CN117788461B - Magnetic resonance image quality evaluation system based on image analysis - Google Patents
Magnetic resonance image quality evaluation system based on image analysis Download PDFInfo
- Publication number
- CN117788461B CN117788461B CN202410203251.3A CN202410203251A CN117788461B CN 117788461 B CN117788461 B CN 117788461B CN 202410203251 A CN202410203251 A CN 202410203251A CN 117788461 B CN117788461 B CN 117788461B
- Authority
- CN
- China
- Prior art keywords
- value
- image
- matrix
- module
- sub
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000013441 quality evaluation Methods 0.000 title claims abstract description 27
- 238000010191 image analysis Methods 0.000 title claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 31
- 238000000034 method Methods 0.000 claims abstract description 24
- 230000005540 biological transmission Effects 0.000 claims abstract description 17
- 238000001303 quality assessment method Methods 0.000 claims abstract 4
- 239000011159 matrix material Substances 0.000 claims description 68
- 238000011156 evaluation Methods 0.000 claims description 29
- 230000011218 segmentation Effects 0.000 claims description 29
- 238000012549 training Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 3
- 238000004088 simulation Methods 0.000 claims description 2
- 238000011166 aliquoting Methods 0.000 claims 1
- 210000004185 liver Anatomy 0.000 abstract description 5
- 238000002595 magnetic resonance imaging Methods 0.000 abstract description 4
- 230000000694 effects Effects 0.000 description 4
- 238000003745 diagnosis Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 208000019423 liver disease Diseases 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Landscapes
- Magnetic Resonance Imaging Apparatus (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a magnetic resonance image quality evaluation system based on image analysis, which belongs to the technical field of liver magnetic resonance imaging quality evaluation and comprises a platform end, a user end and a server; the user terminal comprises an acquisition module, a confidentiality module and a transmission module; the acquisition module is used for acquiring a target image and generating a reference image corresponding to the target image; the method comprises the steps that a target image and a reference image are sent to a confidentiality module for confidentiality processing of the target image and the reference image, and the processed target image and the processed reference image are sent to a platform end through a transmission module; the transmission module is used for data transmission between the user terminal and the platform terminal; the platform end comprises a first sub-item module, a second sub-item module and a quality evaluation module; the first sub-term module is used for analyzing the target image and the reference image and calculating a first sub-term value; the second sub-term module is used for analyzing the target image and calculating a second sub-term value; the quality assessment module is used for assessing the quality score of the target image.
Description
Technical Field
The invention belongs to the technical field of liver magnetic resonance imaging quality evaluation, and particularly relates to a magnetic resonance image quality evaluation system based on image analysis.
Background
Liver magnetic resonance imaging is a noninvasive, non-radiative imaging examination method, and is widely applied to diagnosis and monitoring of liver diseases. However, liver MRI presents several technical challenges in practical applications, particularly the interference of various noise and artifacts. These include, but are not limited to, motion artifacts, metal artifacts, and image distortions caused by magnetic field inhomogeneities. These problems can severely impact the quality of the image, making diagnosis and disease assessment difficult for the physician. The real-time and dynamic evaluation of the imaging quality of magnetic resonance images is the focus of the quality control work of the radiology department of hospitals. Currently, hospitals rely primarily on specialized radiologists or reviewers to evaluate images. They mainly perform subjective scoring according to factors such as signal-to-noise ratio, definition, detail performance and the like of the image. While this method can evaluate image quality, it involves great subjectivity and requires a lot of human resources. Furthermore, such subjective assessment methods may lead to consistency differences between the assessment results.
Therefore, based on the invention, a magnetic resonance image quality evaluation system based on image analysis is provided.
Disclosure of Invention
In order to solve the problems of the scheme, the invention provides a magnetic resonance image quality evaluation system based on image analysis.
The aim of the invention can be achieved by the following technical scheme:
a magnetic resonance image quality evaluation system based on image analysis comprises a platform end, a user end and a server; the platform end, the user end and the server are in communication connection;
the user terminal comprises an acquisition module, a confidentiality module and a transmission module;
The acquisition module is used for acquiring a corresponding magnetic resonance image, marking the magnetic resonance image as a target image and generating a reference image corresponding to the target image; and sending the target image and the reference image to a security module.
The security module is used for performing security processing on the received target image and the reference image, and sending the processed target image and reference image to the platform end through the transmission module.
Further, the method for performing security processing on the target image and the reference image includes:
Marking the received target image and reference image as images to be processed; presetting a secret item and corresponding identification characteristics; setting a secret processing mode corresponding to each secret item;
Identifying the image to be processed according to each secret item and the corresponding identification characteristic, and identifying the corresponding privacy information; matching corresponding secret processing modes for each piece of private information, and processing the private information according to the obtained secret processing modes; and obtaining the target image and the reference image after the security processing.
The transmission module is used for data transmission between the user terminal and the platform terminal.
The platform end comprises a first sub-item module, a second sub-item module and a quality evaluation module;
the first sub-term module is used for comparing and analyzing the target image and the reference image and calculating a corresponding first sub-term value.
Further, the method for calculating the first component value includes:
performing equivalent processing and gray scale processing on the target image and the reference image; obtaining a corresponding target gray level image and a reference gray level image;
Recognizing gray values of pixels in the target gray image and the reference gray image; generating a corresponding target gray matrix and a reference gray matrix according to the gray value of each pixel and the corresponding position; marking the target gray matrix and the reference gray matrix as respectively And/>; I=1, 2, … …, n being a positive integer; j=1, 2, … …, m being a positive integer;
Calculating a corresponding difference matrix according to a formula R 3=R1-R2;
Wherein: r 1 is a target gray matrix; r 2 is a reference gray matrix; r 3 is a difference matrix;
According to the formula Calculating a corresponding first sub-term value; wherein: SUB value is the first SUB-term value; the R 3||F represents the Frobenius norm of the matrix R 3.
The second sub-term module is used for analyzing the target image and calculating a corresponding second sub-term value.
Further, the calculation method of the second component value includes:
Obtaining a target image, identifying the size of the target image, and equally dividing the target image according to the identified size of the target image to obtain K unit blocks, wherein K ' in a formula K ' (K 0.5) is a positive integer, and K ' is more than or equal to 2 and less than or equal to 10; k is a positive integer;
Identifying a contrast representative value, a resolution representative value and a brightness representative value corresponding to each unit block; generating a corresponding contrast matrix, a resolution matrix and a brightness matrix based on the positions of the unit blocks; the contrast matrix, the resolution matrix and the brightness matrix are all k' -order square matrixes;
calculating a corresponding merging matrix according to a formula R s=b1×R4+b2×R5+b3×R6;
Wherein: r s is a merge matrix; r 4 is a contrast matrix; r 5 is a resolution matrix; r 6 is a luminance matrix; b 1、b2、b3 are all proportional coefficients, and the value range is 0<b 1≤1,0<b2≤1,0<b3 to be less than or equal to 1;
Marking a merge matrix as ; The elements in the merge matrix are labeled s qp; q=1, 2, … …, k 'and k' is a positive integer; p=1, 2, … …, k 'with k' being a positive integer;
According to the formula Calculating a corresponding second component value;
Wherein: SRB value is the second component value.
Further, the method for equally dividing the target image comprises the following steps:
performing simulation segmentation on the target image according to the value range of k', obtaining 9 groups of segmented images, and marking a single image in each group of segmented images as a unit block;
Evaluating contrast representative values, resolution representative values and brightness representative values corresponding to the unit blocks in each group of the divided images; when any contrast representative value, resolution representative value and brightness representative value are 0 in each group of divided images, eliminating the divided images of the corresponding group;
calculating the segmentation evaluation value of each group of segmentation images through a preset segmentation evaluation formula, identifying the segmentation image corresponding to the highest segmentation evaluation value, and segmenting the target image according to the segmentation mode corresponding to the segmentation image.
Further, the evaluation method of the contrast representative value includes:
Identifying the value of K corresponding to the segmented image, and equally dividing each unit block according to K to obtain K fine blocks; identifying the contrast corresponding to each subdivision block, and marking the subdivision blocks as single contrast;
According to the formula Calculating a contrast representative value of the corresponding unit block;
Wherein: CON t is the contrast representative value of the corresponding cell block; t represents a corresponding cell block, t=1, 2, … …, K; v represents the corresponding subdivision block, v=1, 2, … …, K; DC tv represents the single contrast of the corresponding subdivision block; g (A) is a judgment model, and A is the area of the subdivision block.
Further, the resolution representative value obtaining method includes:
Identifying the value of K corresponding to the segmented image, and equally dividing each unit block according to K to obtain K fine blocks; identifying the corresponding resolution of each subdivision block, and marking the subdivision block as single resolution;
According to the formula Calculating a resolution representative value of the corresponding unit block;
wherein: RE t is the resolution representative value of the corresponding unit block; t represents a corresponding cell block, t=1, 2, … …, K; v represents the corresponding subdivision block, v=1, 2, … …, K; DR tv represents the single resolution of the corresponding subdivision block; g (A) is a judgment model, and A is the area of the subdivision block.
Further, the method for obtaining the luminance representative value includes:
Identifying the value of K corresponding to the segmented image, and equally dividing each unit block according to K to obtain K fine blocks; identifying the brightness corresponding to each subdivision block, and marking the brightness as single brightness; obtaining corresponding reference brightness, and calculating the brightness rate of the corresponding sub-blocks according to the brightness rate = single brightness/reference brightness;
According to the formula Calculating a brightness representative value of the corresponding unit block;
Wherein: BRIGH t is the luminance representative value of the corresponding cell block; t represents a corresponding cell block, t=1, 2, … …, K; v represents the corresponding subdivision block, v=1, 2, … …, K; DB tv represents a single luminance of the corresponding sub-block; g (A) is a judgment model, and A is the area of the subdivision block.
Further, the expression of the judgment model is as follows。
Further, the segmentation evaluation formula is as follows;
Wherein: a sse is a segmentation evaluation value; b 1、b2、b3 are all proportional coefficients, and the value range is 0<b 1≤1,0<b2≤1,0<b3 to be less than or equal to 1; t represents a corresponding cell block, t=1, 2, … …, K; CON t is the contrast representative value of the corresponding cell block; RE t is the resolution representative value of the corresponding unit block; BRIGH t is a luminance representative value of the corresponding cell block.
The quality evaluation module is used for evaluating the quality score of the target image, acquiring a first polynomial value and a second polynomial value, and calculating a corresponding comprehensive value according to a formula P rior=b4×SUBvalue+b5×SRBvalue;
Wherein: p rior is a comprehensive value; SUB value is the first SUB-term value; SRB value is the second component value; b 4、b5 are all proportional coefficients, and the value range is 0<b 4≤1,0<b5 to be less than or equal to 1;
acquiring corresponding training data, and analyzing the training data to acquire a corresponding comprehensive value set; forming a plurality of scoring coordinates according to quality scoring and a comprehensive value set corresponding to the training data; setting a corresponding evaluation function according to each grading coordinate;
substituting the integrated value into an evaluation function to calculate a corresponding quality score; and sending the quality evaluation to a user side.
Compared with the prior art, the invention has the beneficial effects that:
Through the mutual coordination among the modules, the intelligent evaluation of the magnetic resonance image is realized, the corresponding quality score is obtained, and the effect of resource investment can be exerted to the greatest extent through the separation between the platform end and the user end, so that the cost investment of a hospital is greatly reduced; one-to-many is realized, and services are provided for a plurality of household hospitals; moreover, the platform side is used for management and operation, so that more specialized staff can ensure the normal operation of the system; the privacy module is arranged to ensure the privacy of the patient and prevent the leakage of the privacy of the patient; the advantages and disadvantages of the image processing algorithm can be known by evaluating the quality of the liver magnetic resonance image, the image processing flow is further optimized, and the image processing effect and efficiency are improved; meanwhile, a more accurate and reliable diagnosis basis can be provided for doctors, and the doctors are helped to make more scientific and reasonable clinical decisions.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive effort to a person skilled in the art.
FIG. 1 is a schematic block diagram of the present invention;
FIG. 2 is a flow chart of the method of the present invention.
Detailed Description
The technical solutions of the present invention will be clearly and completely described in connection with the embodiments, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in fig. 1 to 2, a magnetic resonance image quality evaluation system based on image analysis includes a platform end, a user end and a server; the platform end, the user end and the server are in communication connection.
The platform end is used for application, management and maintenance of the platform side.
The user side is used by each registered hospital user.
In order to realize the quality evaluation of the magnetic resonance image, corresponding equipment and software are required to be matched, and for the needs and the reconstruction of hospitals, if the hospitals are independently put into independent use, the effects of all input resources are difficult to be exerted, and the resource waste occurs; therefore, the relationship between the platform side and the user is adopted for management, the utilization effect of input resources can be exerted to the greatest extent, and the cost input of hospitals is greatly reduced; one-to-many is realized, and services are provided for a plurality of household hospitals; and the platform side is used for management and operation, so that the system can be ensured to operate normally by more professional staff.
The user terminal comprises an acquisition module, a confidentiality module and a transmission module;
The acquisition module is used for acquiring corresponding magnetic resonance images when quality evaluation is required, marking the corresponding magnetic resonance images as target images and generating reference images of the target images; and sending the obtained target image and the reference image to a security module.
The reference image refers to a clear, noiseless and undistorted magnetic resonance image similar to the target image, a corresponding image set can be acquired when the target image is acquired, and the collected image is selected for preprocessing, including denoising, correction, normalization and other operations, so as to ensure the quality and stability of the image; a high quality raw magnetic resonance image is selected from the preprocessed images as a reference image. Or other existing means may be used to obtain the reference image.
The security module is used for performing security processing on the received target image and the reference image, and marking the received target image and the received reference image as images to be processed; presetting a secret item and corresponding identification characteristics; namely, the magnetic resonance image possibly comprises data items of privacy information of a patient, and the identification characteristics of the data items in the image are combined, so that the identification characteristics corresponding to the secret items are set;
presetting a corresponding secret processing mode of each secret item, such as various secret modes of deleting corresponding information, adapting information, replacing codes and the like; the method which does not correspond to the quality evaluation of the image to be processed can be selected as a secret processing method according to the requirement.
Identifying the image to be processed according to each preset secret item and the corresponding identification characteristic, and identifying the corresponding privacy information; matching corresponding privacy processing modes for each privacy information, and processing the corresponding privacy information according to the obtained privacy processing modes; the target image and the reference image after the security processing are obtained, and the processed target image and reference image are sent to the platform end through the transmission module.
The transmission module is used for data transmission between the user terminal and the platform terminal.
The platform end comprises a first sub-item module, a second sub-item module and a quality evaluation module;
the first sub-term module is used for comparing and analyzing the target image and the reference image and calculating a corresponding first sub-term value.
The specific calculation method is as follows:
performing equivalent processing and gray processing on the target image and the reference image; wherein the equivalent processing is to process the target image and the reference image to the same size and resolution.
Obtaining a target gray image and a reference gray image which respectively correspond to the processed target image and the reference image;
recognizing gray values of pixels in the target gray image and the reference gray image; generating a corresponding target gray matrix and a corresponding reference gray matrix according to the gray value corresponding to each pixel and the corresponding position; marking the obtained target gray matrix and the reference gray matrix as respectively And/>; I=1, 2, … …, n being a positive integer; j=1, 2, … …, m being a positive integer;
calculating a corresponding difference matrix according to a formula R 3=R1-R2; wherein: r 1 is a target gray matrix; r 2 is a reference gray matrix; r 3 is a difference matrix;
According to the formula Calculating a corresponding first sub-term value; wherein: SUB value is the first SUB-term value; the R 3||F represents the Frobenius norm of the matrix R 3.
The second sub-term module is used for analyzing the target image and calculating a corresponding second sub-term value. The process is as follows:
Obtaining a target image, identifying the size of the target image, and dividing the target image according to the identified size of the target image to obtain K unit blocks, wherein K '= K 0.5 is a positive integer, and K' is more than or equal to 2 and less than or equal to 10;
Identifying a contrast representative value, a resolution representative value and a brightness representative value corresponding to each unit block; generating a corresponding contrast matrix, a resolution matrix and a brightness matrix based on the positions of the unit blocks; the positions of the unit blocks are used as corresponding element bits in the matrix, and corresponding contrast representative values, resolution representative values or brightness representative values are filled to form a corresponding matrix; the contrast matrix, the resolution matrix and the brightness matrix are all k' -order square matrixes; labeled R 4、R5、R6 respectively;
Calculating a corresponding merging matrix according to a formula R s=b1×R4+b2×R5+b3×R6; wherein: r s is a merge matrix; r 4 is a contrast matrix; r 5 is a resolution matrix; r 6 is a luminance matrix; b 1、b2、b3 are all proportional coefficients, and the value range is 0<b 1≤1,0<b2≤1,0<b3 to be less than or equal to 1; the same as the b 1、b2、b3 scaling factor in the segmentation evaluation formula.
Marking a merge matrix as; The elements in the merge matrix are labeled s qp; q=1, 2, … …, k 'and k' is a positive integer; p=1, 2, … …, k 'with k' being a positive integer;
According to the formula Calculating a corresponding second component value;
Wherein: SRB value is the second component value.
The method for segmenting the target image comprises the following steps:
Setting a segmentation evaluation formula, wherein the segmentation evaluation formula is as follows ;
Wherein: a sse is a segmentation evaluation value; b 1、b2、b3 are all proportional coefficients, and the value range is 0<b 1≤1,0<b2≤1,0<b3 to be less than or equal to 1; t represents a corresponding cell block, t=1, 2, … …, K; CON t is the contrast representative value of the corresponding cell block; RE t is the resolution representative value of the corresponding unit block; BRIGH t is a luminance representative value of the corresponding cell block.
Sequentially dividing the target image according to the k' -value range, namely sequentially dividing the target image according to the numbers of 2 2、32、42, … … and the like;
obtaining 9 groups of component images, and marking single images in each group of component images as unit blocks; and evaluating the segmentation evaluation values corresponding to the segmentation images by a segmentation evaluation formula, and selecting the value of K corresponding to the highest segmentation evaluation value for segmentation.
The method for obtaining the contrast representative value comprises the following steps:
identifying the value of K corresponding to the group of the divided images, and equally dividing each unit block according to K to obtain K fine divided blocks; identifying the contrast corresponding to each subdivision block, marking the subdivision block as single contrast, and according to the formula Calculating a contrast representative value of the corresponding unit block; wherein: CON t is the contrast representative value of the corresponding cell block; t represents a corresponding cell block, t=1, 2, … …, K; v represents the corresponding subdivision block, v=1, 2, … …, K; DC tv represents the single contrast of the corresponding subdivision block; g (A) is a judgment model, and A is the area of the fine block; the expression of the judgment model is/>。
The resolution representative value acquisition method comprises the following steps:
Identifying the value of K corresponding to the group of the divided images, and equally dividing each unit block according to K to obtain K fine divided blocks; identifying the corresponding resolution of each subdivision block, marking the resolution as single resolution, and according to the formula Calculating a resolution representative value of the corresponding unit block; wherein: RE t is the resolution representative value of the corresponding unit block; t represents a corresponding cell block, t=1, 2, … …, K; v represents the corresponding subdivision block, v=1, 2, … …, K; DR tv represents the single resolution of the corresponding subdivision block; g (A) is a judgment model, and A is the area of the subdivision block.
The method for acquiring the brightness representative value comprises the following steps:
Identifying the value of K corresponding to the group of the divided images, and equally dividing each unit block according to K to obtain K fine divided blocks; identifying the brightness corresponding to each subdivision block, and marking the brightness as single brightness; obtaining the highest brightness possibly occurring in the magnetic resonance image, marking the highest brightness as reference brightness, and calculating the brightness rate corresponding to the subdivision block according to the formula brightness rate = single brightness/reference brightness;
According to the formula Calculating a brightness representative value of the corresponding unit block; wherein: BRIGH t is the luminance representative value of the corresponding cell block; t represents a corresponding cell block, t=1, 2, … …, K; v represents the corresponding subdivision block, v=1, 2, … …, K; DB tv represents a single luminance of the corresponding sub-block; g (A) is a judgment model, and A is the area of the subdivision block.
The quality evaluation module is used for evaluating the quality score of the target image according to the first and second score values, obtaining the first and second score values, and calculating the corresponding comprehensive value according to a formula P rior=b4×SUBvalue+b5×SRBvalue; wherein: p rior is a comprehensive value; SUB value is the first SUB-term value; SRB value is the second component value; b 4、b5 are all proportional coefficients, and the value range is 0<b 4≤1,0<b5 to be less than or equal to 1;
Acquiring a large number of historical magnetic resonance images, namely, marking the historical magnetic resonance images scored by professionals as training data; analyzing according to the steps to obtain a plurality of historical integrated values, and integrating the historical integrated values and the corresponding quality scores into corresponding score coordinates; after a large number of integration, a plurality of scoring coordinates are obtained, the obtained scoring coordinates are input into a coordinate system, the horizontal axis is a comprehensive value, the vertical axis is a quality score, adjacent scoring coordinates in the coordinate system are connected to form a corresponding evaluation curve, and the evaluation curve is fitted to obtain a corresponding evaluation function;
Inputting the obtained comprehensive value into an evaluation function, and outputting a corresponding quality score;
And sending the obtained quality scores to a transmission module.
The above formulas are all formulas with dimensions removed and numerical values calculated, the formulas are formulas which are obtained by acquiring a large amount of data and performing software simulation to obtain the closest actual situation, and preset parameters and preset thresholds in the formulas are set by a person skilled in the art according to the actual situation or are obtained by simulating a large amount of data.
The above embodiments are only for illustrating the technical method of the present invention and not for limiting the same, and it should be understood by those skilled in the art that the technical method of the present invention may be modified or substituted without departing from the spirit and scope of the technical method of the present invention.
Claims (8)
1. The magnetic resonance image quality evaluation system based on image analysis is characterized by comprising a platform end, a user end and a server; the platform end, the user end and the server are in communication connection;
the user terminal comprises an acquisition module, a confidentiality module and a transmission module;
the acquisition module is used for acquiring a corresponding magnetic resonance image, marking the magnetic resonance image as a target image and generating a reference image corresponding to the target image; transmitting the target image and the reference image to a security module;
the security module is used for performing security processing on the received target image and the reference image, and sending the processed target image and reference image to the platform end through the transmission module;
The transmission module is used for data transmission between the user terminal and the platform terminal;
The platform end comprises a first sub-item module, a second sub-item module and a quality evaluation module;
the first sub-term module is used for comparing and analyzing the target image and the reference image and calculating a corresponding first sub-term value;
the second sub-term module is used for analyzing the target image and calculating a corresponding second sub-term value;
the quality evaluation module is used for evaluating the quality score of the target image, acquiring a first polynomial value and a second polynomial value, and calculating a corresponding comprehensive value according to a formula P rior=b4×SUBvalue+b5×SRBvalue;
Wherein: p rior is a comprehensive value; SUB value is the first SUB-term value; SRB value is the second component value; b 4、b5 are all proportional coefficients, and the value range is 0<b 4≤1,0<b5 to be less than or equal to 1;
acquiring corresponding training data, and analyzing the training data to acquire a corresponding comprehensive value set; forming a plurality of scoring coordinates according to quality scoring and a comprehensive value set corresponding to the training data; setting a corresponding evaluation function according to each grading coordinate;
Substituting the integrated value into an evaluation function to calculate a corresponding quality score; the quality evaluation is sent to a user side;
The calculation method of the first component value comprises the following steps:
performing equivalent processing and gray scale processing on the target image and the reference image; obtaining a corresponding target gray level image and a reference gray level image;
Recognizing gray values of pixels in the target gray image and the reference gray image; generating a corresponding target gray matrix and a reference gray matrix according to the gray value of each pixel and the corresponding position; marking the target gray matrix and the reference gray matrix as respectively And/>; N and m are positive integers;
Calculating a corresponding difference matrix according to a formula R 3=R1-R2;
Wherein: r 1 is a target gray matrix; r 2 is a reference gray matrix; r 3 is a difference matrix;
According to the formula Calculating a corresponding first sub-term value; wherein: SUB value is the first SUB-term value; the R 3||F represents the Frobenius norm of the matrix R 3;
the calculation method of the second component value comprises the following steps:
Obtaining a target image, identifying the size of the target image, and equally dividing the target image according to the identified size of the target image to obtain K unit blocks, wherein K ' in a formula K ' (K 0.5) is a positive integer, and K ' is more than or equal to 2 and less than or equal to 10; k is a positive integer;
Identifying a contrast representative value, a resolution representative value and a brightness representative value corresponding to each unit block; generating a corresponding contrast matrix, a resolution matrix and a brightness matrix based on the positions of the unit blocks; the contrast matrix, the resolution matrix and the brightness matrix are all k' -order square matrixes;
calculating a corresponding merging matrix according to a formula R s=b1×R4+b2×R5+b3×R6;
Wherein: r s is a merge matrix; r 4 is a contrast matrix; r 5 is a resolution matrix; r 6 is a luminance matrix; b 1、b2、b3 are all proportional coefficients, and the value range is 0<b 1≤1,0<b2≤1,0<b3 to be less than or equal to 1;
Marking a merge matrix as ; The elements in the merge matrix are labeled s qp; q=1, 2, … …, k 'and k' is a positive integer; p=1, 2, … …, k 'with k' being a positive integer;
According to the formula Calculating a corresponding second component value;
Wherein: SRB value is the second component value.
2. The image analysis-based magnetic resonance image quality assessment system according to claim 1, wherein the method of performing security processing on the target image and the reference image comprises:
Marking the received target image and reference image as images to be processed; presetting a secret item and corresponding identification characteristics; setting a secret processing mode corresponding to each secret item;
Identifying the image to be processed according to each secret item and the corresponding identification characteristic, and identifying the corresponding privacy information; matching corresponding secret processing modes for each piece of private information, and processing the private information according to the obtained secret processing modes; and obtaining the target image and the reference image after the security processing.
3. The image analysis-based magnetic resonance image quality assessment system of claim 1, wherein the method of aliquoting the target image comprises:
performing simulation segmentation on the target image according to the value range of k', obtaining 9 groups of segmented images, and marking a single image in each group of segmented images as a unit block;
Evaluating contrast representative values, resolution representative values and brightness representative values corresponding to the unit blocks in each group of the divided images; when any contrast representative value, resolution representative value and brightness representative value are 0 in each group of divided images, eliminating the divided images of the corresponding group;
calculating the segmentation evaluation value of each group of segmentation images through a preset segmentation evaluation formula, identifying the segmentation image corresponding to the highest segmentation evaluation value, and segmenting the target image according to the segmentation mode corresponding to the segmentation image.
4. A magnetic resonance image quality evaluation system based on image analysis according to claim 3, characterized in that the evaluation method of the contrast representative value comprises:
Identifying the value of K corresponding to the segmented image, and equally dividing each unit block according to K to obtain K fine blocks; identifying the contrast corresponding to each subdivision block, and marking the subdivision blocks as single contrast;
According to the formula Calculating a contrast representative value of the corresponding unit block;
Wherein: CON t is the contrast representative value of the corresponding cell block; t represents a corresponding cell block, t=1, 2, … …, K; v represents the corresponding subdivision block, v=1, 2, … …, K; DC tv represents the single contrast of the corresponding subdivision block; g (A) is a judgment model, and A is the area of the subdivision block.
5. A magnetic resonance image quality evaluation system based on image analysis according to claim 3, wherein the resolution representative value acquisition method comprises:
Identifying the value of K corresponding to the segmented image, and equally dividing each unit block according to the value to obtain K fine blocks; identifying the corresponding resolution of each subdivision block, and marking the subdivision block as single resolution;
According to the formula Calculating a resolution representative value of the corresponding unit block;
wherein: RE t is the resolution representative value of the corresponding unit block; t represents a corresponding cell block, t=1, 2, … …, K; v represents the corresponding subdivision block, v=1, 2, … …, K; DR tv represents the single resolution of the corresponding subdivision block; g (A) is a judgment model, and A is the area of the subdivision block.
6. A magnetic resonance image quality evaluation system based on image analysis according to claim 3, wherein the method for acquiring the luminance representative value comprises:
Identifying the value of K corresponding to the segmented image, and equally dividing each unit block according to K to obtain K fine blocks; identifying the brightness corresponding to each subdivision block, and marking the brightness as single brightness; obtaining corresponding reference brightness, and calculating the brightness rate of the corresponding sub-blocks according to the brightness rate = single brightness/reference brightness;
According to the formula Calculating a brightness representative value of the corresponding unit block;
Wherein: BRIGH t is the luminance representative value of the corresponding cell block; t represents a corresponding cell block, t=1, 2, … …, K; v represents the corresponding subdivision block, v=1, 2, … …, K; DB tv represents a single luminance of the corresponding sub-block; g (A) is a judgment model, and A is the area of the subdivision block.
7. A magnetic resonance image quality assessment system based on image analysis according to any one of claims 4, 5 or 6, wherein the judgment model is represented by the formula。
8. A magnetic resonance image quality evaluation system based on image analysis according to claim 3, characterized in that the segmentation evaluation formula is;
Wherein: a sse is a segmentation evaluation value; b 1、b2、b3 are all proportional coefficients, and the value range is 0<b 1≤1,0<b2≤1,0<b3 to be less than or equal to 1; t represents a corresponding cell block, t=1, 2, … …, K; CON t is the contrast representative value of the corresponding cell block; RE t is the resolution representative value of the corresponding unit block; BRIGH t is a luminance representative value of the corresponding cell block.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410203251.3A CN117788461B (en) | 2024-02-23 | 2024-02-23 | Magnetic resonance image quality evaluation system based on image analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410203251.3A CN117788461B (en) | 2024-02-23 | 2024-02-23 | Magnetic resonance image quality evaluation system based on image analysis |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117788461A CN117788461A (en) | 2024-03-29 |
CN117788461B true CN117788461B (en) | 2024-05-07 |
Family
ID=90383866
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410203251.3A Active CN117788461B (en) | 2024-02-23 | 2024-02-23 | Magnetic resonance image quality evaluation system based on image analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117788461B (en) |
Citations (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102543793A (en) * | 2012-02-29 | 2012-07-04 | 无锡睿当科技有限公司 | Wafer focusing image quality feedback system and method therefor |
CN104636804A (en) * | 2013-11-07 | 2015-05-20 | 大连东方之星信息技术有限公司 | Data analysis system |
CN105678775A (en) * | 2016-01-13 | 2016-06-15 | 福州大学 | Color correction assessment method based on machine learning |
CN107578403A (en) * | 2017-08-22 | 2018-01-12 | 浙江大学 | The stereo image quality evaluation method of binocular view fusion is instructed based on gradient information |
CN108319911A (en) * | 2018-01-30 | 2018-07-24 | 深兰科技(上海)有限公司 | Biometric identity certification and payment system based on the identification of hand arteries and veins and identity identifying method |
CN109146856A (en) * | 2018-08-02 | 2019-01-04 | 深圳市华付信息技术有限公司 | Picture quality assessment method, device, computer equipment and storage medium |
WO2019047949A1 (en) * | 2017-09-08 | 2019-03-14 | 众安信息技术服务有限公司 | Image quality evaluation method and image quality evaluation system |
CN110491503A (en) * | 2019-08-21 | 2019-11-22 | 山东大学第二医院 | A kind of cholelithiasis intelligent assistance system based on deep learning |
CN110895802A (en) * | 2018-08-23 | 2020-03-20 | 杭州海康威视数字技术股份有限公司 | Image processing method and device |
CN111858746A (en) * | 2020-05-27 | 2020-10-30 | 武汉瞬付科技有限公司 | Personal data storage system based on cloud platform |
CN112215833A (en) * | 2020-10-22 | 2021-01-12 | 江苏云从曦和人工智能有限公司 | Image quality evaluation method, device and computer readable storage medium |
CN112741620A (en) * | 2020-12-30 | 2021-05-04 | 华南理工大学 | Cervical spondylosis evaluation device based on limb movement |
CN112801132A (en) * | 2020-12-28 | 2021-05-14 | 泰康保险集团股份有限公司 | Image processing method and device |
CN113034489A (en) * | 2021-04-16 | 2021-06-25 | 南方医科大学第五附属医院 | Artificial intelligence nasal sinus CT image processing system based on degree of depth learning |
CN115661114A (en) * | 2022-11-09 | 2023-01-31 | 重庆大学 | Full-reference image quality evaluation method based on Conformer and meta learning |
CN116363046A (en) * | 2021-12-28 | 2023-06-30 | 通用电气精准医疗有限责任公司 | Magnetic resonance image quality evaluation method and magnetic resonance imaging system |
WO2023217117A1 (en) * | 2022-05-13 | 2023-11-16 | 北京字跳网络技术有限公司 | Image assessment method and apparatus, and device, storage medium and program product |
CN117152648A (en) * | 2023-10-30 | 2023-12-01 | 泰州爱贝文化传媒有限公司 | Auxiliary teaching picture recognition device based on augmented reality |
CN117409016A (en) * | 2023-12-15 | 2024-01-16 | 华中科技大学同济医学院附属同济医院 | Automatic segmentation method for magnetic resonance image |
CN117558428A (en) * | 2024-01-12 | 2024-02-13 | 华中科技大学同济医学院附属同济医院 | Imaging optimization method and system for liver MRI |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8675957B2 (en) * | 2010-11-18 | 2014-03-18 | Ebay, Inc. | Image quality assessment to merchandise an item |
KR101582478B1 (en) * | 2012-05-03 | 2016-01-19 | 에스케이 텔레콤주식회사 | Image processing apparatus for image haze removal and method using that |
US10916003B2 (en) * | 2018-03-20 | 2021-02-09 | Uber Technologies, Inc. | Image quality scorer machine |
US20200065631A1 (en) * | 2018-08-21 | 2020-02-27 | Jonathan Meyers | Produce Assessment System |
EP3772720B1 (en) * | 2019-08-08 | 2021-09-29 | Siemens Healthcare GmbH | Method and system for image analysis |
JP2023542103A (en) * | 2020-09-14 | 2023-10-05 | ツインナー ゲゼルシャフト ミット ベシュレンクテル ハフツンク | Vehicle surface analysis system |
CN111932532B (en) * | 2020-09-21 | 2021-01-08 | 安翰科技(武汉)股份有限公司 | Method for evaluating capsule endoscope without reference image, electronic device, and medium |
-
2024
- 2024-02-23 CN CN202410203251.3A patent/CN117788461B/en active Active
Patent Citations (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102543793A (en) * | 2012-02-29 | 2012-07-04 | 无锡睿当科技有限公司 | Wafer focusing image quality feedback system and method therefor |
CN104636804A (en) * | 2013-11-07 | 2015-05-20 | 大连东方之星信息技术有限公司 | Data analysis system |
CN105678775A (en) * | 2016-01-13 | 2016-06-15 | 福州大学 | Color correction assessment method based on machine learning |
CN107578403A (en) * | 2017-08-22 | 2018-01-12 | 浙江大学 | The stereo image quality evaluation method of binocular view fusion is instructed based on gradient information |
WO2019047949A1 (en) * | 2017-09-08 | 2019-03-14 | 众安信息技术服务有限公司 | Image quality evaluation method and image quality evaluation system |
CN108319911A (en) * | 2018-01-30 | 2018-07-24 | 深兰科技(上海)有限公司 | Biometric identity certification and payment system based on the identification of hand arteries and veins and identity identifying method |
CN109146856A (en) * | 2018-08-02 | 2019-01-04 | 深圳市华付信息技术有限公司 | Picture quality assessment method, device, computer equipment and storage medium |
CN110895802A (en) * | 2018-08-23 | 2020-03-20 | 杭州海康威视数字技术股份有限公司 | Image processing method and device |
CN110491503A (en) * | 2019-08-21 | 2019-11-22 | 山东大学第二医院 | A kind of cholelithiasis intelligent assistance system based on deep learning |
CN111858746A (en) * | 2020-05-27 | 2020-10-30 | 武汉瞬付科技有限公司 | Personal data storage system based on cloud platform |
CN112215833A (en) * | 2020-10-22 | 2021-01-12 | 江苏云从曦和人工智能有限公司 | Image quality evaluation method, device and computer readable storage medium |
CN112801132A (en) * | 2020-12-28 | 2021-05-14 | 泰康保险集团股份有限公司 | Image processing method and device |
CN112741620A (en) * | 2020-12-30 | 2021-05-04 | 华南理工大学 | Cervical spondylosis evaluation device based on limb movement |
CN113034489A (en) * | 2021-04-16 | 2021-06-25 | 南方医科大学第五附属医院 | Artificial intelligence nasal sinus CT image processing system based on degree of depth learning |
CN116363046A (en) * | 2021-12-28 | 2023-06-30 | 通用电气精准医疗有限责任公司 | Magnetic resonance image quality evaluation method and magnetic resonance imaging system |
WO2023217117A1 (en) * | 2022-05-13 | 2023-11-16 | 北京字跳网络技术有限公司 | Image assessment method and apparatus, and device, storage medium and program product |
CN115661114A (en) * | 2022-11-09 | 2023-01-31 | 重庆大学 | Full-reference image quality evaluation method based on Conformer and meta learning |
CN117152648A (en) * | 2023-10-30 | 2023-12-01 | 泰州爱贝文化传媒有限公司 | Auxiliary teaching picture recognition device based on augmented reality |
CN117409016A (en) * | 2023-12-15 | 2024-01-16 | 华中科技大学同济医学院附属同济医院 | Automatic segmentation method for magnetic resonance image |
CN117558428A (en) * | 2024-01-12 | 2024-02-13 | 华中科技大学同济医学院附属同济医院 | Imaging optimization method and system for liver MRI |
Non-Patent Citations (2)
Title |
---|
Image Quality Assessment for Magnetic Resonance Imaging;Segrey Kastryulin 等;《arXiv:2203.07809v2 [eess.IV]》;20220701;1-13 * |
适用小样本的无参考水下视频质量评价方法;宋巍;刘诗梦;黄冬梅;王文娟;王建;;中国图象图形学报;20200916(第09期);73-85 * |
Also Published As
Publication number | Publication date |
---|---|
CN117788461A (en) | 2024-03-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104424385B (en) | A kind of evaluation method and device of medical image | |
CN112086197B (en) | Breast nodule detection method and system based on ultrasonic medicine | |
Garcia-Hernandez et al. | Analysis of the impact of digital watermarking on computer-aided diagnosis in medical imaging | |
CN106600597A (en) | Non-reference color image quality evaluation method based on local binary pattern | |
CN110415207A (en) | A method of the image quality measure based on image fault type | |
CN106023214B (en) | Image quality evaluating method and system based on central fovea view gradient-structure similitude | |
CN104091309A (en) | Balanced display method and system for flat-plate X-ray image | |
CN111062936B (en) | Quantitative index evaluation method for facial deformation diagnosis and treatment effect | |
CN106202930B (en) | A kind of Lung neoplasm risk forecasting system based on SS-ELM | |
WO2019220825A1 (en) | Chest x-ray image tone scale conversion method, image tone scale conversion program, image tone scale conversion device, server device, and conversion method | |
CN114757839A (en) | Tone mapping method based on macro and micro information enhancement and color correction | |
CN115147636A (en) | Lung disease identification and classification method based on chest X-ray image | |
CN117788461B (en) | Magnetic resonance image quality evaluation system based on image analysis | |
Albiol et al. | Automatic intensity windowing of mammographic images based on a perceptual metric | |
CN113610746A (en) | Image processing method and device, computer equipment and storage medium | |
CN112967254A (en) | Lung disease identification and detection method based on chest CT image | |
US20130308876A1 (en) | Process and apparatus for data registration | |
Chen et al. | A statistical method for evaluation quality of medical images: a case study in bit discarding and image compression | |
US7848551B2 (en) | Method and system for analysis of bone density | |
CN112259199A (en) | Medical image classification model training method, system, storage medium and medical image processing device | |
Smutek | Quality measurement of lossy compression in medical imaging | |
Sunaryo et al. | Preprocessing on Digital Image using Histogram Equalization: An Experiment Study on MRI Brain Image | |
Cui et al. | Medical image quality assessment method based on residual learning | |
Hariharan et al. | An algorithm for the enhancement of chest X-ray images of tuberculosis patients | |
CN111462101A (en) | Staging equipment based on novel coronavirus pneumonia CT detection and using method thereof |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |