Remote data acquisition system and method for early screening of ankylosing spondylitis

Through the remote data acquisition system, multi-dimensional information acquisition and extraction technology is used to solve the problem of low accuracy in early screening of ankylosing spondylitis, improve the accuracy and consistency of the data, and provide doctors with reliable diagnostic support.

CN119028564AActive Publication Date: 2024-11-26THE FIRST MEDICAL CENT CHINESE PLA GENERAL HOSPITAL
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
CN202410991641.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-23
Publication Date
2024-11-26
Estimated Expiration
2044-07-23

AI Technical Summary

Technical Problem

In the prior art, the accuracy of early screening results of ankylosing spondylitis is limited, especially in medical information collection scenarios such as physical examination institutions that are mostly used by nurses, it is difficult to effectively collect lesion information.

Method used

Provide a remote data acquisition system, including a basic information acquisition unit, a posture information acquisition unit and a thoracic expansion information acquisition unit, using the optimized Celestial iteration algorithm and CNN target edge detection algorithm, combined with regional masking technology, the user's symptoms are obtained separately. Descriptive data, posture data, and thoracic dilation data.

Benefits of technology

Through multi-dimensional information collection and extraction, the accuracy and efficiency of early screening of ankylosing spondylitis is improved, and a reliable data foundation is provided to support doctors' diagnosis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119028564A_ABST
    Figure CN119028564A_ABST
Patent Text Reader

Abstract

The invention discloses a remote data acquisition system and method for early screening of ankylosing spondylitis, the system comprises a basic information acquisition unit, an attitude information acquisition unit, a thoracic dilation information acquisition unit and an information extraction unit, the basic information acquisition unit is used for receiving symptom description information of a user; the posture information acquisition unit is used for acquiring continuous posture information of a user; the thoracic dilation information acquisition unit is used for acquiring thoracic dilation information in the breathing process of a user; the information extraction unit is used for extracting symptom description data, posture data and thoracic dilation data corresponding to the user based on the symptom description information, the posture information and the thoracic dilation information, user state information can be accurately obtained in a multi-dimensional mode, and accurate data support is provided for early screening of a doctor.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure generally relates to the field of remote data processing technology, and in particular to a remote data acquisition system and method for early screening of ankylosing spondylitis. Background Art

[0002] Ankylosing spondylitis (AS) is a highly disabling lifelong disability that can be caused by the formation of spinal osteophytes and hip joint destruction. Previous data show that up to 40% of AS patients eventually develop complete spinal fusion or even extreme body distortion.

[0003] In 2020, the National Health Commission stated in its reply to the "Proposal on Strengthening the Prevention and Treatment of Ankylosing Spondylitis" that "strengthening the prevention and control of ankylosing spondylitis is also an important measure to implement the 'Healthy China 2030' Outline and promote the construction of a healthy China". Accurate screening of early patients is an important measure to strengthen prevention and control, and accurate acquisition of early patient lesion information is a prerequisite for achieving effective control of disease progression.

[0004] In the existing technology, disease screening is usually performed only based on the user's pain and limited mobility caused by spinal fusion, and such a single feature source limits the accuracy of early screening results for ankylosing spondylitis. Moreover, since the degree of the lesion is relatively mild, such as causing small changes in the patient's posture, it requires more experienced doctors to accurately detect abnormal lesions, especially in medical information collection scenarios such as physical examination institutions where nurses are mostly responsible for the implementation, it is difficult to collect effective information on ankylosing spondylitis lesions. Summary of the invention

[0005] In view of the above-mentioned defects or deficiencies in the prior art, it is desired to provide a remote data collection system and method for early screening of ankylosing spondylitis, which can accurately obtain user status information in multiple dimensions and provide accurate data support for doctors' early screening.

[0006] In a first aspect, an embodiment of the present application provides a remote data collection system for early screening of ankylosing spondylitis, comprising: a basic information collection unit, a posture information collection unit, a chest expansion information collection unit, and an information extraction unit.

[0007] The basic information collection unit is used to receive symptom description information of the user;

[0008] The posture information collection unit is used to collect continuous posture information of the user;

[0009] The chest expansion information collection unit is used to collect chest expansion information during the user's breathing process;

[0010] The information extraction unit is used to extract the symptom description data, posture data and chest expansion data corresponding to the user based on the symptom description information, the posture information and the chest expansion information respectively.

[0011] In some embodiments, the information extraction unit is further used to:

[0012] Acquire the symptom description information, and perform iterative exploration in the symptom description information using an optimized crayfish iterative algorithm to obtain at least one candidate symptom acquisition strategy;

[0013] Evaluate the at least one candidate symptom acquisition strategy based on a preset fitness function, and select the candidate symptom acquisition strategy with the highest fitness as the target symptom acquisition strategy;

[0014] Based on the target symptom acquisition strategy, the symptom description data is extracted from the symptom description information.

[0015] In some embodiments, the information extraction unit is further used to:

[0016] Obtain initial symptom description information for iterative exploration;

[0017] Iteratively explore the initial symptom description information based on the crayfish iterative algorithm to obtain a symptom acquisition strategy corresponding to the initial symptom description;

[0018] When the number of iterations in which the symptom acquisition strategy remains unchanged continuously reaches a preset number, the symptom acquisition strategy is used as the candidate symptom acquisition strategy, and the new initial symptom description information is returned to be acquired.

[0019] In some embodiments, the posture information acquisition unit includes a first sensor and a second sensor, and the information extraction unit is further used to:

[0020] Respectively acquiring a first image frame acquired by the first sensor and a second image frame acquired by the second sensor at the same moment;

[0021] Performing three-dimensional data fusion on the first image frame and the second image frame, and using a CNN-based target edge detection algorithm to obtain a posture boundary corresponding to the moment;

[0022] The posture data is obtained by performing posture estimation on the image frame at the moment based on the posture boundary.

[0023] In some embodiments, the information extraction unit is further used to:

[0024] In the posture estimation process, a joint loss function is used to perform posture optimization to obtain an optimal posture estimation value as the posture data;

[0025] The joint loss function includes at least a smoothness term, a scene-aware contact term, a pose prior term, and a mesh-to-point term.

[0026] In some embodiments, the information extraction unit is further used to:

[0027] Selecting at least one key area around the thorax; the key area at least includes the intercostal space, the posterior part of the sternum, and the area around the spine;

[0028] For each of the key areas, a pressure-time curve corresponding to each of the key areas is obtained from the chest expansion information using a regional masking technique;

[0029] Based on the pressure-time curve, the pressure peak value and the pressure-time integral corresponding to the key area are determined, and the pressure peak value and the pressure-time integral are used as the chest expansion data.

[0030] In a second aspect, an embodiment of the present application provides a data collection method of a remote data collection system based on early screening of ankylosing spondylitis, the system comprising: a basic information collection unit, a posture information collection unit, a chest expansion information collection unit and an information extraction unit.

[0031] The basic information collection unit is used to receive symptom description information of the user;

[0032] The posture information collection unit is used to collect continuous posture information of the user;

[0033] The chest expansion information collection unit is used to collect chest expansion information during the user's breathing process;

[0034] The method comprises:

[0035] The symptom description data, posture data and chest expansion data corresponding to the user are extracted based on the symptom description information, the posture information and the chest expansion information respectively.

[0036] In a third aspect, an embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the embodiment of the present application when executing the program.

[0037] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the embodiment of the present application.

[0038] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, characterized in that when the computer program is executed by a processor, it implements the method described in the embodiment of the present application.

[0039] The remote data collection system for ankylosing spondylitis provided in the embodiment of the present application can collect information about the user in at least three dimensions, including basic information, posture information, and chest expansion information, effectively solving the problem of inability to diagnose ankylosing spondylitis due to insufficient available data dimensions for early screening. At the same time, using the information extraction unit to extract data from the three dimensions of information respectively can not only improve the accuracy and efficiency of existing calculation methods, but also improve the consistency of data results, providing a reliable data basis for doctors' subsequent diagnosis.

[0040] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0042] Figure 1 A block diagram of a remote data acquisition system for early screening of ankylosing spondylitis provided by an embodiment of the present application is shown;

[0043] Figure 2 A flow chart of a data collection method of a remote data collection system based on early screening of ankylosing spondylitis provided in one embodiment of the present application is shown;

[0044] Figure 3 A flow chart of a data collection method of a remote data collection system based on early screening of ankylosing spondylitis provided in another embodiment of the present application is shown;

[0045] Figure 4 A flow chart of a data collection method of a remote data collection system based on early screening of ankylosing spondylitis provided in another embodiment of the present application is shown;

[0046] Figure 5 A flow chart of a data collection method of a remote data collection system based on early screening of ankylosing spondylitis provided in yet another embodiment of the present application is shown;

[0047] Figure 6 A schematic diagram of the structure of a computer system of an electronic device or server suitable for implementing an embodiment of the present application is shown. DETAILED DESCRIPTION

[0048] The present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0049] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0050] It should also be noted that the acquisition of various information involved in this application complies with relevant legal provisions, for example, obtaining medical records, symptoms and other information.

[0051] In order to further illustrate the technical solution provided by the embodiment of the present application, this is described in detail below in conjunction with the accompanying drawings and specific implementation methods. Although the embodiment of the present application provides the method operation instruction steps shown in the following embodiments or drawings, more or less operation instruction steps may be included in the method based on routine or no creative labor. In the steps where there is no necessary causal relationship logically, the execution order of these steps is not limited to the execution order provided by the embodiment of the present application. The method may be executed in the order of the method shown in the embodiment or drawings or in parallel during the actual processing process or when the device is executed.

[0052] Please refer to Figure 1 , Figure 1 FIG. 1 is a block diagram of a remote data collection system for early screening of ankylosing spondylitis provided by an embodiment of the present application. Figure 1 As shown, the system 100 includes: a basic information acquisition unit 101 , a posture information acquisition unit 102 , a chest expansion information acquisition unit 103 and an information extraction unit 104 .

[0053] The basic information collection unit 101 is used to receive the user's symptom description information.

[0054] The posture information collection unit 102 is used to collect the user's continuous posture information.

[0055] The chest expansion information collection unit 103 is used to collect chest expansion information during the user's breathing process.

[0056] The information extraction unit 104 is used to extract the symptom description data, posture data and chest expansion data corresponding to the user based on the symptom description information, posture information and chest expansion information respectively.

[0057] It should be noted that the basic information collection unit 101 may be a device capable of acquiring user status information. In some embodiments, the basic information collection unit 101 may be a user's own terminal device. For example, the user may directly input his or her own symptom description information, such as his or her own pain perception, degree of limb stiffness, etc., through the terminal device according to the information framework or questions provided by, for example, a consultation platform, so as to achieve the purpose of information collection. In other embodiments, the basic information collection unit 101 may also be an information storage unit of a medical institution, such as a physical examination institution, a community medical institution, etc. When the user undergoes a physical examination or consultation, the relevant medical staff inputs and uploads the user's symptom description information based on the examination or consultation results, so as to achieve the purpose of information collection.

[0058] The posture information acquisition unit 102 is a device for determining the user's posture based on the user's image information, including but not limited to an image acquisition device and / or a spatial information acquisition device. Optionally, the posture information acquisition unit 102 can rely only on the image acquisition device to acquire the user's posture, or can rely only on the spatial information acquisition device to acquire the user's posture, or can rely on both the image acquisition device and the spatial information acquisition device to acquire the user's posture, which is not specifically limited in this application. Among them, the image acquisition device can be an RGB-D camera, and the spatial information acquisition device can be a radar sensor (LiDAR sensor).

[0059] In a feasible embodiment, the posture information collection unit 102 can collect information when the user is unconscious, for example, the posture information collection unit 102 is set at the passage, so that the user can collect posture information without perception during the walking process. Optionally, in order to further improve the standardization of posture information collection, the posture information collection unit 102 can be set to collect posture information of users in a specific area, and when collecting posture information, the user takes a specific posture, such as standing, swinging, etc. Therefore, the posture information collection unit 102 can better capture the changes in the user's posture through the specific posture posed by the user, and provide a more reliable data basis for subsequent posture data extraction.

[0060] The chest expansion information acquisition unit 103 is used to collect the user's chest expansion information. It should be noted that the onset of ankylosing spondylitis can cause patients to have limited chest movement problems, and was included in the ankylosing spondylitis diagnostic criteria in 1984. Based on this, the present application provides a new type of chest expansion information acquisition unit 103, which effectively overcomes the defect that imaging examinations are static data, that is, the chest expansion information that changes during the user's breathing process can be detected, which can not only meet the data requirements for early screening of ankylosing spondylitis for users, but also reduce the cost of inspections, and provide reliable remote data support for early screening of ankylosing spondylitis.

[0061] In a feasible embodiment, the chest expansion information acquisition unit 103 can be a pressure sensing component based on 3D printing technology and hydrogel materials. The 3D printing technology can realize personalized customization of the user, so that the chest expansion information acquisition unit 103 can better fit the user's body and improve the accuracy of chest expansion information collection. The use of 3D printing technology can greatly reduce the customization cost. At the same time, the hydrogel material can ensure that the chest expansion information acquisition unit 103 is not irritating to the skin, thereby improving the comfort of the user when wearing it.

[0062] Specifically, hydrogel is used as a 3D printing material, and the three-dimensional information of the chest expansion information acquisition unit 103 customized for the user is determined by scanning the user's chest, and then 3D printing is performed using a 3D printer to obtain a hydrogel structure that can be attached to the user's chest. Then, based on the need for chest expansion pressure information collection, at least one pressure sensor is set on the inside of the hydrogel structure, so that after the user wears the hydrogel structure, the pressure sensor can be between the hydrogel structure and the user's chest, so that it can not only not be displaced under the attachment pressure of the hydrogel structure to the user's chest, but also can fully withstand the pressure changes caused by the expansion and contraction of the chest during the user's breathing, thereby realizing the collection of chest expansion information.

[0063] The information extraction unit 104 is used to extract the symptom description data, posture data and chest expansion data corresponding to the user based on the symptom description information, posture information and chest expansion information, respectively. Specifically, the information extraction unit 104 uses an optimized crayfish iteration algorithm to extract the symptom description data based on the symptom description information, uses an optimized tracking algorithm to extract the posture data based on the posture information, and provides an algorithm matching the chest expansion information to extract the chest expansion data.

[0064] Therefore, the remote data collection system for early screening of ankylosing spondylitis provided in the embodiment of the present application can collect information about the user in at least three dimensions, including basic information, posture information, and chest expansion information, effectively solving the problem of inability to diagnose ankylosing spondylitis due to insufficient available data dimensions. At the same time, using the information extraction unit to extract data from the three dimensions of information respectively can not only improve the accuracy and efficiency of the existing calculation method, but also improve the consistency of the data results, providing a reliable data basis for the doctor's subsequent diagnosis.

[0065] In a feasible embodiment, the information extraction unit is specifically used to: obtain symptom description information, use the optimized crayfish iterative algorithm to iteratively explore the symptom description information, and obtain at least one candidate symptom acquisition strategy; evaluate at least one candidate symptom strategy based on a preset fitness function, and select the candidate symptom strategy with the highest fitness as the target symptom acquisition strategy; extract symptom description data from the symptom description information based on the target symptom acquisition strategy.

[0066] This application utilizes an optimized crayfish iterative algorithm that can dynamically search for strategies and make adjustments, thereby enhancing the model algorithm's depth of understanding of symptom description information, thereby optimizing the symptom acquisition strategy.

[0067] Specifically, an optimized crayfish iterative algorithm is used to perform iterative exploration in the symptom description information, including: obtaining initial symptom description information for iterative exploration, iteratively exploring the initial symptom description information based on the crayfish iterative algorithm, and obtaining a symptom acquisition strategy corresponding to the initial symptom description; when the number of iterations in which the symptom acquisition strategy has not changed continuously reaches a preset number, the symptom acquisition strategy is used as a candidate symptom acquisition strategy, and new initial symptom description information is returned to obtain.

[0068] The following formula can be used to obtain the initial symptom description information:

[0069] z=M min +(M max -M min )×Q

[0070] Among them, z is the initial description information, M min is the lower bound of the symptom description information, M max is the previous session of symptom description information, and Q is the random reference factor.

[0071] That is to say, in order to expand the search scope of the crayfish iterative algorithm, as much initial symptom description information as possible is obtained for iterative exploration by randomly generating feasible solutions, so as to obtain sufficiently rich candidate symptom acquisition strategies, thereby ensuring the reliability of the target symptom acquisition strategy. In addition, when the number of iterations in which the symptom acquisition strategy has not changed continuously during the iterative exploration process reaches a preset number, the iterative exploration of the symptom acquisition strategy is stopped, which can reduce the exploration pressure of the crayfish iterative algorithm, improve the efficiency of determining the candidate symptom acquisition strategy, and prevent the algorithm from falling into the local optimal solution too early.

[0072] In a specific embodiment, first, the goal of acquiring symptom description data is clarified, such as maximizing the accuracy and coverage of symptom description data, and at the same time, constraints such as time and cost are determined. Then, a fitness function is set according to the level of detail of the symptom description data, the completeness of the diagnostic information, etc. When the symptom description data acquired based on the candidate symptom acquisition strategy is evaluated according to the fitness function, the higher the evaluation result, the higher the quality of the symptom description data acquired by the candidate symptom acquisition strategy, and the higher the accuracy for subsequent diagnosis. A certain number of initial symptom description information is set based on medical common sense, and a part of the initial symptom description information is generated by expanding the above-mentioned random reference factor. The initial symptom description information is iteratively explored in turn using the crayfish iteration algorithm to determine at least one candidate symptom acquisition strategy corresponding to each initial symptom description information, wherein, during the iteration process, if any symptom acquisition strategy stops changing during the iteration process, it is determined that the iteration of the symptom acquisition strategy is completed, determined as a complete candidate symptom acquisition strategy and scored using the fitness function. The candidate symptom strategy with the highest fitness function score among all candidate symptom strategies is used as the target symptom acquisition strategy to acquire the symptom description data corresponding to the user.

[0073] It should be understood that since basic information is usually input by users based on their own feelings, such as pain location, pain time, pain degree, etc., each user is different, and setting a fixed symptom acquisition strategy in advance can easily result in the lack of key symptoms in the symptom description data extracted from some patients, resulting in the problem of being unable to make an effective diagnosis. Based on this, the information extraction unit 104 proposed in this application can use the optimized crayfish iteration algorithm to iteratively explore the symptom description information input by each user, so as to obtain the target symptom acquisition strategy with the highest matching degree to the symptom description information that is most suitable for the user input, so that the symptom description data obtained according to the target symptom acquisition strategy can contain the key information of the user's symptoms as much as possible, avoid the lack of key information, and realize the personalized extraction of symptom description data, without restricting the user's symptom description, and fully guarantee the accuracy and reliability of the symptom description data under the premise of ensuring the user's freedom of expression, and provide a reliable data basis for the doctor's subsequent diagnosis.

[0074] In a feasible embodiment, the posture information acquisition unit includes a first sensor and a second sensor, and the information extraction unit is also used to: respectively obtain a first image frame collected by the first sensor and a second image frame collected by the second sensor at the same moment; perform three-dimensional data fusion on the first image frame and the second image frame, and use a CNN-based target edge detection algorithm to obtain the posture boundary corresponding to the moment; perform posture estimation on the image frame at the moment based on the posture boundary to obtain posture data.

[0075] The first sensor and the second sensor are respectively an image acquisition device and a spatial information acquisition device, such as an RGB-D camera and a LiDAR sensor.

[0076] It should be understood that in order to improve the accuracy of posture estimation using different sensor devices, sensor fusion needs to be performed first, that is, the two-dimensional features captured by the RGB-D camera and the three-dimensional features collected by the LiDAR sensor are deeply associated to better predict the user's actual posture.

[0077] In a feasible embodiment, the three-dimensional features (such as three-dimensional point cloud data) collected by the LiDAR sensor and the two-dimensional features (such as RGB images) captured by the RGB-D camera are input into the CNN algorithm model, and the target edge detection algorithm in the CNN algorithm model is used to extract the posture boundary. Optionally, the extracted posture boundary is a 3D bounding box. Preferably, the posture boundary is the minimum bounding rectangle corresponding to the user. Specifically, by tracking the target of multiple frames of images at consecutive moments, the posture boundary for characterizing the minimum bounding rectangle corresponding to the user is obtained.

[0078] In a feasible embodiment, in order to further reduce the size of the bounding rectangle used to characterize the posture boundary, the projection point of at least one feature point in the three-dimensional point cloud data on the two-dimensional plane is obtained, and the projection point is matched and aligned with the pixel point in the two-dimensional feature. This can effectively reduce the problem that the bounding rectangle corresponding to the three-dimensional point cloud data is larger than the bounding rectangle of the actual target object in the two-dimensional feature image, thereby effectively improving the accuracy of posture estimation and ensuring the consistency of target tracking and the accuracy of posture estimation.

[0079] In a feasible embodiment, when diagnosing ankylosing spondylitis, doctors need to understand the patient's posture data in a static state, and also need to understand the posture information in a moving state, in order to determine the degree of distress caused by spinal fusion and muscle stiffness to the patient's behavior. Therefore, the information extraction unit also needs to track the posture boundaries in continuous image frames to understand the changes in the user's posture during exercise. Among them, in order to improve the accuracy of data association in continuous image frames (including the first image frame and the second image frame), ensure the consistency of target tracking and the accuracy of posture estimation, the application also proposes to use the intersection joint matching method in the time domain to check the posture boundaries detected in the continuous image frames to ensure the accuracy and consistency of posture estimation in the continuous image frames.

[0080] It should be understood that the determination of the gesture boundary can determine the user's motion range. After the gesture boundary in the image frame is determined, gesture estimation is performed through motion prediction to determine the user's gesture data in the image frame.

[0081] In a feasible embodiment, during the posture estimation process, a joint loss function is used to perform posture optimization to obtain an optimal posture estimation value as posture data. The joint loss function includes at least a smoothing term, a scene-aware contact term, a posture prior term, and a grid-to-point term.

[0082] Among them, the smoothing term is used to smooth the human body motion by minimizing the acceleration and angular velocity of the joints. Specifically, the smoothing term focuses on the acceleration of the user's pelvic joints and the other 23 joints relative to the pelvis, while also considering the angular velocity of the user's other joints. Based on this, it is possible to avoid abrupt changes in motion when performing posture estimation in continuous image frames, making the posture estimation in continuous image frames more accurate and more in line with the actual posture of the user when collecting posture information.

[0083] The scene-aware contact item is used to analyze the movement of the limb vertices, such as the movement of the foot vertices. In the embodiment of the present application, when the speed of the foot is analyzed to be less than 0.1 m / s by the movement of the foot vertices in the continuous image frames, the user's foot is considered to be in a stable state. Then, the chamfer distance between the foot and its closest surface is calculated as the loss of scene contact, so as to ensure that the physical contact between the action and the scene is in line with the actual situation.

[0084] The pose prior is used to constrain the pose during the optimization process to be as close to the initial value as possible.

[0085] The mesh-to-point item is used to provide depth prior information using the three-dimensional point cloud data generated by the dynamic radar sensor. By defining the chamfer distance from the human body endpoint to the contact surface, the correspondence between the human body surface and the actual scene is considered to enhance the depth perception capability of posture estimation.

[0086] In a specific embodiment, a gradient descent algorithm is used to gradually minimize the joint loss function, and through an iterative optimization process, the posture parameters of the posture estimation are continuously adjusted until the optimal solution that minimizes the loss function is found as the posture data.

[0087] Therefore, the present application utilizes two-dimensional images and three-dimensional point cloud data to jointly perform posture estimation, which effectively enriches the richness of posture information, fully considers the depth information in the three-dimensional environment, and uses a joint loss function for posture estimation. It can achieve accurate estimation of the user's posture and improve the accuracy of posture data. It can provide more accurate judgment data for early ankylosing spondylitis when the lesions are not obvious, thereby improving the reliability of early screening.

[0088] In a feasible embodiment, the information extraction unit 104 is also used to: select at least one key area around the thorax, and for each key area, use regional masking technology to obtain the pressure-time curve corresponding to each key area from the thorax expansion information; based on the pressure-time curve, determine the pressure peak value and pressure-time integral corresponding to the key area, and use the pressure peak value and pressure-time integral as thorax expansion data.

[0089] It should be understood that in a respiratory cycle, the expansion and contraction of the lungs will change the pressure around the thorax, and different areas will form different pressure ranges and changes based on the respiratory cycle of the lungs. Therefore, dividing the thorax into multiple key areas and extracting thorax expansion data separately can ensure the reliability of data extraction while providing high-precision data guarantee for subsequent doctor diagnosis, that is, effectively preventing the loss of abnormal data caused by operations such as data averaging. Among them, the key areas include at least the intercostal space, the posterior part of the sternum, and the area around the spine.

[0090] In the embodiment of the present application, for the planned multiple key areas, the regional masking technology is used to obtain the pressure-time curve corresponding to each key area from the chest expansion information. Specifically, the chest expansion information acquisition unit can focus on multiple areas around the chest, and when extracting the pressure-time curve, the regional masking technology is used to distinguish the current focus area from other areas, for example, setting other areas to 0, so that the pressure data of the current key area can be highlighted, thereby realizing the partitioned extraction of pressure data, and drawing the pressure-time curve corresponding to the key area according to time.

[0091] For each key area, after obtaining the pressure-time curve, the pressure data is constructed into a pressure matrix according to the acquisition frequency of the chest expansion information acquisition unit, and then the pressure peak value max(x ij ).

[0092] The pressure-time integral is the integral of the chest expansion pressure over time, and is used to characterize the mechanical load generated by chest expansion. Specifically, the pressure-time integral can be obtained using the following formula:

[0093]

[0094] Among them, F is the pressure value, t is the time, and A is the area of ​​the critical area.

[0095] It should be understood that the early lesions of ankylosing spondylitis are prone to occur due to the relatively mild degree of spinal fusion, which has not yet affected the thoracic function and thus has not yet caused a significant change in the pressure value. However, spinal fusion will cause the reaction time of bones and muscles to prolong, that is, the pressure-time integral will increase.

[0096] Therefore, the embodiment of the present application uses the chest expansion information collected by the chest expansion information collection unit with high accuracy to extract chest expansion data, which effectively ensures the reliability of pressure data and provides reliable data support for subsequent medical diagnosis. At the same time, the use of key area partition extraction can effectively avoid the problem of missing key data during global extraction due to the insignificant impact of early lesions, so that local abnormalities under the influence of early lesions can be effectively discovered, thereby improving the accuracy and reliability of early screening. The use of pressure-time integral as chest expansion data for data extraction can identify abnormal chest expansion due to the influence of ankylosing spondylitis earlier, provide a data basis for the early screening of ankylosing spondylitis lesions, and improve the accuracy of early screening of ankylosing spondylitis.

[0097] In summary, the remote data collection system for early screening of ankylosing spondylitis provided in the embodiment of the present application can collect information about the user in at least three dimensions, including basic information, posture information, and chest expansion information, effectively solving the problem of inability to diagnose ankylosing spondylitis due to insufficient available data dimensions. At the same time, using the information extraction unit to extract data from the three dimensions of information respectively can not only improve the accuracy and efficiency of the existing calculation method, but also improve the consistency of the data results, providing a reliable data basis for the doctor's subsequent diagnosis.

[0098] For the several modules or units mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0099] Figure 2 A flow chart of a data collection method of a remote data collection system based on early screening of ankylosing spondylitis provided in one embodiment of the present application is shown.

[0100] like Figure 2 As shown, the method includes:

[0101] Step 201, obtaining symptom description information, posture description information and chest expansion information.

[0102] Step 202 , extracting symptom description data, posture data and chest expansion data corresponding to the user based on the symptom description information, posture description information and chest expansion information respectively.

[0103] In a possible embodiment, if Figure 3 As shown, symptom description data is extracted based on the symptom description information, including:

[0104] Step 301, obtain symptom description information, use the optimized crayfish iterative algorithm to iteratively explore the symptom description information, and obtain at least one candidate symptom acquisition strategy.

[0105] Step 302: Evaluate at least one candidate symptom acquisition strategy based on a preset fitness function, and select the candidate symptom acquisition strategy with the highest fitness as the target symptom acquisition strategy.

[0106] Step 303: extract symptom description data from the symptom description information based on the target symptom acquisition strategy.

[0107] In a possible embodiment, if Figure 4 As shown, the posture data is extracted based on the posture information, including:

[0108] Step 401 : acquiring a first image frame captured by a first sensor and a second image frame captured by a second sensor at the same time.

[0109] Step 402: perform three-dimensional data fusion on the first image frame and the second image frame, and use a CNN-based target edge detection algorithm to obtain a posture boundary corresponding to the moment.

[0110] Step 403: perform posture estimation on the image frame at the moment based on the posture boundary to obtain posture data.

[0111] In a possible embodiment, if Figure 5 As shown, extracting chest expansion data based on chest expansion information includes:

[0112] Step 501, selecting at least one key area around the thorax; the key area at least includes the intercostal space, the back of the sternum, and the area around the spine.

[0113] Step 502: For each key area, a pressure-time curve corresponding to each key area is obtained from the chest expansion information using a regional masking technique.

[0114] Step 503: Based on the pressure-time curve, determine the pressure peak value and pressure-time integral corresponding to the key area, and use the pressure peak value and pressure-time integral as chest expansion data.

[0115] To summarize, the data collection method of the remote data collection system based on the early screening of ankylosing spondylitis provided in the embodiment of the present application extracts data from three dimensions of basic information, posture information and chest expansion information, which can not only improve the accuracy and efficiency of existing calculation methods, but also improve the consistency of data results, thereby providing a reliable data basis for doctors' subsequent diagnosis.

[0116] Reference below Figure 6 , Figure 6A schematic diagram of the structure of a computer system of an electronic device or server suitable for implementing an embodiment of the present application is shown.

[0117] like Figure 6 As shown, the computer system includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 602 or the program loaded from the storage part 608 to the random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation instructions of the system are also stored. The CPU 601, the ROM 602 and the RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0118] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage section 608 as needed.

[0119] In particular, according to an embodiment of the present application, the above reference flow chart Figure 2 The described process can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer readable medium, and the computer program includes a program code for executing the method shown in the flow chart. In such an embodiment, the computer program includes a program code for executing the method shown in the flow chart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, the above-mentioned functions defined in the system of the present application are executed.

[0120] It should be noted that the computer-readable medium shown in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium such as a computer-readable storage medium that can send, propagate or transmit a program for use by or in conjunction with an instruction execution system, device or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0121] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operating instructions of the system, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the aforementioned module, program segment or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, the boxes represented by two connections can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operating instruction, or can be implemented with a combination of dedicated hardware and computer instructions.

[0122] As another aspect, the present application further provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiment, or may exist independently without being assembled into the electronic device. The above computer-readable storage medium stores one or more programs, and when the above programs are used by one or more processors to execute the data collection method based on the remote data collection system for ankylosing spondylitis described in the present application.

[0123] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the aforementioned disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

Claims

1. A remote data acquisition system for early screening of ankylosing spondylitis, characterized in that: include: Basic information collection unit, posture information collection unit, chest expansion information collection unit and information extraction unit, The basic information collection unit is used to receive symptom description information of the user; The posture information collection unit is used to collect continuous posture information of the user; The chest expansion information collection unit is used to collect chest expansion information during the user's breathing process; The information extraction unit is used to extract the symptom description data, posture data and chest expansion data corresponding to the user based on the symptom description information, the posture information and the chest expansion information respectively.

2. The remote data acquisition system for early screening of ankylosing spondylitis according to claim 1, characterized in that: The information extraction unit is further used for: Acquire the symptom description information, and perform iterative exploration in the symptom description information using an optimized crayfish iterative algorithm to obtain at least one candidate symptom acquisition strategy; Evaluate the at least one candidate symptom acquisition strategy based on a preset fitness function, and select the candidate symptom acquisition strategy with the highest fitness as the target symptom acquisition strategy; Based on the target symptom acquisition strategy, the symptom description data is extracted from the symptom description information.

3. The remote data acquisition system for early screening of ankylosing spondylitis according to claim 2, characterized in that: The information extraction unit is further used for: Obtain initial symptom description information for iterative exploration; Iteratively explore the initial symptom description information based on the crayfish iterative algorithm to obtain a symptom acquisition strategy corresponding to the initial symptom description; When the number of iterations in which the symptom acquisition strategy remains unchanged continuously reaches a preset number, the symptom acquisition strategy is used as the candidate symptom acquisition strategy, and the new initial symptom description information is returned to be acquired.

4. The remote data acquisition system for early screening of ankylosing spondylitis according to claim 1, characterized in that: The posture information acquisition unit includes a first sensor and a second sensor, and the information extraction unit is further used for: Respectively acquiring a first image frame acquired by the first sensor and a second image frame acquired by the second sensor at the same moment; Performing three-dimensional data fusion on the first image frame and the second image frame, and using a CNN-based target edge detection algorithm to obtain a posture boundary corresponding to the moment; The posture data is obtained by performing posture estimation on the image frame at the moment based on the posture boundary.

5. The remote data acquisition system for early screening of ankylosing spondylitis according to claim 4, characterized in that: The information extraction unit is further used for: In the posture estimation process, a joint loss function is used to perform posture optimization to obtain an optimal posture estimation value as the posture data; The joint loss function includes at least a smoothness term, a scene-aware contact term, a pose prior term, and a mesh-to-point term.

6. The remote data acquisition system for early screening of ankylosing spondylitis according to claim 1, characterized in that: The information extraction unit is further used for: Selecting at least one key area around the thorax; the key area at least includes the intercostal space, the posterior part of the sternum, and the area around the spine; For each of the key areas, a pressure-time curve corresponding to each of the key areas is obtained from the chest expansion information using a regional masking technique; Based on the pressure-time curve, the pressure peak value and the pressure-time integral corresponding to the key area are determined, and the pressure peak value and the pressure-time integral are used as the chest expansion data.

7. A data collection method based on a remote data collection system for early screening of ankylosing spondylitis, characterized in that: The system includes: a basic information collection unit, a posture information information collection unit, chest expansion information collection unit and information extraction unit, The basic information collection unit is used to receive symptom description information of the user; The posture information collection unit is used to collect continuous posture information of the user; The chest expansion information collection unit is used to collect chest expansion information during the user's breathing process; The method comprises: Acquiring the symptom description information, the posture information and the chest expansion information; The symptom description data, posture data and chest expansion data corresponding to the user are extracted based on the symptom description information, the posture information and the chest expansion information respectively.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the data collection method of the remote data collection system based on early screening of ankylosing spondylitis as claimed in claim 7 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the data collection method of the remote data collection system based on early screening of ankylosing spondylitis as claimed in claim 7 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the data collection method of the remote data collection system based on early screening of ankylosing spondylitis described in claim 7 is implemented.

Citation Information

Patent Citations

  • Early disease prediction method and related equipment

    CN116189878A

  • Dynamic tracking and screening system for physical activity track of ankylosing spondylitis patient

    CN117122307A

  • Spine detection method and system, electronic equipment and medium

    CN117576734A

  • Feature classification method based on multi-strategy improved crayfish algorithm

    CN117911799A

  • Method and system for segmenting medical imaging data according to a skeletal atlas

    US20100128954A1