Visualized intelligent management system for closed reduction of fractures

Abstract

The present invention is applicable to the technical field of human bone surgery, and provides a visual intelligent management system for closed reduction of fractures, including a main control module that receives images from the input end of the system, a cloud database and a local database, a probe, and a pressure sensor and a gyroscope, a projection module and other structures provided on the probe. The present invention matches the current fracture injury site by calling the fracture data and probe data in the database, thereby calling the repair data with a suitable matching degree to provide a treatment reference plan for the current injured site. Based on the bone data and repair data in the database, a laser projection line is projected onto a platform, and a mold taking at the fracture site can also provide a guiding position for the current insertion position of the probe. The instrument user fixes the fracture site with reference to the projected laser straight line to optimize the insertion mode of the probe.

Description

Technical Field

[0001] The present invention relates to the technical field of human bone surgery, and in particular to a visual intelligent management system for closed reduction of fractures. Background Art

[0002] Fracture refers to the complete or partial break in the continuity of bone structure, which is a common trauma in life. In the field of automated human bone surgery, the closed reduction and internal fixation of femoral neck fractures usually requires the results of medical imaging to make judgments. However, traditional closed reduction and internal fixation of femoral neck fractures requires the help of doctors to set relevant points, and doctors and patients need to withstand more radiation doses during the operation. In addition, traditional bone closure surgery lacks automated identification and guidance of the puncture point, end point and puncture direction, as well as detection of probes.

[0003] In summary, the existing technology obviously has inconveniences and defects in practical use, so it is necessary to improve it. In order to solve the above problems, a visual intelligent management system for closed reduction of fractures is proposed. Summary of the invention

[0004] In view of the above-mentioned defects, the purpose of the present invention is to provide a visual intelligent management system for closed reduction of fractures, reduce the number of X-ray fluoroscopy times, improve the accuracy and efficiency of fracture reduction, reduce medical and patient injuries, and use a magnetic locator to collect spatial data of broken bones in real time and move the three-dimensional fracture model rendered in the computer in real time. The position of the broken bones can be monitored in real time, assisting doctors in closed reduction of fractures, improving the accuracy and efficiency of reduction, and providing a visual intelligent management system for closed reduction of fractures.

[0005] In order to achieve the above-mentioned objectives, the present invention provides a visual intelligent management system for closed reduction of fractures, including a main control module for receiving images from the input end of the system, the main control module including a computing unit for performing operations on data and programs, a storage unit for storing data, and a communication unit for transmitting data.

[0006] The database includes a cloud database and a local database. The cloud database is connected to the local database and is used to store image data.

[0007] The execution module comprises a probe, on which a pressure sensor and a gyroscope are arranged. The top of one end of the probe is provided with a pressure sensor, and the inside of the probe is provided with a gyroscope.

[0008] The projection module comprises a platform, and a laser projection component is connected to the platform. The laser projection component is used to project a corresponding probe insertion auxiliary line on the platform.

[0009] According to the visual intelligent management system for closed reduction of fractures of the present invention, the steps for using the visual intelligent management system are as follows:

[0010] S1, establish a local database and a cloud database, calculate and store the CT data of the CT image of the initial fracture state and the CT image of the repaired bone fracture, wherein the CT data includes bone data and repair data.

[0011] S2, performing real-time CT scanning on the injured part on the platform to obtain anteroposterior and lateral views of the bones, uploading the captured anteroposterior and lateral views to the main control module, and matching them through matching features of the bones at the injured part.

[0012] S3, calculating the insertion slope of the probe according to the matched CT image, and projecting the laser onto the injured part using the projection module.

[0013] S4, insert the probe into the injured part along the laser projection, start the gyroscope in the probe and record the initial position of the gyroscope, and calculate the bone reduction status at the injured part based on the initial position of the probe, the three-dimensional model of the fracture and the movement of the probe position.

[0014] According to the visual intelligent management system for closed reduction of fractures of the present invention, the cloud database and the local database collect several CT images of different bones, and establish the collected CT images Datasets and Dataset, where The data set is CT images of the initial fracture state of different bones. The data set is the corresponding CT image of the bone fracture repair, that is, the CT image taken after the bone fracture is repaired by using a probe or a steel nail, and the main control module is used to Datasets and The images in the dataset are subjected to image processing such as dilation and edge processing to generate corresponding bone data and repair data.

[0015] According to the visual intelligent management system for closed reduction of fractures of the present invention, the vertical distance between the straight lines on both sides , the straight lines on both sides and the matching features form Plane area, calculate The number of pixels of the black area in the plane area or the minimum envelope rectangle, the broken part is regarded as a rectangle, and the break distance of the broken part is set to , the number of pixels corresponding to the fracture distance is , the total number of pixels in the broken part is ,pass , calculate the number of pixels corresponding to the fracture distance, , and thus the fracture distance is calculated .

[0016] According to the visual intelligent management system for closed reduction of fractures of the present invention, The probe positions in the repaired CT images of bone fractures in the dataset are manually annotated, and the probe straight line is generated by least squares fitting using the annotated probe position area , let the straight line parallel to the backbone be the 0° position line, the straight line perpendicular to the backbone be the 90° position line, the 90° position line pointing out of the human body be the +90° position line, and the position line pointing into the human body be the -90° position line, and obtain the inclination angles of several probe straight lines relative to the fracture position .

[0017] According to the visual intelligent management system for closed reduction of fractures of the present invention, the position coordinates of the first characteristic point of the bone are set as , the position coordinates of the second feature point of the skeleton are , the upper left corner of the image is defined as the coordinate origin , the initial matrix is ​​formed by the vectors of the feature points and the origin ,

[0018] ,

[0019] Assume that the position coordinates of the first feature point of the broken bone in the current CT image are , the position coordinates of the second feature point of the skeleton are , the upper left corner of the image is defined as the coordinate origin , the initial matrix is ​​formed by the feature points of the current CT image and the vector of the origin , to achieve matching,

[0020] .

[0021] According to the visual intelligent management system for closed reduction of fractures of the present invention, the intersection of the probe and the backbone edge of the image in the database is assumed to be , let the position of the probe corresponding to the current image be ,

[0022] ,

[0023] Let the transformation matrix .

[0024] The present invention provides a visual intelligent management system for closed reduction of fractures, which has the following beneficial effects:

[0025] 1. Match the current fracture injury site by calling the fracture data and probe data in the database, and then select the repair data with appropriate matching degree to provide a treatment reference plan for the current injury site. Project the laser projection line on the platform based on the bone data and repair data in the database. The mold taking of the fracture site can also provide guidance for the current probe insertion position. The instrument user can fix the fracture site according to the projected laser straight line reference to assist in optimizing the probe insertion mode.

[0026] 2. Using the position of the probe at both ends of the fracture, through the position coordinates of the probe and the initial angle of penetration into the bone, the fracture state and three-dimensional model of the bone at the fracture site in the initial state of the component, when the bone at the fracture site is repositioned, the probe moves accordingly with the movement of the bone, and then generates a new three-dimensional model of the bone after the fracture is closed based on the original one according to the three-dimensional coordinates of the moved probe, thereby realizing the judgment of the closure state of the bone according to the position of the probe, avoiding and reducing the workload caused by multiple shots of the fracture site and reducing the radiation amount caused to patients and doctors. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a flowchart of the steps used by the visual intelligent management system for closed reduction of fractures of the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0029] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0030] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0031] See also Figure 1The present invention provides a visual intelligent management system for closed reduction of fractures, which includes a main control module for receiving images from the input end of the system and obtaining output projection data based on the input calculations. The main control module includes a computing unit for performing operations on data and programs, a storage unit for storing data, and a communication unit for transmitting data.

[0032] The database includes a cloud database and a local database. The cloud database is connected to the local database. The database matches the image at the input end and compares it with the local database first. When the matching rate is lower than the set matching value, the image at the input end is uploaded to the cloud database and the data in the cloud database is called for matching. During the matching process, a new number is created for the uploaded image at the input end and stored in the cloud database, and at the same time, the image restoration image of the subsequent input end is merged into the above number.

[0033] The execution module comprises a probe, on which a pressure sensor and a gyroscope are arranged. The top of one end of the probe is provided with a pressure sensor, and a gyroscope is arranged inside the probe. The gyroscope is used to measure the current rotation state of the probe.

[0034] The projection module comprises a platform, on which a laser projection component is connected, and the laser projection component is used to project a corresponding probe insertion auxiliary line on the platform.

[0035] The projection module receives the projection data calculated by the main control module, and projects the projection data onto the injured part of the patient on the platform through the laser projection component.

[0036] The data entry module contains relevant data of the image in the database, including the height of the patient corresponding to the image and the length of the circumference of the injured part.

[0037] Furthermore, the visual intelligent management system of the present invention is used in the following steps:

[0038] S1, establish a local database and a cloud database, calculate and store the CT data of the CT image of the initial fracture state and the CT image of the repaired bone fracture, wherein the CT data includes bone data and repair data.

[0039] S2, performing real-time CT scanning on the injured part on the platform to obtain anteroposterior and lateral views of the bones, uploading the captured anteroposterior and lateral views to the main control module, and matching them through matching features of the bones at the injured part.

[0040] S3, calculating the insertion slope of the probe according to the matched CT image, and projecting the laser onto the injured part using the projection module.

[0041] S4, insert the probe into the injured part along the laser projection, start the gyroscope in the probe and record the initial position of the gyroscope, and calculate the bone reduction status at the injured part based on the initial position of the probe, the three-dimensional model of the fracture and the movement of the probe position.

[0042] Among them, the gyroscope uses the MPU6050 chip produced by InvenSense, which includes a three-axis gyroscope and a three-axis accelerometer. The chip outputs acceleration and angular velocity signals through IIC. The maximum sensitivity of the gyroscope is 131LSBs / ° / sec, that is, 1LSB corresponds to 0.0076° of the MPU6050 chip.

[0043] Among them, a cloud database and a local database are established to collect several CT images of different bones, and the collected CT images are established Datasets and Dataset, where The data set is CT images of the initial fracture state of different bones. The data set is the corresponding CT image of the bone fracture repair, that is, the CT image taken after the bone fracture is repaired by using a probe or a steel nail, and the main control module is used to Datasets and The images in the dataset are subjected to image processing such as dilation and edge processing to generate corresponding bone data and repair data.

[0044] Furthermore, the bone data includes at least the bone name, bone position information, and the inclination of the bone fracture; the repair data includes at least the number of probes used and the probe positions; and each probe is serially numbered to facilitate generation of a position array corresponding to the probe.

[0045] Furthermore, the bone position information includes the position coordinates, tilt angle and number of bone pixels of the bone corresponding to the current bone serial number, and the probe position includes the position coordinates, tilt angle and pixel information corresponding to the probe corresponding to the current probe serial number.

[0046] Among them, The probe positions in the corresponding bone fracture repair CT images in the dataset are manually annotated, and the probe straight line is generated by least squares fitting using the annotated probe position area , let the straight line parallel to the backbone be the 0° position line, the straight line perpendicular to the backbone be the 90° position line, the 90° position line pointing out of the human body be the +90° position line, and the position line pointing into the human body be the -90° position line, and obtain the inclination angles of several probe straight lines relative to the fracture position .

[0047] According to the initial fracture state of the bone, the CT images of the frontal and lateral views corresponding to the fracture state of the bone are collected, the images are binarized, and the shapes of the bone ends on both sides of the fractured bone are matched with the shapes of the bone ends on both sides of the normal bone to locate the position of the fractured bone. The fractured bone image and the normal bone image are processed in a uniform proportion according to the number of pixels of the matching features, and the center position point and the inclination angle of the fractured bone are determined by connecting the shape feature points of the bone ends on both sides of the fractured bone. The length of the connecting line of the bone ends on both sides of the bone is set to , the corresponding number of pixels is ,

[0048] Furthermore, by inputting the name and left and right position information of the broken bone, such as "femur, left", on the interface of the visual intelligent management system for closed reduction of fractures, the name and left and right position information of the bone are used to match bones with the same bone name in the database.

[0049] Furthermore, the CT image of the initial fracture state of the bone is demarcated into an ROI area to obtain a single bone image in a fracture state. The straight lines on both sides of the bone are extracted from the single bone image in a fracture state. The vertical distance between the straight lines on both sides is calculated based on the straight lines on both sides of the bone. The vertical distance between the straight lines on both sides is set as , the corresponding number of pixels is .

[0050] Among them, the vertical distance between the two sides of the straight line , the straight lines on both sides and the matching features form Plane area, calculate The black area in the plane area (i.e. The fractured part of the plane area fracture appears black in the binary image, and the bone appears white) is the number of pixels or the minimum envelope rectangle. The fractured part is regarded as a rectangle, and the fracture distance of the fractured part is , the number of pixels corresponding to the fracture distance is , the total number of pixels in the broken part is ,pass , calculate the number of pixels corresponding to the fracture distance, , and thus the fracture distance is calculated .

[0051] The ratio of the actual distance to the pixel distance is set to , .

[0052] Furthermore, the fracture level ,

[0053] If the fracture level , using the projection module to project the corresponding projection straight line to the injured part;

[0054] If the fracture level , after manual judgment, the projection straight line can be manually adjusted and projected again;

[0055] If the fracture level or fracture grade , it is judged that the current bone fracture state does not meet the judgment standard of the fracture closed reduction visual intelligent management system, and manual intervention is required to judge and repair.

[0056] Preferably, a corresponding straight line is generated using the straight lines on both sides to generate a straight line parallel to the backbone, and the parallel straight line is intersected with the edge of the bone fracture to obtain two one-to-one corresponding points. The number of pixel points of the average fracture distance is calculated by calculating the distance between several two points.

[0057] The fracture distance d and bone data of the calculated current bone data are matched with the data in the local database. When the matching rate a between the data in the local database and the current bone data is greater than the initial matching rate, the repair data corresponding to the data in the local database is called, and the insertion slope of the probe is calculated based on the repair data to provide the instrument user with the corresponding data and projection curve of the probe insertion.

[0058] Among them, The dataset is a manual annotation of the probe positions in the corresponding bone fracture repair CT image, and the probe straight line is generated by least squares fitting using the annotated probe position area. , let the straight line parallel to the backbone be the 0° position line, the straight line perpendicular to the backbone be the 90° position line, the 90° position line pointing out of the human body be the +90° position line, and the position line pointing into the human body be the -90° position line, and obtain the inclination angles of several probe straight lines relative to the fracture position .

[0059] The ratio of the broken bone to the normal bone is optimized according to the vertical distance between the straight lines on both sides of the bone, and is optimized and matched according to the length of the circumference around the injured position.

[0060] Among them, the position coordinates of the first feature point of the skeleton are , the position coordinates of the second feature point of the skeleton are , the upper left corner of the image is defined as the coordinate origin , the initial matrix is ​​formed by the vectors of the feature points and the origin ,

[0061] .

[0062] Assume that the position coordinates of the first feature point of the broken bone in the current CT image are , the position coordinates of the second feature point of the skeleton are , the upper left corner of the image is defined as the coordinate origin , the initial matrix is ​​formed by the feature points of the current CT image and the vector of the origin ,

[0063] .

[0064] Let the transformation matrix , calculated by the transformation matrix ,

[0065] ,

[0066] .

[0067] Assume that the intersection of the probe and the backbone edge of the image in the database is ,

[0068] Suppose the position of the probe corresponding to the current image is ,

[0069] ,

[0070] Through the above calculation, the probe position straight line about the fracture site in the database is converted into a projection straight line received by the current projection module, and the projection module projects on the platform according to the above converted coordinates.

[0071] The monitoring module includes a probe, on which a pressure sensor and a gyroscope are arranged. The end of the probe inserted into the bone is provided with a pressure sensor. When the pressure sensor on the probe contacts the skin, the gyroscope in the probe is triggered to start working.

[0072] Using the probe positions obtained above, let the insertion slope of the first probe be , the insertion slope of the second probe is The fracture site is placed on the platform, and a projection module is used to perform laser projection on the fracture site. The projection module projects the fracture site according to the insertion slope of the first probe. and the insertion slope of the second probe A corresponding laser projection straight line is generated, and the projection straight line is projected at the fracture position. The instrument user inserts the probe into the fracture position along the projection straight line, so as to provide prompts to the instrument user through projection. According to the patient's fracture position and fracture degree, a reference mode and auxiliary treatment function are provided based on the bone data and repair data in the cloud database, which provides auxiliary reference for the instrument user and improves the visualization of the treatment process.

[0073] Furthermore, the main control module or sub-item control module and electrical integration are composed of various operators and chips, and are built using the CPU central processing unit system architecture. The communication system is also integrated into the electrical centralized system. The communication module further uses 4G / 5G / GPRS / WIFI / LORA / NB IOT / BIU / ZIGBEE wireless communication and CAN / 485 / Ethernet wired communication bus to send commands, report data, receive commands, transmit information, voice interpretation, networking, etc., and interfaces are reserved to accommodate new communication modules and other communication modules in the future.

[0074] The specific implementation process of the present invention is:

[0075] First, the injured part of the patient is placed on the platform, the injured part is photographed by the CT on the platform and the CT image is uploaded to the database, the laser projection line is projected onto the platform based on the bone data and repair data in the database, and the data in the database is used for calculation, and the probe position straight line about the fracture part in the database is converted into the projection straight line received by the current projection module. The projection module projects on the platform according to the above-mentioned converted coordinates, and the instrument user fixes the fracture with reference to the projected laser straight line.

[0076] The present invention provides a visual intelligent management system for closed reduction of fractures, including a main control module for receiving images at the input end of a system, a cloud database and a local database, a probe, and pressure sensors and gyroscopes provided on the probe, a projection module and other structures that cooperate with each other to reduce the number of fluoroscopy times, improve the accuracy and efficiency of fracture reduction, and reduce medical and patient injuries. A magnetic locator is used to collect spatial data of broken bones in real time and a three-dimensional fracture model rendered in a computer is moved in real time, so that the position of the broken bones can be monitored in real time, and doctors can be assisted in closed reduction of fractures to improve reduction accuracy and efficiency. The fracture data and probe data in the database are called to match the current fracture injury site, so that repair data with a suitable matching degree is called to provide a treatment reference plan for the current injured site. Based on the bone data and repair data in the database, a laser projection line is projected onto a platform, and a mold of the fracture site can also be used to provide a guide position for the current probe insertion position. The instrument user fixes the fracture site based on the reference of the projected laser straight line to optimize the insertion mode of the probe.

[0077] Of course, the present invention may have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art may make various corresponding changes and modifications based on the present invention, but these corresponding changes and modifications should all fall within the scope of protection of the claims attached to the present invention.

Claims

1. A visual intelligent management system for closed reduction of fractures, characterized in that: include: A main control module, used for receiving images from an input end of the system, the main control module comprising a computing unit for computing data and programs, a storage unit for storing data, and a communication unit for transmitting data; A database, including a cloud database and a local database, wherein the cloud database is connected to the local database and is used to store image data; The execution module comprises a probe, on which a pressure sensor and a gyroscope are arranged, a pressure sensor is arranged on the top of one end of the probe, and a gyroscope is arranged inside the probe; A projection module, comprising a platform, to which a laser projection assembly is connected, and the laser projection assembly is used to project a corresponding probe insertion auxiliary line on the platform; The steps for using the visual intelligent management system are as follows: S1, establishing a local and cloud database, calculating and storing CT data of the CT image of the initial fracture state and the CT image of the repaired bone fracture, wherein the CT data includes bone data and repair data; S2, taking real-time CT anteroposterior and lateral views of the injured part on the platform, and uploading the taken anteroposterior and lateral views to the main control module for matching; S3, calculating the insertion slope of the probe according to the matched CT image, and projecting the laser onto the injured part using a projection module; S4, inserting the probe into the injured part along the laser projection, starting the gyroscope in the probe and recording the initial position of the gyroscope, and calculating the bone reduction at the injured part according to the initial position of the probe and the three-dimensional model of the fracture site and the movement of the probe position; Extract the straight lines on both sides of the bone from the CT image of the initial fracture state, calculate the vertical distance between the straight lines on both sides according to the straight lines on both sides of the bone, the vertical distance s1 between the straight lines on both sides, and the distance between the straight lines on both sides and the matching features constitute Plane area, calculate The number of pixels or the minimum envelope rectangle of the black area in the plane area. The broken part is regarded as a rectangle. The break distance of the broken part is , the number of pixels corresponding to the fracture distance is , the total number of pixels in the broken part is ,pass , calculate the number of pixels corresponding to the fracture distance, , and thus the fracture distance is calculated ; The ratio of the actual distance to the pixel distance is set to k, ; Set fracture level If the fracture level , using the projection module to project the corresponding projection straight line to the injured part; If the fracture level , after manual judgment, the projection straight line can be manually adjusted and projected again; Fracture grade or fracture grade , it is judged that the current bone fracture state does not meet the judgment standard of the fracture closed reduction visual intelligent management system, and manual intervention is required to judge and repair.

2. The visual intelligent management system for closed reduction of fractures according to claim 1, characterized in that: The cloud database and the local database collect several CT images of different bones, and establish an inj_img data set and an inp_img data set with the collected CT images, wherein the inj_img data set is the CT image images of the initial fracture state of different bones, and the inp_img data set is the repaired CT image images of the corresponding bone fractures, that is, the CT images taken after the bone fractures are repaired using a probe or a steel nail. The main control module is used to expand and edge process the images in the inj_img data set and the inp_img data set to generate corresponding bone data and repair data.

3. The visual intelligent management system for closed reduction of fractures according to claim 1, characterized in that: The inp_img dataset is manually annotated for the probe positions in the corresponding repaired CT images of bone fractures, and the probe straight line ki_probe is generated by least squares fitting using the annotated probe position area. The straight line parallel to the backbone is set as the 0° position line, the straight line perpendicular to the backbone is set as the 90° position line, the 90° position line pointing outside the human body is set as the +90° position line, and the position line pointing inside the human body is set as the -90° position line, and the inclination angles ai_probe of several probe straight lines relative to the fracture position are obtained.

4. According to the visual intelligent management system for closed reduction of fractures according to claim 1, it is characterized in that: Assume the position coordinates of the first feature point of the skeleton are , the position coordinates of the second feature point of the skeleton are , the upper left corner of the image is defined as the coordinate origin , the initial matrix is ​​formed by the vectors of the feature points and the origin , ， Assume that the position coordinates of the first feature point of the broken bone in the current CT image are , the position coordinates of the second feature point of the skeleton are , the upper left corner of the image is defined as the coordinate origin , the initial matrix is ​​formed by the feature points of the current CT image and the vector of the origin , ， Let the transformation matrix , calculated by the transformation matrix , ， Suppose the intersection point between the probe and the backbone edge of the image in the database is , let the position of the probe corresponding to the current image be , ， The linear coordinates of the probe position of the fracture site in the database are converted into projection linear coordinates received by the projection module, and the projection module projects the projection linear coordinates.

Images