An intelligent recognition and analysis system for urological stone images

Abstract

The present invention discloses an intelligent recognition and analysis system for urological stone images, which relates to the field of image processing technology. The present invention forms a core architecture of an intelligent recognition and analysis system for urological stone images through the mutual cooperation of three algorithm units. The comprehensive image score value Ism is calculated by the comprehensive image score value algorithm unit, which can more comprehensively and accurately evaluate the image quality, reduce misdiagnosis and missed diagnosis caused by a single factor, help the intelligent recognition and analysis system to better distinguish stones from surrounding tissues, improve the detection rate of stones, and better perform image recognition. The adjusted stone segmentation threshold Ast is calculated by the adjusted stone segmentation threshold algorithm unit, that is, the stone segmentation threshold can be dynamically adjusted according to factors such as image quality, which helps to more accurately identify the boundary and range of the stone, improve the accuracy of stone segmentation, and provide a scientific and reliable data basis for subsequent diagnosis and treatment.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to an intelligent recognition and analysis system for urological stone images. Background Art

[0002] Urology is a hospital department that mainly diagnoses and treats diseases of the "surgical" part of the urinary system. It mainly treats various urinary diseases. Digital images of urology are generally presented in the form of CT (computed tomography), MRI (magnetic resonance imaging), ultrasound, etc.

[0003] The existing intelligent recognition system for urological stone images still uses traditional image processing segmentation algorithms to segment and identify stones in images. This conventional image segmentation and recognition algorithm is relatively simple, and its segmentation threshold is usually fixed, making it difficult to adapt to different scanning conditions (such as low dose, low scanning layer thickness, and obese patients). Dynamic adjustment of the segmentation threshold is prone to missing smaller stones, which is not conducive to use.

[0004] Therefore, there is an urgent need for an intelligent recognition and analysis system for urological stone images to solve the above problems. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent recognition and analysis system for urological stone images to solve the problems raised in the above background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an intelligent recognition and analysis system for urological stone images, comprising:

[0007] A data collection module is used to obtain the patient's first imaging data through the metadata of the imaging device;

[0008] The data preprocessing module is used to decode and preprocess the data information in the database to obtain the parameters involved in the calculation in the calculation processing module;

[0009] Calculation processing module, the specific calculation processing steps are as follows:

[0010] S1, substitute the parameter values ​​obtained after decoding preprocessing into the comprehensive image score value algorithm unit to calculate the comprehensive image score value Ism, and take the ideal value of the parameter to calculate the maximum value of the comprehensive image score value Ism max ;

[0011] S2, set the standard threshold value Y1 of image quality in the database as Ism max / 2, when the comprehensive image score value Ism>Y1, the comprehensive image score value Ism is input as an input parameter into the adjusted stone segmentation threshold algorithm unit to calculate the adjusted stone segmentation threshold Ast;

[0012] S3, inputting the adjusted stone segmentation threshold Ast into the segmentation algorithm of the intelligent stone image recognition system to perform stone image segmentation and recognition;

[0013] S4, when the comprehensive image score value Ism is less than Y1, the comprehensive image score value Ism is input as an input parameter into the adjusted image scanning layer thickness algorithm unit to calculate the adjusted image scanning layer thickness Thk again ;

[0014] S5, according to the adjusted image scanning layer thickness Thk again , perform a second scan of the patient's urological stone images and re-calculate the comprehensive image scoring value algorithm unit.

[0015] Optionally, obtaining the patient's first imaging data through the metadata of the imaging device specifically includes:

[0016] Obtain the image scanning layer thickness Thk, radiation dose Ed, and stenosis diameter Dus through the metadata of the imaging device;

[0017] The image signal-to-noise ratio (SNR) and CT value gradient (CTg) are obtained through the built-in image processing software of the imaging device;

[0018] The patient's BMI value is obtained from the patient's electronic medical record and uploaded to the database.

[0019] Optionally, the calculation and processing module includes a comprehensive image scoring value algorithm unit, an adjusted stone segmentation threshold algorithm unit, and an adjusted image scanning layer thickness algorithm unit.

[0020] Optionally, the calculation logic of the comprehensive image scoring value algorithm unit is as follows:

[0021] S11, the influence of the image signal-to-noise ratio (SNR) on the comprehensive image score Ism is amplified by the power function;

[0022] S12, by coupling the comprehensive influence of the scanning layer thickness THK and the CT value gradient CTg on the comprehensive image score value Ism, the low contrast detectable value after balancing the spatial resolution is obtained;

[0023] S13, normalizing the influence item of the patient's BMI value on the comprehensive imaging score value Ism by using the reference value 25 to obtain the influence item of the patient's BMI value on the comprehensive imaging score value Ism;

[0024] S14, multiplying the influence of the image signal-to-noise ratio (SNR) on the comprehensive image score Ism, the low contrast detectable value after balancing the spatial resolution, and the patient's BMI on the comprehensive image score Ism to obtain the comprehensive image score Ism.

[0025] Optionally, the calculation logic of the adjusted stone segmentation threshold algorithm unit is as follows:

[0026] S21, the maximum value of the comprehensive image score Ism max The influence of the comprehensive image score Ism on the stone segmentation threshold was normalized to a value range of 0 to 1;

[0027] S22, the coupling effect of image blur and noise is obtained by dividing the radiation dose Ed by the image scanning layer thickness Thk;

[0028] S23, scaling the coupling effect of image blur and noise by a logarithmic function;

[0029] S24, multiplying the coupled influence term of image blur and noise after logarithmic function scaling by the influence term of the comprehensive image score value Ism on the stone segmentation threshold and the stone basic segmentation threshold Bst to obtain the adjusted stone segmentation threshold Ast.

[0030] Optionally, the calculation logic of the adjusted image scanning layer thickness algorithm unit is as follows:

[0031] S41, mapping the adjustment requirement of the ureteral stenosis degree to the layer thickness through the hyperbolic tangent function;

[0032] S42, the maximum value of the comprehensive image score value Ism max The comprehensive image score value Ism is adjusted to the image scanning layer thickness Thk again The influence items are standardized;

[0033] S43, the adjusted image scanning layer thickness Thk is calculated by multiplying the image scanning layer thickness Thk by the standardized comprehensive image score value Ism. again The influencing factors and the adjustment requirement of the layer thickness due to the degree of ureteral stenosis are used to obtain the adjusted image scan layer thickness.

[0034] Optionally, the decoding preprocessing includes data cleaning and data standardization.

[0035] Optionally, the equipment used in the data collection module includes ImageJ image processing software.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. The present invention forms the core architecture of an intelligent recognition and analysis system for urological stone images through the mutual cooperation of three groups of algorithm units. The adjusted stone segmentation threshold algorithm unit comprehensively considers the basic stone segmentation threshold Bst, the comprehensive image score value Ism, the radiation dose Ed and the image scanning layer thickness Thk to calculate the adjusted stone segmentation threshold Ast. It can dynamically adjust the stone segmentation threshold in the image segmentation algorithm of the intelligent recognition system of stone images under different scanning conditions (such as low dose, low scanning layer thickness, and obese patients), which helps the system to more accurately identify the boundaries and ranges of stones, improve the accuracy of stone segmentation, and enhance the recognition accuracy of stone images, providing a scientific and reliable data basis for subsequent diagnosis and treatment.

[0038] 2. The present invention uses a comprehensive image scoring value algorithm unit to comprehensively consider the image signal-to-noise ratio SNR, image scanning layer thickness Thk, and CT value gradient CTg to calculate the comprehensive image scoring value Ism, which can more comprehensively and accurately identify and evaluate the patient's stone images, and reduce misdiagnosis and missed diagnosis caused by a single factor, so that the intelligent recognition and analysis system can better identify and distinguish stones and surrounding tissues in the image, thereby improving the detection rate of stones. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a flowchart of an intelligent recognition and analysis system for urological stone images;

[0040] Figure 2 The figure is a schematic diagram of the overall structure of an intelligent recognition and analysis system for urological stone images. DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0042] For example 1, please refer to Figures 1 to 2 The present invention provides an intelligent recognition and analysis system for urological stone images, comprising the following steps:

[0043] The data collection module is used to obtain the patient's first imaging data through the metadata of the imaging device, including:

[0044] Obtain the image scanning layer thickness Thk, radiation dose Ed, and stenosis diameter Dus through the metadata of the imaging device;

[0045] The image signal-to-noise ratio (SNR) and CT value gradient (CTg) are obtained through the built-in image processing software of the imaging device;

[0046] Obtain the patient's BMI value from the patient's electronic medical record and upload it to the database;

[0047] The data preprocessing module is used to decode and preprocess the data information in the database to obtain the parameters involved in the calculation in the calculation processing module;

[0048] Computational processing module,

[0049] It is used to substitute the parameter values ​​obtained after decoding preprocessing into the comprehensive image score value algorithm unit to calculate the comprehensive image score value Ism, and take the ideal value of the parameter to calculate the maximum value of the comprehensive image score value Ism max , and upload them to the database together;

[0050] The standard threshold value Y1 of image quality is set as Ism in the database. max / 2, when the comprehensive image score value Ism>Y1, the comprehensive image score value Ism is input as an input parameter into the adjusted stone segmentation threshold algorithm unit to calculate the adjusted stone segmentation threshold Ast;

[0051] Inputting the adjusted stone segmentation threshold Ast into the segmentation algorithm of the intelligent recognition system for stone images to perform stone image segmentation and recognition;

[0052] When the comprehensive image score value Ism is less than Y1, the comprehensive image score value Ism is input as an input parameter into the adjusted image scanning layer thickness algorithm unit to calculate the adjusted image scanning layer thickness Thk again ;

[0053] According to the adjusted image scanning layer thickness Thk again , perform a second scan of the patient's urological stone images and re-calculate the comprehensive image scoring value algorithm unit.

[0054] Specific image segmentation and recognition is a relatively mature existing technology in the field of image processing, and will not be described in detail here.

[0055] In this embodiment:

[0056] The present invention forms the core architecture of an intelligent recognition and analysis system for urological stone images through the mutual cooperation of three algorithm units. The comprehensive image scoring value algorithm unit comprehensively considers influencing factors such as image signal-to-noise ratio (SNR), image scanning layer thickness (Thk), and CT value gradient (CTg) to comprehensively calculate the comprehensive image scoring value (Ism). This can more comprehensively and accurately evaluate image quality, reduce misdiagnosis and missed diagnosis caused by a single factor, and help the intelligent recognition and analysis system better distinguish stones from surrounding tissues, improve the detection rate of stones, and better perform image recognition. The adjusted stone segmentation threshold algorithm unit comprehensively considers the basic stone segmentation threshold (Bst), the comprehensive image scoring value (Ism), the radiation dose (Ed), and the image scanning layer thickness (Thk) to calculate the adjusted stone segmentation threshold (Ast). That is, the stone segmentation threshold can be dynamically adjusted according to factors such as image quality, which helps to more accurately identify the boundaries and ranges of stones, improve the accuracy of stone segmentation, and provide a scientific and reliable data basis for subsequent diagnosis and treatment.

[0057] See also Figures 1 to 2 , the comprehensive image scoring value algorithm unit is as follows:

[0058] ;

[0059] in:

[0060] Ism represents the comprehensive image score value:

[0061] SNR stands for image signal-to-noise ratio, which is the ratio of the image signal average to the standard deviation of the background noise and is obtained using the ImageJ image processing software built into the imaging device;

[0062] Thk represents the image scanning slice thickness, which is the slice thickness when the imaging device scans and is obtained from the metadata of the imaging device;

[0063] CTg stands for CT value gradient, which is the CT value gradient between the stone and the surrounding tissue, and is obtained through the built-in image processing software of the imaging device;

[0064] BMI represents the patient's BMI value, which was obtained from the patient's electronic medical record;

[0065] In the formula calculation:

[0066] This part reflects the nonlinear effect of the image signal-to-noise ratio (SNR) on the comprehensive image score value through the power function. As the image signal-to-noise ratio (SNR) increases, the calculated comprehensive image score value (Ism) increases, but the growth rate gradually slows down. It is used to slow down the excessive effect on the comprehensive image score value (Ism) when the image signal-to-noise ratio (SNR) is too large.

[0067] The combined effect of coupled scan slice thickness THK and CT value gradient CTg on the integrated image score Ism is used to balance spatial resolution and low contrast detectability. Specifically:

[0068] As the molecular scanning layer thickness THK decreases, This part of the value will increase and increase the calculated comprehensive image score value Ism;

[0069] When the CT value gradient CTG is high, it means that the image edge is clear, the denominator is close to 1, and the value of this coupling term is close to the value of the scanning layer thickness THK;

[0070] When the CT value gradient CTG is low, it means that the image is blurred, the denominator increases, this part of the coupling term decreases, and the calculated comprehensive image score value Ism decreases. In this case, in order to ensure image quality, the system will select a thinner slice thickness when the CT value gradient CTG is low;

[0071] When the patient's BMI value is greater than or equal to 25, it is considered to be overweight, and artifacts will appear during image capture, reducing image quality. The standard value of the BMI value is set to 25 at the numerator. This section normalizes the effect of the patient's BMI on the comprehensive imaging score using a baseline value of 25. Specifically:

[0072] When the patient's BMI value is equal to 25, The value of this part is 1, which means that the patient's current weight will not affect the calculation of the comprehensive imaging score Ism;

[0073] When the patient's BMI is greater than 25, This part of the value is greater than 1. As the patient's BMI value increases, The value of this part will be reduced to between 0 and 1, thereby reducing the calculated comprehensive image score value Ism;

[0074] When the patient's BMI is less than 25, Some values ​​are less than 1. As the patient's BMI decreases, The value of this part will increase, thereby increasing the calculated comprehensive image score value Ism.

[0075] In this embodiment, the image signal-to-noise ratio (SNR) reflects the clarity of the image. The higher the image signal-to-noise ratio (SNR), the less noise in the image, and the clearer the stone features. The image scan slice thickness (Thk) affects the observation of stone details. A thinner slice thickness can provide more detailed stone morphological information. The CT value gradient (CTg) reflects the density difference between the stone and surrounding tissue, which helps to accurately identify the stone. The patient's BMI value is related to the absorption and scattering of X-rays by body tissue and can affect image quality. The comprehensive image scoring algorithm unit comprehensively considers the image signal-to-noise ratio (SNR), image scan slice thickness (Thk), CT value gradient (CTg), and the patient's BMI value (BMI) to calculate the comprehensive image scoring value (Ism). This can more comprehensively and accurately assess image quality and reduce misdiagnosis and missed diagnosis caused by a single factor. The appearance of urological stones on images may vary depending on factors such as stone composition, size, and location. The comprehensive image scoring value (Ism) serves as a comprehensive indicator in the system database, which can help the intelligent recognition and analysis system better distinguish stones from surrounding tissue, improve the stone detection rate, and better perform image recognition.

[0076] See also Figures 1 to 2 , the adjusted stone segmentation threshold algorithm unit is as follows:

[0077] ;

[0078] in:

[0079] Ast represents the adjusted stone segmentation threshold;

[0080] Bst stands for the basic segmentation threshold for stones, which is the preset threshold for identifying stones in the image segmentation algorithm;

[0081] Ism represents the comprehensive imaging score value;

[0082] Ism max Represents the maximum value of the comprehensive image score, which is the theoretical upper limit of image quality and represents the image score value under ideal conditions;

[0083] Ed represents the radiation dose, which is obtained from the metadata of the imaging device;

[0084] Thk represents the image scanning layer thickness;

[0085] α represents the compensation coefficient, α∈(0.5,3) can be self-adjusted in the adjusted stone segmentation threshold algorithm unit, specifically:

[0086] When the radiation dose Ed used in the CT examination is low, the compensation coefficient α will be larger (2-3) to increase the calculated adjusted stone segmentation threshold Ast. This is used to reduce noise misjudgment in the image when the radiation dose Ed is low, thereby avoiding the missed detection of small stones.

[0087] In the formula calculation:

[0088] This part is achieved by the maximum value of the comprehensive image score Ism max , the effect of the comprehensive image score Ism on the stone segmentation threshold is normalized to the mathematical range of 0 to 1. Specifically:

[0089] As the comprehensive image score value Ism increases, the closer the comprehensive image score value Ism is to the maximum image score value Ism max , This term is approximately equal to 1, indicating that the image quality of the patient is close to the ideal value and there is no need to adjust the stone segmentation threshold;

[0090] As the comprehensive image score Ism decreases, If this item increases, it means that the image quality of the patient is poor and the stone segmentation threshold needs to be increased to avoid missing small stones in the image.

[0091] Regarding the image scan slice thickness Thk: As the image scan slice thickness THK increases, the spatial resolution will be reduced and the stone edge will be blurred. Therefore, the threshold needs to be increased to reduce false positives.

[0092] Regarding radiation dose Ed: Reducing the radiation dose ED will increase the noise image during image capture, so the threshold needs to be increased to suppress noise misjudgment;

[0093] This part actually represents the coupling effect of image blur and noise by dividing the radiation dose Ed by the image scanning layer thickness Thk. Specifically:

[0094] When the scanning layer thickness Thk increases and the radiation dose Ed decreases, The ratio of this part increases significantly, and the calculated adjusted stone segmentation threshold value represents an increase in the adjusted stone segmentation threshold value Ast, which means that the spatial resolution in the image capture is low and there is a lot of noise, and the segmentation threshold value needs to be significantly compensated. The calculated adjusted stone segmentation threshold value Ast increases;

[0095] It is worth noting that the logarithmic function can be used here to The coupled image of this part of the image blur and noise is mapped to the logarithmic space, so that the threshold compensation is The increase of this part is gradual rather than linear.

[0096] In this embodiment:

[0097] In the intelligent recognition and analysis system of urological stone images, in traditional stone image segmentation algorithms, the stone style threshold is usually fixed and cannot adapt to different scanning conditions (such as low dose, low scanning layer thickness, and obese patients), which can easily lead to false positives or missed detections. However, the adjusted stone segmentation threshold algorithm unit comprehensively considers the basic stone segmentation threshold Bst, the comprehensive image score value Ism, the radiation dose Ed, and the image scanning layer thickness Thk to calculate the adjusted stone segmentation threshold Ast. That is, the stone segmentation threshold can be dynamically adjusted according to factors such as image quality, which helps to more accurately identify the boundaries and range of stones, improve the accuracy of stone segmentation, and provide a scientific and reliable data basis for subsequent diagnosis and treatment.

[0098] By automatically calculating the adjusted stone segmentation threshold Ast through the intelligent recognition and analysis system of stone images, the time and energy required for doctors to manually adjust the segmentation threshold can be reduced, making the image analysis process more efficient. Doctors can obtain accurate stone information more quickly, improving the efficiency of medical work. In addition, by incorporating the radiation dose Ed into the calculation of the adjusted stone segmentation threshold algorithm unit, the radiation exposure to patients can be reduced while ensuring image quality, which helps protect the health of patients and reduce potential risks caused by radiation.

[0099] See also Figures 1 to 2 , the adjusted image scanning layer thickness algorithm unit is as follows:

[0100] ;

[0101] in:

[0102] Thk again Represents the thickness of the image scan layer after adjustment;

[0103] Thk represents the image scanning layer thickness;

[0104] Dus stands for the stricture diameter, which is the diameter of the stricture segment of the patient's ureter and is obtained from the metadata of the imaging device;

[0105] Dus ref The normal value of the diameter of the stricture segment is the diameter of the stricture segment of the ureter in a normal person without urinary diseases.

[0106] Ism represents the comprehensive imaging score value;

[0107] Ism max Represents the maximum value of the comprehensive image score

[0108] β represents the zero division constant, which is 0.1;

[0109] In the formula calculation:

[0110] The smaller the diameter Dus of the patient's stenosis segment, the greater the potential risk of stones to the patient. Therefore, thinner CT slices are needed for scanning to improve image quality and capture small stones more clearly, thereby reducing the potential risk of small stones to the patient. This part reflects the adjustment requirements of the layer thickness according to the degree of ureteral stenosis through the hyperbolic tangent function. Specifically:

[0111] When the diameter of the stenosis segment Dus is close to the normal value of the stenosis segment diameter Dus ref , The value of this part of the function approaches 0. This part of the value is close to 1, which means that the diameter Dus of the patient's stenosis segment is normal and the slice thickness does not need to be adjusted;

[0112] When the diameter of the stenosis segment Dus is smaller than the normal value Dus ref , The value of this part of the function decreases. This part of the value is reduced to reduce the calculated adjusted image scanning layer thickness Thk again , used to capture finer image details;

[0113] The maximum value of the partial pass comprehensive image score Ism max , the comprehensive image score value Ism is adjusted to the image scanning layer thickness Thk again The impact of is standardized, specifically:

[0114] As the comprehensive image score value Ism increases, when the comprehensive image score value Ism approaches the maximum value of the comprehensive image score value Ism max When , it means the image quality is good. The value of this part is reduced, thereby reducing the calculated adjusted image scanning layer thickness Thk again As the comprehensive image score Ism decreases, it means the image quality is poor, so the calculated adjusted image scanning layer thickness Thk is increased. again , which means that thinner scanning layers are needed to capture finer image details.

[0115] In this embodiment:

[0116] The adjusted image scan layer thickness algorithm unit comprehensively considers the patient's ureteral stricture diameter Dus and the comprehensive image score value Ism to calculate the adjusted image scan layer thickness Thk again On the one hand, according to the adjusted image scanning layer thickness Thk again, it can retake urological stone images for patients whose image quality is lower than the standard threshold Y1, realize the automatic adjustment of image scanning layer thickness Thk, reduce the time and energy required for doctors to manually adjust scanning parameters, and improve the efficiency of medical work. This allows doctors to focus more on the diagnosis and treatment of patients, and improves the quality and level of medical services;

[0117] On the other hand, this algorithm can adjust the image scan layer thickness according to the different patient conditions. For patients with smaller stenosis diameters, the intelligent recognition and analysis system for urological stone images can select a thinner scan layer thickness to more clearly capture small stones, thereby reducing the risk of missed diagnosis due to blurred images. Specifically:

[0118] When the diameter Dus of the patient's stenosis segment is small, the adjusted image scanning layer thickness algorithm unit will force a reduction in layer thickness through the hyperbolic tangent function (tanh) (for example, when Dus = 1mm, the layer thickness will be compressed to 30% of the original value), significantly improving the image resolution at the stenosis and avoiding blurred details caused by excessive layer thickness, thereby improving image quality and diagnostic accuracy.

[0119] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent recognition and analysis system for urological stone images, characterized in that: include: A data collection module is used to obtain the patient's first imaging data through the metadata of the imaging device; The data preprocessing module is used to decode and preprocess the data information in the database to obtain the parameters involved in the calculation in the calculation processing module; Calculation processing module, the specific calculation processing steps are as follows: S1, substitute the parameter values ​​obtained after decoding preprocessing into the comprehensive image score value algorithm unit to calculate the comprehensive image score value Ism, and take the ideal value of the parameter to calculate the maximum value of the comprehensive image score value Ism max ; S2, set the standard threshold value Y1 of image quality in the database as Ism max / 2, when the comprehensive image score value Ism>Y1, the comprehensive image score value Ism is input as an input parameter into the adjusted stone segmentation threshold algorithm unit to calculate the adjusted stone segmentation threshold Ast; S3, inputting the adjusted stone segmentation threshold Ast into the segmentation algorithm of the intelligent stone image recognition system to perform stone image segmentation and recognition; S4, when the comprehensive image score value Ism is less than Y1, the comprehensive image score value Ism is input as an input parameter into the adjusted image scanning layer thickness algorithm unit to calculate the adjusted image scanning layer thickness Thk again ; S5, according to the adjusted image scanning layer thickness Thk again , perform a second scan of the patient's urological stone images and re-calculate the comprehensive image scoring value algorithm unit; The calculation logic of the adjusted stone segmentation threshold algorithm unit is as follows: S21, the maximum value of the comprehensive image score Ism max The influence of the comprehensive image score Ism on the stone segmentation threshold was normalized to a value range of 0 to 1; S22, the coupling effect of image blur and noise is obtained by dividing the radiation dose Ed by the image scanning layer thickness Thk; S23, scaling the coupling effect of image blur and noise by a logarithmic function; S24, multiplying the coupled influence term of image blur and noise after logarithmic function scaling by the influence term of the comprehensive image score value Ism on the stone segmentation threshold and the stone basic segmentation threshold Bst to obtain the adjusted stone segmentation threshold Ast.

2. The intelligent recognition and analysis system for urological stone images according to claim 1, characterized in that: The patient's first imaging data is obtained through the metadata of the imaging device, including: Obtain the image scanning layer thickness Thk, radiation dose Ed, and stenosis diameter Dus through the metadata of the imaging device; The image signal-to-noise ratio (SNR) and CT value gradient (CTg) are obtained through the built-in image processing software of the imaging device; The patient's BMI value is obtained from the patient's electronic medical record and uploaded to the database.

3. The intelligent recognition and analysis system for urological stone images according to claim 1 is characterized by: The calculation and processing module includes a comprehensive image scoring value algorithm unit, an adjusted stone segmentation threshold algorithm unit, and an adjusted image scanning layer thickness algorithm unit.

4. The intelligent recognition and analysis system for urological stone images according to claim 3, characterized in that: The calculation logic of the comprehensive image scoring value algorithm unit is as follows: S11, the influence of the image signal-to-noise ratio (SNR) on the comprehensive image score Ism is amplified by the power function; S12, by coupling the comprehensive influence of the scanning layer thickness Thk and the CT value gradient CTg on the comprehensive image score value Ism, the low contrast detectable value after balancing the spatial resolution is obtained; S13, normalizing the influence item of the patient's BMI value on the comprehensive imaging score value Ism by using the reference value 25 to obtain the influence item of the patient's BMI value on the comprehensive imaging score value Ism; S14, multiplying the influence of the image signal-to-noise ratio (SNR) on the comprehensive image score Ism, the low contrast detectable value after balancing the spatial resolution, and the patient's BMI on the comprehensive image score Ism to obtain the comprehensive image score Ism.

5. The intelligent recognition and analysis system for urological stone images according to claim 4, characterized in that: The calculation logic of the adjusted image scanning layer thickness algorithm unit is as follows: S41, mapping the adjustment requirement of the ureteral stenosis degree to the layer thickness through the hyperbolic tangent function; S42, the maximum value of the comprehensive image score value Ism max The comprehensive image score value Ism is adjusted to the image scanning layer thickness Thk again The influence items are standardized; S43, the adjusted image scanning layer thickness Thk is calculated by multiplying the image scanning layer thickness Thk by the standardized comprehensive image score value Ism. again The influencing factors and the adjustment requirement of the layer thickness due to the degree of ureteral stenosis are used to obtain the adjusted image scan layer thickness.

6. The intelligent recognition and analysis system for urological stone images according to claim 1, characterized in that: The decoding preprocessing includes data cleaning and data standardization.

7. The intelligent recognition and analysis system for urological stone images according to claim 1, characterized in that: The equipment used in the data collection module includes ImageJ image processing software.

Images