Method for detecting benign and malignant lung lesions using MRI imaging

Abstract

The present invention relates to the field of image enhancement technology, and specifically to a method for detecting benign and malignant lung lesions using an MRI imaging group. The method analyzes the image distribution and image prominence state of lung MRI images, determines the recognizability of the lung area in the lung MRI image; determines the confidence of the recognizability according to the change in the recognizability during the respiratory movement detection process; determines the contrast-benefit ratio of the lung MRI image according to the change trend of the lung MRI image sequence; determines the contrast stretch coefficient of the lung MRI image in combination with the confidence of the recognizability and the contrast-benefit ratio; and performs contrast enhancement on the lung MRI image based on the contrast stretch coefficient to obtain an enhanced lung detection image. The lung detection image obtained by the present invention makes the boundary between the lesion and the normal tissue clearer, and improves the detection quality of the lung MRI image.

Description

Technical Field

[0001] The present invention relates to the technical field of image enhancement, and in particular to a method for detecting benign or malignant lung lesions using an MRI imaging group. Background Art

[0002] Early detection of lung lesions is of great significance to the user's prognosis, especially in the management of diseases such as lung cancer, pneumonia, and tuberculosis. In recent years, with the rapid development of imaging technology, magnetic resonance imaging (MRI) has gradually become an important tool for the diagnosis of lung diseases. MRI has the characteristics of being non-invasive, radiation-free, and having excellent soft tissue contrast, and is particularly suitable for the evaluation of lung lesions. In addition, MRI performs well in observing the blood supply, tissue structure, and metabolic changes of lesions, and can provide more information support.

[0003] Existing methods for lung detection using MRI imaging groups mainly include image acquisition, image post-processing, feature extraction, classification and diagnosis. Although the existing lung detection methods of MRI imaging groups have improved the lung detection capabilities to a certain extent, there are still some shortcomings, such as: respiratory movement of the lungs may cause image blur, thereby affecting the clarity and visibility of the lesions; existing methods often rely on empirical selection in feature extraction, which may miss some valuable information and lead to decreased classification performance; and in the process of contrast stretching of MRI images using the histogram equalization algorithm, some detail information may be lost, resulting in certain errors and inaccuracies in the detection and classification of regions in MRI images. Summary of the invention

[0004] In order to solve the technical problem that some detail information may be lost in the process of contrast stretching of MRI images using a histogram equalization algorithm, resulting in certain errors and inaccuracies in the detection and classification of regions in MRI images, the purpose of the present invention is to provide a method for detecting benign and malignant lung lesions using an MRI imaging group, and the technical scheme adopted is as follows:

[0005] In a first aspect, an embodiment of the present invention provides a method for detecting benign or malignant lung lesions using an MRI imaging group, the method comprising:

[0006] Obtain lung MRI images;

[0007] Analyzing the image distribution and image prominence of the lung MRI image to determine the recognizability of the lung area in the lung MRI image;

[0008] Determine the confidence level of the recognizability according to the change of the recognizability of the image during the respiratory movement detection process;

[0009] According to the changing trend of the lung MRI image sequence, the contrast-benefit ratio of the lung MRI image is determined;

[0010] The contrast stretch coefficient of the lung MRI image is determined in combination with the confidence of the identifiability and the contrast-benefit ratio; based on the contrast stretch coefficient, the lung MRI image is contrast enhanced to obtain an enhanced lung detection image.

[0011] Preferably, analyzing the image distribution and image prominence state of the lung MRI image to determine the recognizability of the lung area in the lung MRI image includes:

[0012] Obtain the middle grayscale value of the grayscale histogram of the lung MRI image, which is recorded as the grayscale representation value;

[0013] Analyze the numerical stability of the grayscale values ​​of the pixels in the lung MRI images and determine the grayscale discrete values;

[0014] Determining image distribution parameters of the lung MRI image according to the grayscale representation value and the grayscale discrete value;

[0015] Identifying the edge line of the lung MRI image; and determining the image prominence parameter of the lung MRI image according to the overall gradient of the edge line;

[0016] The image distribution parameter and the image prominence parameter are combined to determine the recognizability of the lung region in the lung MRI image.

[0017] Preferably, analyzing the numerical stability of the grayscale values ​​of the pixels of the lung MRI image to determine the grayscale discrete value includes:

[0018] The standard deviation of the grayscale values ​​of the pixels in the lung MRI image is taken as the grayscale discrete value.

[0019] Preferably, determining the image distribution parameter of the lung MRI image according to the grayscale representation value and the grayscale discrete value includes:

[0020] The mean of the grayscale values ​​of the pixels of the lung MRI image and the absolute value of the difference between the grayscale representation values ​​are calculated as image difference values; and the product of the image difference value and the grayscale discrete value is used as the image distribution parameter of the lung MRI image.

[0021] Preferably, determining the confidence level of the recognizability according to the change of the recognizability of the image during the respiratory movement detection process includes:

[0022] Acquire lung MRI image sequences within a preset respiratory cycle;

[0023] Arrange the acquisition time of the lung MRI images in the lung MRI image sequence in chronological order to obtain a time sequence;

[0024] The recognizability of the lung area in the lung MRI image is sorted according to the order of acquisition time to obtain a recognizability sequence;

[0025] Calculating the correlation coefficient between the time sequence and the recognizability sequence to obtain the recognizability correlation coefficient;

[0026] Using a magnetic sensitive device to collect magnetic induction intensity corresponding to a preset breathing cycle; sorting the magnetic induction intensity in the order of collection time to obtain a magnetic induction intensity sequence;

[0027] Calculating the correlation coefficient between the time sequence and the magnetic induction intensity sequence to obtain the magnetic induction intensity correlation coefficient;

[0028] The confidence of the recognizability is obtained by combining the recognizability correlation coefficient, the magnetic induction intensity correlation coefficient, the recognizability sequence and the magnetic induction intensity sequence.

[0029] Preferably, the combining of the identifiability correlation coefficient, the magnetic induction intensity correlation coefficient, the identifiability sequence and the magnetic induction intensity sequence to obtain the confidence level of the identifiability includes:

[0030] The sequence elements of the recognizability sequence and the magnetic induction intensity sequence are standardized, and the mean square error of the standardized sequence elements is calculated; the absolute value of the difference between the recognizability correlation coefficient and the magnetic induction intensity correlation coefficient is calculated as the initial confidence; and the product value of the initial confidence and the mean square error is used as the confidence of the recognizability.

[0031] Preferably, determining the contrast-benefit ratio of the lung MRI image according to the change trend of the lung MRI image sequence includes:

[0032] Acquire lung MRI image sequences within a preset respiratory cycle;

[0033] The mean value of the pixel difference of the pixel points at the same position of adjacent lung MRI images in the lung MRI image sequence is calculated to obtain the signal difference;

[0034] Identify suspected lesion areas and normal areas in lung MRI images, calculate the average of the differences between the grayscale means of the suspected lesion areas and the grayscale means of the normal areas in adjacent lung MRI images in the lung MRI image sequence, and obtain the regional differences;

[0035] The contrast-benefit ratio of lung MRI images was determined by combining signal differences and regional differences.

[0036] Preferably, the combining of signal difference and regional difference to determine the contrast-benefit ratio of lung MRI images comprises:

[0037] The negative correlation mapping value of the product of signal difference and regional difference is used as the contrast-benefit ratio of lung MRI images.

[0038] Preferably, the step of combining the confidence of the identifiability and the contrast-benefit ratio to determine the contrast stretch factor of the lung MRI image comprises:

[0039] The confidence of recognizability is used as the numerator and the contrast-benefit ratio is used as the denominator. The ratio corresponding to the numerator and denominator is normalized to obtain the contrast stretch coefficient of the lung MRI image.

[0040] Preferably, the step of contrast enhancing the lung MRI image based on the contrast stretch coefficient to obtain an enhanced lung detection image includes:

[0041] Identify suspected lesion areas in lung MRI images;

[0042] An arbitrary grayscale value of the suspected lesion area is used as the grayscale value to be adjusted, and the sum of the contrast stretching coefficient and the preset constant adjustment value is used as the adjustment coefficient of the grayscale value to be adjusted during histogram equalization. For grayscale values ​​other than the grayscale values ​​of the suspected lesion area, the histogram equalization operation is directly performed to enhance the lung detection image.

[0043] In a second aspect, a system for detecting benign and malignant lung lesions using an MRI imaging group is provided, the system comprising the following modules:

[0044] An image acquisition module, used to acquire lung MRI images;

[0045] A static analysis module, used to analyze the image distribution and image prominence state of the lung MRI image, and determine the recognizability of the lung area in the lung MRI image;

[0046] A dynamic analysis module is used to determine the confidence level of the recognizability according to the change of the image recognizability during the respiratory motion detection process;

[0047] A contrast analysis module, used to determine the contrast-benefit ratio of the lung MRI image according to the change trend of the lung MRI image sequence;

[0048] The enhancement module is used to determine the contrast stretch coefficient of the lung MRI image in combination with the confidence of the identifiability and the contrast benefit ratio; based on the contrast stretch coefficient, the lung MRI image is contrast enhanced to obtain an enhanced lung detection image.

[0049] In a third aspect, an embodiment of the present invention provides an electronic device, including a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, various possible embodiments of the first aspect are implemented.

[0050] In a fourth aspect, an embodiment of the present invention provides a computer program product, which includes: a computer program code, which, when executed on a computer, enables the computer to execute the method in the first aspect or any possible implementation of the first aspect.

[0051] In a fifth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, enables the computer to execute various possible implementations of the first aspect.

[0052] The embodiments of the present invention have at least the following beneficial effects:

[0053] The invention relates to the field of image enhancement technology. In order to solve the problem that the respiratory movement of the lungs may cause image blurring during the detection process of the existing detection method, thereby affecting the clarity and visibility of the lesion, and some detail information may be lost during the contrast stretching of the lung MRI image using the histogram equalization algorithm, which causes certain errors and inaccuracies in the detection and classification. The method first obtains the recognizability of the lung area through the image grayscale distribution state in the lung MRI image and the overall image prominence state in the image, and then calculates the confidence of the recognizability of the lung area in combination with the respiratory movement changes during the monitoring process; then obtains the contrast benefit ratio of the image data through the change trend of the lung MRI image in the dynamic lung MRI image sequence, and finally calculates the contrast stretching coefficient by combining the confidence and contrast benefit ratio obtained by the dynamic and static frame image analysis, and contrast enhancement is performed on the collected lung lesion MRI image to obtain the lung detection image, so that the boundary between the lesion and the normal tissue is clearer, the quality of the lung MRI image is further improved, and reliable data support is also provided for subsequent diagnosis and treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1A method flow chart of a method for detecting benign and malignant lung lesions using an MRI imaging group provided by one embodiment of the present invention;

[0056] Figure 2 A system block diagram of a system for detecting benign and malignant lung lesions using an MRI imaging group provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation method, structure, characteristics and effects of the method for detecting benign and malignant lung lesions using MRI imaging group proposed by the present invention are described in detail as follows in combination with the accompanying drawings and preferred embodiments.

[0058] In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0059] Among them, in the description of the embodiments of the present invention, unless otherwise specified, " / " means or, for example, A / B can mean A or B: "and / or" in the text is only a way to describe the association relationship of associated objects, indicating that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present invention, "multiple" refers to two or more than two.

[0060] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features.

[0061] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0062] The embodiments of the present invention are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided by the embodiments of the present invention are also applicable to similar technical problems.

[0063] The specific scheme of the method for detecting benign and malignant lung lesions using MRI imaging group provided by the present invention is described in detail below with reference to the accompanying drawings.

[0064] See also Figure 1 , which shows a flowchart of a method for detecting benign and malignant lung lesions using an MRI imaging group provided by an embodiment of the present invention, the method comprising the following steps:

[0065] Step S100, obtaining a lung MRI image.

[0066] First, the user's medical and allergy history is collected to assess suitability for a magnetic resonance imaging (MRI) examination, and then the examination process is explained to the user, especially the importance of remaining still and cooperating with breathing instructions.

[0067] Ensure that the MRI equipment is in optimal working condition, perform necessary calibration and quality checks, and select the best scanning sequence and parameters based on clinical needs and specific lesion characteristics. It should be noted that if contrast agents are required, evaluate their practicality and prepare for injection, and ensure that the user is not allergic to contrast agents.

[0068] Properly position the user on the scanning table, ensuring comfort and immobilization to reduce movement; use the target area to set the appropriate scan range to ensure full coverage.

[0069] Start the MRI scanning program, pay attention to testing and monitoring the image quality, and obtain the user's lung MRI image; store the obtained lung MRI image data in a secure database system and make a backup.

[0070] Step S200, analyzing the image distribution and image prominence state of the lung MRI image to determine the recognizability of the lung area in the lung MRI image.

[0071] During the lung MRI image processing, the benign and malignant nature of lung lesions is accurately assessed based on the shape characteristics, size, position, and texture characteristics of the lesion area in the lung image. However, if the image data contrast is poor, it may affect the identification and feature extraction of the lesion area, thereby affecting the accuracy of detecting the benign and malignant nature of the lesion.

[0072] Therefore, according to the acquired lung MRI image data, the lung MRI image is preprocessed to obtain the grayscale histogram of the lung MRI image; the grayscale value in the middle of the grayscale histogram of the lung MRI image is obtained and recorded as the grayscale representation value. It should be noted that the preprocessing includes basic preprocessing operations such as denoising the lung MRI image, which can be processed by the implementer according to the actual situation, or the grayscale histogram of the lung MRI image can be directly obtained without preprocessing.

[0073] The numerical stability of the grayscale values ​​of the pixels of the lung MRI image is analyzed to determine the grayscale discrete value. Specifically, the standard deviation of the grayscale values ​​of the pixels of the lung MRI image is used as the grayscale discrete value.

[0074] The image distribution parameters of the lung MRI image are determined according to the grayscale characterization value and the grayscale discrete value. Specifically, the absolute value of the difference between the mean grayscale value of the pixel points of the lung MRI image and the grayscale characterization value is calculated as the image difference value; and the product value of the image difference value and the grayscale discrete value is used as the image distribution parameter of the lung MRI image.

[0075] The edge detection is performed on the lung MRI image to identify the edge line of the lung MRI image. It should be noted that the method of performing edge detection on the image to obtain the edge line is a well-known technology for those skilled in the art and will not be described in detail here.

[0076] According to the overall gradient of the edge line, the image highlighting parameter of the lung MRI image is determined. Specifically, the gradient mean of all edge points on all edge lines is used as the image highlighting parameter of the lung MRI image.

[0077] The recognizability of the lung region in the lung MRI image is determined by combining the image distribution parameter and the image prominence parameter. Specifically, the product value of the image distribution parameter and the image prominence parameter is used as the recognizability of the lung region in the lung MRI image.

[0078] The calculation formula of the recognizability is:

[0079] ;

[0080] Among them, R is the recognizability; is the mean gray value of the pixels in the lung MRI image; is the grayscale representation value; is the grayscale discrete value; is the mean gradient of the g-th edge line; is the image highlighting parameter; G is the number of edge lines of lung MRI images; is the image distribution parameter.

[0081] Image distribution parameters The smaller the product of , the closer the average gray value is to the middle value and the smaller the standard deviation is, which may indicate poor contrast. It represents the average gradient value of all edge lines. The smaller the average gradient value, the lower the overall gradient intensity in the lung MRI image, indicating that the edges and details in the lung MRI image are not clear. The larger the product of , the higher the recognizability of the lung area in the image, that is, the larger the recognizability R.

[0082] Step S300, determining the confidence of the recognizability according to the change of the recognizability of the image during the respiratory movement detection process. It should be noted that the image during the respiratory movement detection process mentioned here is the lung MRI image during the respiratory movement detection process.

[0083] However, the quality of lung MRI images is easily affected by respiratory motion, which may introduce errors in the actual detection process, resulting in inaccurate detection results. This requires further analysis of the interference caused by motion artifacts. Specifically, the interference of respiratory motion on image recognizability is analyzed.

[0084] In the process of acquiring lung MRI images, the user is required to maintain a certain still state or cooperate with breathing instructions, but the human breathing state cannot fully meet the requirements of acquiring images, that is, the quality of the image may be affected by the user's breathing state, that is, the image recognizability needs to be verified.

[0085] During the process of the user acquiring lung MRI images, a magnetic sensitive device is used to record the changes in the magnetic field during the process. It can be understood that the magnetic sensitive device is used to collect the magnetic induction intensity during the process, and the changes in the magnetic induction intensity are used to represent the changes in breathing. That is, the magnetic sensitive device can reflect the breathing state by recording the changes in the magnetic field, and then set the time corresponding to the first peak value and the second peak value of the magnetic induction intensity as a preset breathing cycle. The implementer can also manually set a preset breathing cycle.

[0086] A lung MRI image sequence within a preset respiratory cycle is acquired, wherein the lung MRI image sequence is composed of lung MRI images and is an image sequence arranged according to acquisition time.

[0087] The acquisition times of the lung MRI images in the lung MRI image sequence are sorted in order to obtain a time sequence; the recognizability of the lung area in the lung MRI image is sorted in order of the acquisition time to obtain a recognizability sequence; and the correlation coefficient between the time sequence and the recognizability sequence is calculated to obtain a recognizability correlation coefficient.

[0088] The magnetic induction intensity within the corresponding preset respiratory cycle is collected by a magnetic sensitive device; the magnetic induction intensity is sorted in the order of the collection time to obtain a magnetic induction intensity sequence; the correlation coefficient of the moment sequence and the magnetic induction intensity sequence is calculated to obtain the magnetic induction intensity correlation coefficient. In the embodiment of the present invention, the correlation coefficient is the Pearson correlation coefficient, and in other embodiments, the implementer can set other methods for obtaining the correlation coefficient. It should be noted that when the data within a preset respiratory cycle is difficult to determine the correlation coefficient, it can also be set to analyze and calculate the correlation coefficient of the moment sequence and the identifiability sequence corresponding to the lung MRI image sequence within a preset number of preset respiratory cycles, and the correlation coefficient of the moment sequence and the magnetic induction intensity sequence. In the embodiment of the present invention, the value of the preset number is 5, and in other embodiments, the implementer can also adjust the value according to the actual situation.

[0089] The confidence of the recognizability is obtained by combining the recognizability correlation coefficient, the magnetic induction intensity correlation coefficient, the recognizability sequence and the magnetic induction intensity sequence.

[0090] The sequence elements of the identifiability sequence and the magnetic induction intensity sequence are standardized, and the mean square error of the standardized sequence elements is calculated. It should be noted that the standard deviation of the two sequence elements is performed here in order to unify the dimensions of the two sequences. The absolute value of the difference between the identifiability correlation coefficient and the magnetic induction intensity correlation coefficient is calculated as the initial confidence; the product of the initial confidence and the mean square error is used as the confidence of the identifiability.

[0091] The calculation formula of the confidence w of the identifiability is: ;in, is the identifiability correlation coefficient; is the magnetic induction intensity correlation coefficient; is the initial confidence; is the mean square error of the normalized sequence elements.

[0092] Initial confidence The smaller the value of is, the more synchronized the change of the identifiability of the lung area in the lung MRI image is with the change of respiratory movement, that is, the quality of the lung MRI image may be more affected by respiratory movement; The larger the product of , the higher the confidence that the lung MRI image is poorly recognizable, that is, the quality of the acquired lung MRI image is low. The contrast of the lung MRI image can be further measured in combination with the dynamic change process.

[0093] Step S400, determining the contrast-benefit ratio of the lung MRI image according to the change trend of the lung MRI image sequence.

[0094] In the process of acquiring lung MRI images, a dynamic lung MRI image sequence is obtained. The dynamic lung MRI image sequence refers to each frame of dynamic lung MRI images in a preset respiratory cycle mentioned above. The contrast state of the lung MRI image can be further effectively identified through the changes in the dynamic lung MRI image sequence. Combined with the recognizability of the above-mentioned lung MRI image, the quality of the lung MRI image can be further analyzed, and better data support can be provided for subsequent lung MRI image processing.

[0095] After acquiring a lung MRI image sequence within a preset respiratory cycle, the mean value of the pixel difference of pixel points at the same position of adjacent lung MRI images in the lung MRI image sequence is calculated to obtain a signal difference.

[0096] Taking the i-th lung MRI image and the i+1-th lung MRI image as examples, the corresponding signal difference The calculation formula is: ; M is the number of pixels in the lung MRI image; is the pixel value of the pixel at the mth position in the i-th lung MRI image; is the pixel value of the pixel at the mth position in the i+1th lung MRI image.

[0097] The signal difference reflects the signal difference between different tissues between two adjacent frames in the dynamic lung MRI image sequence; the smaller the signal difference, the smaller the change in signal intensity of the same tissue or different tissues in adjacent lung MRI images, which means that the boundary between tissues in the dynamic lung MRI image is not obvious or blurred, indicating that the contrast of the lung MRI image is poor.

[0098] Furthermore, the suspected lesion areas and normal areas in the lung MRI images are identified. A large amount of lung lesion image data in the medical library is used to extract the features of the image data, such as shape, texture, signal strength, etc. The large amount of lung lesion image data in the medical library is manually annotated, and a deep learning model is trained. The U-net deep learning model is used for automatic learning, and the suspected lesion areas in the lung MRI images are identified and marked.

[0099] The average value of the difference between the grayscale mean of the suspected lesion area and the grayscale mean of the normal area in the adjacent lung MRI images in the lung MRI image sequence is calculated to obtain the regional difference;

[0100] The contrast-benefit ratio of the lung MRI image is determined by combining the signal difference and the regional difference. In some embodiments, the negative correlation mapping value of the product of the signal difference and the regional difference is used as the contrast-benefit ratio of the lung MRI image.

[0101] The calculation formula of the contrast ratio s is: ; Where exp is an exponential function with a natural constant as the base; is the signal difference; is the difference between the grayscale mean of the suspected lesion area and the grayscale mean of the normal area in the adjacent lung MRI images of group b; It is the average value of the difference between the grayscale mean of the suspected lesion area and the grayscale mean of the normal area in adjacent lung MRI images in the lung MRI image sequence, that is, the regional difference.

[0102] Regional differences The smaller the value, the smaller the grayscale change in the lung MRI image sequence or almost no change, which means that the contrast of the lung MRI image is low; The larger the product of , the larger the signal difference between tissues in the lung MRI image and the larger the grayscale change in the lung MRI image sequence, which means that the contrast in the image is higher, that is, the contrast-benefit ratio of the lung MRI image is higher, that is, the larger the contrast-benefit ratio s is.

[0103] Step S500, determining a contrast stretch coefficient of the lung MRI image in combination with the confidence of the identifiability and the contrast-benefit ratio; and performing contrast enhancement on the lung MRI image based on the contrast stretch coefficient to obtain an enhanced lung detection image.

[0104] In the process of detecting benign and malignant lung lesions using MRI imaging, the image contrast in the lung MRI images actually collected may be poor, which may cause key details in the image to not stand out. In addition, key details may be lost during contrast stretching using histogram equalization, which requires more accurate contrast stretching adjustment.

[0105] The contrast quality of the collected MRI image of lung lesions is further evaluated based on the data obtained in step S100 to step S400, and then the contrast stretch coefficient of the image data is further obtained to achieve contrast enhancement.

[0106] The confidence of recognizability is used as the numerator and the contrast-benefit ratio is used as the denominator. The ratio corresponding to the numerator and denominator is normalized to obtain the contrast stretch coefficient of the lung MRI image.

[0107] The calculation formula of the comparative stretch coefficient z is: ; Where th is the hyperbolic tangent function, which is used to normalize the data; w is the confidence level of identifiability; s is the contrast-benefit ratio.

[0108] The larger the product of , the higher the confidence that the lung MRI image is poorly recognizable and the lower the contrast ratio of the lung MRI image. That is, the contrast stretching coefficient z of the actual MRI image obtained by combining the dynamic and static lung MRI image features is larger, and the quality of the acquired lung MRI image needs to be further improved to make the characteristics of the lesion area more obvious.

[0109] Based on the contrast stretching coefficient, the lung MRI image is contrast enhanced to obtain an enhanced lung detection image. Specifically: after the contrast stretching coefficient of the lung MRI image is obtained, the grayscale value range in the suspected lesion area is obtained through the suspected lesion area marked in the image; any grayscale value of the suspected lesion area is used as the grayscale value to be adjusted, and the sum of the contrast stretching coefficient and the preset constant adjustment value is used as the adjustment coefficient of the grayscale value to be adjusted during histogram equalization. For grayscale values ​​other than the grayscale value of the suspected lesion area, a histogram equalization operation is directly performed to obtain an enhanced lung detection image. In the embodiment of the present invention, the value of the preset constant adjustment value is 1. It is understandable that the adaptive histogram equalization operation is performed on the lung MRI image, that is, in the process of performing histogram equalization on the grayscale in the marked suspected lesion area, the adjustment coefficient 1+z is used as the adjustment coefficient of all grayscales in the suspected lesion area, and the preset constant adjustment value is added to ensure that the contrast enhancement operation with the contrast stretching coefficient z as the adjustment value is performed on the basis of the traditional histogram equalization operation; it is understandable that when z is 0, the corresponding adjustment coefficient is 1. At this time, for the grayscale value to be adjusted corresponding to z being 0, the default histogram equalization operation is followed. It is also equivalent to widening the grayscale in the original image during the traditional histogram equalization processing.

[0110] It can be understood that the grayscale value to be adjusted is first obtained after conventional histogram equalization, the difference between the grayscale value to be adjusted and the grayscale value after histogram equalization is used as the initial stretch value, the product of the adjustment coefficient 1+z and the initial stretch value is used as the adaptive stretch value; the sum of the grayscale value to be adjusted and the adaptive stretch value is used as the grayscale value of the grayscale value to be adjusted after contrast enhancement. It should be noted that the grayscale value here can also be understood as the grayscale level during histogram equalization. For example, for the initial gray level 50, after conventional histogram equalization, the gray level 50 is equalized to gray level 70. The adjustment coefficient corresponding to the initial gray level 50 is 1.2, which is equivalent to widening the gray level 50 by 20. After conventional histogram equalization, the initial gray level 50 has a widening value of 20, and the conventional histogram equalization is gray level 70. The corresponding gray level of gray level 50 adjusted by the adjustment coefficient 1.2 is 50+20×1.2=50+24=74. Then the gray level of the remaining non-suspected lesion areas can be operated according to the default histogram equalization operation, and then the lung MRI image after contrast stretching is obtained, which is the enhanced lung detection image.

[0111] As a preferred embodiment of the present invention, the shape features of the lesions, such as area, perimeter, irregular shape, texture features, etc., are extracted from the contrast-enhanced lung detection image to detect and classify the benign and malignant lesions, and provide reliable data support for subsequent diagnosis and treatment. For example, the area of ​​the suspected lesion area in the lung detection image can be identified to detect the lung lesions of the user corresponding to the lung MRI image.

[0112] See also Figure 2 , which shows a system block diagram of a system for detecting benign and malignant lung lesions using an MRI imaging group provided by an embodiment of the present invention, the system includes the following modules:

[0113] An image acquisition module, used to acquire lung MRI images;

[0114] A static analysis module, used to analyze the image distribution and image prominence state of the lung MRI image, and determine the recognizability of the lung area in the lung MRI image;

[0115] A dynamic analysis module, used to determine the confidence level of recognizability according to the change of the recognizability of the image during the respiratory motion detection process;

[0116] A contrast analysis module, used to determine the contrast-benefit ratio of the lung MRI image according to the change trend of the lung MRI image sequence;

[0117] The enhancement module is used to determine the contrast stretch coefficient of the lung MRI image in combination with the confidence of the identifiability and the contrast benefit ratio; based on the contrast stretch coefficient, the lung MRI image is contrast enhanced to obtain an enhanced lung detection image.

[0118] Optionally, the transmission medium may be a wired link, such as but not limited to coaxial cable, optical fiber, and digital subscriber line, or a wireless link, such as but not limited to Wireless Fidelity (WIFI), Bluetooth, and mobile device network.

[0119] It should be noted that the device provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above.

[0120] An embodiment of the present invention provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the computer device can execute any of the above-mentioned methods for detecting benign or malignant lung lesions using MRI imaging groups.

[0121] In addition, an embodiment of the present invention also protects a device, which may include a memory and a processor, wherein the memory stores an executable program code, and the processor is used to call and execute the executable program code to execute the method for detecting benign and malignant lung lesions using an MRI imaging group provided in an embodiment of the present invention.

[0122] The embodiment of the present invention can divide the functional modules of the device according to the above method example. For example, each functional module can be corresponded, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0123] In the case of dividing each module according to each function, the device may also include a signal uploading module, a determining module, an adjusting module, etc. It should be noted that all relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module, which will not be repeated here.

[0124] It should be understood that the device provided in the embodiment of the present invention is used to execute the above-mentioned method for detecting benign and malignant lung lesions using the MRI imaging group, and thus can achieve the same effect as the above-mentioned implementation method.

[0125] In the case of an integrated unit, the device may include a processing module and a storage module. When the device is applied to a device, the processing module may be used to control and manage the actions of the device. The storage module may be used to support the device to execute mutual program codes, etc. The processing module may be a processor or a controller, which may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of the present invention. The processor may also be a combination that implements a computing function, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc. The storage module may be a memory.

[0126] In addition, the device provided by the embodiment of the present invention can specifically be a chip, a component or a module, and the chip may include a connected processor and a memory; wherein the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute the method for detecting benign and malignant lung lesions using the MRI imaging group provided in the above embodiment.

[0127] An embodiment of the present invention also provides a computer-readable storage medium, which stores computer program code. When the computer program code is run on a computer, the computer executes the above-mentioned related method steps to implement the method for detecting benign and malignant lung lesions using MRI imaging groups provided in the above embodiment.

[0128] An embodiment of the present invention also provides a computer program product. When the computer program product is run on a computer, the computer executes the above-mentioned related steps to implement the method for detecting benign and malignant lung lesions using MRI imaging groups provided in the above embodiment.

[0129] Among them, the device, computer-readable storage medium, computer program product or chip provided in the embodiments of the present invention are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be repeated here. Through the description of the above implementation methods, technicians in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways.

[0130] The device embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division. There may be other divisions in actual implementation. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection of devices or units, which may be electrical, mechanical or other forms.

[0131] It should also be noted that, in this article, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0132] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0133] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0134] The above contents are only specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered within the protection scope of the present invention.

Claims

1. A method for detecting benign or malignant lung lesions using MRI imaging, characterized in that: The method comprises the following steps: Obtain lung MRI images; Analyzing the image distribution and image prominence of the lung MRI image to determine the recognizability of the lung area in the lung MRI image; Determining the confidence level of the recognizability according to the change of the recognizability of the lung area during the respiratory movement detection process; According to the changing trend of the lung MRI image sequence, the contrast-benefit ratio of the lung MRI image is determined; Determining a contrast stretch coefficient of the lung MRI image in combination with the confidence of the identifiability and the contrast-benefit ratio; performing contrast enhancement on the lung MRI image based on the contrast stretch coefficient to obtain an enhanced lung detection image; Among them, the method for obtaining the confidence of the recognizability is as follows: obtaining a lung MRI image sequence within a preset respiratory cycle; sorting the acquisition moments of the lung MRI images in the lung MRI image sequence in order to obtain a moment sequence; sorting the recognizability of the lung area in the lung MRI image in order of the acquisition moments to obtain a recognizability sequence; calculating the correlation coefficient between the moment sequence and the recognizability sequence to obtain a recognizability correlation coefficient; using a magnetic sensitive device to collect the magnetic induction intensity within the corresponding preset respiratory cycle; sorting the magnetic induction intensity in order of the acquisition moments to obtain a magnetic induction intensity sequence; calculating the correlation coefficient between the moment sequence and the magnetic induction intensity sequence to obtain a magnetic induction intensity correlation coefficient; combining the recognizability correlation coefficient, the magnetic induction intensity correlation coefficient, the recognizability sequence and the magnetic induction intensity sequence to obtain a confidence of the recognizability; Among them, the method for obtaining the contrast-benefit ratio is: obtaining a lung MRI image sequence within a preset respiratory cycle; calculating the mean of pixel differences of pixel points at the same position of adjacent lung MRI images in the lung MRI image sequence to obtain signal differences; identifying suspected lesion areas and normal areas in the lung MRI images, and calculating the average of the differences between the grayscale means of the suspected lesion areas and the grayscale means of the normal areas of adjacent lung MRI images in the lung MRI image sequence to obtain regional differences; combining the signal differences and regional differences to determine the contrast-benefit ratio of the lung MRI images.

2. The method for detecting benign and malignant lung lesions using MRI imaging according to claim 1, characterized in that: The analyzing the image distribution and image prominence state of the lung MRI image to determine the recognizability of the lung area in the lung MRI image includes: Obtain the middle grayscale value of the grayscale histogram of the lung MRI image, which is recorded as the grayscale representation value; Analyze the numerical stability of the grayscale values ​​of the pixels in the lung MRI images and determine the grayscale discrete values; Determining image distribution parameters of the lung MRI image according to the grayscale representation value and the grayscale discrete value; Identifying the edge line of the lung MRI image; and determining the image prominence parameter of the lung MRI image according to the overall gradient of the edge line; The image distribution parameter and the image prominence parameter are combined to determine the recognizability of the lung region in the lung MRI image.

3. The method for detecting benign and malignant lung lesions using MRI imaging according to claim 2, characterized in that: The step of analyzing the numerical stability of the grayscale values ​​of the pixels of the lung MRI image and determining the grayscale discrete values ​​includes: The standard deviation of the grayscale values ​​of the pixels in the lung MRI image is used as the grayscale discrete value.

4. The method for detecting benign and malignant lung lesions using MRI imaging according to claim 2, characterized in that: Determining the image distribution parameter of the lung MRI image according to the grayscale representation value and the grayscale discrete value includes: The mean of the grayscale values ​​of the pixels of the lung MRI image and the absolute value of the difference between the grayscale representation values ​​are calculated and used as the image difference value; the product of the image difference value and the grayscale discrete value is used as the image distribution parameter of the lung MRI image.

5. The method for detecting benign and malignant lung lesions using MRI imaging according to claim 1, characterized in that: The confidence level of the recognizability is obtained by combining the recognizability correlation coefficient, the magnetic induction intensity correlation coefficient, the recognizability sequence and the magnetic induction intensity sequence, including: The sequence elements of the recognizability sequence and the magnetic induction intensity sequence are standardized, and the mean square error of the standardized sequence elements is calculated; the absolute value of the difference between the recognizability correlation coefficient and the magnetic induction intensity correlation coefficient is used as the initial confidence; and the product value of the initial confidence and the mean square error is used as the confidence of the recognizability.

6. The method for detecting benign and malignant lung lesions using MRI imaging according to claim 1, characterized in that: The method of combining the signal difference and the regional difference to determine the contrast-benefit ratio of the lung MRI image includes: The negative correlation mapping value of the product of signal difference and regional difference is used as the contrast-benefit ratio of lung MRI images.

7. The method for detecting benign and malignant lung lesions using MRI imaging according to claim 1, characterized in that: The step of combining the confidence of the identifiability and the contrast-benefit ratio to determine the contrast stretch factor of the lung MRI image includes: The confidence of recognizability is used as the numerator and the contrast-benefit ratio is used as the denominator. The ratio corresponding to the numerator and denominator is normalized to obtain the contrast stretch coefficient of the lung MRI image.

8. The method for detecting benign and malignant lung lesions using MRI imaging according to claim 1, characterized in that: The step of contrast enhancing the lung MRI image based on the contrast stretch coefficient to obtain an enhanced lung detection image includes: Identify suspected lesion areas in lung MRI images; An arbitrary grayscale value of the suspected lesion area is used as the grayscale value to be adjusted, and the sum of the contrast stretching coefficient and the preset constant adjustment value is used as the adjustment coefficient of the grayscale value to be adjusted during histogram equalization. For grayscale values ​​other than the grayscale values ​​of the suspected lesion area, a histogram equalization operation is directly performed to obtain an enhanced lung detection image.

Images