the restricted expression of the artifact identification correlation coefficient is as follows:

[0038]

[0039] In the formula, represents the artifact recognition correlation coefficient of the mth MRI image, m=1,2,...,n, m represents the number of the MRI image, n represents the total number of MRI images, xw m represents the phase change influencing factor of the mth MRI image, KJ m represents the spatial resolution influencing factor of the mth MRI image, γ m Indicates the echo frequency, CB m represents the magnetic field strength of the mth MRI image, t m represents the acquisition time of the mth MRI image, FOV m represents the imaging field of view of the mth MRI image, SN mrepresents the number of sampling points in the horizontal sampling direction of the mth MRI image, R0 represents the preset phase change threshold, J0 represents the preset spatial resolution threshold, and e represents a natural constant.

[0040] The algorithm of this embodiment combines the artifact recognition data for comprehensive analysis to obtain the artifact recognition correlation coefficient. The artifact recognition data in the algorithm of this embodiment does not exist independently, but is interrelated. The higher the echo frequency, the greater the magnetic field strength, the longer the acquisition time, and the larger the imaging field of view, do not mean that the artifact recognition correlation coefficient is higher. It may increase the complexity and noise level of the signal, thereby increasing the possibility of artifacts. The influence of the number of sampling points in the horizontal sampling direction should also be considered comprehensively. When the number of sampling points is more, the data processing time and calculation complexity may also increase, and the artifact recognition correlation coefficient may decrease. The magnetic field strength directly affects the magnetic susceptibility and relaxation time of hydrogen protons, and then affects the intensity and frequency of the echo signal. The larger the field of view, the more data acquisition is required to cover the entire area, which usually leads to an increase in acquisition time. While keeping the field of view size unchanged, increasing the number of sampling points can improve the detail resolution of the image, but it will also increase the data processing volume and acquisition time. The parameters of the algorithm of this embodiment need to consider the impact on the results together.

[0041] Specifically, assuming that the phase change influence factor xw m The range is 0-2, and the spatial resolution impact factor KJ m The range is 0-1, such as Figure 2 As shown in FIG. 1 , it is a statistical diagram of the change of the artifact recognition correlation coefficient provided by the embodiment of the present invention. Figure 2 It can be seen that when the spatial resolution influence factor KJ m When it is 0, the influence factor xw changes with the phase m The artifact recognition correlation coefficient gradually increases. When the phase change influence factor xw m When it is 0, as the spatial resolution influence factor KJ m The artifact recognition correlation coefficient gradually increases, and the accuracy of prenatal screening for fetal spinal neural tube defects is improved.

[0042] Among them, the specific process of classifying MRI images based on the artifact recognition correlation coefficient and the preset artifact recognition threshold obtained from the preset database to obtain artifact images and non-artifact images is as follows: determine whether the artifact recognition correlation coefficient is higher than the preset artifact recognition threshold. When the artifact recognition correlation coefficient is higher than the preset artifact recognition threshold, the corresponding MRI image is classified as an artifact image and the artifact image is numbered; otherwise, the corresponding MRI image is classified as a non-artifact image, and the non-artifact image is numbered. The artifact image represents the MRI image corresponding to the artifact recognition correlation coefficient higher than the preset artifact recognition threshold, and the non-artifact image represents the MRI image corresponding to the artifact recognition correlation coefficient not higher than the preset artifact recognition threshold.

[0043] In this embodiment, by comparing the artifact recognition correlation coefficient with the preset artifact recognition threshold, the MRI image can be classified into an artifact image and a non-artifact image. This helps to identify images that may affect the diagnosis, and can ensure that images that are affected by artifacts as little as possible are used in the screening of fetal spinal neural tube defects, thereby improving the accuracy of diagnosis. By improving the quality of MRI images, the accuracy of diagnosis of fetal spinal neural tube defects is improved. Artifact images are numbered in ascending order of 1, 2, 3, etc., and non-artifact images are numbered in ascending order of 1, 2, 3, etc., and only need to ensure that the numbers are not repeated or omitted; the accuracy of prenatal screening for fetal spinal neural tube defects is improved.

[0044] The specific process of obtaining the signal quality evaluation value from the relevant data is as follows: the reference signal related data includes a preset MRI signal-to-noise ratio threshold and a preset average signal quantity threshold; the MRI signal-to-noise ratio influencing factor is obtained by combining the preset MRI signal-to-noise ratio threshold, MRI signal intensity and MRI noise intensity; the MRI average signal quantity influencing factor is obtained by combining the preset average signal quantity threshold, MRI signal intensity, MRI noise intensity and the number of MRI signal acquisition repetitions; the signal quality evaluation value is obtained by combining the artifact recognition correlation coefficient corresponding to the acquired artifact image, the MRI signal-to-noise ratio influencing factor and the MRI average signal quantity influencing factor; the specific process of obtaining the MRI signal-to-noise ratio influencing factor is as follows: W1, a ratio operation is performed on the MRI signal intensity and the MRI noise intensity of the artifact image; W2, a ratio operation is performed on the preset MRI signal-to-noise ratio threshold and the result of the ratio operation in W1 A sum operation is performed; W3, the result of the sum operation in W2 is ratioed with the preset MRI signal-to-noise ratio threshold to obtain the MRI signal-to-noise ratio influence factor, and the MRI signal-to-noise ratio influence factor is represented by the result of the ratio operation between the result of the sum operation in W2 and the preset MRI signal-to-noise ratio threshold; the specific process of obtaining the MRI average signal quantity influence factor is as follows: A1, the MRI signal intensity of the artifact image, the MRI noise intensity and the number of repeated MRI signal acquisitions are multiplied; A2, the preset average signal quantity threshold and the result of the product operation in A1 are summed; A3, the result of the sum operation in A2 is ratioed with the preset average signal quantity threshold to obtain the MRI average signal quantity influence factor, and the MRI average signal quantity influence factor is represented by the result of the ratio operation between the result of the sum operation in A2 and the preset average signal quantity threshold.

[0045] The constrained expression of the signal quality evaluation value is as follows:

[0046]

[0047]

[0048] Where XZ i represents the signal quality evaluation value of the i-th artifact image, i=1,2,...,g, i represents the number of the artifact image, g represents the total number of artifact images, and the total number of artifact images is less than or equal to the total number of MRI images. Represents the artifact recognition correlation coefficient of the i-th artifact image, SNR i represents the MRI signal-to-noise ratio influencing factor of the i-th artifact image, PN i S1 represents the influence factor of the average MRI signal quantity of the i-th artifact image, i represents the MRI signal intensity of the i-th artifact image, S2 i represents the MRI noise intensity of the i-th artifact image, N irepresents the number of repetitions of MRI signal acquisition of the i-th artifact image, S0 represents the preset MRI signal-to-noise ratio threshold, and S1 represents the preset average signal quantity threshold.

[0049] In this embodiment, when the artifact recognition correlation coefficient is higher than the preset artifact recognition threshold, the corresponding artifact images are numbered in the order of 1, 2, 3, etc., and it is only necessary to ensure that the numbers are not repeated or omitted. In the formula, That is, it is the corresponding artifact recognition correlation coefficient calculated before image classification. The preset MRI signal-to-noise ratio threshold is represented by the average value of the signal-to-noise ratio data in the MRI images of the historical time period in the preset database. The preset average signal quantity threshold is represented by the average value of the average signal quantity data in the MRI images of the historical time period in the preset database.

[0050] The algorithm of this embodiment combines signal-related data analysis to obtain a signal quality assessment value. The signal-related data in the algorithm of this embodiment does not exist independently, but is interrelated. The higher the signal strength and the more repeated times of MRI signal acquisition do not mean that the signal quality assessment value is higher. The influence of MRI noise intensity should also be considered comprehensively. When noise is superimposed on the real signal, the signal-to-noise ratio and contrast of the image may be reduced, and the signal quality assessment value may decrease. By increasing the number of repetitions, more signals can be accumulated and the influence of random noise can be reduced. Increasing the number of repetitions may also increase the scanning time, and the artifact recognition correlation coefficient may increase. The parameters of the algorithm of this embodiment need to be considered together and their impact on the results.

[0051] Specifically, assuming that the MRI signal-to-noise ratio factor SNR i The range is 0-5, and the MRI average signal number influence factor PN i The range is 0-5, artifact identification correlation coefficient The range is 0-1, as shown in Table 1, which is a statistical table of changes in signal quality evaluation values ​​provided in an embodiment of the present invention:

[0052] Table 1 Statistics of changes in signal quality evaluation values

[0053]

[0054] As can be seen from Table 1, with the gradual increase of the MRI signal-to-noise ratio influencing factor, the gradual increase of the MRI average signal number influencing factor, the gradual decrease of the artifact recognition correlation coefficient, the gradual increase of the signal quality assessment value, which means that the signal quality corresponding to the artifact image gradually increases. The higher the signal quality, the clearer the image can be, which is helpful for more accurate diagnosis of fetal spinal neural tube defects; the accuracy of prenatal screening for fetal spinal neural tube defects is improved.

[0055] Among them, the specific process of judging whether to adjust the signal quality based on the signal quality evaluation value and the preset signal quality threshold obtained from the preset database is as follows: SS1, judging whether the signal quality evaluation value is not lower than the preset signal quality threshold. When the signal quality evaluation value is not lower than the preset signal quality threshold, no signal quality adjustment is performed, otherwise SS2 is executed; SS2, performing median filtering processing, when the signal quality evaluation value corresponding to the monitored artifact image is not lower than the preset signal quality threshold, and the artifact recognition correlation coefficient is not higher than the preset artifact recognition threshold, stop executing, otherwise execute SS3, the median filtering processing is used to replace the value of each pixel in the artifact image with the median value in its neighborhood; SS3; performing wavelet transform processing, when the signal quality evaluation value corresponding to the monitored artifact image is not lower than the preset signal quality threshold, and the artifact recognition correlation coefficient is not higher than the preset artifact recognition threshold, stop executing, otherwise it is fed back to the preset personnel to obtain the coding quality evaluation value, and the wavelet transform processing is used to remove the noise coefficient and reconstruct the image.

[0056] In this embodiment, wavelet transform can separate the noise and real signal in the image, thereby removing the noise coefficient and reconstructing a clearer image. Median filtering can better retain the edge information of the image while removing noise, which is particularly important for MRI image analysis. In MRI images, artifacts may be generated due to various reasons (such as movement, magnetic field inhomogeneity, etc.). Median filtering can help reduce the impact of these artifacts and improve image quality. The neighborhood is usually a square window, called a kernel or mask, which slides on the image and is applied to each pixel of the image. Through this adjustment process, the diagnostic accuracy of MRI images in fetal spinal neural tube malformation screening can be improved, the image quality can be improved, and the risk of misdiagnosis and missed diagnosis can be reduced; the accuracy of prenatal screening for fetal spinal neural tube malformations is improved.

[0057] Among them, the specific process of obtaining the coding quality evaluation value according to the coding direction related data corresponding to the artifact image and the reference coding direction data obtained from the preset database is as follows: the reference coding direction data includes a preset frequency coding direction field of view threshold and a preset phase coding direction field of view threshold; the frequency coding influence factor is obtained by combining the preset frequency coding direction field of view threshold, MRI bandwidth and MRI frequency resolution; the phase coding influence factor is obtained by combining the preset phase coding direction field of view threshold, MRI bandwidth and MRI phase resolution; the coding quality evaluation value is obtained by combining the artifact recognition correlation coefficient, frequency coding influence factor and phase coding influence factor corresponding to the acquired artifact image; the specific process of obtaining the frequency coding influence factor is as follows: Q1, the MRI bandwidth and MRI frequency of the artifact image The resolution is ratio-operated; Q2, the result of the ratio operation in Q1 is summed with the preset frequency encoding direction field of view threshold; Q3, the result of the sum operation in Q2 is ratio-operated with the preset frequency encoding direction field of view threshold to obtain the frequency encoding influence factor, and the frequency encoding influence factor is represented by the result of the ratio operation in Q3; the specific process of obtaining the phase encoding influence factor is as follows: E1, the MRI bandwidth of the artifact image and the MRI phase resolution are ratio-operated; E2, the result of the ratio operation in E1 is summed with the preset phase encoding direction field of view threshold; E3, the result of the sum operation in E2 is ratio-operated with the preset phase encoding direction field of view threshold to obtain the phase encoding influence factor, and the phase encoding influence factor is represented by the result of the ratio operation in E3.

[0058] The restricted expression of the encoding quality evaluation value is as follows:

[0059]

[0060] Where BM i represents the coding quality evaluation value of the i-th artifact image, i=1,2,...,g, i represents the number of the artifact image, g represents the total number of artifact images, and the total number of artifact images is less than or equal to the total number of MRI images. represents the artifact recognition correlation coefficient of the i-th artifact image, PL i represents the frequency coding influence factor of the i-th artifact image, XW i represents the phase encoding influence factor of the i-th artifact image, B i represents the MRI bandwidth of the i-th artifact image, F i represents the MRI frequency resolution of the i-th artifact image, ρ i represents the MRI phase resolution of the i-th artifact image, BM0 represents the preset frequency encoding direction field of view threshold, and BM1 represents the preset phase encoding direction field of view threshold.

[0061] In this embodiment, when the artifact recognition correlation coefficient is higher than the preset artifact recognition threshold, the corresponding artifact images are numbered in the order of 1, 2, 3, etc., and it is only necessary to ensure that the numbers are not repeated or omitted. In the formula, That is, the corresponding artifact recognition correlation coefficient calculated before image classification, the preset frequency encoding direction field of view threshold is represented by the average value of the frequency encoding direction field of view in the historical time period MRI images in the preset database, and the preset phase encoding direction field of view threshold is represented by the average value of the phase encoding direction field of view in the historical time period MRI images in the preset database;

[0062] The algorithm of this embodiment combines the analysis of encoding direction related data to obtain the encoding quality evaluation value. The encoding direction related data in the algorithm of this embodiment does not exist independently, but has mutual correlation. The larger the MRI bandwidth does not mean the higher the encoding quality evaluation value. The influence of MRI frequency resolution and MRI phase resolution should also be considered comprehensively. The higher the frequency resolution, the more detailed information can be captured, but it may cause longer scanning time and increased noise. The higher the phase resolution, the better the spatial resolution of the image, but it may cause data redundancy and longer scanning time, and the encoding quality evaluation value may decrease. The parameters of the algorithm of this embodiment need to consider the impact on the results together. The accuracy of prenatal screening for fetal spinal neural tube malformations is improved.

[0063] Among them, the specific process of judging whether to adjust the coding quality based on the coding quality evaluation value and the preset coding quality threshold obtained from the preset database is as follows: WW1, judging whether the coding quality evaluation value is not lower than the preset coding quality threshold. When the coding quality evaluation value is not lower than the preset coding quality threshold, no coding quality adjustment is performed, otherwise WW2 is executed; WW2, frequency coding direction resolution improvement processing is performed, and the execution is stopped when the coding quality evaluation value corresponding to the monitored artifact image is not lower than the preset coding quality threshold and the artifact recognition correlation coefficient is not higher than the preset artifact recognition threshold, otherwise WW3 is executed. The frequency coding direction resolution improvement processing means improving the resolution in the frequency coding direction through a non-Cartesian sampling trajectory; WW3, phase coding direction resolution improvement processing is performed, and the execution is stopped when the coding quality evaluation value corresponding to the monitored artifact image is not lower than the preset coding quality threshold and the artifact recognition correlation coefficient is not higher than the preset artifact recognition threshold, otherwise feedback is given to the preset personnel to re-acquire the MRI image. The phase coding direction resolution improvement processing means increasing the number of phase coding steps through a sampling method.

[0064] In this embodiment, the frequency encoding direction resolution improvement process is performed, which is usually achieved through a non-Cartesian sampling trajectory, such as increasing the number of sampling points to improve the resolution in the frequency encoding direction. If the quality requirements are still not met after the frequency encoding direction resolution improvement process, the phase encoding direction resolution improvement process is continued to be performed to improve the resolution in the phase encoding direction; thus, the accuracy of prenatal screening for fetal spinal neural tube defects is improved.

[0065] Among them, the specific process of prenatal screening for fetal spinal neural tube defects is as follows: determine whether the coding quality evaluation value is not lower than the preset coding quality threshold, when the coding quality evaluation value is not lower than the preset coding quality threshold, obtain the artifact image corresponding to the coding quality evaluation value not lower than the preset coding quality threshold; combine the artifact image and non-artifact image corresponding to the obtained coding quality evaluation value not lower than the preset coding quality threshold to perform image fusion, and the image fusion is used to perform weighted averaging processing on the artifact image and non-artifact image corresponding to the coding quality evaluation value not lower than the preset coding quality threshold by a weighted averaging method to obtain a fused image; and feed back the obtained fused image to the preset personnel for prenatal screening for fetal spinal neural tube defects.

[0066] In this embodiment, image fusion is usually achieved through a weighted averaging method. For each pixel position, the weighted average of all images is calculated, that is, the images are weighted according to their quality, and then the weighted average is calculated to generate a fused image. The weights should be based on objective image quality indicators, such as artifact recognition correlation coefficients, signal quality assessment values, encoding quality assessment values, etc. The higher the quality of the image, the higher the weight may be assigned to ensure that the fused image reflects the true fetal spinal neural tube structure as much as possible; the accuracy of prenatal screening for fetal spinal neural tube malformations is improved.

[0067] There are a few points to note:

[0068] (1) The drawings of the embodiments of the present invention only relate to the structures related to the embodiments of the present invention, and other structures may refer to the general design.

[0069] (2) For the sake of clarity, in the drawings used to describe the embodiments of the present invention, the thickness of the layers or regions is exaggerated or reduced, that is, these drawings are not drawn according to the actual scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being "on" or "under" another element, the element may be "directly" "on" or "under" the other element or there may be intermediate elements.

[0070] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to obtain new embodiments.

[0071] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for prenatal screening of fetal spinal neural tube defects based on MRI, characterized in that: The following steps are involved: S1, acquiring corresponding MRI image data according to the acquired MRI image, wherein the MRI image is used to display fetal information at different angles and different levels, and the MRI image data includes artifact recognition data, signal-related data, and encoding direction-related data; S2, obtaining an artifact recognition correlation coefficient according to the artifact recognition data and reference artifact recognition data obtained from a preset database, and performing image classification on the MRI image based on the artifact recognition correlation coefficient and a preset artifact recognition threshold obtained from the preset database to obtain an artifact image and a non-artifact image, wherein the artifact recognition correlation coefficient is used to evaluate the possibility of generating motion artifacts; S3, obtaining a signal quality evaluation value according to signal-related data corresponding to the artifact image and reference signal-related data obtained from a preset database, and determining whether to perform signal quality adjustment based on the signal quality evaluation value and a preset signal quality threshold obtained from the preset database, wherein the signal quality evaluation value is used to evaluate the signal quality corresponding to the artifact image; S4, obtaining a coding quality evaluation value according to coding direction related data corresponding to the artifact image and reference coding direction data obtained from a preset database, and judging whether to perform coding quality adjustment and prenatal screening for fetal spinal neural tube defects based on the coding quality evaluation value and a preset coding quality threshold obtained from the preset database, wherein the coding quality evaluation value is used to evaluate the quality of the coding method corresponding to the artifact image.

2. The method for prenatal screening of fetal spinal neural tube defects based on MRI according to claim 1, characterized in that: The artifact identification data includes echo frequency, magnetic field strength, acquisition time, imaging field of view and number of sampling points in horizontal sampling direction; The reference signal related data includes MRI signal intensity, MRI noise intensity and MRI signal acquisition repetition times; The encoding direction related data includes MRI bandwidth, MRI frequency resolution and MRI phase resolution.

3. The method for prenatal screening of fetal spinal neural tube defects based on MRI according to claim 1, characterized in that: The specific process of obtaining the artifact recognition correlation coefficient according to the artifact recognition data and the reference artifact recognition data obtained from the preset database is as follows: The reference artifact identification data includes a preset phase change threshold and a preset spatial resolution threshold; Acquire a phase change influencing factor by combining reference artifact recognition data, echo frequency, magnetic field strength, and acquisition time, wherein the phase change influencing factor is used to reflect the degree of influence of the phase change in the MRI image on the artifact recognition correlation coefficient; The spatial resolution impact factor is obtained by combining the preset spatial resolution threshold, the imaging field of view and the number of sampling points in the horizontal sampling direction, wherein the spatial resolution impact factor is used to reflect the influence of the spatial resolution in the MRI image on the artifact recognition correlation coefficient; The artifact identification correlation coefficient is obtained by combining the acquired phase change influencing factor and spatial resolution influencing factor.

4. The method for prenatal screening of fetal spinal neural tube defects based on MRI according to claim 3, characterized in that: The restriction expression of the artifact identification correlation coefficient is as follows: In the formula, represents the artifact recognition correlation coefficient of the mth MRI image, m=1,2,...,n, m represents the number of the MRI image, n represents the total number of MRI images, xw m represents the phase change influencing factor of the mth MRI image, KJ m represents the spatial resolution influencing factor of the mth MRI image, γ m Indicates the echo frequency, CB m represents the magnetic field strength of the mth MRI image, t m represents the acquisition time of the mth MRI image, FOV m represents the imaging field of view of the mth MRI image, SN m represents the number of sampling points in the horizontal sampling direction of the mth MRI image, R0 represents the preset phase change threshold, J0 represents the preset spatial resolution threshold, and e represents a natural constant.

5. The method for prenatal screening of fetal spinal neural tube defects based on MRI according to claim 1, characterized in that: The specific process of classifying the MRI image based on the artifact recognition correlation coefficient and the preset artifact recognition threshold obtained from the preset database to obtain the artifact image and the non-artifact image is as follows: Determine whether the artifact recognition correlation coefficient is higher than a preset artifact recognition threshold. When the artifact recognition correlation coefficient is higher than the preset artifact recognition threshold, classify the corresponding MRI image as an artifact image and number the artifact image. Otherwise, the corresponding MRI image is classified as a non-artifact image, and the non-artifact image is numbered.

6. The method for prenatal screening of fetal spinal neural tube defects based on MRI according to claim 1, characterized in that: The specific process of obtaining the signal quality evaluation value according to the signal related data corresponding to the artifact image and the reference signal related data obtained from the preset database is as follows: The reference signal related data includes a preset MRI signal-to-noise ratio threshold and a preset average signal quantity threshold; The MRI signal-to-noise ratio influencing factor is obtained by combining the preset MRI signal-to-noise ratio threshold, MRI signal intensity, and MRI noise intensity; The influence factor of the MRI average signal quantity is obtained by combining the preset average signal quantity threshold, MRI signal intensity, MRI noise intensity and the number of repetitions of MRI signal acquisition; The signal quality evaluation value is obtained by combining the artifact recognition correlation coefficient corresponding to the acquired artifact image, the MRI signal-to-noise ratio influencing factor and the MRI average signal quantity influencing factor; The specific process of obtaining the MRI signal-to-noise ratio influencing factor is as follows: W1, ratio operation between the MRI signal intensity and the MRI noise intensity of the artifact image; W2, summing the preset MRI signal-to-noise ratio threshold and the result of the ratio operation in W1; W3, performing a ratio operation on the result of the sum operation in W2 and the preset MRI signal-to-noise ratio threshold to obtain the MRI signal-to-noise ratio influencing factor; The specific process of obtaining the influencing factor of the average MRI signal quantity is as follows: A1, perform product operation on the MRI signal intensity of the artifact image, the MRI noise intensity, and the number of repetitions of MRI signal acquisition; A2, performing a sum operation on the preset average signal quantity threshold and the result of the product operation in A1; A3, performing a ratio operation on the result of the sum operation in A2 and the preset average signal quantity threshold to obtain an MRI average signal quantity influence factor.

7. The method for prenatal screening of fetal spinal neural tube defects based on MRI according to claim 1, characterized in that: The specific process of determining whether to adjust the signal quality based on the signal quality evaluation value and the preset signal quality threshold obtained from the preset database is as follows: SS1, judging whether the signal quality evaluation value is not lower than the preset signal quality threshold, if the signal quality evaluation value is not lower than the preset signal quality threshold, no signal quality adjustment is performed, otherwise SS2 is executed; SS2, performing median filtering processing, when the signal quality evaluation value corresponding to the monitored artifact image is not lower than the preset signal quality threshold, and the artifact recognition correlation coefficient is not higher than the preset artifact recognition threshold, stop executing, otherwise execute SS3; SS3; perform wavelet transform processing, and stop executing when the signal quality evaluation value corresponding to the monitored artifact image is not lower than the preset signal quality threshold, and the artifact recognition correlation coefficient is not higher than the preset artifact recognition threshold. Otherwise, feedback is given to the preset personnel to obtain the coding quality evaluation value.

8. The method for prenatal screening of fetal spinal neural tube defects based on MRI according to claim 2, characterized in that: The specific process of obtaining the coding quality evaluation value according to the coding direction related data corresponding to the artifact image and the reference coding direction data obtained from the preset database is as follows: The reference encoding direction data includes a preset frequency encoding direction field of view threshold and a preset phase encoding direction field of view threshold; The frequency encoding influencing factor is obtained by combining the preset frequency encoding direction field of view threshold, MRI bandwidth and MRI frequency resolution; The phase encoding influencing factor is obtained by combining the preset phase encoding direction field of view threshold, MRI bandwidth and MRI phase resolution; Acquire a coding quality evaluation value by combining an artifact recognition correlation coefficient, a frequency coding influence factor, and a phase coding influence factor corresponding to the acquired artifact image; The specific process of obtaining the frequency coding impact factor is as follows: Q1, ratio calculation of MRI bandwidth and MRI frequency resolution of artifact image; Q2, summing the result of the ratio operation in Q1 and the preset frequency encoding direction visual field threshold; Q3, performing a ratio operation on the result of the sum operation in Q2 and the preset frequency coding direction field of view threshold to obtain a frequency coding influence factor; The specific process of obtaining the phase encoding influence factor is as follows: E1, ratio calculation of MRI bandwidth and MRI phase resolution of artifact image; E2, performing a summation operation on the result of the ratio operation in E1 and the preset phase encoding direction visual field threshold; E3, performing a ratio operation on the result of the sum operation of E2 and the preset phase encoding direction field of view threshold to obtain a phase encoding influence factor.

9. The method for prenatal screening of fetal spinal neural tube defects based on MRI according to claim 1, characterized in that: The specific process of judging whether to adjust the coding quality based on the coding quality evaluation value and the preset coding quality threshold obtained from the preset database is as follows: WW1, judging whether the coding quality evaluation value is not lower than the preset coding quality threshold, and not performing coding quality adjustment when the coding quality evaluation value is not lower than the preset coding quality threshold, otherwise executing WW2; WW2, performing a frequency coding direction resolution improvement process, when the coding quality evaluation value corresponding to the monitored artifact image is not lower than the preset coding quality threshold and the artifact recognition correlation coefficient is not higher than the preset artifact recognition threshold, stop executing, otherwise execute WW3, the frequency coding direction resolution improvement process means improving the resolution in the frequency coding direction through a non-Cartesian sampling trajectory; WW3, performs resolution improvement processing in the phase encoding direction. When the encoding quality evaluation value corresponding to the monitored artifact image is not lower than the preset encoding quality threshold and the artifact recognition correlation coefficient is not higher than the preset artifact recognition threshold, the execution is stopped. Otherwise, feedback is given to the preset personnel to re-acquire the MRI image. The resolution improvement processing in the phase encoding direction means increasing the number of phase encoding steps through the oversampling method.

10. The method for prenatal screening of fetal spinal neural tube defects based on MRI according to claim 1, characterized in that: The specific process of prenatal screening for fetal spinal neural tube defects is as follows: Determine whether the coding quality evaluation value is not lower than a preset coding quality threshold, and when the coding quality evaluation value is not lower than the preset coding quality threshold, obtain an artifact image corresponding to when the coding quality evaluation value is not lower than the preset coding quality threshold; Performing image fusion by combining the artifact image and the non-artifact image corresponding to the acquired encoding quality evaluation value not less than a preset encoding quality threshold; The acquired fusion images are fed back to preset personnel for prenatal screening of fetal spinal neural tube defects.

Images