Image signal-to-noise ratio evaluation method, system and device, medium and program product

Abstract

The invention provides a signal-to-noise ratio evaluation method, system and device of an image, a medium and a program product. The signal-to-noise ratio evaluation method comprises the steps of obtaining an image of magnetic resonance function imaging before denoising; performing discrete wavelet transform on the image of the magnetic resonance functional imaging before denoising to obtain a high-frequency coefficient matrix and a low-frequency coefficient matrix; performing de-noising and discrete wavelet inverse transformation processing on the high-frequency coefficient matrix and the low-frequency coefficient matrix by using a soft threshold function to obtain a de-noised magnetic resonance functional imaging image; and acquiring the signal-to-noise ratio of the magnetic resonance functional imaging based on the image of the magnetic resonance functional imaging before denoising and the image of the magnetic resonance functional imaging after denoising. According to the invention, the automatic and precise evaluation of the signal-to-noise ratio of the magnetic resonance functional imaging image is realized through the discrete wavelet transform method, and the evaluation efficiency and accuracy of the signal-to-noise ratio are improved.

Description

Technical Field

[0001] The present disclosure relates to the field of image information technology, and in particular to a method, system, device, medium and program product for evaluating the signal-to-noise ratio of an image. Background Art

[0002] Medical image quality assessment is crucial in medical practice because it directly impacts the accuracy of clinical diagnoses. Medical images are used to identify and monitor diseases, so ensuring their quality is a prerequisite for accurate diagnosis. Medical image quality assessment typically includes the following metrics: signal-to-noise ratio (SNR), peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and root mean square error (RMSE). Other visual subjective quality metrics include clarity, contrast, and color. SNR is crucial in image quality assessment, as it directly impacts the accuracy of clinical diagnoses. SNR analysis helps provide more reliable and clearer medical images, providing strong support for medical practice.

[0003] SNR represents the ratio between image signal and noise. An image with high SNR shows clearer details, while an image with low SNR produces blur and noise. Its calculation formula is shown in the following formula (1):

[0004]

[0005] Among them, SI represents the signal intensity of the image. In the measurement of image signal intensity, it is divided into the overall signal intensity of the image and the signal intensity SI of a specific organ. organ Noise represents the noise of the image. It is often measured by selecting tissue with uniform signal in the image, and its standard deviation is selected as the noise. In practice, the standard deviation of the signal intensity of air is often selected as the noise, sd air Represents the standard deviation of air signals. Since noise is defined as signal fluctuations in uniform tissue, the standard deviation of air signal intensity is often used as noise in actual operations.

[0006] Currently, the primary method for measuring the SNR of medical images relies on manually selecting and delineating regions of interest (ROIs). To improve measurement accuracy, different regions are often selected for measurement, further increasing the measurement workload. Due to the significant randomness in ROI selection by different measurers, the results across studies are difficult to compare. Furthermore, since most parametric maps reconstructed by fMRI (functional magnetic resonance imaging) remove the signal from the surrounding air, this limits methods that rely on measuring the standard deviation of the air signal intensity as a proxy for noise. Summary of the Invention

[0007] The technical problem to be solved by the present disclosure is to overcome the defects of low efficiency and low accuracy in the conventional manual measurement of the SNR of medical images in the prior art, and to provide a method, system, device, medium and program product for evaluating the signal-to-noise ratio of an image.

[0008] The present disclosure solves the above technical problems through the following technical solutions:

[0009] A first aspect of the present disclosure provides a method for evaluating a signal-to-noise ratio of an image, the method comprising:

[0010] Acquire functional magnetic resonance imaging images before denoising;

[0011] Performing discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain a high-frequency coefficient matrix and a low-frequency coefficient matrix;

[0012] performing denoising and inverse discrete wavelet transform processing on the high-frequency coefficient matrix and the low-frequency coefficient matrix using a soft threshold function to obtain a denoised functional magnetic resonance imaging image;

[0013] A signal-to-noise ratio of functional magnetic resonance imaging is acquired based on the functional magnetic resonance imaging image before denoising and the functional magnetic resonance imaging image after denoising.

[0014] Preferably, the step of performing discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain a high-frequency coefficient matrix and a low-frequency coefficient matrix comprises:

[0015] Performing a multi-level two-dimensional discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain high-frequency coefficient matrices and low-frequency coefficient matrices at each level;

[0016] The step of performing denoising and inverse discrete wavelet transform processing on the high-frequency coefficient matrix and the low-frequency coefficient matrix using a soft threshold function to obtain a denoised functional magnetic resonance imaging image comprises:

[0017] The high-frequency coefficient matrices at each level and the low-frequency coefficient matrices at each level are subjected to denoising and discrete wavelet inverse transformation processing by using a soft threshold function to obtain a denoised functional magnetic resonance imaging image.

[0018] Preferably, the step of performing a multi-level two-dimensional discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain high-frequency coefficient matrices and low-frequency coefficient matrices at each level comprises:

[0019] Performing a multi-level two-dimensional discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain a first-level high-frequency coefficient matrix and a first-level low-frequency coefficient matrix;

[0020] Performing discrete wavelet transform on the primary low-frequency coefficient matrix to obtain a secondary high-frequency coefficient matrix and a secondary low-frequency coefficient matrix;

[0021] Performing discrete wavelet transform on the secondary low-frequency coefficient matrix to obtain a tertiary high-frequency coefficient matrix and a tertiary low-frequency coefficient matrix;

[0022] Performing discrete wavelet transform on the three-level low-frequency coefficient matrix to obtain an N-level high-frequency coefficient matrix and an N-level low-frequency coefficient matrix;

[0023] Among them, N is greater than 3.

[0024] Preferably, the step of performing denoising and inverse discrete wavelet transform processing on the high-frequency coefficient matrices and the low-frequency coefficient matrices of each level using a soft threshold function to obtain a denoised functional magnetic resonance imaging image comprises:

[0025] Obtaining the median of the noise signal strength corresponding to the high-frequency coefficient matrix at each level;

[0026] Obtaining thresholds of the high-frequency coefficient matrices at each level based on the median and a preset value;

[0027] Obtaining a target noise signal whose noise signal strength of each level of high frequency coefficient matrix is less than a preset signal strength;

[0028] denoising the target noise signal using the thresholds of the high-frequency coefficient matrices at each level to obtain denoised high-frequency coefficient matrices at each level;

[0029] A denoised functional magnetic resonance imaging image is acquired based on the denoised high-frequency coefficient matrices at each level and the N-level low-frequency coefficient matrices obtained after inverse discrete wavelet transform processing.

[0030] Preferably, the expression for obtaining the signal-to-noise ratio of functional magnetic resonance imaging based on the functional magnetic resonance imaging image before denoising and the functional magnetic resonance imaging image after denoising is:

[0031]

[0032] Where WSNR represents the signal-to-noise ratio of functional magnetic resonance imaging, N represents the noise signal or the total number of samples or pixels in the image, and X i Y represents the functional magnetic resonance imaging image before denoising. i The image shows the denoised functional magnetic resonance imaging.

[0033] Preferably, the number of stages of the discrete wavelet transform is the same as the number of stages of the inverse discrete wavelet transform.

[0034] A second aspect of the present disclosure provides a signal-to-noise ratio evaluation system for an image, the signal-to-noise ratio evaluation system comprising:

[0035] A first acquisition module is used to acquire a functional magnetic resonance imaging image before noise removal;

[0036] A first processing module is configured to perform discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain a high-frequency coefficient matrix and a low-frequency coefficient matrix;

[0037] a second processing module, configured to perform denoising and inverse discrete wavelet transform processing on the high-frequency coefficient matrix and the low-frequency coefficient matrix using a soft threshold function to obtain a denoised functional magnetic resonance imaging image;

[0038] The second acquisition module is configured to acquire a signal-to-noise ratio of the functional magnetic resonance imaging based on the functional magnetic resonance imaging image before denoising and the functional magnetic resonance imaging image after denoising.

[0039] Preferably, the first processing module is used to perform a multi-level two-dimensional discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain high-frequency coefficient matrices and low-frequency coefficient matrices at each level;

[0040] The second processing module is used to perform denoising and inverse discrete wavelet transform processing on the high-frequency coefficient matrices and the low-frequency coefficient matrices at each level using a soft threshold function to obtain a denoised functional magnetic resonance imaging image.

[0041] Preferably, the first processing module includes:

[0042] a first processing unit, configured to perform a multi-level two-dimensional discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain a first-level high-frequency coefficient matrix and a first-level low-frequency coefficient matrix;

[0043] a second processing unit, configured to perform discrete wavelet transform on the primary low-frequency coefficient matrix to obtain a secondary high-frequency coefficient matrix and a secondary low-frequency coefficient matrix;

[0044] a third processing unit, configured to perform discrete wavelet transform on the secondary low-frequency coefficient matrix to obtain a tertiary high-frequency coefficient matrix and a tertiary low-frequency coefficient matrix;

[0045] a fourth processing unit, configured to perform discrete wavelet transform on the three-level low-frequency coefficient matrix to obtain an N-level high-frequency coefficient matrix and an N-level low-frequency coefficient matrix;

[0046] Among them, N is greater than 3.

[0047] Preferably, the second processing module includes:

[0048] A first acquiring unit, configured to acquire the median of the noise signal strength corresponding to each level of the high frequency coefficient matrix;

[0049] A second acquiring unit, configured to acquire the thresholds of the high-frequency coefficient matrices at each level based on the median and a preset value;

[0050] A third acquisition unit is used to acquire a target noise signal whose noise signal strength of each level of high-frequency coefficient matrix is less than a preset signal strength;

[0051] a fifth processing unit, configured to perform denoising on the target noise signal using the thresholds of the high-frequency coefficient matrices at each level to obtain denoised high-frequency coefficient matrices at each level;

[0052] The sixth processing unit is used to obtain a denoised functional magnetic resonance imaging image based on the denoised high-frequency coefficient matrices at each level and the N-level low-frequency coefficient matrices obtained after inverse discrete wavelet transform processing.

[0053] Preferably, the expression of the signal-to-noise ratio of functional magnetic resonance imaging is:

[0054]

[0055] Where WSNR represents the signal-to-noise ratio of functional magnetic resonance imaging, N represents the noise signal or the total number of samples or pixels in the image, and X i Y represents the functional magnetic resonance imaging image before denoising. i The image shows the denoised functional magnetic resonance imaging.

[0056] Preferably, the number of stages of the discrete wavelet transform is the same as the number of stages of the inverse discrete wavelet transform.

[0057] A third aspect of the present disclosure provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and used to run on the processor, wherein when the processor executes the computer program, the image signal-to-noise ratio evaluation method described in the first aspect is implemented.

[0058] A fourth aspect of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for evaluating the signal-to-noise ratio of an image as described in the first aspect.

[0059] A fifth aspect of the present disclosure provides a computer program product, comprising a computer program, which, when executed by a processor, implements the method for evaluating the signal-to-noise ratio of an image as described in the first aspect.

[0060] On the basis of conforming to the common sense in this field, the above-mentioned preferred conditions can be arbitrarily combined to obtain the preferred embodiments of the present disclosure.

[0061] The positive progress of this disclosure is:

[0062] The present disclosure realizes automated and precise evaluation of the signal-to-noise ratio of functional magnetic resonance imaging images through the discrete wavelet transform method, thereby improving the evaluation efficiency and accuracy of the signal-to-noise ratio. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a flowchart of the method for evaluating the signal-to-noise ratio of an image provided in Example 1 of the present disclosure.

[0064] Figure 2 This is a module diagram of the image signal-to-noise ratio evaluation system provided in Example 2 of the present disclosure.

[0065] Figure 3 This is a schematic structural diagram of an electronic device for implementing the method for evaluating the signal-to-noise ratio of an image according to embodiment 3 of the present disclosure. DETAILED DESCRIPTION

[0066] The present disclosure is further illustrated below by way of examples, but the present disclosure is not limited to the scope of the examples.

[0067] In the embodiments of the present disclosure, prefixes such as "first" and "second" are used only to distinguish different description objects, and have no limiting effect on the position, order, priority, quantity or content of the described objects. In the embodiments of the present disclosure, the use of prefixes such as ordinal numbers to distinguish description objects does not constitute a limitation on the described objects. For the statement of the described objects, please refer to the description in the context of the claims or embodiments, and no unnecessary limitations should be constituted due to the use of such prefixes. In addition, in the description of this embodiment, unless otherwise specified, the meaning of "plurality" is two or more.

[0068] In the embodiments of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0069] Example 1

[0070] Figure 1 This is a flow chart of a method for evaluating the signal-to-noise ratio of an image provided in Example 1 of the present disclosure, as shown in FIG. Figure 1 As shown, the signal-to-noise ratio evaluation method includes:

[0071] S1, obtaining functional magnetic resonance imaging images before denoising;

[0072] S2, performing discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain a high-frequency coefficient matrix and a low-frequency coefficient matrix;

[0073] S3, performing denoising and inverse discrete wavelet transform processing on the high-frequency coefficient matrix and the low-frequency coefficient matrix using a soft threshold function to obtain a denoised functional magnetic resonance imaging image;

[0074] S4. Acquire a signal-to-noise ratio of the functional magnetic resonance imaging based on the functional magnetic resonance imaging image before denoising and the functional magnetic resonance imaging image after denoising.

[0075] In this embodiment, the expression for obtaining the signal-to-noise ratio of the functional magnetic resonance imaging based on the functional magnetic resonance imaging image before denoising and the functional magnetic resonance imaging image after denoising is shown in formula (2):

[0076]

[0077] Where WSNR (Wavelet-Based Signal-to-Noise Ratio, WSNR) represents the signal-to-noise ratio of functional magnetic resonance imaging, N represents the total number of samples or pixels in the noise signal or image, and X represents the total number of samples or pixels in the image. i Y represents the functional magnetic resonance imaging image before denoising. i The image shows the denoised functional magnetic resonance imaging.

[0078] This embodiment realizes the automated and precise evaluation of the signal-to-noise ratio of functional magnetic resonance imaging images by using the discrete wavelet transform method, thereby improving the evaluation efficiency and accuracy of the signal-to-noise ratio.

[0079] In an optional embodiment, step S2 includes:

[0080] S21, performing a multi-level two-dimensional discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain high-frequency coefficient matrices and low-frequency coefficient matrices at each level;

[0081] Step S3 includes:

[0082] S31. Utilizing a soft threshold function, denoising and performing inverse discrete wavelet transform processing on high-frequency coefficient matrices at all levels and low-frequency coefficient matrices at all levels are performed to obtain a denoised functional magnetic resonance imaging image.

[0083] In the specific implementation process, the fMRI image is first subjected to a multi-level two-dimensional discrete wavelet transform (2D Discrete Wavelet Transform, 2D DWT) to obtain the low-frequency coefficient matrix (Approximation Coefficient Matrix, ACM) and the high-frequency coefficient matrix (Detailed Coefficient Matrix, DCM) at each level. Then, the soft threshold function of formula (3) is used to calculate the threshold of the high-frequency coefficient matrix at each level.

[0084]

[0085] Among them, soft(x,T) represents the threshold of the high-frequency coefficient matrix at each level, x represents the coefficient of the wavelet transform, T represents the pre-selected threshold, which is defined as T = σ2log(N), N is the number of signal samples, σ is the standard deviation of the noise, and |x| represents the absolute value of the wavelet coefficient.

[0086] In an optional embodiment, step S21 includes:

[0087] S211, performing a multi-level two-dimensional discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain a first-level high-frequency coefficient matrix and a first-level low-frequency coefficient matrix;

[0088] S212, performing discrete wavelet transform on the first-level low-frequency coefficient matrix to obtain a second-level high-frequency coefficient matrix and a second-level low-frequency coefficient matrix;

[0089] S213, performing discrete wavelet transform on the secondary low-frequency coefficient matrix to obtain a tertiary high-frequency coefficient matrix and a tertiary low-frequency coefficient matrix;

[0090] S214, performing discrete wavelet transform on the three-level low-frequency coefficient matrix to obtain an N-level high-frequency coefficient matrix and an N-level low-frequency coefficient matrix;

[0091] Among them, N is greater than 3.

[0092] In an optional embodiment, step S31 includes:

[0093] S311, obtaining the median of the noise signal strength corresponding to each level of high-frequency coefficient matrix;

[0094] S312, obtaining thresholds of high-frequency coefficient matrices at each level based on the median and a preset value;

[0095] S313, obtaining a target noise signal whose noise signal strength of each level of high-frequency coefficient matrix is less than a preset signal strength;

[0096] S314, performing denoising processing on the target noise signal using the thresholds of the high-frequency coefficient matrices at each level to obtain denoised high-frequency coefficient matrices at each level;

[0097] S315 , acquiring a denoised functional magnetic resonance imaging image based on the denoised high-frequency coefficient matrices of each level and the N-level low-frequency coefficient matrices obtained after inverse discrete wavelet transform processing.

[0098] In this embodiment, the number of discrete wavelet transform stages is the same as the number of inverse discrete wavelet transform stages. The specific number of stages is determined empirically according to the noise level and signal strength of the image.

[0099] In a specific implementation process, for example, taking N as 4, the functional magnetic resonance imaging image f(k) before denoising is first input and decomposed using a multi-level two-dimensional discrete wavelet transform (2D DWT) to obtain a first-level high-frequency coefficient matrix (CD1) and a first-level low-frequency coefficient matrix (CA1). Then, a second discrete wavelet transform (DWT) is performed on CA1 to decompose the first-level high-frequency coefficient matrix (CD2) and a second-level low-frequency coefficient matrix (CA2). Then, a third discrete wavelet transform (DWT) is performed on CA2 to decompose the first-level high-frequency coefficient matrix (CD3) and a third-level low-frequency coefficient matrix (CA3). Then, a fourth discrete wavelet transform (DWT) is performed on CA3 to decompose the fourth-level high-frequency coefficient matrix (CD4) and a fourth-level low-frequency coefficient matrix (CA4). Then, a soft threshold function is used to remove noise and perform inverse discrete wavelet transform (IDWT) on the first-level high-frequency coefficient matrix (CD1), the second-level high-frequency coefficient matrix (CD2), the third-level high-frequency coefficient matrix (CD3), the fourth-level high-frequency coefficient matrix (CD4), and the fourth-level low-frequency coefficient matrix (CA4), respectively. Specifically, the median of the signal intensity corresponding to each level of the high-frequency coefficient matrix is first found, and then the median is divided by a preset value (for example, the preset value can be 0.6745, or it can be set to other values according to actual conditions) to calculate the median absolute deviation, that is, the threshold of each level of the high-frequency coefficient matrix; then, the threshold of each level of the high-frequency coefficient matrix is used to set the target noise signal portion of the high-frequency coefficient matrix at each level whose noise signal intensity is less than the preset signal intensity to 0 (that is, the threshold of each level of the high-frequency coefficient matrix is used to set the low signal intensity portion of the high-frequency coefficient matrix at each level to 0) to eliminate noise, and then the high-frequency coefficient matrix at each level after noise elimination and the final low-frequency coefficient matrix (for example, the N-level low-frequency coefficient matrix) are used to reconstruct the denoised magnetic resonance functional imaging image Y i .

[0100] Compared to traditional manual SNR measurement methods, the signal-to-noise ratio (SNR) assessment technique based on discrete wavelet transform (DWT) in this embodiment offers the advantages of greater efficiency, accuracy, and repeatability. It addresses the difficulty in selecting homogeneous signal regions in fMRI images due to the lack of background air signal intensity, improves the repeatability of SNR assessment, and addresses the inherent subjectivity, low repeatability, time-consuming measurement, and inefficiency of traditional medical fMRI image SNR assessment. This technique improves the accuracy, repeatability, and efficiency of medical fMRI image SNR assessment, and has significant application value for the analysis of medical magnetic resonance images, particularly in the absence of background signal in parametric images after fMRI reconstruction. Accurate image quality assessment not only improves diagnostic accuracy and treatment efficacy, but also ensures the reliability of scientific research and the effectiveness of clinical trials, while also supporting the development of automated image processing technology.

[0101] Example 2

[0102] Corresponding to the aforementioned embodiment of a method for evaluating the signal-to-noise ratio of an image, the present disclosure further provides an embodiment of a system for evaluating the signal-to-noise ratio of an image.

[0103] Figure 2 This is a module diagram of an image signal-to-noise ratio evaluation system provided in Example 2 of the present disclosure, such as Figure 2 As shown, the evaluation system includes: a first acquisition module 21, a first processing module 22, a second processing module 23, and a second acquisition module 24;

[0104] A first acquisition module 21 is used to acquire a functional magnetic resonance imaging image before noise removal;

[0105] A first processing module 22 is configured to perform discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain a high-frequency coefficient matrix and a low-frequency coefficient matrix;

[0106] The second processing module 23 is used to perform denoising and inverse discrete wavelet transform on the high-frequency coefficient matrix and the low-frequency coefficient matrix using a soft threshold function to obtain a denoised functional magnetic resonance imaging image;

[0107] The second acquisition module 24 is configured to acquire a signal-to-noise ratio of the functional magnetic resonance imaging based on the functional magnetic resonance imaging image before denoising and the functional magnetic resonance imaging image after denoising.

[0108] In this embodiment, the expression for obtaining the signal-to-noise ratio of the functional magnetic resonance imaging based on the functional magnetic resonance imaging image before denoising and the functional magnetic resonance imaging image after denoising is shown in formula (2) in embodiment 1.

[0109] This embodiment realizes the automated and precise evaluation of the signal-to-noise ratio of functional magnetic resonance imaging images by using the discrete wavelet transform method, thereby improving the evaluation efficiency and accuracy of the signal-to-noise ratio.

[0110] In an optional embodiment, the first processing module is configured to perform a multi-level two-dimensional discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain high-frequency coefficient matrices and low-frequency coefficient matrices at each level;

[0111] The second processing module is used to perform denoising and inverse discrete wavelet transform processing on high-frequency coefficient matrices and low-frequency coefficient matrices at all levels using a soft threshold function to obtain a denoised functional magnetic resonance imaging image.

[0112] In the specific implementation process, firstly, a multi-level two-dimensional discrete wavelet transform (2D Discrete Wavelet Transform, 2D DWT) is performed on the fMRI image to obtain low-frequency coefficient matrices (ACM) and high-frequency coefficient matrices (DCM) at each level respectively; then, the soft threshold function of formula (3) in Example 1 is used to calculate the threshold of each high-frequency coefficient matrix;

[0113] In an optional embodiment, the first processing module includes:

[0114] The first processing unit is used to perform a multi-level two-dimensional discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain a first-level high-frequency coefficient matrix and a first-level low-frequency coefficient matrix;

[0115] The second processing unit is used to perform discrete wavelet transform on the first-level low-frequency coefficient matrix to obtain a second-level high-frequency coefficient matrix and a second-level low-frequency coefficient matrix;

[0116] a third processing unit, configured to perform discrete wavelet transform on the secondary low-frequency coefficient matrix to obtain a tertiary high-frequency coefficient matrix and a tertiary low-frequency coefficient matrix;

[0117] a fourth processing unit, configured to perform discrete wavelet transform on the three-level low-frequency coefficient matrix to obtain an N-level high-frequency coefficient matrix and an N-level low-frequency coefficient matrix;

[0118] Among them, N is greater than 3.

[0119] In an optional embodiment, the second processing module includes:

[0120] A first acquisition unit is used to obtain the median of the noise signal strength corresponding to each level of high-frequency coefficient matrix;

[0121] A second obtaining unit is used to obtain the threshold value of each level of high frequency coefficient matrix based on the median and the preset value;

[0122] A third acquisition unit is used to acquire a target noise signal whose noise signal strength of each level of high-frequency coefficient matrix is less than a preset signal strength;

[0123] a fifth processing unit, configured to perform denoising on the target noise signal using the thresholds of the high-frequency coefficient matrices at each level to obtain denoised high-frequency coefficient matrices at each level;

[0124] The sixth processing unit is used to obtain a denoised functional magnetic resonance imaging image based on the denoised high-frequency coefficient matrices at each level and the N-level low-frequency coefficient matrices obtained after inverse discrete wavelet transform processing.

[0125] In this embodiment, the number of discrete wavelet transform stages is the same as the number of inverse discrete wavelet transform stages. The specific number of stages is determined empirically according to the noise level and signal strength of the image.

[0126] In a specific implementation process, for example, taking N as 4, the functional magnetic resonance imaging image f(k) before denoising is first input and decomposed using a multi-level two-dimensional discrete wavelet transform (2D DWT) to obtain a first-level high-frequency coefficient matrix (CD1) and a first-level low-frequency coefficient matrix (CA1). Then, a second discrete wavelet transform (DWT) is performed on CA1 to decompose the first-level high-frequency coefficient matrix (CD2) and a second-level low-frequency coefficient matrix (CA2). Then, a third discrete wavelet transform (DWT) is performed on CA2 to decompose the first-level high-frequency coefficient matrix (CD3) and a third-level low-frequency coefficient matrix (CA3). Then, a fourth discrete wavelet transform (DWT) is performed on CA3 to decompose the fourth-level high-frequency coefficient matrix (CD4) and a fourth-level low-frequency coefficient matrix (CA4). Then, a soft threshold function is used to remove noise and perform inverse discrete wavelet transform (IDWT) on the first-level high-frequency coefficient matrix (CD1), the second-level high-frequency coefficient matrix (CD2), the third-level high-frequency coefficient matrix (CD3), the fourth-level high-frequency coefficient matrix (CD4), and the fourth-level low-frequency coefficient matrix (CA4), respectively. Specifically, the median of the signal intensity corresponding to each level of the high-frequency coefficient matrix is first found, and then the median is divided by a preset value (for example, the preset value can be 0.6745, or it can be set to other values according to actual conditions) to calculate the median absolute deviation, that is, the threshold of each level of the high-frequency coefficient matrix; then, the threshold of each level of the high-frequency coefficient matrix is used to set the target noise signal portion of the high-frequency coefficient matrix at each level whose noise signal intensity is less than the preset signal intensity to 0 (that is, the threshold of each level of the high-frequency coefficient matrix is used to set the low signal intensity portion of the high-frequency coefficient matrix at each level to 0) to eliminate noise, and then the high-frequency coefficient matrix at each level after noise elimination and the final low-frequency coefficient matrix (for example, the N-level low-frequency coefficient matrix) are used to reconstruct the denoised magnetic resonance functional imaging image Y i .

[0127] Compared to traditional manual SNR measurement methods, the signal-to-noise ratio (SNR) assessment technique based on discrete wavelet transform (DWT) in this embodiment offers the advantages of greater efficiency, accuracy, and repeatability. It addresses the difficulty in selecting homogeneous signal regions in fMRI images due to the lack of background air signal intensity, improves the repeatability of SNR assessment, and addresses the inherent subjectivity, low repeatability, time-consuming measurement, and inefficiency of traditional medical fMRI image SNR assessment. This technique improves the accuracy, repeatability, and efficiency of medical fMRI image SNR assessment, and has significant application value for the analysis of medical magnetic resonance images, particularly in the absence of background signal in parametric images after fMRI reconstruction. Accurate image quality assessment not only improves diagnostic accuracy and treatment efficacy, but also ensures the reliability of scientific research and the effectiveness of clinical trials, while also supporting the development of automated image processing technology.

[0128] Since the system embodiments generally correspond to the method embodiments, reference will be made to the description of the method embodiments for relevant details. The system embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separate, and the components of the units may or may not be physical units, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the disclosed solution.

[0129] Example 3

[0130] Figure 3 This is a structural schematic diagram of an electronic device shown in Example 3 of the present disclosure. The electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor. When the processor executes the computer program, it implements the image signal-to-noise ratio evaluation method described in any of the above embodiments. Figure 3 The electronic device 90 shown is only an example and should not limit the functionality and scope of use of the embodiments of the present disclosure.

[0131] like Figure 3 As shown, the electronic device 90 may be a general-purpose computing device, such as a server device. Components of the electronic device 90 may include, but are not limited to, the at least one processor 91, the at least one memory 92, and a bus 93 connecting different system components (including the memory 92 and the processor 91).

[0132] The bus 93 includes a data bus, an address bus, and a control bus.

[0133] The memory 92 may include a volatile memory, such as a random access memory (RAM) 921 and / or a cache memory 922 , and may further include a read-only memory (ROM) 923 .

[0134] The memory 92 may also include a program tool 925 (or utility) having a set (at least one) of program modules 924, such program modules 924 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include an implementation of a network environment.

[0135] The processor 91 executes various functional applications and data processing by running the computer program stored in the memory 92, such as the image signal-to-noise ratio evaluation method provided in any of the above embodiments.

[0136] The electronic device 90 can also communicate with one or more external devices 94 (e.g., keyboards, pointing devices, etc.). Such communication can be performed through an input / output (I / O) interface 95. In addition, the electronic device 90 can also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 96. Figure 3 As shown, the network adapter 96 communicates with other modules of the electronic device 90 via the bus 93. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 90, including but not limited to microcode, device drivers, redundant processors, external disk drive arrays, RAID (RAID) systems, tape drives, and data backup storage systems.

[0137] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0138] Example 4

[0139] Embodiment 4 of the present disclosure further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the image signal-to-noise ratio evaluation method provided in any of the above embodiments.

[0140] The readable storage medium may include, but is not limited to, a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0141] Example 5

[0142] Embodiment 5 of the present disclosure further provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-mentioned methods for evaluating the signal-to-noise ratio of an image.

[0143] The program code for executing the computer program product of the present disclosure may be written in any combination of one or more programming languages, and the program code may be executed entirely on the user device, partially on the user device, as a standalone software package, partially on the user device and partially on a remote device, or entirely on the remote device.

[0144] While specific embodiments of the present disclosure have been described above, those skilled in the art will appreciate that these are merely illustrative and that the scope of protection of the present disclosure is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present disclosure, and such changes and modifications are intended to fall within the scope of protection of the present disclosure.

Claims

1. A method for evaluating the signal-to-noise ratio of an image, characterized in that: The signal-to-noise ratio evaluation method comprises: Acquire functional magnetic resonance imaging images before denoising; Performing discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain a high-frequency coefficient matrix and a low-frequency coefficient matrix; performing denoising and inverse discrete wavelet transform processing on the high-frequency coefficient matrix and the low-frequency coefficient matrix using a soft threshold function to obtain a denoised functional magnetic resonance imaging image; A signal-to-noise ratio of functional magnetic resonance imaging is acquired based on the functional magnetic resonance imaging image before denoising and the functional magnetic resonance imaging image after denoising.

2. The method for evaluating the signal-to-noise ratio of an image according to claim 1, wherein: The step of performing discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain a high-frequency coefficient matrix and a low-frequency coefficient matrix comprises: Performing a multi-level two-dimensional discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain high-frequency coefficient matrices and low-frequency coefficient matrices at each level; The step of performing denoising and inverse discrete wavelet transform processing on the high-frequency coefficient matrix and the low-frequency coefficient matrix using a soft threshold function to obtain a denoised functional magnetic resonance imaging image comprises: The high-frequency coefficient matrices at each level and the low-frequency coefficient matrices at each level are subjected to denoising and discrete wavelet inverse transformation processing by using a soft threshold function to obtain a denoised functional magnetic resonance imaging image.

3. The method for evaluating the signal-to-noise ratio of an image according to claim 2, wherein: The step of performing a multi-level two-dimensional discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain high-frequency coefficient matrices and low-frequency coefficient matrices at each level comprises: Performing a multi-level two-dimensional discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain a first-level high-frequency coefficient matrix and a first-level low-frequency coefficient matrix; Performing discrete wavelet transform on the primary low-frequency coefficient matrix to obtain a secondary high-frequency coefficient matrix and a secondary low-frequency coefficient matrix; Performing discrete wavelet transform on the secondary low-frequency coefficient matrix to obtain a tertiary high-frequency coefficient matrix and a tertiary low-frequency coefficient matrix; Performing discrete wavelet transform on the three-level low-frequency coefficient matrix to obtain an N-level high-frequency coefficient matrix and an N-level low-frequency coefficient matrix; Among them, N is greater than 3.

4. The method for evaluating the signal-to-noise ratio of an image according to claim 3, wherein: The step of performing denoising and inverse discrete wavelet transform processing on the high-frequency coefficient matrices and the low-frequency coefficient matrices of each level using a soft threshold function to obtain a denoised functional magnetic resonance imaging image comprises: Obtaining the median of the noise signal strength corresponding to the high-frequency coefficient matrix at each level; Obtaining thresholds of the high-frequency coefficient matrices at each level based on the median and a preset value; Obtaining a target noise signal whose noise signal strength of each level of high frequency coefficient matrix is less than a preset signal strength; denoising the target noise signal using the thresholds of the high-frequency coefficient matrices at each level to obtain denoised high-frequency coefficient matrices at each level; A denoised functional magnetic resonance imaging image is acquired based on the denoised high-frequency coefficient matrices at each level and the N-level low-frequency coefficient matrices obtained after inverse discrete wavelet transform processing.

5. The method for evaluating the signal-to-noise ratio of an image according to claim 1, wherein: The expression for obtaining the signal-to-noise ratio of the functional magnetic resonance imaging based on the functional magnetic resonance imaging image before denoising and the functional magnetic resonance imaging image after denoising is: Where WSNR represents the signal-to-noise ratio of functional magnetic resonance imaging, N represents the noise signal or the total number of samples or pixels in the image, and X i Y represents the functional magnetic resonance imaging image before denoising. i The image shows the denoised functional magnetic resonance imaging.

6. The method for evaluating the signal-to-noise ratio of an image according to claim 1, wherein: The number of stages of the discrete wavelet transform is the same as the number of stages of the inverse discrete wavelet transform.

7. A system for evaluating the signal-to-noise ratio of an image, characterized in that: The signal-to-noise ratio evaluation system comprises: A first acquisition module is used to acquire a functional magnetic resonance imaging image before noise removal; A first processing module is configured to perform discrete wavelet transform on the functional magnetic resonance imaging image before denoising to obtain a high-frequency coefficient matrix and a low-frequency coefficient matrix; a second processing module, configured to perform denoising and inverse discrete wavelet transform processing on the high-frequency coefficient matrix and the low-frequency coefficient matrix using a soft threshold function to obtain a denoised functional magnetic resonance imaging image; The second acquisition module is configured to acquire a signal-to-noise ratio of the functional magnetic resonance imaging based on the functional magnetic resonance imaging image before denoising and the functional magnetic resonance imaging image after denoising.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and configured to run on the processor, wherein: When the processor executes the computer program, the method for evaluating the signal-to-noise ratio of an image according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for evaluating the signal-to-noise ratio of an image according to any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for evaluating the signal-to-noise ratio of an image according to any one of claims 1 to 6 is implemented.

Images