Artificial intelligence-based method and system for measuring glomerular basement membrane thickness under electron microscope

Abstract

The present invention belongs to the technical field of glomerular basement membrane thickness measurement, and specifically relates to a method and system for measuring glomerular basement membrane thickness under an electron microscope based on artificial intelligence. The invention can accurately measure the thickness of the glomerular basement membrane based on the continuously collected electron microscope image data of the glomerular basement membrane, avoid the influence of accidental factors, and provide an important reference basis for the early diagnosis and treatment of kidney diseases. Through automated image processing and analysis, the accuracy and efficiency of glomerular basement membrane thickness detection are improved, and the errors of human operation are reduced. In addition, the kidney status can be automatically evaluated according to the measurement results, and timely intervention measures can be provided in abnormal situations, making the diagnosis and treatment of kidney diseases more accurate and timely, and also helping doctors to formulate more accurate treatment plans, and bringing better treatment effects and quality of life to patients.

Description

Technical Field

[0001] The invention belongs to the technical field of glomerular basement membrane thickness measurement, and in particular relates to a method and system for measuring glomerular basement membrane thickness under an electron microscope based on artificial intelligence. Background Art

[0002] In recent years, artificial intelligence has developed rapidly, bringing many opportunities for medical progress. As the prevalence of kidney disease increases year by year worldwide, early detection, early diagnosis, and early treatment are critical to improving the prognosis of kidney disease. Artificial intelligence provides effective assistance in the diagnosis, treatment, prognosis, and prediction of kidney disease, which can reduce the workload of kidney disease diagnosis and treatment and improve the efficiency of kidney disease management. The glomerular basement membrane (GBM) is located between the podocytes and the specialized fenestrated endothelial cells on the glomerular capillaries. During glomerulogenesis, the matrix synthesized by immature podocytes and endothelial cells continues to develop and fuse. In glomerular filtration, GBM plays an important role as a mechanical barrier and a charge barrier. With the development of the body, GBM continues to thicken. The thickness of children is generally 275~315 nm, and the thickness increases to 330~460nm when growing and developing to adults. The thickness of the glomerular basement membrane under electron microscopy is the key basis for the diagnosis of various kidney diseases. Many diseases are manifested by changes in the thickness of the glomerular basement membrane. The diagnosis of thin basement membrane nephropathy (TBMN) relies on transmission electron microscopy, which is manifested as diffuse thinning of the GBM under electron microscopy. Diabetic nephropathy is mainly manifested by diffuse and uniform thickening of the glomerular GBM under electron microscopy. The main pathological features of membranous nephropathy, which has a higher incidence in the population, are significant thickening of the glomerular GBM and subepithelial immunoglobulin deposits. Anti-basement membrane nephropathy is mainly manifested by rupture of the glomerular basement membrane. Various genetically related gene mutations that encode extracellular matrix proteins on the GBM can cause basement membrane changes. Among them, Alport syndrome is a common hereditary kidney disease related to the basement membrane, which is mainly manifested by stratified changes in the glomerular basement membrane.

[0003] However, the current measurement method of the glomerular basement membrane is still to measure after random sampling under an electron microscope, which cannot accurately and comprehensively reflect the thickness of the glomerular basement membrane. Human subjective judgment will inevitably have large errors, which may lead to misdiagnosis and misjudgment. Based on this, this program proposes a method that can quickly identify the basement membrane in continuous projection electron microscope images and accurately measure the thickness of the basement membrane, providing objective diagnostic descriptions for renal pathologists and improving the accuracy of pathological diagnosis of kidney diseases. Summary of the invention

[0004] The purpose of the present invention is to provide an artificial intelligence-based electron microscope glomerular basement membrane thickness measurement method and system, which can accurately measure the thickness of the glomerular basement membrane, thereby providing more accurate data support for the diagnosis of kidney diseases.

[0005] The technical solution adopted by the present invention is as follows:

[0006] The method for measuring the thickness of the glomerular basement membrane under an electron microscope based on artificial intelligence includes:

[0007] Acquiring electron microscope image data of the glomerulus, performing feature extraction on the electron microscope image data, outputting a regional feature image of the glomerular basement membrane, and then preprocessing the regional feature image of the glomerular basement membrane to obtain basement image data;

[0008] Extracting the edge curve of the glomerular basement membrane according to the regional characteristic image of the glomerular basement membrane, calculating the thickness of the glomerular basement membrane according to the edge curve, and recording the thickness of the glomerular basement membrane as a reference characteristic parameter;

[0009] Evaluating and processing the reference characteristic parameters, and determining the kidney status according to the evaluation results, wherein the kidney status includes a normal state and an abnormal state;

[0010] In the normal state, it is indicated that the thickness of the glomerular basement membrane is within a normal range, and continuous monitoring is performed, and the thickness of the glomerular basement membrane within the normal range is recorded in a test log;

[0011] In the abnormal state, it indicates that the thickness of the glomerular basement membrane exceeds the normal range, and the detection deviation of the glomerular basement membrane is output, and the kidney abnormality level is determined based on the detection deviation.

[0012] In a preferred embodiment, the step of preprocessing the regional characteristic image of the glomerular basement membrane to obtain basement image data includes:

[0013] Acquiring electron microscope image data of the glomerulus, and performing denoising processing on the electron microscope image data to eliminate noise interference in the electron microscope image data;

[0014] Performing grayscale conversion on the denoised electron microscope image data to convert the electron microscope image data into a grayscale image;

[0015] Performing edge detection on the grayscale image to highlight edge information of the glomerular basement membrane and simultaneously enhancing edge features of the glomerular basement membrane;

[0016] The glomerular basement membrane is segmented according to the edge features of the glomerular basement membrane, and the glomerular basement membrane is separated from the background of the grayscale image to form a regional feature image, which is recorded as basement image data.

[0017] In a preferred embodiment, the step of calculating the thickness of the glomerular basement membrane according to the edge curve and recording the thickness of the glomerular basement membrane as a reference characteristic parameter comprises:

[0018] constructing a virtual coordinate system according to the base image data, and calibrating the inflection point coordinates of the edge inflection point in the edge curve in the virtual coordinate system;

[0019] Constructing a plurality of closed curves according to every three edge inflection points, wherein the closed curves are circular;

[0020] Selecting a closed curve corresponding to the outer circle of the glomerular basement membrane and having the largest radius, and marking it as a standard outer circle curve;

[0021] Selecting a closed curve corresponding to the inner circle of the glomerular basement membrane and having the smallest radius, and marking it as a standard inner circle curve;

[0022] The thickness of the glomerular basement membrane is calculated based on the standard outer circle curve and the standard inner circle curve, and recorded as a reference characteristic parameter.

[0023] In a preferred embodiment, the step of calculating the thickness of the glomerular basement membrane based on the standard outer circle curve and the standard inner circle curve and recording it as a reference characteristic parameter comprises:

[0024] A plurality of sampling nodes are arranged on the standard outer circle curve, wherein the plurality of sampling nodes are distributed in an equidistant manner in a circular shape on the standard outer circle curve;

[0025] Taking each of the sampling nodes as a reference, a virtual line segment is extended toward the center of the standard inner circle curve, and the intersection point between the virtual line segment and the standard inner circle curve is recorded as a measurement intersection point;

[0026] The distance between the measured intersection point and the sampling node under the same virtual line segment was calculated as the thickness of the glomerular basement membrane and recorded synchronously as the benchmark characteristic parameter.

[0027] In a preferred embodiment, the step of evaluating the reference characteristic parameters and determining the kidney status according to the evaluation results includes:

[0028] Obtaining all the reference characteristic parameters;

[0029] Obtaining an evaluation interval, and comparing the benchmark characteristic parameter with the evaluation interval;

[0030] If all the reference characteristic parameters fall within the evaluation interval, it indicates that the thickness of the glomerular basement membrane is normal, and the corresponding kidney state is recorded as a normal state;

[0031] If there are any baseline characteristic parameters that do not fall within the evaluation interval, the number of baseline characteristic parameters that do not fall within the evaluation interval is counted and recorded as post-evaluation condition parameters, and when the post-evaluation condition parameters exceed the preset evaluation threshold, the corresponding kidney state is recorded as an abnormal state, otherwise, it is still recorded as a normal state.

[0032] In a preferred embodiment, in the normal state, the historical detected thickness of the glomerular basement membrane is retrieved from the detection log and recorded as a historical condition parameter;

[0033] Obtaining an evaluation function, inputting the historical condition parameter into the evaluation function, and recording the output result of the evaluation function as a trend condition parameter;

[0034] Obtaining edge values ​​of the evaluation interval, wherein the edge values ​​include an upper limit value and a lower limit value of the evaluation interval;

[0035] Based on the edge value and trend condition parameters, the predicted safety duration of the glomerular basement membrane under the normal state is calculated and output.

[0036] In a preferred embodiment, the step of outputting the detection deviation of the glomerular basement membrane comprises:

[0037] Obtaining all baseline characteristic parameters that do not fall within the evaluation interval under the abnormal state, and recording them as abnormal condition parameters;

[0038] Performing difference processing on the abnormal condition parameter and the edge value of the evaluation interval to obtain an initial thickness deviation, wherein the initial thickness deviation includes a first thickness deviation higher than an upper limit value of the evaluation interval and a second thickness deviation lower than a lower limit value of the evaluation interval;

[0039] A measurement function is obtained, and the first thickness deviation and the second thickness deviation are respectively input into the measurement function, and the output result of the measurement function is recorded as the detection deviation of the glomerular basement membrane.

[0040] In a preferred embodiment, the step of determining the grade of kidney abnormality based on the detected deviation comprises:

[0041] Obtaining the detection deviation of the glomerular basement membrane under the abnormal state, and recording it as a grading condition parameter;

[0042] Obtaining evaluation intervals, wherein each evaluation interval corresponds to an abnormality level;

[0043] The grading condition parameters are compared with the evaluation interval, and the kidney abnormality grade corresponding to the glomerular basement membrane is matched and output.

[0044] The present invention also provides an artificial intelligence-based electron microscope glomerular basement membrane thickness measurement system, using the artificial intelligence-based electron microscope glomerular basement membrane thickness measurement method, comprising:

[0045] An image acquisition module, the image acquisition module is used to obtain electron microscope image data of the glomerulus, perform feature extraction on the electron microscope image data, output a regional feature image of the glomerular basement membrane, and then preprocess the regional feature image of the glomerular basement membrane to obtain basement image data;

[0046] A basement membrane thickness calculation module, the basement membrane thickness calculation module is used to extract the edge curve of the glomerular basement membrane according to the regional characteristic image of the glomerular basement membrane, calculate the thickness of the glomerular basement membrane according to the edge curve, and record the thickness of the glomerular basement membrane as a reference characteristic parameter;

[0047] A state evaluation module, the state evaluation module is used to evaluate the reference characteristic parameters and determine the kidney state according to the evaluation result, wherein the kidney state includes a normal state and an abnormal state;

[0048] A normal intervention module, wherein the normal intervention module is used to indicate that the thickness of the glomerular basement membrane is within a normal range under the normal state, and to continuously monitor and record the thickness of the glomerular basement membrane within the normal range into a detection log;

[0049] The abnormal intervention module is used to indicate that the thickness of the glomerular basement membrane exceeds the normal range in the abnormal state, and output the detection deviation of the glomerular basement membrane, and then determine the abnormality level of the kidney based on the detection deviation.

[0050] And, an electronic device, the electronic device comprising:

[0051] at least one processor;

[0052] and a memory communicatively coupled to the at least one processor;

[0053] Wherein, the memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the above-mentioned artificial intelligence-based electron microscope glomerular basement membrane thickness measurement method.

[0054] The technical effects achieved by the present invention are:

[0055] The present invention is based on continuously collected electron microscope image data of the glomerular basement membrane, and can accurately measure the thickness of the glomerular basement membrane, avoiding the influence of accidental factors, and providing an important reference for the early diagnosis and treatment of kidney disease. Through automated image processing and analysis, the accuracy and efficiency of glomerular basement membrane thickness detection are improved, and the errors of human operation are reduced. In addition, the present invention can automatically evaluate the kidney status according to the measurement results, and provide timely intervention measures in abnormal situations, making the diagnosis and treatment of kidney disease more accurate and timely, and also helping doctors to formulate more accurate treatment plans, and bringing better treatment effects and quality of life to patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic flow chart of the method of the present invention;

[0057] Figure 2 It is a schematic diagram of the system module of the present invention;

[0058] Figure 3 It is a schematic diagram of the structure of an electronic device of the present invention. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0060] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0061] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present invention. The term "in a preferred embodiment" that appears in different places in this specification does not refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0062] With the continuous advancement of artificial intelligence technology, its application in the medical field is becoming more and more extensive. The method for measuring the thickness of glomerular basement membrane under an electron microscope of the present invention can achieve accurate measurement and status evaluation of the thickness of glomerular basement membrane through artificial intelligence technology. This method not only improves the accuracy and efficiency of detection, but also can detect kidney lesions in time, providing strong support for clinical diagnosis and treatment.

[0063] See also Figure 1 As shown, the present invention provides a method for measuring the thickness of glomerular basement membrane under an electron microscope based on artificial intelligence, comprising:

[0064] S1, acquiring electron microscope image data of glomeruli, performing feature extraction on the electron microscope image data, outputting a regional feature image of the glomerular basement membrane, and then preprocessing the regional feature image of the glomerular basement membrane to obtain basement image data;

[0065] In step S1, when measuring the thickness of the glomerular basement membrane, it is first necessary to obtain electron microscope image data of the glomerulus. The electron microscope image data is obtained by taking pictures with an electron microscope and can clearly show the microscopic structure of the glomerulus. After the electron microscope image data is output, the electron microscope image data will be preprocessed synchronously. The preprocessing includes operations such as image enhancement and denoising to improve the image quality for subsequent processing. Feature extraction is performed on the preprocessed electron microscope image data. By using deep learning algorithms, such as convolutional neural networks (CNN), regional feature images of the glomerular basement membrane can be extracted from the electron microscope image data. Next, the regional feature images of the glomerular basement membrane are further preprocessed to obtain clearer and more accurate basement image data.

[0066] Specifically, the step of preprocessing the regional characteristic image of the glomerular basement membrane to obtain basement image data includes:

[0067] Acquire electron microscope image data of the glomerulus, and perform denoising on the electron microscope image data to eliminate noise interference in the electron microscope image data;

[0068] Performing grayscale conversion on the denoised electron microscope image data, converting the electron microscope image data into a grayscale image;

[0069] Perform edge detection on the grayscale image to highlight the edge information of the glomerular basement membrane and simultaneously enhance the edge features of the glomerular basement membrane;

[0070] The glomerular basement membrane is segmented according to its edge features, and the glomerular basement membrane is separated from the background of the grayscale image to form a regional feature image, which is recorded as basement image data.

[0071] In the above, in order to effectively pre-process the regional characteristic image of the glomerular basement membrane in order to obtain high-quality basement image data, it is first necessary to obtain the electron microscope image data of the glomerulus. These data usually contain rich details, but may also be interfered by various noises. In order to ensure the accuracy of subsequent processing, the electron microscope image data is first denoised to eliminate noise interference in the image, which usually involves the use of various image processing algorithms, such as median filtering, Gaussian filtering, etc., to ensure that the image quality is significantly improved, and then the denoised electron microscope image data will be converted to grayscale. Grayscale conversion is the process of converting color images into grayscale images, which helps to simplify image data and make it easier for subsequent processing. Grayscale images only contain brightness information, not color information, which enables image processing algorithms to run more efficiently. Through grayscale conversion, electron microscope image data can be converted into grayscale images, laying the foundation for further image analysis and processing. On the basis of grayscale images, edge detection will be performed to highlight the edge information of the glomerular basement membrane and identify important features and structures in the image. By using edge detection algorithms such as Sobel operator, Canny edge detection, etc., the edge features of the glomerular basement membrane can be effectively enhanced to make it more obvious and clear in the image. Finally, regional segmentation will be performed based on the edge features of the glomerular basement membrane. The purpose of regional segmentation is to separate the glomerular basement membrane from the background of the grayscale image to form a clear regional feature image. It usually involves image segmentation algorithms such as threshold segmentation and regional growing to ensure that the glomerular basement membrane is accurately identified and separated. Through regional segmentation, the regional feature image of the glomerular basement membrane can be recorded as basement image data to provide corresponding image materials for subsequent analysis and research.

[0072] S2. extracting the edge curve of the glomerular basement membrane based on the regional feature image of the glomerular basement membrane, calculating the thickness of the glomerular basement membrane based on the edge curve, and recording the thickness of the glomerular basement membrane as a reference feature parameter;

[0073] In step S2, the edge curve of the glomerular basement membrane is extracted based on the regional feature image of the glomerular basement membrane. The edge curve extraction can be achieved using an edge detection algorithm, such as a Canny edge detector. By extracting the edge curve, the boundary of the glomerular basement membrane can be accurately located. Then, the thickness of the glomerular basement membrane can be calculated based on the extracted edge curve. The calculated glomerular basement membrane thickness will be recorded as a benchmark feature parameter for subsequent evaluation and analysis.

[0074] Specifically, the step of calculating the thickness of the glomerular basement membrane according to the edge curve and recording the thickness of the glomerular basement membrane as a reference characteristic parameter includes:

[0075] A virtual coordinate system is constructed according to the base image data, and the inflection point coordinates of the edge inflection point in the edge curve are calibrated in the virtual coordinate system;

[0076] Constructing multiple closed curves based on every three edge inflection points, wherein the closed curves are circular;

[0077] Screening a closed curve corresponding to the outer circle of the glomerular basement membrane and with the largest radius, and marking it as the standard outer circle curve;

[0078] Screening a closed curve corresponding to the inner circle of the glomerular basement membrane and with the smallest radius, and marking it as the standard inner circle curve;

[0079] According to the standard outer circle curve and the standard inner circle curve, the thickness of the glomerular basement membrane is calculated and recorded as the benchmark characteristic parameter;

[0080] In the above, in order to accurately measure the thickness of the glomerular basement membrane, it is first necessary to construct a virtual coordinate system based on the basement image data. In the virtual coordinate system, the edge inflection points in the edge curve are accurately calibrated, and the coordinates of the inflection points in the edge curve are recorded. The inflection points are points with a greater degree of curvature in the edge curve. They can be selected manually or automatically identified using an edge detection algorithm. Then, multiple closed curves are constructed based on every three edge inflection points. The shape of the closed curve here should be circular, which is used to simulate the outer and inner circles of the glomerular basement membrane. Based on this method, it can be ensured that the shape of the closed curve matches the actual glomerular basement membrane structure. After constructing multiple closed curves, the corresponding The corresponding screening work specifically needs to screen out the closed curves corresponding to the outer circle of the glomerular basement membrane, and find out the one with the largest radius, and calibrate this curve as the standard outer circle curve, which represents the maximum possible range of the outer circle of the glomerular basement membrane. Similarly, it is also necessary to screen out the closed curves corresponding to the inner circle of the glomerular basement membrane, and find out the one with the smallest radius, and calibrate this curve as the standard inner circle curve, which represents the minimum possible range of the inner circle of the glomerular basement membrane. Finally, calculations will be performed based on the standard outer circle curve and the standard inner circle curve. By measuring the distance between the standard outer circle curve and the standard inner circle curve, the thickness of the glomerular basement membrane can be obtained and recorded as a benchmark characteristic parameter.

[0081] The step of calculating the thickness of the glomerular basement membrane according to the standard outer circle curve and the standard inner circle curve and recording it as a reference characteristic parameter includes:

[0082] A plurality of sampling nodes are set on the standard outer circle curve, wherein the plurality of sampling nodes are distributed in an annular manner and at equal intervals on the standard outer circle curve;

[0083] Taking each sampling node as a reference, extend a virtual line segment to the center of the standard inner circle curve, and record the intersection point between the virtual line segment and the standard inner circle curve as the measurement intersection point;

[0084] The distance between the measured intersection point and the sampling node under the same virtual line segment is calculated as the thickness of the glomerular basement membrane and recorded synchronously as the benchmark characteristic parameter;

[0085] As mentioned above, in order to accurately measure and record the thickness of the glomerular basement membrane, it is first necessary to evenly set multiple sampling nodes on the standard outer circle curve. The sampling nodes are distributed in a circular shape on the standard outer circle curve and are equidistant from each other. In this way, the distribution of the sampling nodes is ensured to be uniform and representative. Then, taking each sampling node as the starting point, a virtual line segment is extended toward the center of the standard inner circle curve, passing through the standard outer circle curve, and intersecting with the standard inner circle curve. In this way, the intersection between the virtual line segment and the standard inner circle curve can be determined and recorded as the measurement intersection. Then, the position coordinates of these intersections are recorded, and finally the straight-line distance between the measurement intersection and the corresponding sampling node under the same virtual line segment is calculated and recorded as the thickness of the glomerular basement membrane as a benchmark characteristic parameter.

[0086] S3, evaluating and processing the baseline characteristic parameters, and determining the kidney status according to the evaluation results, wherein the kidney status includes a normal state and an abnormal state;

[0087] In step S3, after the baseline characteristic parameters are output, the baseline characteristic parameters need to be evaluated to determine the state of the kidney. The kidney state can be divided into a normal state and an abnormal state. The specific process is as follows:

[0088] Get all benchmark feature parameters;

[0089] Obtaining an evaluation interval, and comparing the benchmark characteristic parameter with the evaluation interval;

[0090] If all the baseline characteristic parameters fall within the evaluation interval, it indicates that the thickness of the glomerular basement membrane is normal, and the corresponding kidney status is recorded as normal;

[0091] If there are baseline characteristic parameters that do not fall within the evaluation interval, the number of baseline characteristic parameters that do not fall within the evaluation interval is counted and recorded as a post-evaluation condition parameter, and if the post-evaluation condition parameter exceeds the preset evaluation threshold, the corresponding kidney state is recorded as an abnormal state, otherwise, it is still recorded as a normal state;

[0092] In the above, in order to accurately evaluate the health status of the kidneys, we first need to fully obtain all relevant baseline characteristic parameters, and then determine a preset evaluation interval. The evaluation interval is based on a large amount of clinical data and medical research, and can reflect the parameter range of whether the kidney function is normal or not. The obtained baseline characteristic parameters are compared with the evaluation interval accordingly, and the health status of the kidneys can be preliminarily judged. If all baseline characteristic parameters fall within the evaluation interval, it means that the baseline characteristic parameters are within the normal range and the thickness of the glomerular basement membrane is also at a normal level. At this time, it can be determined that the kidneys are in a normal state. However, if some baseline characteristic parameters fail to fall within the evaluation interval, it indicates that there may be some abnormal conditions (the glomerular basement membrane is too thick or too thin). At this time, it is necessary to count the number of baseline characteristic parameters that fail to fall within the evaluation interval and record them as post-evaluation condition parameters. If the number of post-evaluation condition parameters exceeds the preset evaluation threshold, it indicates that there may be some problems with the kidneys and further examination and treatment are required. At this time, the corresponding kidney state will be recorded as an abnormal state. On the contrary, if the number of post-evaluation condition parameters does not exceed the preset evaluation threshold, it can be considered that the kidneys are still in a normal state and this result will be recorded.

[0093] S4. Under normal conditions, it indicates that the thickness of the glomerular basement membrane is within the normal range, and is continuously monitored, and the thickness of the glomerular basement membrane within the normal range is recorded in the test log;

[0094] Under normal conditions, it indicates that the thickness of the glomerular basement membrane is within the normal range. At this time, continuous monitoring should be carried out, and the thickness of the glomerular basement membrane within the normal range should be recorded in the test log for long-term tracking and analysis. Under normal conditions, the historical test thickness of the glomerular basement membrane is called from the test log and recorded as a historical condition parameter;

[0095] Obtain an evaluation function, input the historical condition parameter into the evaluation function, and record the output result of the evaluation function as a trend condition parameter;

[0096] Obtaining edge values ​​of the evaluation interval, the edge values ​​include the upper limit value and the lower limit value of the evaluation interval;

[0097] According to the edge value and trend condition parameters, the predicted safe time of the glomerular basement membrane under normal conditions is calculated and output;

[0098] Specifically, under normal conditions, it is necessary to extract the historical test thickness data of the glomerular basement membrane from the test log and record it as a historical condition parameter. The historical condition parameter will be used as a reference benchmark for subsequent analysis and evaluation. Then, it is necessary to obtain the preset evaluation function, where the expression of the evaluation function is: , where Q represents the trend condition parameter, T represents the time length of the detection log record, and n represents the number of records of the historical condition parameter. S i and S i-1 Represents the historical condition parameters of adjacent records. The output of the evaluation function will be recorded as a trend condition parameter to assist in understanding the changing trend of the glomerular basement membrane thickness, thereby providing a basis for future predictions. In order to evaluate more accurately, it is also necessary to obtain the edge value of the evaluation interval. The edge value includes the upper and lower limits of the evaluation interval. Finally, the corresponding calculation will be performed based on the edge value and the trend condition parameter, so as to predict the duration of the glomerular basement membrane in normal condition under normal conditions. The corresponding standard function is: , where t represents the predicted safety time, S b represents the edge value, S d Indicates the thickness of the glomerular basement membrane under the current node. The predicted safety time will provide a reference for clinicians to help them assess the stability of their patients' kidney health and arrange follow-up and further examination plans accordingly. If the predicted safety time is short, more frequent monitoring and evaluation may be required to ensure that any potential health problems are discovered and treated in a timely manner.

[0099] S5. In an abnormal state, it indicates that the thickness of the glomerular basement membrane exceeds the normal range, and the detection deviation of the glomerular basement membrane is output, and the abnormality level of the kidney is determined based on the detection deviation.

[0100] In step S5, in an abnormal state, it indicates that the thickness of the glomerular basement membrane exceeds the normal range. At this time, it is necessary to output the detection deviation of the glomerular basement membrane, that is, the deviation between the actual measurement value and the normal range. According to the size of the detection deviation, the abnormality level of the kidney can be determined, such as mild abnormality, moderate abnormality or severe abnormality, which helps doctors to evaluate the severity of kidney disease and formulate corresponding treatment plans;

[0101] The step of outputting the detection deviation of the glomerular basement membrane includes:

[0102] Obtain all baseline characteristic parameters that do not fall within the evaluation interval under abnormal conditions and record them as abnormal condition parameters;

[0103] Performing difference processing on the abnormal condition parameter and the edge value of the evaluation interval to obtain an initial thickness deviation, wherein the initial thickness deviation includes a first thickness deviation higher than an upper limit value of the evaluation interval and a second thickness deviation lower than a lower limit value of the evaluation interval;

[0104] Obtaining a measurement function, and inputting the first thickness deviation and the second thickness deviation into the measurement function respectively, and recording the output result of the measurement function as the detection deviation of the glomerular basement membrane;

[0105] Specifically, first, it is necessary to obtain all the baseline characteristic parameters that do not fall into the evaluation interval under abnormal conditions and record the abnormal condition parameters, and then perform difference processing on the edge values ​​of the abnormal condition parameters and the evaluation interval. The evaluation interval is usually defined by the parameter values ​​under normal physiological conditions, and the abnormal condition parameters represent the situation of deviation from the normal state. By performing difference processing, the initial thickness deviation can be obtained. The initial thickness deviation involves the thickness deviation of the glomerular basement membrane in continuous projection electron microscope images. The initial thickness deviation includes two parts: the first thickness deviation, that is, the part higher than the upper limit of the evaluation interval, and the second thickness deviation, that is, the part lower than the lower limit of the evaluation interval. Finally, it is necessary to obtain a preset measurement function. The measurement function can calculate the detection deviation of the glomerular basement membrane according to the input parameters. By taking the thickness deviation of the glomerular basement membrane under a large number of electron microscope images as the basis, the influence of accidental factors can be avoided. Specifically, the first thickness deviation and the second thickness deviation are respectively input into the measurement function. After calculation, the output result of the measurement function is recorded as the final glomerular basement membrane detection deviation, wherein the expression of the measurement function is: , where h represents the detection deviation of the glomerular basement membrane, and m represents the amount of the first thickness deviation or the second thickness deviation. f j It represents the first thickness deviation or the second thickness deviation. The detection deviation reflects the degree to which the thickness of the glomerular basement membrane deviates from the normal physiological state under abnormal conditions, and has important reference value for clinical diagnosis and treatment.

[0106] Secondly, the steps to determine the grade of renal abnormality based on the detection deviation include:

[0107] Obtain the detection deviation of the glomerular basement membrane under abnormal conditions and record it as a grading condition parameter;

[0108] Obtaining evaluation intervals, where each evaluation interval corresponds to an abnormality level;

[0109] The grading condition parameters are compared with the evaluation interval, and the glomerular basement membrane corresponding to the kidney abnormality grade is matched and output.

[0110] The detailed steps for determining the grade of renal abnormality based on the test deviation are as follows:

[0111] Specifically, it is first necessary to obtain the detection bias of the glomerular basement membrane under abnormal conditions as the basis for subsequent grading, and record it as a grading condition parameter for subsequent comparison and analysis. Then, it is necessary to determine the evaluation interval. The evaluation interval refers to a series of pre-set numerical ranges. Each evaluation interval corresponds to a specific kidney abnormality level, which is based on a large amount of clinical data and medical research and can accurately reflect the health of the kidneys. Finally, the grading condition parameters are compared with the evaluation interval to determine which evaluation interval the detection bias of the glomerular basement membrane belongs to, thereby matching and outputting the corresponding kidney abnormality level, so as to help doctors better understand the patient's kidney health status and formulate corresponding treatment plans.

[0112] See also Figure 2 , an artificial intelligence-based electron microscope glomerular basement membrane thickness measurement system, using the above-mentioned artificial intelligence-based electron microscope glomerular basement membrane thickness measurement method, including:

[0113] An image acquisition module is used to obtain electron microscope image data of the glomerulus, perform feature extraction on the electron microscope image data, output a regional feature image of the glomerular basement membrane, and then pre-process the regional feature image of the glomerular basement membrane to obtain basement image data;

[0114] A basement membrane thickness calculation module is used to extract the edge curve of the glomerular basement membrane based on the regional feature image of the glomerular basement membrane, calculate the thickness of the glomerular basement membrane based on the edge curve, and record the thickness of the glomerular basement membrane as a reference feature parameter;

[0115] A state evaluation module, which is used to evaluate the baseline characteristic parameters and determine the kidney state according to the evaluation results, wherein the kidney state includes a normal state and an abnormal state;

[0116] Normal intervention module: The normal intervention module is used to indicate that the thickness of the glomerular basement membrane is within the normal range under normal conditions, and to continuously monitor and record the thickness of the glomerular basement membrane within the normal range into the detection log;

[0117] The abnormal intervention module is used in abnormal conditions, indicating that the thickness of the glomerular basement membrane exceeds the normal range, and outputs the detection deviation of the glomerular basement membrane, and then determines the abnormality level of the kidney based on the detection deviation.

[0118] In the above, the system includes an image acquisition module, a basement membrane thickness calculation module, a state evaluation module, a normal intervention module and an abnormal intervention module. The image acquisition module is responsible for acquiring the electron microscope image data of the glomerulus. Specifically, through high-precision imaging equipment, the detailed image of the glomerulus is captured, and preliminary processing is performed simultaneously. Subsequently, the basement membrane area in the electron microscope image data is extracted to generate a regional feature image containing the glomerular basement membrane. After further preprocessing, the regional feature image finally obtains clear basement image data, which provides a basis for subsequent analysis and calculation. The basement membrane thickness calculation module uses image processing and pattern recognition technology to extract edge curves from the regional feature images of the glomerular basement membrane. By analyzing the edge curves, the thickness of the glomerular basement membrane can be accurately calculated, and the calculation results are recorded as benchmark feature parameters. The state evaluation module is responsible for receiving the benchmark feature parameters from the basement membrane thickness calculation module in the system, and performing corresponding evaluation processing on the benchmark feature parameters. By comparing the benchmark feature parameters with the parameter values ​​within the normal range, the state evaluation module can determine whether the kidney is in a normal state. Kidney status is divided into normal and abnormal states. The system will give a clear diagnosis based on the evaluation results. Under normal conditions, the normal intervention module will play a role and conduct continuous monitoring, and record the data within the normal range in the detection log, so that doctors and researchers can access these data at any time for long-term tracking and evaluation of kidney health. The abnormal intervention module is activated when an abnormal state is detected. When the thickness of the glomerular basement membrane exceeds the normal range, an alarm will be issued, and the detection deviation of the glomerular basement membrane will be output, and the abnormal level of the kidney will be further determined, thereby providing doctors with more detailed diagnostic basis. Doctors can then take appropriate treatment measures according to the abnormal level to ensure that the patient's kidney health is maintained in a timely and effective manner.

[0119] See also Figure 3 , an electronic device, the electronic device comprising:

[0120] at least one processor;

[0121] and a memory communicatively coupled to the at least one processor;

[0122] Wherein, the memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the above-mentioned artificial intelligence-based electron microscope glomerular basement membrane thickness measurement method.

[0123] The processor of the above electronic device can be any type of central processing unit (CPU) or graphics processing unit (GPU), which can process complex image data and execute advanced algorithms. The memory of the electronic device can be random access memory (RAM) or non-volatile memory, such as a solid state drive (SSD), which is used to store program code and temporary data. The computer program contains a series of instructions that guide the processor to perform specific tasks, such as image acquisition, edge detection, thickness calculation and state evaluation. The electronic device can also include an operator, input device and output device to support the operation of the entire measurement system. The operator of the electronic device is responsible for performing mathematical and logical operations to ensure the accuracy of data processing. The input device allows the user to interact with the system, such as inputting instructions and parameters through a keyboard, mouse or touch screen, and the output device is used to display the measurement results and system status, such as a display, printer, etc.

[0124] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method 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, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0125] The above is only a preferred embodiment of the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications should also be considered as the protection scope of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention shall be implemented according to the conventional means in the art unless otherwise specified and limited.

Claims

1. A method for measuring glomerular basement membrane thickness under an electron microscope based on artificial intelligence, characterized in that: include: Acquiring electron microscope image data of the glomerulus, performing feature extraction on the electron microscope image data, outputting a regional feature image of the glomerular basement membrane, and then preprocessing the regional feature image of the glomerular basement membrane to obtain basement image data; Extracting the edge curve of the glomerular basement membrane according to the regional characteristic image of the glomerular basement membrane, calculating the thickness of the glomerular basement membrane according to the edge curve, and recording the thickness of the glomerular basement membrane as a reference characteristic parameter; Evaluating the reference characteristic parameters and determining the kidney status according to the evaluation results, wherein the kidney status includes a normal state and an abnormal state; In the normal state, it is indicated that the thickness of the glomerular basement membrane is within a normal range, and continuous monitoring is performed, and the thickness of the glomerular basement membrane within the normal range is recorded in a test log; In the abnormal state, it indicates that the thickness of the glomerular basement membrane exceeds the normal range, and the detection deviation of the glomerular basement membrane is output, and the abnormal grade of the kidney is determined based on the detection deviation; The step of calculating the thickness of the glomerular basement membrane according to the edge curve and recording the thickness of the glomerular basement membrane as a reference characteristic parameter includes: constructing a virtual coordinate system according to the base image data, and calibrating the inflection point coordinates of the edge inflection point in the edge curve in the virtual coordinate system; Constructing a plurality of closed curves according to every three edge inflection points, wherein the closed curves are circular; Selecting a closed curve corresponding to the outer circle of the glomerular basement membrane and having the largest radius, and marking it as a standard outer circle curve; Selecting a closed curve corresponding to the inner circle of the glomerular basement membrane and having the smallest radius, and marking it as a standard inner circle curve; Calculating the thickness of the glomerular basement membrane according to the standard outer circle curve and the standard inner circle curve, and recording it as a reference characteristic parameter; The step of calculating the thickness of the glomerular basement membrane based on the standard outer circle curve and the standard inner circle curve and recording it as a reference characteristic parameter comprises: A plurality of sampling nodes are arranged on the standard outer circle curve, wherein the plurality of sampling nodes are distributed in an equidistant manner in a circular shape on the standard outer circle curve; Taking each of the sampling nodes as a reference, a virtual line segment is extended toward the center of the standard inner circle curve, and the intersection point between the virtual line segment and the standard inner circle curve is recorded as a measurement intersection point; The distance between the measured intersection point and the sampling node under the same virtual line segment was calculated as the thickness of the glomerular basement membrane and recorded synchronously as the benchmark characteristic parameter.

2. The method for measuring glomerular basement membrane thickness under an electron microscope based on artificial intelligence according to claim 1, characterized in that: The step of preprocessing the regional characteristic image of the glomerular basement membrane to obtain basement image data comprises: Acquiring electron microscope image data of the glomerulus, and performing denoising processing on the electron microscope image data to eliminate noise interference in the electron microscope image data; Performing grayscale conversion on the denoised electron microscope image data to convert the electron microscope image data into a grayscale image; Performing edge detection on the grayscale image to highlight edge information of the glomerular basement membrane and simultaneously enhancing edge features of the glomerular basement membrane; The region is segmented according to the edge features of the glomerular basement membrane, the glomerular basement membrane is separated from the background of the grayscale image, a regional feature image is formed, and the image is recorded as basement image data.

3. The method for measuring glomerular basement membrane thickness under an electron microscope based on artificial intelligence according to claim 1, characterized in that: The step of evaluating the baseline characteristic parameters and determining the kidney status according to the evaluation results includes: Obtaining all the reference characteristic parameters; Obtaining an evaluation interval, and comparing the benchmark characteristic parameter with the evaluation interval; If all the reference characteristic parameters fall within the evaluation interval, it indicates that the thickness of the glomerular basement membrane is normal, and the corresponding kidney state is recorded as a normal state; If there are any baseline characteristic parameters that do not fall within the evaluation interval, the number of baseline characteristic parameters that do not fall within the evaluation interval is counted and recorded as post-evaluation condition parameters, and when the post-evaluation condition parameters exceed the preset evaluation threshold, the corresponding kidney state is recorded as an abnormal state, otherwise, it is still recorded as a normal state.

4. The method for measuring glomerular basement membrane thickness under an electron microscope based on artificial intelligence according to claim 3, characterized in that: In the normal state, the historical detected thickness of the glomerular basement membrane is retrieved from the detection log and recorded as a historical condition parameter; Obtaining an evaluation function, inputting the historical condition parameter into the evaluation function, and recording the output result of the evaluation function as a trend condition parameter; Obtaining edge values ​​of the evaluation interval, wherein the edge values ​​include an upper limit value and a lower limit value of the evaluation interval; Based on the edge value and trend condition parameters, the predicted safety duration of the glomerular basement membrane under the normal state is calculated and output.

5. The method for measuring glomerular basement membrane thickness under an electron microscope based on artificial intelligence according to claim 4, characterized in that: The step of outputting the detection deviation of the glomerular basement membrane comprises: Obtaining all baseline characteristic parameters that do not fall within the evaluation interval under the abnormal state, and recording them as abnormal condition parameters; Performing difference processing on the abnormal condition parameter and the edge value of the evaluation interval to obtain an initial thickness deviation, wherein the initial thickness deviation includes a first thickness deviation higher than an upper limit value of the evaluation interval and a second thickness deviation lower than a lower limit value of the evaluation interval; A measurement function is obtained, and the first thickness deviation and the second thickness deviation are respectively input into the measurement function, and the output result of the measurement function is recorded as the detection deviation of the glomerular basement membrane.

6. The method for measuring glomerular basement membrane thickness under an electron microscope based on artificial intelligence according to claim 1, characterized in that: The step of determining the level of kidney abnormality based on the detected deviation comprises: Obtaining the detection deviation of the glomerular basement membrane under the abnormal state, and recording it as a grading condition parameter; Obtaining evaluation intervals, wherein each evaluation interval corresponds to an abnormality level; The grading condition parameters are compared with the evaluation interval, and the kidney abnormality grade corresponding to the glomerular basement membrane is matched and output.

7. The artificial intelligence-based electron microscope glomerular basement membrane thickness measurement system is characterized by: The method for measuring the thickness of the glomerular basement membrane under an electron microscope based on artificial intelligence according to any one of claims 1 to 6 comprises: An image acquisition module, the image acquisition module is used to obtain electron microscope image data of the glomerulus, perform feature extraction on the electron microscope image data, output a regional feature image of the glomerular basement membrane, and then preprocess the regional feature image of the glomerular basement membrane to obtain basement image data; A basement membrane thickness calculation module, the basement membrane thickness calculation module is used to extract the edge curve of the glomerular basement membrane according to the regional characteristic image of the glomerular basement membrane, calculate the thickness of the glomerular basement membrane according to the edge curve, and record the thickness of the glomerular basement membrane as a reference characteristic parameter; A state evaluation module, the state evaluation module is used to evaluate the reference characteristic parameters and determine the kidney state according to the evaluation result, wherein the kidney state includes a normal state and an abnormal state; A normal intervention module, wherein the normal intervention module is used to indicate that the thickness of the glomerular basement membrane is within a normal range under the normal state, and to continuously monitor and record the thickness of the glomerular basement membrane within the normal range into a detection log; The abnormal intervention module is used to indicate that the thickness of the glomerular basement membrane exceeds the normal range in the abnormal state, and output the detection deviation of the glomerular basement membrane, and then determine the abnormality level of the kidney based on the detection deviation.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; Wherein, the memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the artificial intelligence-based electron microscope glomerular basement membrane thickness measurement method described in any one of claims 1 to 6.

Images