An osteoporosis detection system based on nuclear magnetic resonance

Abstract

The present invention discloses an osteoporosis detection system based on nuclear magnetic resonance, which relates to the technical field of intelligent detection. It includes a pre-operation module, an image processing module, an analysis module, and an accurate module. Train an image segmentation model to obtain intervertebral disc images, segment the preprocessed images to obtain region-of-interest images, judge the first osteoporosis grade of the region-of-interest, input the region-of-interest images to the expert platform to obtain the second osteoporosis grade, and perform the second operation. Through the collaborative work of the pre-operation module, the image processing module, the analysis module, and the accurate module, the present invention improves the accuracy and efficiency of detection. Through the image segmentation model and the analysis module, a more comprehensive osteoporosis assessment is provided. By introducing the expert platform, the accuracy and reliability of the detection results are ensured. Using MRI technology for osteoporosis detection avoids radiation exposure and is a non-invasive and safe detection method.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent detection, and in particular to an osteoporosis detection system based on nuclear magnetic resonance. Background Art

[0002] In recent years, the osteoporosis detection technology based on nuclear magnetic resonance is constantly developing and improving, and has broad application prospects. MRI technology has clinical value in the early diagnosis of osteoporosis, the prediction of fracture risk and the follow-up after treatment. MRI technology has the advantages of being non-invasive, radiation-free and having high tissue resolution. Compared with the assessment of osteoporosis based solely on bone mineral density (BMD), MRI technology can assess factors such as bone marrow composition, microcirculation, and bone microstructure, thereby more comprehensively reflecting bone strength.

[0003] At present, a Chinese invention patent with publication number CN119014815A discloses an osteoporosis detection method based on multi-mode photoacoustics and a detection system thereof. The method establishes an inversion model of the physical properties of bone tissue based on an inverse solution model, inputs the bone measurement photoacoustic wave signal parameters into the inversion model, outputs the measured bone tissue physical property information data, inputs the osteoporosis clinical standard data, compares the osteoporosis clinical standard data with the measurement data, and gives a bone diagnosis result. However, the related technology does not establish and analyze the model based on historical detection data, lacks the scientific nature of auxiliary diagnosis, does not eliminate invalid data, and does not eliminate the errors of the acquisition part, which is easy to cause wrong judgments, lacks the accuracy of judgment, does not identify and delete redundant data, is not conducive to the rapid responsiveness of the system, and has certain limitations. Summary of the invention

[0004] The technical problem solved by the present invention is: the related technology does not establish and analyze the model based on historical detection data, lacks the scientific nature of auxiliary diagnosis, does not eliminate invalid data and eliminate errors in the acquisition part, which easily leads to wrong judgments and lacks the accuracy of judgments. It does not identify and delete redundant data, is not conducive to the rapid responsiveness of the system, and has certain limitations.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: an osteoporosis detection system based on nuclear magnetic resonance, comprising a pre-operation module, an image processing module, an analysis module and a precision module;

[0006] The pre - operation module sets scanning parameters, sets a positioning line and a positioning frame, identifies the posture of the subject, performs a first operation according to the subject's posture, obtains historical scan images, identifies the spinal part in the historical scan images, marks feature points in the spinal part according to the spatial order, obtains a spinal cord curve image based on the feature points and the Hough transform, crops the corresponding historical intervertebral disc region image according to the spinal cord curve image, trains an image segmentation model based on the historical scan images and the historical intervertebral disc region images, and obtains an intervertebral disc image according to the image segmentation model;

[0007] The image processing module obtains the image after the first operation is completed, pre - processes the image, inputs the pre - processed image into the image segmentation model to obtain an intervertebral disc region image, segments the pre - processed image according to the intervertebral disc region image to obtain a region of interest image, and numbers the region of interest image;

[0008] The analysis module obtains relevant data of the region of interest, sets relevant data thresholds, judges the first osteoporosis grade of the region of interest according to the relevant data thresholds, and sets a label for the first osteoporosis grade according to the corresponding number;

[0009] The precision module calls the expert platform, inputs the region of interest image to the expert platform to obtain a second osteoporosis grade, compares the second osteoporosis grade with the first osteoporosis grade, and performs a second operation according to the comparison result.

[0010] As a preferred solution of the osteoporosis detection system based on nuclear magnetic resonance according to the present invention, wherein: the scanning parameters include the radio - frequency pulse interval time, echo time, slice thickness, slice gap, matrix size and scanning time, and the scanning parameters are obtained according to historical detection experience.

[0011] As a preferred solution of the osteoporosis detection system based on nuclear magnetic resonance according to the present invention, wherein: the setting logic of the positioning line and the positioning frame includes:

[0012] Obtain the ground image of the internal space of the device, draw the center line of the internal space ground, which is represented as the connection line of the mid - points of the width of the ground, set the center line as the positioning line, draw a plane passing through the center line and perpendicular to the ground image, denoted as the first plane;

[0013] Obtain the height inside the device, set the height of the ground as the starting height, set the height inside the device as the ending height, draw a plane parallel to the ground, perpendicular to the first plane and not passing through the first plane, denoted as the second plane, and set the second plane as the positioning frame.

[0014] As a preferred solution of an osteoporosis detection system based on nuclear magnetic resonance according to the present invention, wherein: the first operation includes sending a posture correction signal and a device start signal, and the posture correction signal includes a translation correction signal and a rotation correction signal;

[0015] When the device senses an occluder, it obtains an image of the subject through machine vision, identifies the posture of the subject, and starts the positioning operation when the posture of the subject is lying flat;

[0016] In response to the start of the positioning operation, the first detection is started, and the detection logic of the first detection includes:

[0017] Identify the edge line of the subject's image, segment the image inside the edge line parallel to the positioning line, calculate the number of pixels belonging to the inside of the edge line on both sides of the positioning line, calculate the difference between the number of pixels belonging to the inside of the edge line on both sides of the positioning line, which is denoted as the first difference, set the first value as the first difference threshold, when the first difference is less than the first value, start the second detection, and perform up and down positioning of the subject according to the positioning frame. When the first difference is greater than or equal to the first value, send a translation correction signal, and the subject performs horizontal translation according to the translation posture correction signal until the first difference is less than the first value, then stop sending the translation correction signal.

[0018] As a preferred solution of an osteoporosis detection system based on nuclear magnetic resonance according to the present invention, wherein: the detection logic of the second detection includes:

[0019] Identify the height of the subject, and obtain two points symmetrical to the positioning line and corresponding to the height of the subject on the positioning frame;

[0020] When the heights of the two points symmetrical to the positioning line and corresponding to the subject on the positioning frame are the same, send a start scan signal. When the heights of the two points symmetrical to the positioning line and corresponding to the subject on the positioning frame are different, send a rotation correction signal, and the subject performs a flipping angle according to the rotation correction signal until the heights of the two points symmetrical to the positioning line and corresponding to the subject on the positioning frame are the same, then stop sending the rotation correction signal.

[0021] As a preferred solution of an osteoporosis detection system based on nuclear magnetic resonance according to the present invention, wherein: the recognition logic of the spinal part includes:

[0022] Retrieve the reference image of the human spine, extract the shape feature quantity of the reference image of the human spine, denoted as the first feature quantity, extract the shape feature quantity of the historical scan image, denoted as the second feature quantity, calculate the similarity between the first feature quantity and the second feature quantity through the cosine similarity formula, set the second value as the similarity threshold, when the similarity is greater than or equal to the second value, set the corresponding part in the historical scan image as the spine part, otherwise, jump to the second feature quantity of the next part;

[0023] The cropping logic of the historical intervertebral disc region image includes:

[0024] Mark feature points in the spine part according to the spatial order, and the spatial order is represented as from top to bottom in the plane graph;

[0025] Obtain the spinal cord curve image according to the feature points and the Hough transform, and obtain the amplitude of the spinal cord curve image;

[0026] Set the third value and the fourth value as the amplitude thresholds, set the part of the spinal cord curve corresponding to the amplitude between the third value and the fourth value as the historical intervertebral disc curve part, set the part of the historical scan image corresponding to the historical intervertebral disc curve as the historical intervertebral disc region, and crop the corresponding historical intervertebral disc region image;

[0027] Train an image segmentation model according to the historical scan image and the historical intervertebral disc region image, and obtain the intervertebral disc image according to the image segmentation model.

[0028] As a preferred solution of the osteoporosis detection system based on nuclear magnetic resonance according to the present invention, wherein: the training logic of the image segmentation model includes:

[0029] Use the historical scan image and the historical intervertebral disc region image as the sample set, divide the sample set into a training set and a test set according to the first ratio, and the first ratio is obtained through training experience;

[0030] Input the training set into the neural network for training to obtain the image segmentation model, and test the image segmentation model according to the test set until the accuracy reaches the first accuracy, and then stop the test;

[0031] The image segmentation model takes the scan image as the input quantity, the intervertebral disc region image as the output quantity, the actual intervertebral disc region image as the training target, and the distance between the geometric center of the intervertebral disc region image and the geometric center of the actual intervertebral disc region image as the accuracy standard.

[0032] As a preferred solution of an osteoporosis detection system based on nuclear magnetic resonance according to the present invention, wherein: the preprocessing includes enhancing contrast processing and denoising processing, inputting the preprocessed image into an image segmentation model to obtain an intervertebral disc region image, deleting the intervertebral disc region image from the preprocessed image, and determining the side length of the segmentation frame according to the greatest common divisor of the boundary lines of the image after deletion. The segmentation frame is square. Starting from the upper left corner of the image after deletion, segment the preprocessed image according to the segmentation frame to obtain a region of interest image;

[0033] Number the region of interest images, and the numbers of the region of interest images are natural numbers.

[0034] As a preferred solution of an osteoporosis detection system based on nuclear magnetic resonance according to the present invention, wherein: the analysis module obtains relevant data of the region of interest, and the relevant data includes bone marrow fat fraction and fat area ratio;

[0035] Set the fifth value and the sixth value as relevant data thresholds, and judge the first osteoporosis grade of the region of interest according to the relevant data thresholds. The first osteoporosis grade includes an automatic first grade, an automatic second grade, and an automatic third grade. The judgment logic of the first osteoporosis grade is:

[0036] When the bone marrow fat fraction is less than or equal to the fifth value and the fat area ratio is less than or equal to the sixth value, set the first osteoporosis grade as the automatic first grade. When the bone marrow fat fraction is greater than the fifth value and the fat area ratio is less than or equal to the sixth value, set the first osteoporosis grade as the automatic second grade. When the bone marrow fat fraction is greater than the fifth value and the fat area ratio is greater than the sixth value, set the first osteoporosis grade as the automatic third grade;

[0037] The severity of osteoporosis represented by the automatic first grade, automatic second grade, and automatic third grade shows an increasing trend.

[0038] As a preferred solution of an osteoporosis detection system based on nuclear magnetic resonance according to the present invention, wherein: the precise module retrieves the expert platform, inputs the region of interest image into the expert platform to obtain the second osteoporosis grade, and the second osteoporosis grade includes an artificial first grade, an artificial second grade, and an artificial third grade;

[0039] Compare the second osteoporosis grade with the first osteoporosis grade to obtain a comparison result, and perform a second operation according to the comparison result. The second operation includes adjusting the relevant data threshold and sending a detection completion signal;

[0040] When the first osteoporosis grade is the same as the second osteoporosis grade, set the second operation to send a completion detection signal and output the corresponding number. When the first osteoporosis grade is different from the second osteoporosis grade, set the second operation to adjust the relevant data threshold;

[0041] When the second operation is to adjust the relevant data threshold, set the seventh value and the eighth value to the change variables of the fifth value and the sixth value respectively, continuously increase or decrease the fifth value and the sixth value, and continuously obtain the corresponding first osteoporosis grade until the adjusted first osteoporosis grade is the same as the second osteoporosis grade, then stop adjusting the relevant data threshold.

[0042] Advantages of the present invention: Through the collaborative work of the pre-operation module, image processing module, analysis module and precision module, an automated process from scan parameter setting to image segmentation and data analysis is achieved, improving the accuracy and efficiency of detection. It can perform personalized detection according to the posture of the subject and historical scan images to ensure the accuracy of the detection results. By identifying the posture of the subject and the spinal part, the system can more accurately locate and analyze the intervertebral disc area. Through the image segmentation model and analysis module, the system can obtain relevant data of the region of interest and set relevant data thresholds to judge the osteoporosis grade, providing a more comprehensive osteoporosis assessment. The precision module retrieves the expert platform, inputs the image of the region of interest, obtains the second osteoporosis grade, and compares the second osteoporosis grade with the first osteoporosis grade to ensure the accuracy and reliability of the detection results. Using MRI technology for osteoporosis detection avoids radiation exposure and is a non-invasive and safe detection method. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a schematic diagram of the basic process of an osteoporosis detection system based on nuclear magnetic resonance provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] To make the above objects, features and advantages of the present invention more obvious and understandable, the following detailed description of the specific embodiments of the present invention will be made in conjunction with the accompanying drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them.

[0045] Example, referring to Figure 1 , which is an embodiment of the present invention, provides an osteoporosis detection system based on nuclear magnetic resonance, including a pre-operation module, an image processing module, an analysis module and a precision module;

[0046] The pre-operation module sets scanning parameters, sets a positioning line and a positioning frame, identifies the subject's posture, performs a first operation according to the subject's posture, obtains historical scan images, identifies the spinal part in the historical scan images, marks feature points in the spinal part according to the spatial order, obtains a spinal cord curve image according to the feature points and the Hough transform, crops the corresponding historical intervertebral disc region image according to the spinal cord curve image, trains an image segmentation model according to the historical scan images and the historical intervertebral disc region images, and obtains an intervertebral disc image according to the image segmentation model;

[0047] The image processing module obtains the image after the first operation is completed, preprocesses the image, inputs the preprocessed image into the image segmentation model to obtain an intervertebral disc region image, segments the preprocessed image according to the intervertebral disc region image to obtain a region of interest image, and numbers the region of interest image;

[0048] The analysis module obtains the relevant data of the region of interest, sets a relevant data threshold, determines the first osteoporosis grade of the region of interest according to the relevant data threshold, and sets a label of the first osteoporosis grade according to the corresponding number;

[0049] The precision module calls the expert platform, inputs the region of interest image to the expert platform to obtain a second osteoporosis grade, compares the second osteoporosis grade with the first osteoporosis grade, and performs a second operation according to the comparison result.

[0050] Through the collaborative work of the pre-operation module, the image processing module, the analysis module and the precision module, the present invention realizes an automated process from scanning parameter setting to image segmentation and data analysis, improves the accuracy and efficiency of detection, can perform personalized detection according to the subject's posture and historical scan images, ensures the accuracy of the detection result. By identifying the subject's posture and the spinal part, the system can more accurately locate and analyze the intervertebral disc region. Through the image segmentation model and the analysis module, the system can obtain the relevant data of the region of interest and set the relevant data threshold, so as to judge the osteoporosis grade, providing a more comprehensive osteoporosis assessment. The precision module calls the expert platform, inputs the region of interest image, obtains a second osteoporosis grade, and compares the second osteoporosis grade with the first osteoporosis grade to ensure the accuracy and reliability of the detection result. Using MRI technology for osteoporosis detection avoids radiation exposure and is a non-invasive and safe detection method.

[0051] The scanning parameters include radio frequency pulse interval time, echo time, slice thickness, slice spacing, matrix size and scanning time, and the scanning parameters are obtained according to historical detection experience.

[0052] In specific implementation, fast spin echo sagittal T1-weighted imaging (TR 645 ms), fast spin echo sagittal T2WI (TR 3500 ms), fast spin echo sagittal FS-T2WI (TR 3000 ms), fast spin echo sagittal T1-weighted imaging (TE 9.6 ms), fast spin echo sagittal T2WI (TE 127 ms), fast spin echo sagittal FS-T2WI (TE 87 ms), slice thickness is set to 4 mm, slice gap is set to 1 mm, matrix is set to 300×300, scanning time for T1WI sagittal plane is 1 minute and 9 seconds, and scanning time for T2WI sagittal plane is 1 minute and 19 seconds.

[0053] The setting logic of the positioning line and the positioning frame includes:

[0054] Obtain the ground image of the internal space of the device, draw the midline of the internal space ground, where the midline is represented as the connection line of the midpoints of the width of the ground, set the midline as the positioning line, draw a plane passing through the midline and perpendicular to the ground image, denoted as the first plane;

[0055] Obtain the height inside the device, set the height of the ground as the starting height, set the height inside the device as the ending height, draw a plane parallel to the ground, perpendicular to the first plane and not passing through the first plane, denoted as the second plane, and set the second plane as the positioning frame.

[0056] In specific implementation, by obtaining the ground image of the internal space of the device and drawing the midline, it is ensured that the positioning line accurately reflects the midpoint of the ground width. This precise positioning helps to ensure the symmetry and consistency of the scan. The logic of setting the positioning line and the positioning frame simplifies the operation process, enabling technicians to quickly and accurately set the scan parameters, reducing the operation time and potential errors. By drawing the second plane parallel to the ground and perpendicular to the first plane as the positioning frame, the accuracy and integrity of the scan area are ensured. This setting helps to improve the scan efficiency and ensure that all necessary areas are covered. By ensuring that the positioning frame does not pass through the first plane, the errors and artifacts that may occur during the scan are reduced, thereby improving the accuracy and reliability of the image.

[0057] The first operation includes sending a posture correction signal and a device start signal, where the posture correction signal includes a translation correction signal and a rotation correction signal;

[0058] When the device senses an obstacle, obtain the image of the subject through machine vision, identify the subject's posture, and when the subject's posture is lying flat, start the positioning operation;

[0059] In response to starting the positioning operation, start the first detection, and the detection logic of the first detection includes:

[0060] Identify the edge line of the subject's image, segment the image inside the edge line parallel to the positioning line, calculate the number of pixels belonging to the inside of the edge line on both sides of the positioning line, calculate the difference between the number of pixels belonging to the inside of the edge line on both sides of the positioning line, denoted as the first difference, set the first value as the first difference threshold. When the first difference is less than the first value, start the second detection, perform up and down positioning of the subject according to the positioning box. When the first difference is greater than or equal to the first value, send a translation correction signal, and the subject performs horizontal translation according to the translation posture correction signal until the first difference is less than the first value, then stop sending the translation correction signal.

[0061] In specific implementation, by sending posture correction signals, including translation correction signals and rotation correction signals, the system can accurately correct the posture of the subject. This correction ensures the correct posture of the subject during the scanning process, thereby improving the accuracy and reliability of the scanning. When the device senses an obstacle, it obtains the subject's image through machine vision and identifies the subject's posture. If the subject's posture is lying flat, start the positioning operation. This automated positioning operation reduces manual intervention and improves the efficiency and accuracy of the operation. The detection logic of the first detection includes identifying the edge line of the subject's image and segmenting the image inside the edge line parallel to the positioning line. By calculating the number of pixels belonging to the inside of the edge line on both sides of the positioning line and their difference, the system can effectively determine whether further detection or correction is required. This efficient detection logic ensures the speed and accuracy of the detection process.

[0062] The detection logic of the second detection includes:

[0063] Identify the subject's height, and obtain two points symmetric about the positioning line and corresponding to the subject's height on the positioning box;

[0064] When the subject's heights corresponding to the two points symmetric about the positioning line and distributed on the positioning box are the same, send a start scanning signal. When the subject's heights corresponding to the two points symmetric about the positioning line and distributed on the positioning box are different, send a rotation correction signal, and the subject performs a flip angle according to the rotation correction signal until the subject's heights corresponding to the two points symmetric about the positioning line and distributed on the positioning box are the same, then stop sending the rotation correction signal.

[0065] In specific implementation, by identifying the height of the subject and obtaining two points symmetric to the positioning line, the system can accurately determine the position of the subject in the positioning frame. This accurate height recognition helps to ensure the accuracy of scanning. The detection logic ensures the symmetry of the subject by comparing the heights of the subject corresponding to the two symmetric points on the positioning line in the positioning frame. This symmetry correction is a key step to ensure the accuracy of the scanning result. When the heights of the subject corresponding to the two symmetric points on the positioning line are the same, the system will automatically send a signal to start scanning. This automated scanning start reduces manual intervention and improves scanning efficiency. When the heights of the subject corresponding to the two symmetric points on the positioning line are different, the system will send a rotation correction signal, and the subject will correct the flipping angle according to the signal. This dynamic rotation correction mechanism ensures the continuous correction of the subject's posture until the optimal scanning state is reached.

[0066] The recognition logic of the spinal part includes:

[0067] Retrieve the reference image of the human spine, extract the shape feature quantity of the reference image of the human spine, denoted as the first feature quantity, extract the shape feature quantity of the historical scan image, denoted as the second feature quantity, calculate the similarity between the first feature quantity and the second feature quantity through the cosine similarity formula, set the second value as the similarity threshold. When the similarity is greater than or equal to the second value, set the corresponding part in the historical scan image as the spinal part; otherwise, jump to the second feature quantity of the next part.

[0068] The cropping logic of the historical intervertebral disc area image includes:

[0069] Mark feature points in the spinal part according to the spatial order, and the spatial order is represented as from top to bottom in the plane graph.

[0070] Obtain the spinal cord curve image according to the feature points and the Hough transform, and obtain the amplitude of the spinal cord curve image.

[0071] Set the third value and the fourth value as the amplitude thresholds, set the part of the spinal cord curve corresponding to the amplitude between the third value and the fourth value as the historical intervertebral disc curve part, set the part of the historical scan image corresponding to the historical intervertebral disc curve as the historical intervertebral disc area, and crop the corresponding historical intervertebral disc area image.

[0072] Train an image segmentation model according to the historical scan image and the historical intervertebral disc area image, and obtain the intervertebral disc image according to the image segmentation model.

[0073] In specific implementation, by retrieving the reference image of the human spine and extracting the shape feature quantities, the system can accurately identify the spine part in the historical scan image. This feature quantity-based recognition method improves the accuracy of spine recognition. The cosine similarity formula is used to calculate the similarity between the first feature quantity and the second feature quantity, and a similarity threshold is set. This method ensures that only when the similarity reaches a certain standard will the part in the image be recognized as the spine, thus reducing the possibility of misrecognition. The third value and the fourth value are set as amplitude thresholds to ensure that only the part corresponding to the spinal cord curve with an amplitude distributed between these two thresholds is set as the historical intervertebral disc curve part. This threshold setting helps to accurately crop the intervertebral disc area. An image segmentation model is trained based on the historical scan image and the historical intervertebral disc area image. This data-driven model training method improves the accuracy and efficiency of intervertebral disc image segmentation.

[0074] The training logic of the image segmentation model includes:

[0075] Taking the historical scan image and the historical intervertebral disc area image as a sample set, dividing the sample set into a training set and a test set according to a first ratio, and the first ratio is obtained through training experience;

[0076] Inputting the training set into a neural network for training to obtain an image segmentation model, and testing the image segmentation model according to the test set until the accuracy rate reaches the first accuracy rate, then stopping the test;

[0077] The image segmentation model takes the scan image as the input quantity, the intervertebral disc area image as the output quantity, the actual intervertebral disc area image as the training target, and reducing the distance between the geometric center of the intervertebral disc area image and the geometric center of the actual intervertebral disc area image as the accuracy standard.

[0078] The preprocessing includes contrast enhancement processing and denoising processing. Inputting the preprocessed image into the image segmentation model to obtain the intervertebral disc area image, deleting the intervertebral disc area image in the preprocessed image, and determining the side length of the segmentation box according to the greatest common divisor of the boundary lines of the image after deletion. The segmentation box is a square. Starting from the upper left corner of the image after deletion, segment the preprocessed image according to the segmentation box to obtain the region of interest image;

[0079] Number the region of interest images, and the numbers of the region of interest images are natural numbers.

[0080] In specific implementation, by using historical scanned images and historical intervertebral disc region images as a sample set, the system can train an image segmentation model based on actual data. This data-driven method improves the accuracy and generalization ability of the model. The sample set is divided into a training set and a test set according to a first ratio obtained through training experience. This reasonable data set division ensures the independence of model training and testing, avoiding overfitting. A neural network is used for model training. The powerful non-linear mapping ability of the neural network enables the image segmentation model to capture complex image features, improving the accuracy of segmentation. The image segmentation model takes the scanned image as the input quantity, the intervertebral disc region image as the output quantity, and the actual intervertebral disc region image as the training target. This precise control of the output quantity ensures a clear training target for the model and improves the pertinence of the model.

[0081] The analysis module obtains relevant data of the region of interest, and the relevant data includes the bone marrow fat fraction and the fat area ratio.

[0082] Set the fifth value and the sixth value as the relevant data thresholds, and judge the first osteoporosis grade of the region of interest according to the relevant data thresholds. The first osteoporosis grade includes an automatic first grade, an automatic second grade, and an automatic third grade. The judgment logic of the first osteoporosis grade is as follows:

[0083] When the bone marrow fat fraction is less than or equal to the fifth value and the fat area ratio is less than or equal to the sixth value, set the first osteoporosis grade as the automatic first grade; when the bone marrow fat fraction is greater than the fifth value and the fat area ratio is less than or equal to the sixth value, set the first osteoporosis grade as the automatic second grade; when the bone marrow fat fraction is greater than the fifth value and the fat area ratio is greater than the sixth value, set the first osteoporosis grade as the automatic third grade.

[0084] The severity of osteoporosis represented by the automatic first grade, the automatic second grade, and the automatic third grade shows an increasing trend.

[0085] In specific implementation, relevant data of the region of interest can be obtained, such as bone marrow fat fraction and fatty area ratio. These data are crucial for evaluating the severity of osteoporosis. By setting the fifth value and the sixth value as relevant data thresholds, the system can determine the osteoporosis grade of the region of interest according to these thresholds. This clear threshold setting helps to achieve automated grade judgment. According to the comparison of the bone marrow fat fraction and the fatty area ratio with the thresholds, the system automatically classifies the osteoporosis grade into the automatic first grade, the automatic second grade, and the automatic third grade. This automated grade judgment reduces manual intervention and improves the efficiency and accuracy of judgment. The severity of osteoporosis represented by the automatic first grade, the automatic second grade, and the automatic third grade shows an increasing trend. This increasing representation method helps doctors and patients clearly understand the severity of the condition.

[0086] The precise module retrieves the expert platform, inputs the image of the region of interest to the expert platform, and obtains the second osteoporosis grade, where the second osteoporosis grade includes the manual first grade, the manual second grade, and the manual third grade;

[0087] Compare the second osteoporosis grade with the first osteoporosis grade to obtain a comparison result, and perform a second operation according to the comparison result. The second operation includes adjusting the relevant data thresholds and sending a completion detection signal;

[0088] When the first osteoporosis grade is the same as the second osteoporosis grade, set the second operation to send a completion detection signal and output the corresponding number. When the first osteoporosis grade is different from the second osteoporosis grade, set the second operation to adjust the relevant data thresholds;

[0089] When the second operation is to adjust the relevant data thresholds, set the seventh value and the eighth value as the change variables of the fifth value and the sixth value respectively, continuously increase or decrease the fifth value and the sixth value, and continuously obtain the corresponding first osteoporosis grade until the first osteoporosis grade after adjustment is the same as the second osteoporosis grade, then stop adjusting the relevant data thresholds.

[0090] In specific implementation, the precise module retrieves the expert platform and inputs the image of the region of interest to obtain the second osteoporosis grade. This method of retrieving the expert platform helps to improve the accuracy and reliability of osteoporosis grade judgment. When the first osteoporosis grade is different from the second osteoporosis grade, the system will perform the operation of adjusting the relevant data thresholds. This dynamic adjustment mechanism helps to optimize the judgment criteria of the system, improve the adaptability and accuracy of the system, and the continuous threshold adjustment method helps to achieve precise grade matching.

[0091] Through the collaborative work of the pre-operation module, image processing module, analysis module, and precision module, the present invention realizes an automated process from scanning parameter setting to image segmentation and data analysis, improves the accuracy and efficiency of detection, can perform personalized detection according to the subject's posture and historical scan images, ensures the accuracy of the detection results. By identifying the subject's posture and spinal part, the system can more accurately locate and analyze the intervertebral disc area. Through the image segmentation model and analysis module, the system can obtain relevant data of the region of interest and set relevant data thresholds to determine the osteoporosis grade, providing a more comprehensive osteoporosis assessment. The precision module retrieves the expert platform, inputs the image of the region of interest, obtains the second osteoporosis grade, and compares the second osteoporosis grade with the first osteoporosis grade to ensure the accuracy and reliability of the detection results. Using MRI technology for osteoporosis detection avoids radiation exposure and is a non-invasive and safe detection method.

[0092] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. Among them, the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, a magnetic disk, or an optical disc. These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured article including an instruction device, and the instruction device implements the functions specified in one process Figure 1 one process or multiple processes and / or boxes Figure 1 specified in one box or multiple boxes.

[0093] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. An osteoporosis detection system based on nuclear magnetic resonance, characterized in that, It includes a pre-operation module, an image processing module, an analysis module, and an accurate module; The pre-operation module sets scanning parameters, sets a positioning line and a positioning frame, identifies the subject's posture, performs a first operation according to the subject's posture. The first operation includes sending a posture correction signal and a start device signal, obtaining historical scan images, identifying the spinal part in the historical scan images, marking feature points in the spinal part according to the spatial order, obtaining a spinal cord curve image according to the feature points and the Hough transform, cropping the corresponding historical intervertebral disc area image according to the spinal cord curve image, training an image segmentation model according to the historical scan images and the historical intervertebral disc area images, and obtaining an intervertebral disc image according to the image segmentation model; The image processing module obtains the image after the first operation is completed, preprocesses the image, inputs the preprocessed image into the image segmentation model to obtain an intervertebral disc area image, segments the preprocessed image according to the intervertebral disc area image to obtain a region of interest image, and numbers the region of interest image; The analysis module obtains the relevant data of the region of interest. The relevant data includes the bone marrow fat fraction and the fatification area ratio, sets the relevant data threshold, judges the first osteoporosis grade of the region of interest according to the relevant data threshold, and sets the label of the first osteoporosis grade according to the corresponding number; The accurate module calls the expert platform, inputs the region of interest image into the expert platform to obtain the second osteoporosis grade, compares the second osteoporosis grade with the first osteoporosis grade, and performs a second operation according to the comparison result. The second operation includes adjusting the relevant data threshold and sending a completion detection signal.

2. The osteoporosis detection system based on nuclear magnetic resonance according to claim 1, wherein: The scanning parameters include the radiofrequency pulse interval time, the echo time, the slice thickness, the slice spacing, the matrix size, and the scanning time, and the scanning parameters are obtained according to historical detection experience.

3. An osteoporosis detection system based on nuclear magnetic resonance according to claim 1, characterized in that: The setting logic of the positioning line and the positioning frame includes: Obtain the ground image of the internal space of the device, draw the midline of the internal space ground, and the midline is represented as the connection of the midpoints of the width of the ground. Set the midline as the positioning line, draw a plane passing through the midline and perpendicular to the ground image, and denote it as the first plane; Obtain the height of the internal part of the device, set the height of the ground as the starting height, set the height of the internal part of the device as the ending height, draw a plane parallel to the ground, perpendicular to the first plane and not passing through the first plane, and denote it as the second plane. Set the second plane as the positioning frame.

4. The osteoporosis detection system based on nuclear magnetic resonance according to claim 1, characterized in that: The posture correction signal includes a translation correction signal and a rotation correction signal; When the device senses an obstacle, obtain the subject's image through machine vision, identify the subject's posture, and start the positioning operation when the subject's posture is lying flat; In response to starting the positioning operation, start the first detection. The detection logic of the first detection includes: Identify the edge line of the subject's image, segment the image inside the edge line parallel to the positioning line, calculate the number of pixels belonging to the inside of the edge line on both sides of the positioning line, calculate the difference between the number of pixels belonging to the inside of the edge line on both sides of the positioning line, denoted as the first difference, set the first value as the first difference threshold. When the first difference is less than the first value, start the second detection, perform up and down positioning of the subject according to the positioning frame. When the first difference is greater than or equal to the first value, send a translation correction signal, and the subject performs horizontal translation according to the translation posture correction signal until the first difference is less than the first value, then stop sending the translation correction signal.

5. The osteoporosis detection system based on nuclear magnetic resonance according to claim 4, wherein: The detection logic of the second detection includes: Identify the height of the subject, and obtain two points symmetric about the positioning line and corresponding to the subject's height on the positioning frame; When the subject's heights corresponding to the two points symmetric about the positioning line and distributed on the positioning frame are the same, send a start scanning signal. When the subject's heights corresponding to the two points symmetric about the positioning line and distributed on the positioning frame are different, send a rotation correction signal, and the subject performs a flipping angle according to the rotation correction signal until the subject's heights corresponding to the two points symmetric about the positioning line and distributed on the positioning frame are the same, then stop sending the rotation correction signal.

6. The osteoporosis detection system based on nuclear magnetic resonance according to claim 1, characterized in that: The recognition logic of the spinal part includes: Retrieve the human spinal reference image, extract the shape feature quantity of the human spinal reference image, denoted as the first feature quantity, extract the shape feature quantity of the historical scan image, denoted as the second feature quantity, calculate the similarity between the first feature quantity and the second feature quantity through the cosine similarity formula, set the second value as the similarity threshold. When the similarity is greater than or equal to the second value, set the corresponding part in the historical scan image as the spinal part, otherwise, jump to the second feature quantity of the next part; The clipping logic of the historical intervertebral disc region image includes: Mark feature points in the spinal part according to the spatial order, and the spatial order is represented as from top to bottom in the plane graph; Obtain the spinal cord curve image according to the feature points and the Hough transform, and obtain the amplitude of the spinal cord curve image; Set the third value and the fourth value as the amplitude thresholds, set the part of the spinal cord curve corresponding to the amplitude between the third value and the fourth value as the historical intervertebral disc curve part, set the part of the historical scan image corresponding to the historical intervertebral disc curve as the historical intervertebral disc region, and clip the corresponding historical intervertebral disc region image; Train an image segmentation model according to the historical scan image and the historical intervertebral disc region image, and obtain the intervertebral disc image according to the image segmentation model.

7. The osteoporosis detection system based on nuclear magnetic resonance according to claim 1, characterized in that: The training logic of the image segmentation model includes: Use the historical scan image and the historical intervertebral disc region image as the sample set, divide the sample set into a training set and a test set according to the first ratio, and the first ratio is obtained through training experience; Input the training set into the neural network for training to obtain the image segmentation model, and test the image segmentation model according to the test set until the accuracy rate reaches the first accuracy rate, then stop the test; The image segmentation model takes a scanned image as the input, a disc region image as the output, and an actual disc region image as the training target, with the accuracy criterion being to minimize the distance between the geometric centers of the disc region image and the actual disc region image.

8. The osteoporosis detection system based on nuclear magnetic resonance according to claim 1, wherein: The preprocessing includes contrast enhancement processing and denoising processing. The preprocessed image is input into the image segmentation model to obtain a disc region image. The disc region image is deleted from the preprocessed image, and the side length of the segmentation box is determined based on the greatest common divisor of the boundary lines of the image after deletion. The segmentation box is square. Starting from the upper left corner of the image after deletion, the preprocessed image is segmented according to the segmentation box to obtain a region of interest image. The region of interest images are numbered, and the numbers of the region of interest images are natural numbers.

9. The osteoporosis detection system based on nuclear magnetic resonance according to claim 1, wherein: The analysis module obtains relevant data of the region of interest. The fifth value and the sixth value are set as relevant data thresholds. According to the relevant data thresholds, the first osteoporosis grade of the region of interest is judged. The first osteoporosis grade includes an automatic first grade, an automatic second grade, and an automatic third grade. The judgment logic of the first osteoporosis grade is as follows: When the bone marrow fat fraction is less than or equal to the fifth value and the fattening area ratio is less than or equal to the sixth value, the first osteoporosis grade is set as the automatic first grade. When the bone marrow fat fraction is greater than the fifth value and the fattening area ratio is less than or equal to the sixth value, the first osteoporosis grade is set as the automatic second grade. When the bone marrow fat fraction is greater than the fifth value and the fattening area ratio is greater than the sixth value, the first osteoporosis grade is set as the automatic third grade. The severity of osteoporosis represented by the automatic first grade, automatic second grade, and automatic third grade shows an increasing trend.

10. A nuclear magnetic resonance-based osteoporosis detection system according to claim 9, characterized in that: The precision module calls the expert platform, inputs the region of interest image into the expert platform, and obtains the second osteoporosis grade. The second osteoporosis grade includes an artificial first grade, an artificial second grade, and an artificial third grade. The second osteoporosis grade is compared with the first osteoporosis grade to obtain a comparison result, and a second operation is performed according to the comparison result. When the first osteoporosis grade is the same as the second osteoporosis grade, the second operation is set to send a completion detection signal and output the corresponding number. When the first osteoporosis grade is different from the second osteoporosis grade, the second operation is set to adjust the relevant data thresholds. When the second operation is to adjust the relevant data thresholds, the seventh value and the eighth value are respectively set as the change variables of the fifth value and the sixth value. The fifth value and the sixth value are continuously increased or decreased, and the corresponding first osteoporosis grades are continuously obtained until the first osteoporosis grade after adjustment is the same as the second osteoporosis grade, at which time the adjustment of the relevant data thresholds is stopped.

Images