Wound measurement system and method based on depth point cloud image calculation

Abstract

The invention relates to a wound measurement system and method based on depth point cloud image calculation, and the method comprises the steps: carrying out the three-dimensional scanning of a wound part of a patient through a three-dimensional ultrasonic system, carrying out the image analysis and wound edge contour processing, and obtaining the three-dimensional data of a wound; and based on a three-dimensional reconstruction technology, constructing a wound model and performing three-dimensional data solution on the wound model. A nurse requests and checks the wound measurement data of the patient included in the solved wound model through the nursing end, intelligent and automatic measurement can be achieved, the length, width and depth of the wound are obtained, and comprehensive assessment of the wound is ensured. By constructing the three-dimensional ultrasonic wound model of the wound part, the wound shape and three-dimensional parameters of the patient can be visually and visually displayed, the recovery degree of the wound of the patient can be visually evaluated from the three-dimensional structure of the wound, manual measurement work of measurement personnel on the wound can be completely liberated, and the automatic measurement level is achieved.

Description

Technical Field

[0001] The present disclosure relates to the field of medical measurement technology, and in particular to a wound measurement system based on depth point cloud image calculation, an application method thereof, and an electronic device. Background Art

[0002] The recovery of wound patients needs to be evaluated through wound measurement data. Currently, most hospitals use manual measurement of wound diameter, length, depth, wound size, etc.

[0003] In the prior art, there are also some patents on wound area measurement tools and clinical nursing tools, such as CN201710107921.1 and CN201710150890.8. There are also patents based on infrared wound surface imaging, such as CN201810397230.4, which uses infrared to perform three-dimensional imaging of the wound surface and then measure the wound on the imaging. However, these prior arts have the following technical defects:

[0004] The existing manual measurement methods of wound depth, diameter, etc. in hospitals, as well as the measuring tools mentioned in patents such as CN201710107921.1 and CN201710107921.8, are all measuring rulers that require manual touching of the wound. Medical staff are prone to touching the wound during the measurement process, causing the wound to be injured again and infected. And due to the irregular shape of the wound and the presence of a certain degree of ulceration, simple tools cannot meet the accuracy requirements. It is difficult for medical staff to accurately measure the surface contour area of ​​the wound and the exact ulceration depth of the wound, and there is great difficulty in measurement.

[0005] The method mentioned in the patent based on infrared sore surface imaging invented by patent CN201810397230.4 uses a depth camera and an infrared laser to perform three-dimensional imaging of the wound, and then measures it on the three-dimensional imaging. It is indeed more convenient and simpler than traditional measurement tools. However, this method does not involve the process of removing the background before three-dimensional imaging. Direct imaging will inevitably bring in part of the background, which will inevitably affect the three-dimensional imaging. Then, due to the irregularity and complexity of the wound contour, it will also bring certain difficulties for users to measure the wound area, volume, etc. on 3D imaging. The user needs to manually draw the contour before accurately calculating the surface area and volume of the wound. The operation method is still cumbersome, and the measurement process will also cause some measurement errors to a certain extent.

[0006] In addition, traditional technology cannot visually evaluate the degree of recovery of a patient's wound based on the three-dimensional structure of the wound. The measurement technology is not smart enough and difficult to implement. It cannot completely free measurement personnel from the manual measurement of the wound and cannot reach the level of automated measurement. Summary of the invention

[0007] In order to solve the above problems, the present application proposes a wound measurement system based on depth point cloud image calculation, an application method and an electronic device thereof.

[0008] In one aspect, the present application provides a wound measurement system based on depth point cloud image calculation, the system comprising:

[0009] A three-dimensional ultrasound system is used to perform a three-dimensional scan of the wound area of ​​the wound patient, obtain the corresponding three-dimensional scanning data of the wound and send it to the backend server;

[0010] The backend server is used to perform image analysis and wound edge contour processing on the three-dimensional scanning data of the wound to obtain three-dimensional data of the wound; based on the three-dimensional reconstruction technology, the three-dimensional data of the wound is used to construct a wound model that matches the wound, and three-dimensional data solution is performed on the wound model;

[0011] The nursing end is used to request and view the solved wound model from the background server, and obtain the patient's wound measurement data from the solved wound model;

[0012] The three-dimensional ultrasound system and the nursing end are respectively connected to the background server for communication.

[0013] As an optional implementation scheme of the present application, optionally, the background server includes:

[0014] processor;

[0015] An image analysis module, used for analyzing the three-dimensional scanning data of the wound to obtain a slice image set consisting of a plurality of two-dimensional slice images of the wound;

[0016] An edge contour detection module is used to traverse and detect the wound edge contour of each two-dimensional slice in the slice image set based on a preset edge contour recognition model, and mark the wound edge contour image on each two-dimensional slice to obtain three-dimensional data of the wound composed of a plurality of the wound edge contour images;

[0017] A three-dimensional modeling module, used to perform three-dimensional reconstruction on the three-dimensional data of the wound, construct a three-dimensional model of the corresponding wound, and obtain a wound model matching the wound;

[0018] Mysql database, used to store the slice image set of the wound patient and the model file corresponding to the wound model, and bound to the wound patient's consultation ID;

[0019] The image analysis module, edge contour detection module, three-dimensional modeling module and Mysql database are respectively connected to the processor for communication.

[0020] As an optional implementation scheme of the present application, optionally, the background server further includes:

[0021] A data port is used for data transmission, including: receiving the three-dimensional scanning data of the patient's wound collected by the three-dimensional ultrasound system;

[0022] A preprocessing module, used for preprocessing the three-dimensional scanning data of the wound;

[0023] A digital conversion module, used for converting the three-dimensional scanning data of the wound into digital signals;

[0024] The data port, preprocessing module and digital conversion module are respectively connected to the processor for communication.

[0025] As an optional implementation scheme of the present application, optionally, the background server further includes:

[0026] A parameter-solving module, used for solving model parameters of the wound model based on a voxel model method, including the length, width, depth, area and / or volume of the wound model;

[0027] The parameter seeking module is communicatively connected to the processor.

[0028] As an optional implementation scheme of the present application, optionally, the background server further includes:

[0029] A visualization processing module, used for performing visualization processing and rendering on the wound model to generate a visualization model of the wound model;

[0030] The visualization processing module is communicatively connected to the processor.

[0031] As an optional implementation scheme of the present application, optionally, the background server further includes:

[0032] A feature recognition module, used to randomly recognize wound feature points or feature surfaces on a plurality of wound edge contour images based on a convolutional neural network, and extract the corresponding wound feature points or feature surfaces;

[0033] A wound detection module, used to detect the wound feature points or feature surfaces on the wound model, determine whether the wound model matches the entity features in the three-dimensional data of the wound, and output a corresponding determination result;

[0034] The feature recognition module and the wound detection module are respectively connected to the processor for communication.

[0035] On the other hand, the present application proposes an application method of a wound measurement system based on deep point cloud image calculation, comprising the following steps:

[0036] S1. Collect three-dimensional scanning data of the wound of the patient through a three-dimensional ultrasound system and send it to a backend server;

[0037] S2. On the background server, the three-dimensional scanning data of the wound is subjected to image analysis and wound edge contour processing to obtain three-dimensional data of the wound; based on three-dimensional reconstruction technology, the three-dimensional data of the wound is used to construct a wound model matching the wound, and three-dimensional data of the wound model is solved;

[0038] S3. The nurse requests and checks the solved wound model from the background server through the nursing terminal, and obtains the wound measurement data of the wound patient from the solved wound model.

[0039] In another aspect, the present application further provides an electronic device, comprising:

[0040] processor;

[0041] a memory for storing processor-executable instructions;

[0042] Wherein, the processor is configured to implement the above-mentioned application method when executing the executable instructions.

[0043] Technical effects of the present invention:

[0044] Based on the implementation of the present invention, the present application performs a three-dimensional scan of the patient's wound area through a three-dimensional ultrasound system, obtains the corresponding three-dimensional scanning data of the wound and sends it to the background server; the background server performs image analysis and wound edge contour processing on the three-dimensional scanning data of the wound to obtain the three-dimensional data of the wound; based on the three-dimensional reconstruction technology, the three-dimensional data of the wound is used to construct a wound model, and the three-dimensional data of the wound model is solved. Allowing nurses to request and view the patient's wound measurement data contained in the solved wound model through the nursing end can realize intelligent and automated measurement, obtain the length, width and depth of the wound, and ensure a comprehensive assessment of the wound.

[0045] By constructing a three-dimensional ultrasonic wound model of the wound site, the patient's wound shape and three-dimensional parameters can be displayed visually and intuitively, and the degree of recovery of the patient's wound can be visually evaluated based on the three-dimensional structure of the wound. This can completely free the measurement personnel from the manual measurement of the wound and achieve the level of automated measurement.

[0046] Further features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments, features, and aspects of the disclosure and, together with the description, serve to explain the principles of the disclosure.

[0048] Figure 1 It is shown as a schematic diagram of the application system composition of the present invention;

[0049] Figure 2 A schematic diagram of the functional architecture of a backend server of the present invention is shown;

[0050] Figure 3 A functional architecture diagram of another backend server of the present invention is shown;

[0051] Figure 4 A functional architecture diagram of another backend server of the present invention is shown;

[0052] Figure 5 A medical ultrasound image schematic diagram of a wound (marked defect site) occurring on the body surface of the biceps brachii according to the present invention is shown;

[0053] Figure 6 A three-dimensional schematic diagram of a wound model of the present invention is shown;

[0054] Figure 7 Shown is a schematic diagram of a wound model of the present invention;

[0055] Figure 8 It is a schematic diagram showing the application of the electronic device of the present invention. DETAILED DESCRIPTION

[0056] Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0057] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0058] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can also be implemented without certain specific details. In some examples, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the main purpose of the present disclosure.

[0059] In this embodiment, the various hardware devices used can be the image acquisition devices / systems and their application devices commonly used in clinical practice. For example, the three-dimensional ultrasound system can use the clinical three-dimensional color ultrasound probe and the supporting background software. The background server can use the three-dimensional color ultrasound application server (the latter installs the corresponding three-dimensional color ultrasound processing software on the hospital background server), and the nursing end can be an EDA or PDA terminal device. For the communication between each device and the background server, please refer to the existing clinical communication mode.

[0060] Example 1

[0061] like Figure 1 As shown, in one aspect, the present application proposes a wound measurement system based on depth point cloud image calculation, the system comprising:

[0062] A three-dimensional ultrasound system is used to perform a three-dimensional scan of the patient's wound area, obtain the corresponding three-dimensional scanning data of the wound and send it to the backend server;

[0063] The backend server is used to perform image analysis and wound edge contour processing on the three-dimensional scanning data of the wound to obtain three-dimensional data of the wound; based on the three-dimensional reconstruction technology, the three-dimensional data of the wound is used to construct a wound model that matches the wound, and three-dimensional data solution is performed on the wound model;

[0064] The nursing end is used to request and view the solved wound model from the background server, and obtain wound measurement data of the wound patient from the solved wound model;

[0065] The three-dimensional ultrasound system and the nursing end are respectively connected to the background server for communication.

[0066] The present invention mainly utilizes three-dimensional ultrasound (3D ultrasound): 3D ultrasound is a technology that uses multiple two-dimensional images to construct a three-dimensional image of the surface of a real object. It uses computer software to combine multiple two-dimensional scan images into a three-dimensional image to provide more comprehensive and realistic anatomical structure information.

[0067] Principle of 3D ultrasound: 3D ultrasound uses high-frequency sound waves to pass through human tissues and generates 2D slice images from echo signals. Then, computer software constructs a 3D model based on these 2D slice images to present a real 3D structure.

[0068] For the attached Figure 5 The surface wound shown (on the cross section, the wound is displayed as a contour line) is scanned using three-dimensional ultrasound and a three-dimensional model of the corresponding wound is constructed, which enables nurses to understand the condition and parameters of the patient's wound more intuitively and vividly.

[0069] The three-dimensional ultrasound system can scan the wound area of ​​the patient to obtain the three-dimensional data of the wound area and transmit it to the background server. The three-dimensional data is processed on the background server to generate a corresponding three-dimensional model, and the three-dimensional reconstruction technology is used to reconstruct the wound model that matches the wound. The model is used to simulate the shape of the wound, so that the nurse can view the patient's wound model through the terminal device and intuitively understand its parameters.

[0070] Using the above method, the wound model generated by three-dimensional reconstruction technology can match the patient's wound. In this way, nurses can understand the patient's wound changes and corresponding size parameters, such as depth, width, length, volume, etc. through the wound model, and understand the wound morphology more intuitively, while avoiding psychological discomfort caused by contact and viewing.

[0071] like Figure 2 As shown, as an optional implementation scheme of the present application, optionally, the background server includes:

[0072] processor;

[0073] An image analysis module, used for analyzing the three-dimensional scanning data of the wound to obtain a slice image set consisting of a plurality of two-dimensional slice images of the wound;

[0074] An edge contour detection module is used to traverse and detect the wound edge contour of each two-dimensional slice in the slice image set based on a preset edge contour recognition model, and mark the wound edge contour image on each two-dimensional slice to obtain three-dimensional data of the wound composed of a plurality of the wound edge contour images;

[0075] A three-dimensional modeling module, used to perform three-dimensional reconstruction on the three-dimensional data of the wound, construct a three-dimensional model of the corresponding wound, and obtain a wound model matching the wound;

[0076] Mysql database, used to store the slice image set of the wound patient and the model file corresponding to the wound model, and bound to the wound patient's consultation ID;

[0077] The image analysis module, edge contour detection module, three-dimensional modeling module and Mysql database are respectively connected to the processor for communication.

[0078] The background processor is mainly a processing board such as GPU, which is used for image processing.

[0079] To construct a 3D ultrasound model, it is necessary to first process a large number of 2D images obtained from a 3D scanning sequence, cut and separate the corresponding parts, and perform 3D synthesis on the cut and separated 2D images. Using 3D modeling software, the 2D ultrasound images are converted into 3D models of the corresponding parts.

[0080] In this solution, the three-dimensional image of the wound can be extracted through image segmentation and edge feature detection methods to obtain the wound edge contour on each two-dimensional slice. Subsequently, the wound edge contours on the slices scanned at different angles are combined to perform two-dimensional image synthesis to construct a contour three-dimensional model of the wound site. The surface of the model is consistent with the body surface, so it is only necessary to construct a corresponding three-dimensional contour model based on the wound edge contour. The model surface can be pre-processed during subsequent three-dimensional modeling, and the generated three-dimensional contour can be closed to simulate the patient's body surface skin (such as Figure 6 contour model shown).

[0081] The specific steps for 3D model reconstruction can be referred to as follows:

[0082] 1. Preparation stage: First, make sure the patient is in a suitable position so that the ultrasound probe can clearly scan the wound. At the same time, the doctor needs to understand the patient's medical history, wound condition and other information in order to better collect data.

[0083] 2. Data collection: Use a 3D ultrasound probe to scan the wound. During the scanning process, the ultrasound probe emits ultrasound waves and receives the reflected signals, which are converted into digital data and stored in the computer. In order to obtain more comprehensive wound information, the doctor may need to scan from multiple angles and depths.

[0084] 3. Data processing: The collected data needs to go through a series of processing before it can be used to build a three-dimensional model (the principle of three-dimensional ultrasound: three-dimensional ultrasound uses high-frequency sound waves to pass through human tissues, and the echo signals generate two-dimensional slice images. Then, the computer software builds a three-dimensional model based on these two-dimensional slice images to present a real three-dimensional structure). This includes steps such as removing noise, smoothing data, and extracting features. Through these processes, the accuracy and reliability of the data can be ensured.

[0085] 4. Build a 3D model: After the data processing is completed, special software can be used to convert the 2D ultrasound image into a 3D model. This process involves complex image processing and computer graphics technology. By adjusting the parameters and details of the model, it can be made to more realistically reflect the morphology and structure of the wound.

[0086] Image segmentation and edge contour feature extraction: In this step, an improved fully convolutional neural network learning algorithm is used to segment the data set and extract the edge contour features of each part of the wound.

[0087] Accurate edge contour extraction is the key to subsequent 3D reconstruction.

[0088] 3D reconstruction of wound model: After extracting the features of each wound site, 3D modeling of each wound site begins. The constructed single wound site model (wound outline on each 2D slice) is synthesized to construct a complete 3D wound model.

[0089] like Figure 6 As shown, it is a schematic diagram of the three-dimensional structure of the constructed wound model. By synthesizing the wound contours on the two-dimensional slices obtained under different scanning angles of three-dimensional ultrasound, the corresponding wound model can be constructed and generated. The wound model is mainly generated by three-dimensional synthesis of the wound contours on different slices, so as to obtain the contour model of the wound. The upper part of the model is open, but when generating the three-dimensional model, the upper opening is smoothly transitioned (processed in the three-dimensional modeling software), and the model opening is closed to simulate the skin on the surface of the body, so as to obtain a complete model.

[0090] The constructed wound model can be matched with the patient's wound site.

[0091] In the three-dimensional reconstruction of the wound model, the three-dimensional reconstruction software used is not limited in this embodiment, for example, finite element simulation software can be used.

[0092] like Figure 3 As shown, as an optional implementation scheme of the present application, optionally, the background server further includes:

[0093] A data port is used for data transmission, including: receiving the three-dimensional scanning data of the wound of the wound patient collected by the three-dimensional ultrasound system;

[0094] A preprocessing module, used for preprocessing the three-dimensional scanning data of the wound;

[0095] A digital conversion module, used for converting the three-dimensional scanning data of the wound into digital signals;

[0096] The data port, preprocessing module and digital conversion module are respectively connected to the processor for communication.

[0097] The data port is the data communication interface of the backend server, which can receive data, such as a USB interface or a wireless antenna.

[0098] Preprocessing, such as noise removal, data smoothing, feature extraction, etc. Through these processes, the accuracy and reliability of the data can be ensured.

[0099] The digital conversion module is used to convert the ultrasonic probe's transmitted ultrasonic wave and received reflected signal into digital data and store it in the database of the backend server. It is mainly an A / D digital-to-analog conversion circuit module.

[0100] like Figure 4 As shown, the present invention also has the functions of model parameterization, visualization and detection.

[0101] As an optional implementation scheme of the present application, optionally, the background server further includes:

[0102] A parameter-solving module, used for solving model parameters of the wound model based on a voxel model method, including the length, width, depth, area and / or volume of the wound model;

[0103] The parameter seeking module is communicatively connected to the processor.

[0104] The constructed wound model is an irregular model, so the user can smooth the model in the 3D modeling software, and level and smooth the irregular areas on the outer side of the wound model to facilitate 3D measurement.

[0105] In order to facilitate measurement, the present invention combines the voxel model method to solve the model parameters of the wound model.

[0106] Voxel Model, also known as voxel model, is a method for representing discrete data in three-dimensional space. In three-dimensional modeling and image processing, a voxel is a small cubic unit in three-dimensional space, similar to a pixel in a two-dimensional image.

[0107] To use the voxel model method to solve the size of a 3D model, you first need to determine the voxel resolution of the model (i.e. the size of each voxel) and the boundaries of the model in voxel space. The following is a simplified description of the steps:

[0108] 1. Determine the voxel resolution (set by the user):

[0109] The voxel resolution determines the size of each voxel in 3D space. For example, if the voxel resolution is 1 cm, each voxel represents a 1 cm x 1 cm x 1 cm cube.

[0110] 2. Determine the model boundaries:

[0111] Determine the minimum and maximum coordinates of the 3D model in voxel space. This usually involves scanning or analyzing the model in space to find the extent of the voxels it occupies.

[0112] 3. Calculate the model size:

[0113] Once the bounding voxels of the model are determined, the size of the model can be found by calculating the difference between the largest and smallest coordinates. For example, if the smallest coordinate of the model on the X-axis is X_min and the largest coordinate is X_max, then the size of the model on the X-axis is X_max-X_min. Similarly, the size on the Y-axis and Z-axis can be calculated.

[0114] 4. Consider accuracy and error:

[0115] Since the voxel model is discrete, the calculated size may be limited by the voxel resolution. This means that the calculated size may not be completely accurate, but rather a close approximation of the actual size. If higher accuracy is required, consider using a smaller voxel resolution.

[0116] 5. Programming implementation (if using Python):

[0117] In Python, you can use libraries such as NumPy to work with three-dimensional arrays (similar to voxel space) and use simple math to calculate the dimensions of the model. Here is a simplified pseudocode example:

[0118] Python

[0119] Assume there is a 3D array (voxel space) and the bounding voxel index of the model,

[0120] voxel_space is a three-dimensional NumPy array representing the voxel space.

[0121] min_index is a tuple containing three elements, indicating the minimum index of the model in the voxel space.

[0122] max_index is a tuple containing three elements, indicating the maximum index of the model in the voxel space.

[0123] Calculate the model size,

[0124] voxel_size=1.0 assumes that each voxel is 1 cm in size

[0125] model_dimensions=(

[0126] (max_index[0]-min_index[0]+1)voxel_size, X-axis size,

[0127] (max_index[1]-min_index[1]+1)voxel_size, Y-axis size,

[0128] (max_index[2]-min_index[2]+1)voxel_size Z-axis size,

[0129] ),

[0130] print(f"Model size:{model_dimensions}").

[0131] In the above manner, for the calculation of length, depth or area, the corresponding parameters can be calculated by the index of the boundary element of the corresponding dimension. For example, the depth of the model can be determined by establishing an index between the highest point and the lowest point of the model to determine the depth of the wound.

[0132] The following is a method for evaluating wound area using the voxel model, which estimates the surface area based on the boundaries and adjacent relationships of the voxels and estimates the surface area of ​​the voxel-based three-dimensional model:

[0133] 1. Define voxels: First, you need a 3D grid where each grid cell represents a voxel. Voxels can be cubes or other shapes, but cubes are the most common.

[0134] 2. Mark boundary elements: Traverse all elements and mark those that have at least one face (or edge, corner) exposed outside the model. These elements will be used to calculate the surface area.

[0135] 3. Calculate the surface area of ​​the boundary elements: For each boundary element, calculate the area of ​​its exposed face. If the element is a cube, then the area of ​​each face is fixed (the square of the side length). However, you need to consider which faces are exposed.

[0136] 4. Accumulated surface area: The exposed surfaces of all boundary elements are accumulated to obtain the estimated surface area of ​​the entire model (in this embodiment, it is only necessary to mark each boundary element where the upper surface of the wound model is located and calculate the surface area).

[0137] You can also use libraries such as NumPy to process three-dimensional arrays (representing a grid of voxels) and use loops and conditional statements to mark bounding voxels and calculate surface areas.

[0138] In addition to the above methods, you can also use the following methods:

[0139] Solving the 3D parameters of an irregular 3D model usually involves multiple steps, which may include model acquisition, data preprocessing, feature extraction, parameter estimation, etc. The following is a general process and possible methods:

[0140] 1. Model acquisition (see the above wound model for details):

[0141] If the model exists physically, 3D scanning technology (such as laser scanning, structured light scanning, etc.) can be used to obtain its three-dimensional point cloud data.

[0142] If the model is digital (such as created in CAD software), it can be exported directly to a 3D file format (such as STL, OBJ, PLY, etc.).

[0143] 2. Data preprocessing:

[0144] Clean the data: remove noise, outliers, and non-model components.

[0145] Meshing: If the data is in the form of a point cloud, it may be necessary to convert it into a mesh form (such as a triangular mesh) for subsequent processing.

[0146] Normalize: Scale the model to a uniform scale or move the model to the coordinate origin.

[0147] 3. Feature extraction:

[0148] Geometric features: Extract basic geometric parameters of the model such as length, width, height, volume, and surface area.

[0149] Shape features: Use shape descriptors (such as shape context, shape distribution, etc.) to capture the shape characteristics of the model.

[0150] Topological features: If the model is in mesh form, its topological information such as the number of vertices, edges, and faces can be extracted.

[0151] 4. Parameter estimation:

[0152] If the geometric parameters of the model (such as size) need to be estimated, they can be calculated directly from the preprocessed data.

[0153] If more complex parameters need to be estimated (such as the pose and position of the model), more advanced methods may need to be used, such as the least squares method, the iterative closest point (ICP) algorithm, the RANSAC algorithm, etc.

[0154] 5. Implementation:

[0155] In Python, you can use multiple libraries to process 3D models and perform parameter estimation. For example, `numpy` and `scipy` can be used for numerical calculations and optimization, `meshpy` and `trimesh` can be used to process 3D mesh data, and `Open3D` and `PCL (Python-PCL)` can be used to process 3D point cloud data.

[0156] As an optional implementation scheme of the present application, optionally, the background server further includes:

[0157] A visualization processing module, used for performing visualization processing and rendering on the wound model to generate a visualization model of the wound model;

[0158] The visualization processing module is communicatively connected to the processor.

[0159] Visualization can be directly rendered in 3D modeling applications to generate corresponding visualization models:

[0160] After modeling is completed, make necessary optimizations and adjustments, such as cleaning unnecessary patches and points, adjusting textures and materials to reduce the file size and rendering time of the model. Make appropriate rendering settings to achieve the best display effect.

[0161] If data needs to be mapped into the three-dimensional model, adding the data to the model may include mapping the data onto coordinate axes in the three-dimensional space or representing the data in the form of a three-dimensional object.

[0162] Adjust the visual effects of the model as needed, including changing color, transparency, rotating and scaling the model, adding labels and annotations, etc. Ensure that the visualization results are readable and easy to understand.

[0163] Use a rendering engine to render the model as an image or animation, and add effects such as shadows, reflections, and anti-aliasing to improve visual quality. Output the rendered model as a picture, video, or interactive application to share or use with others.

[0164] The visualization processing module can use the visualization components in the 3D modeling application, and the user can complete the required visualization configuration.

[0165] As an optional implementation scheme of the present application, optionally, the background server further includes:

[0166] A feature recognition module, used to randomly recognize wound feature points or feature surfaces on a plurality of wound edge contour images based on a convolutional neural network, and extract the corresponding wound feature points or feature surfaces;

[0167] A wound detection module, used to detect the wound feature points or feature surfaces on the wound model, determine whether the wound model matches the entity features in the three-dimensional data of the wound, and output a corresponding determination result;

[0168] The feature recognition module and the wound detection module are respectively connected to the processor for communication.

[0169] like Figure 6As shown, although the wound model has been constructed, in order to determine whether the wound model matches the patient's wound, the contours on the wound slices collected by the previous scan can be identified and used for model contour detection on the wound model. If the wound model matches the patient's wound, the wound model and the wound contours displayed on the wound slices will overlap and match.

[0170] Therefore, the present application identifies the wound contour (feature points or feature contour lines or contour features) on several wound edge contour images and matches them with the wound model to determine whether the wound model matches the wound contour on the wound edge contour image. A match indicates consistency.

[0171] Specifically, several slices are randomly selected, and the contour features in the slices are identified and marked using a convolutional neural network to obtain the corresponding wound contour (refer to the principle of convolutional neural network).

[0172] The wound contour is moved along the outer side of the wound model. When the wound contour matches the outer side of the wound model at a certain position, it means that the contour surface of the wound model is consistent with the wound and has a high degree of matching with the patient's wound.

[0173] Random sampling is required here, with at least 3 tests.

[0174] Obviously, those skilled in the art should understand that the implementation of all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above-mentioned embodiments. Those skilled in the art can understand that the implementation of all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above-mentioned embodiments. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory (Flash Memory), a hard disk (Hard Disk Drive, abbreviated as: HDD) or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above-mentioned types of memory.

[0175] Example 2

[0176] Based on the implementation principle of Example 1, on the other hand, the present application proposes an application method of a wound measurement system based on depth point cloud image calculation, comprising the following steps:

[0177] S1. Collect three-dimensional scanning data of the wound of the wound patient through a three-dimensional ultrasound system and send it to a backend server;

[0178] S2. On the background server, the three-dimensional scanning data of the wound is subjected to image analysis and wound edge contour processing to obtain three-dimensional data of the wound; based on three-dimensional reconstruction technology, the three-dimensional data of the wound is used to construct a wound model matching the wound, and three-dimensional data of the wound model is solved;

[0179] S3. The nurse requests and checks the solved wound model from the background server through the nursing terminal, and obtains the wound measurement data of the wound patient from the solved wound model.

[0180] Please understand in conjunction with Example 1 for details.

[0181] The modules or steps of the present invention described above can be implemented by a general-purpose computing system, they can be concentrated on a single computing system, or distributed on a network composed of multiple computing systems, and optionally, they can be implemented by a program code executable by a computing system, so that they can be stored in a storage system and executed by the computing system, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0182] Example 3

[0183] like Figure 8 As shown, further, in another aspect, the present application also proposes an electronic device, including:

[0184] processor;

[0185] a memory for storing processor-executable instructions;

[0186] Wherein, the processor is configured to implement the application method described in Example 2 when executing the executable instructions.

[0187] The electronic device of the embodiment of the present disclosure includes a processor and a memory for storing instructions executable by the processor, wherein the processor is configured to implement the application method described in the above embodiment 2 when executing the executable instructions.

[0188] Here, it should be noted that the number of processors can be one or more. At the same time, the electronic device of the embodiment of the present disclosure may also include an input system and an output system. Among them, the processor, memory, input system and output system may be connected through a bus or in other ways, which are not specifically limited here.

[0189] The memory is a computer-readable storage medium that can be used to store software programs, computer executable programs, and various modules, such as the program or module corresponding to the application method of the embodiment of the present disclosure. The processor executes various functional applications and data processing of the electronic device by running the software programs or modules stored in the memory.

[0190] The input system can be used to receive input numbers or signals. The signal can be a key signal related to user settings and function control of the device / terminal / server. The output system can include display devices such as display screens.

[0191] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A wound measurement system based on deep point cloud image calculation, characterized in that: The system comprises: A three-dimensional ultrasound system is used to perform a three-dimensional scan of the patient's wound area, obtain the corresponding three-dimensional scanning data of the wound and send it to the backend server; The backend server is used to perform image analysis and wound edge contour processing on the three-dimensional scanning data of the wound to obtain three-dimensional data of the wound; based on the three-dimensional reconstruction technology, the three-dimensional data of the wound is used to construct a wound model that matches the wound, and three-dimensional data solution is performed on the wound model; The nursing end is used to request and view the solved wound model from the background server, and obtain the patient's wound measurement data from the solved wound model; The three-dimensional ultrasound system and the nursing end are respectively connected to the background server for communication.

2. The wound measurement system based on depth point cloud image calculation according to claim 1, characterized in that: The backend server comprises: processor; An image analysis module, used for analyzing the three-dimensional scanning data of the wound to obtain a slice image set consisting of a plurality of two-dimensional slice images of the wound; An edge contour detection module is used to traverse and detect the wound edge contour of each two-dimensional slice in the slice image set based on a preset edge contour recognition model, and mark the wound edge contour image on each two-dimensional slice to obtain three-dimensional data of the wound composed of a plurality of the wound edge contour images; A three-dimensional modeling module, used to perform three-dimensional reconstruction on the three-dimensional data of the wound, construct a three-dimensional model of the corresponding wound, and obtain a wound model matching the wound; Mysql database, used to store the slice image set of the wound patient and the model file corresponding to the wound model, and bound to the wound patient's consultation ID; The image analysis module, edge contour detection module, three-dimensional modeling module and Mysql database are respectively connected to the processor for communication.

3. The wound measurement system based on depth point cloud image calculation according to claim 2, characterized in that: The backend server further includes: A data port is used for data transmission, including: receiving the three-dimensional scanning data of the wound of the wound patient collected by the three-dimensional ultrasound system; A preprocessing module, used for preprocessing the three-dimensional scanning data of the wound; A digital conversion module, used for converting the three-dimensional scanning data of the wound into digital signals; The data port, preprocessing module and digital conversion module are respectively connected to the processor for communication.

4. The wound measurement system based on depth point cloud image calculation according to claim 2, characterized in that: The backend server further includes: A parameter-solving module, used for solving model parameters of the wound model based on a voxel model method, including the length, width, depth, area and / or volume of the wound model; The parameter seeking module is communicatively connected to the processor.

5. The wound measurement system based on depth point cloud image calculation according to claim 2, characterized in that: The backend server further includes: A visualization processing module, used for performing visualization processing and rendering on the wound model to generate a visualization model of the wound model; The visualization processing module is communicatively connected to the processor.

6. The wound measurement system based on depth point cloud image calculation according to claim 2, characterized in that: The backend server further includes: A feature recognition module, used to randomly recognize wound feature points or feature surfaces on a plurality of wound edge contour images based on a convolutional neural network, and extract the corresponding wound feature points or feature surfaces; A wound detection module, used to detect the wound feature points or feature surfaces on the wound model, determine whether the wound model matches the entity features in the three-dimensional data of the wound, and output a corresponding determination result; The feature recognition module and the wound detection module are respectively connected to the processor for communication.

7. An application method of the wound measurement system based on depth point cloud image calculation according to any one of claims 1 to 6, characterized in that: The steps include: S1. Collect three-dimensional scanning data of the wound of the patient through a three-dimensional ultrasound system and send it to a backend server; S2. On the background server, the three-dimensional scanning data of the wound is subjected to image analysis and wound edge contour processing to obtain three-dimensional data of the wound; based on three-dimensional reconstruction technology, the three-dimensional data of the wound is used to construct a wound model matching the wound, and three-dimensional data of the wound model is solved; S3. The nurse requests and checks the solved wound model from the background server through the nursing terminal, and obtains the wound measurement data of the wound patient from the solved wound model.

8. An electronic device, characterized in that include: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to implement the application method described in claim 7 when executing the executable instructions.

Images