Orthopedic case image reconstruction method and device

Abstract

The invention discloses a reconstruction method and device for an orthopedic case image. The method comprises the following steps: determining that a skeleton of the orthopedic case image meets a reconstruction requirement according to diagnosis requirement information; according to the diagnosis demand information, calling an edge detection algorithm to identify a bone tissue contour and a focus tissue contour of the orthopedic case image, and intercepting the image to obtain a tissue contour image; determining focus feature severity based on tissue feature parameters of bone tissues in the tissue contour image, and rendering the tissue contour image according to the focus feature severity to obtain a tissue rendering image; and obtaining a gray value parameter according to the tissue rendering image, and carrying out pixel enhancement processing on the tissue of the tissue rendering image to obtain an orthopedic reconstructed image. The lesion feature severity is determined based on the tissue feature parameters of the image, and the image is reconstructed according to the lesion feature severity, so that a doctor can intuitively determine the lesion condition of the patient according to the image color and brightness, and the diagnosis efficiency of the doctor is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric digital data processing, and in particular to a reconstruction method and device for orthopedic case images. BACKGROUND

[0002] In the research of orthopedic diagnosis and treatment, medical images are increasingly used. By collecting orthopedic case images of patients, doctors can be assisted in diagnosing diseases and conducting research.

[0003] In the prior art, there are several methods for assisting doctors in diagnosis and research through orthopedic case images: A medical image lesion region positioning method based on deep learning (patent number CN202411234634.3). The operation of positioning the image includes: obtaining a to-be-tested medical image; uniforming the format and size of the to-be-tested medical image to obtain a standardized image; denoising the standardized image to obtain a denoised image; converting the denoised image into an RBG image; inputting the RBG image into a trained lesion region positioning model to obtain a positioning result; the lesion region positioning model is constructed based on a U-Net network. The lesion region positioning model is used to position the lesions of the orthopedic image, so that the doctor can view the lesions in the image according to the positioning result.

[0004] An orthopedic image auxiliary detection system based on big data (patent number CN202510835293.3). The detection system includes: a fuzzy processing module: performing clear processing on the to-be-detected orthopedic image to obtain a clear orthopedic image; an abnormality analysis module: based on a big data training set, a classification-positioning cascade model is integrated, and the clear orthopedic image is subjected to abnormality determination, if the determination is abnormal, the abnormal region in the clear orthopedic image is positioned to generate an abnormal image block; a disease recognition module: based on a big data training orthopedic disease classification model, difficult diseases are screened in the training process, the abnormal image block is subjected to disease recognition through the orthopedic disease classification model, if it cannot be recognized, the abnormal image block is marked as a difficult image block, triggering difficult recognition; a difficult recognition module: if difficult recognition is triggered, the abnormal image block is matched with similar images in the constructed structured difficult image library as auxiliary diagnosis basis. The classification-positioning cascade model is used to recognize the disease of the image block, and the doctor is provided with auxiliary diagnosis according to the recognition result.

[0005] An image processing method, device and electronic equipment (patent number CN202111197929.4), the method comprises: acquiring a medical image to be processed; determining the target image processing logic corresponding to the medical image according to the medical image, wherein the target image processing logic is generated based on the operation characteristics of the target image processing object; based on the target image processing logic, the medical image is processed to obtain a processed medical image, wherein the processing process conforms to the operation characteristics, one or more steps in the existing target image processing logic are replaced by revising the image processing logic, and the processed medical image is processed based on the replaced target image processing logic, if the corresponding step in each target image processing logic is executed, the processed medical image and the result of the image processing object operation meet the preset accuracy matching condition, then the corresponding next step in the target image processing logic is executed. Wherein, the processing logic includes marking, enlarging and arranging the image according to the doctor's personal habit. Through the above-mentioned logic processing, the doctor's diagnosis is assisted.

[0006] An orthopedic image auxiliary detection method and system (patent number CN202410666190.4). The auxiliary detection method comprises: acquiring the bone image of a patient; according to the gray histogram of the bone image, the bone image is segmented at several different gray levels to obtain an initial bone region; the bone density of each pixel point in the initial bone region is obtained; the gray value sequence and the bone density sequence of the local region of each pixel point in the initial bone region are obtained; according to the correlation coefficient and the standard correlation coefficient between the gray value sequence and the bone density sequence of the local region, the possibility of each pixel point in the initial bone region belonging to the bone region is obtained; the standard correlation coefficient is used to describe the average correlation between the gray value and the bone density after the bone image is segmented at all gray levels; according to the possibility of each pixel point in the initial bone region belonging to the bone region, the bone region of the bone image is obtained. The bone region of the image is determined by recognizing the bone density of the pixel point, so that the doctor can diagnose according to the detected bone region.

[0007] A method for processing distal radius fracture images based on Faster-RCNN (patent number CN202011387794.3). This method includes acquiring X-ray images of the distal radius fracture to be processed; preprocessing the acquired X-ray images; annotating the distal radius fracture region using the LabelImg annotation tool; selecting Faster-RCNN as the image segmentation tool, and improving the original Faster-RCNN feature extraction layer by replacing the original VGG-16 feature extraction network with the deeper ResNet101, which has better feature extraction performance; performing target detection on the preprocessed distal radius fracture image using Faster-RCNN to obtain the region of interest (ROI) of the distal radius fracture lesion; outputting the coordinates of the ROI obtained in the above steps, and using the output coordinates to complete the ROI segmentation task. The model segments the image to extract the lesion region for physician review.

[0008] While existing technologies can locate lesions in images and extract lesion areas for doctors to view using various models, image processing methods, and image detection methods, fracture lesions often appear as very fine transparent or dense lines in images. During diagnosis, doctors often need to review images repeatedly to determine the actual condition of the lesion. This is especially true for less experienced doctors, who may even need to consult the diagnostic report in conjunction with the images to confirm the fracture diagnosis. This not only increases the time required for diagnosis but also affects the efficiency of the diagnostic process. Summary of the Invention

[0009] This invention provides a method and apparatus for reconstructing orthopedic case images, which can shorten diagnostic time and improve diagnostic efficiency by reconstructing images.

[0010] A first aspect of this invention provides a method for reconstructing orthopedic case images, the method comprising: The orthopedic case images and diagnostic requirement information are acquired separately. After determining that the skeleton of the orthopedic case image meets the reconstruction requirements based on the skeletal keywords in the diagnostic requirement information, the format of the orthopedic case image is adjusted according to the diagnostic requirement information to obtain a format-adjusted image. Based on the bone keywords in the diagnostic requirements information, the edge detection algorithm is invoked to identify the bone tissue contour and lesion tissue contour of the format-adjusted image, and the image corresponding to the contour is cropped to obtain the tissue contour image; Tissue feature parameters of bone tissue are extracted from the tissue contour image. The severity of the lesion is determined based on the values ​​of the tissue feature parameters. The contour color value and contour brightness value are determined according to the severity of the lesion. The edges of the tissue contour image are rendered using the contour color value and the contour brightness value to obtain a tissue rendering image. The tissue feature parameters include the fracture length, number of fractures, and fracture displacement angle of the bone tissue. Grayscale parameters are obtained from the tissue rendering image, and pixel enhancement processing is performed on the tissue in the tissue rendering image according to the grayscale parameters to obtain an orthopedic reconstruction image. The grayscale parameters include the grayscale values ​​of the lesion area within the treatment tissue and the grayscale values ​​of the adjacent areas of the lesion area.

[0011] A second aspect of the present invention provides an apparatus for reconstructing orthopedic case images, the apparatus comprising: The acquisition module is used to acquire orthopedic case images and diagnostic requirement information respectively, and after determining that the skeleton of the orthopedic case image meets the reconstruction requirements based on the skeletal keywords in the diagnostic requirement information, the orthopedic case image is formatted according to the diagnostic requirement information to obtain a formatted image. The recognition module is used to call the edge detection algorithm to recognize the bone tissue contour and lesion tissue contour of the format-adjusted image based on the bone keywords of the diagnostic requirement information, and to extract the image corresponding to the contour to obtain the tissue contour image. The rendering module is used to extract tissue feature parameters of bone tissue from the tissue contour image, determine the severity of lesion tissue based on the value of the tissue feature parameters, determine contour color value and contour brightness value according to the severity of lesion tissue, and render the edge of the tissue contour image using the contour color value and contour brightness value to obtain a tissue rendering image. The tissue feature parameters include fracture length, number of fractures and fracture displacement angle of bone tissue. An enhancement module is used to obtain grayscale value parameters from the tissue rendering image, and perform pixel enhancement processing on the tissue in the tissue rendering image according to the grayscale value parameters to obtain an orthopedic reconstruction image. The grayscale value parameters include the grayscale value of the lesion area within the treatment tissue and the grayscale value of the adjacent area of ​​the lesion area.

[0012] A third aspect of the present invention 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 executes the computer program to implement the method for reconstructing orthopedic case images as described above.

[0013] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer-executable program for causing a computer to perform the orthopedic case image reconstruction method as described above.

[0014] A fifth aspect of the present invention provides a computer program product, including a computer program or instructions, characterized in that, when the computer program or instructions are executed by a device, they implement the method for reconstructing orthopedic case images as described above.

[0015] Compared with existing technologies, the reconstruction method and apparatus for orthopedic case images provided in this invention have the following advantages: This invention can acquire orthopedic case images and diagnostic requirement information. After determining that the skeleton in the orthopedic case image meets the reconstruction requirements based on the skeletal keywords in the diagnostic requirement information, it identifies the skeletal tissue contour and lesion tissue contour of the orthopedic case image according to the diagnostic requirement information, and extracts the image corresponding to the contour to obtain a tissue contour image. It determines the severity of the lesion tissue based on the values ​​of the tissue feature parameters of the tissue contour image, and determines the contour color value and contour brightness value based on the severity of the lesion tissue. It renders the edge of the tissue contour image using the contour color value and contour brightness value to obtain a tissue rendering image. It obtains grayscale value parameters from the tissue rendering image, and performs pixel enhancement processing on the tissue of the tissue rendering image based on the grayscale value parameters to obtain the orthopedic reconstructed image. This invention can extract lesions and bone tissue from orthopedic case images, determine the severity of the lesions based on the size of the tissue's characteristic parameters, and then render the tissue outline according to the color and brightness corresponding to the severity of the lesions. The rendered image is then pixel-enhanced to highlight the lesions and bones, allowing doctors to quickly and intuitively determine the actual condition of the lesions based on the color and brightness of the reconstructed image. This eliminates the need for repeated image reviews and reliance on test reports for diagnosis, effectively improving the efficiency of doctors in diagnosing and studying lesions. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a method for reconstructing orthopedic case images according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of an orthopedic case image reconstruction device provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an orthopedic case image reconstruction device provided in another embodiment of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] To address the aforementioned issues, the following specific embodiments will provide a detailed description and explanation of an orthopedic case image reconstruction method and apparatus provided in this application.

[0019] To address the technical problem of noise in existing orthopedic case images, resulting in blurred lesion areas, this paper refers to... Figure 1 The diagram shows a flowchart of a method for reconstructing orthopedic case images according to an embodiment of the present invention.

[0020] In one embodiment, the orthopedic case image reconstruction method is applicable to a hospital's data management platform. After acquiring the patient's orthopedic case image, the data management platform reconstructs and enhances it to highlight the lesion area, making the lesion area clearly visible in the image. This facilitates diagnosis by doctors based on the reconstructed image and improves the efficiency of doctors' diagnosis and research.

[0021] As an example, the method for reconstructing the orthopedic case images may include: S11. Acquire orthopedic case images and diagnostic requirement information respectively. After determining that the skeleton of the orthopedic case image meets the reconstruction requirements based on the skeletal keywords in the diagnostic requirement information, adjust the format of the orthopedic case image according to the diagnostic requirement information to obtain a format-adjusted image.

[0022] In one embodiment, images of the patient's orthopedic medical records can be acquired first, followed by information on the doctor's diagnostic needs.

[0023] The patient's orthopedic medical records images can be taken at the hospital where the doctor being consulted works, or they can be taken at another hospital. For example, after receiving an orthopedic medical record image and diagnosis at hospital A, the patient takes the image to hospital B to consult with another doctor. Specifically, the patient's orthopedic medical record image is an X-ray image.

[0024] Hospital B's data management platform can scan patients' orthopedic medical records to obtain orthopedic medical images. For example, a medical film scanner converts captured orthopedic films into digital images. This process is primarily based on optical capture, photoelectric conversion, and digital processing technologies. Its core steps and principles are as follows: The first step, optical capture: The film is illuminated by a light source (such as an LED or laser) from the scanner. The transmitted light is focused through a lens group to ensure that the image details on the film are accurately captured. This step requires adjusting the light source intensity and focus position to match the density range of the film.

[0025] The second step is photoelectric conversion: the focused light signal is received by a photosensitive element (such as CCD or CMOS) and converted into an analog electrical signal. Common photoelectric conversion elements such as CCD (charge-coupled device) or CMOS (complementary metal-oxide-semiconductor) can be used, and their sensitivity directly affects image quality.

[0026] The third step is analog-to-digital conversion (ADC): the analog electrical signal output by the photoelectric conversion element is quantized into discrete digital values ​​by the analog-to-digital converter (ADC). The brightness and color information of each pixel are encoded into binary data (such as RGB values). This process determines the dynamic range and color depth of the image.

[0027] The fourth step is image processing and output: the digitized data can be denoised, color corrected, and format converted by software to finally generate standard digital image files (such as JPEG and PNG). Some scanners support multi-frame synthesis or high dynamic range (HDR) processing to improve image quality.

[0028] In another optional embodiment, it can also be connected to a Hospital Information System (HIS) to obtain the patient's orthopedic medical records images. The Hospital Information System is an information platform for comprehensive hospital management, capable of storing various patient test information and extracting orthopedic medical records images from it.

[0029] The diagnostic needs of doctors can include diagnostic information input by the doctor consulting the patient, such as the need for image reconstruction and the doctor's personal requirements and habits for viewing images. For example, it may be necessary to extract the location and length of the fracture, the density of the bone, and the image must be in .jpg format for viewing. This could involve converting the bone in the image to a 9:16 ratio and highlighting or changing the colors of the image.

[0030] After acquiring orthopedic case images and diagnostic requirements, it is necessary to determine whether the skeleton in the orthopedic case images meets the reconstruction requirements. In one embodiment, the reconstruction requirement may be whether the skeleton in the orthopedic case image represents the patient's diseased bones. For example, if the orthopedic case image uploaded by the patient is incorrect, and an image of a non-diseased bone is uploaded to the data management platform. Patient A consulted about a femur fracture but mistakenly uploaded a lumbar spine image taken at hospital B to this hospital. Using an incorrect image for reconstruction would waste computing resources and delay subsequent diagnostic processing by doctors.

[0031] Alternatively, reconstruction requirements could include whether the clarity of the bones in the orthopedic case images meets the required clarity standards, whether the angle from which the orthopedic case images were taken is a preset shooting angle, or whether the time interval between the time the orthopedic case image was taken and the current time is within a preset time. This is because if the time interval is too long, the patient's affected bones may have changed, resulting in a significant discrepancy between the reconstructed and current images. Reconstruction requirements could also include image format, as the equipment used to take orthopedic case images at other hospitals may differ from that at our hospital, and the output image format may also differ.

[0032] In practice, the refactoring requirement can be one or more of the examples above, and can be adjusted according to actual needs.

[0033] In a preferred embodiment, the most important aspect is that the skeleton in the orthopedic case image is the same as the affected skeleton of the patient during the current consultation. If the bones in the image are different from the affected skeleton consulted by the patient, there is no reconstruction meaning. A preset orthopedic word segmentation model can be used to segment the diagnostic requirement information, and then bone keywords can be filtered from multiple segmented words according to a preset bone dictionary. Then, the corresponding bone template image is found according to the bone keywords of the diagnostic requirement information, and the similarity between the bone template image and the orthopedic case image is calculated. If the similarity between the two meets the threshold, the skeleton of the orthopedic case image is determined to meet the reconstruction requirements; otherwise, the reconstruction requirements are not met.

[0034] Alternatively, a skeletal classification model can be used to classify orthopedic case images to determine the category corresponding to the bones in the orthopedic case images. Then, it can be determined whether the keywords corresponding to the category are the same as the keywords of the bones. If they are the same, it is determined that the bones of the orthopedic case images meet the reconstruction requirements; otherwise, they do not meet the reconstruction requirements.

[0035] After determining that the skeleton in the orthopedic case image meets the reconstruction requirements, the diagnostic requirement information can be segmented using a preset image format word segmentation model or by calling a word segmentation algorithm. Then, format keywords are selected from multiple word segments according to a preset format dictionary. Finally, the orthopedic case image is format-converted according to the format keywords so that the orthopedic case image can meet the doctor's needs.

[0036] On the one hand, it allows orthopedic case images to be converted into an image format familiar to doctors; on the other hand, it ensures the images meet the hospital's data requirements. This is because orthopedic case images may have been taken at other hospitals, where equipment differs, potentially leading to discrepancies. Format adjustment allows the images to meet the hospital's requirements and facilitates storage and processing by the data management platform. Furthermore, for fracture-related diseases, it's generally necessary to examine the morphology and direction of the fracture line, as well as the fracture type and degree of displacement. For example, X-ray anteroposterior and lateral views can determine the separation, angularity, or rotational displacement of the fracture ends, such as the typical "silver fork" deformity of Colles' fracture. Since image format conversion does not affect the morphology and direction of the fracture line, fracture type, or degree of displacement within the image, it maintains image integrity and avoids image deviations caused by conversion. This invention, through the aforementioned series of processes, can convert orthopedic case images into format-adjusted images.

[0037] S12. Based on the bone keywords of the diagnostic requirements information, call the edge detection algorithm to identify the bone tissue contour and lesion tissue contour of the format-adjusted image, and crop the image corresponding to the contour to obtain the tissue contour image.

[0038] After obtaining the formatted image, it is necessary to identify the lesion area in the formatted image, and then adjust and reconstruct the lesion area to highlight it.

[0039] In one operation mode, the Canny edge detection algorithm can be invoked based on the skeletal keywords of the diagnostic needs information to identify the contours of diseased bone tissue and lesion tissue in orthopedic case images.

[0040] Since the outline of skeletal tissue is generally clear, the Canny edge detection algorithm can be directly used to identify the outline of diseased skeletal tissue in orthopedic case images. However, the outline of lesion tissue is generally blurry. Therefore, a preset lesion localization model can be used first to locate the lesion tissue. After determining the location of the lesion tissue, the Canny edge detection algorithm can then be used to identify the outline of diseased lesion tissue in orthopedic case images.

[0041] It should be noted that the lesion tissue being located specifically refers to bone fracture lines, bone spurs, deformed or displaced bones, etc.

[0042] In one embodiment, using an orthopedic case image of the radius as an example, the Canny edge detection algorithm can be used to process the orthopedic case image to identify the contour boundary of the radius. Specifically, this involves calculating the grayscale gradient magnitude of each pixel in the orthopedic case image, filtering pixels with grayscale gradient magnitudes greater than a preset threshold, combining these pixels, and then connecting adjacent boundary pixels to obtain the contour boundary of the radius. Next, the image coordinates of the contour boundary can be obtained, and the image can be cropped according to these coordinates to obtain the tissue contour image corresponding to the bone tissue contour.

[0043] For relatively blurry lesion contours, after determining the location of the lesion, the Canny edge detection algorithm can be used to process orthopedic case images to identify the lesion contour boundary. Specifically, this involves calculating the grayscale gradient amplitude of each pixel in the orthopedic case image, filtering pixels with grayscale gradient amplitudes below a preset threshold, combining these pixels, and then connecting them to obtain the lesion contour boundary. Next, the image coordinates of the lesion contour boundary can be obtained, and the image can be cropped according to these coordinates to obtain the tissue contour image corresponding to the lesion contour.

[0044] It should be noted that when connecting pixels, one can start from the pixel of each adjacent boundary and search for the edge points among its eight neighboring pixels, forming a continuous contour line through recursive traversal. For common skeletal contours in orthopedic images (such as the femur and tibia contours), connecting pixels can effectively separate the boundaries of different bones and avoid contour confusion.

[0045] Preferably, morphological closing operations can also be performed on the extracted outer contour boundary of the skeleton to fill small gaps in the contour and ensure the integrity of the outer contour of the skeleton.

[0046] In addition, as can be seen from the previous analysis, the image may contain different kinds of noise. In an optional embodiment, when performing edge detection on the orthopedic case image, the orthopedic case image is first processed by Gaussian filtering to reduce noise interference.

[0047] S13. Extract tissue feature parameters of bone tissue from the tissue contour image, determine the severity of lesion tissue based on the value of the tissue feature parameters, and determine contour color value and contour brightness value according to the severity of lesion tissue. Render the edge of the tissue contour image using the contour color value and the contour brightness value to obtain a tissue rendering image. The tissue feature parameters include the fracture length, number of fractures, and fracture displacement angle of the bone tissue.

[0048] After acquiring tissue contour images, these images can be analyzed to extract parameters of bone tissue and lesion tissue, yielding tissue feature parameters. These tissue feature parameters are essential for physicians reviewing orthopedic case images. For example, in cases of fractures, each orthopedic surgeon must observe bone structure and morphology, bone curvature angle, cortical continuity, bone density changes, fracture line length, number of fracture lines, and number of bone fragments during imaging diagnosis.

[0049] To better meet the needs of attending physicians, parameter requirements can be extracted from diagnostic needs information, and then several parameters can be extracted based on these parameter requirements to obtain tissue characteristic parameters.

[0050] Next, the numerical values ​​of the tissue feature parameters can be determined, the severity of the lesion feature can be assessed based on the numerical values, and the edges of the tissue contour image can be rendered based on the severity of the lesion feature to obtain a tissue rendering image.

[0051] After outlining the tissue contour image, rendering the edges of the tissue contour image can further highlight the bones and lesions in the orthopedic case image, thereby improving the clarity of the bones and lesions in the image and making it easier for doctors to view.

[0052] In fracture-related diseases, each doctor needs to examine the same information from the images. Therefore, the information each doctor needs to examine in fracture-related diseases can be predetermined, with each information item representing an information item. Combining multiple information items yields a pre-defined list of bone items. Each information item in the pre-defined list corresponds to a tissue feature parameter. Then, based on each item in the pre-defined list, the corresponding feature parameter value can be extracted from the tissue contour image to obtain the tissue feature parameters.

[0053] In one embodiment, tissue feature parameters may include the fracture length, number of fractures, and fracture displacement angle of the bone tissue. The severity of the user's lesion is determined based on the fracture length, number of fractures, and fracture displacement angle. Then, the contour color value and contour brightness value required for rendering the image are determined based on the severity of the lesion. Finally, the edges of the tissue contour image are rendered according to the contour color value and contour brightness value to obtain the tissue rendering image.

[0054] In one embodiment, to determine the severity of lesions within an image for each tissue feature parameter, as an example, the step of extracting tissue feature parameters of bone tissue from the tissue contour image, determining the severity of lesions based on the values ​​of the tissue feature parameters, determining contour color values ​​and contour brightness values ​​based on the severity of lesions, and rendering the edges of the tissue contour image using the contour color values ​​and contour brightness values ​​to obtain a tissue rendering image may include the following sub-steps: S131. Determine the comparison result between each of the tissue feature parameters and the corresponding preset first feature threshold.

[0055] S132. Determine the severity of the corresponding lesion tissue based on the three comparison results, and find the corresponding color value based on the severity of the lesion tissue to obtain the contour color value.

[0056] S133. Determine the contour brightness value corresponding to the contour color value based on the severity of each lesion tissue.

[0057] S134. Render the edges of the tissue contour of the tissue contour image according to the contour color value and the contour brightness value to obtain a tissue rendering image.

[0058] In one embodiment, a preset first feature threshold can be pre-set for each tissue feature parameter. Simultaneously, a corresponding color value can be set for the severity of each lesion feature.

[0059] In practice, each tissue feature parameter is compared with its corresponding preset first feature threshold to obtain the comparison result. Then, the severity of the lesion feature is determined based on the comparison result between the parameter and the threshold. Subsequently, the color value corresponding to the severity of the lesion feature is obtained, and the color value corresponding to the severity of the lesion feature is used as the contour color value.

[0060] Next, the contour brightness value corresponding to the contour color value can be determined based on the severity of the lesion. Finally, the edges of the tissue contour image are rendered based on the contour color value and contour brightness value to obtain the tissue rendering image.

[0061] Optionally, during rendering, you can render only the edges of the tissue outline or render the entire tissue; the specific method can be adjusted according to actual needs.

[0062] Color and brightness can be used to visually assess the severity of lesions, allowing doctors to understand the condition of lesions based on the color and brightness of the image, thereby improving the efficiency of diagnosis and assisting in diagnosis.

[0063] In an optional embodiment, different colors can be used to correspond to different lesion severity characteristics, and the specific colors can be adjusted according to actual needs.

[0064] Since different tissue feature parameters have different comparison results with the corresponding preset first feature threshold, in order to determine the severity of lesions based on different comparison results, as an example, the step of determining the corresponding severity of lesions based on the three comparison results and finding the corresponding color value based on the severity of lesions to obtain the contour color value may include the following sub-steps: S1321. If each comparison result is that the tissue feature parameter is less than or equal to the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be mild, and the first color value corresponding to the mild severity of the lesion feature is searched in the preset color list to obtain the outline color value.

[0065] S1322. If any of the comparison results is that the tissue feature parameter is greater than the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be moderate, and the second color value corresponding to the moderate severity of the lesion feature is searched in the preset color list to obtain the outline color value.

[0066] S1323. If any two of the comparison results are greater than the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be severe, and the third color value corresponding to the severity of the lesion feature is searched in the preset color list to obtain the outline color value.

[0067] S1324. If each comparison result shows that the tissue feature parameter is greater than the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be urgent, and the fourth color value corresponding to the severity of the lesion feature being urgent is searched in the preset color list to obtain the contour color value.

[0068] In one embodiment, the tissue characteristic parameters include the fracture length, number of fractures, and fracture displacement angle of the bone tissue. For each tissue characteristic parameter, a corresponding preset first characteristic threshold can be set, which can be adjusted according to actual needs.

[0069] Simultaneously, a corresponding color value can be assigned to the severity of each lesion feature, and a preset color value list can be constructed for each lesion feature severity to find the color value corresponding to the severity of the lesion feature from the preset color value list.

[0070] In a practical example, the preset first feature threshold for fracture length can be set to 5mm, the preset first feature threshold for the number of fractures to 3, and the preset first feature threshold for fracture displacement angle to 10 degrees. The first color value is green, the second color value is yellow, the third color value is orange, and the fourth color value is red. Taking RGB mode as an example, the color value for green is 0,255,0, and the color value for red is 255,0,0.

[0071] In practical applications, if the obtained fracture length is 3mm, the number of fractures is 2, the fracture displacement angle is 3 degrees, and all three tissue feature parameters are less than or equal to their corresponding preset first feature thresholds, then the severity of the lesion feature is determined to be mild, and the first color value corresponding to mild is found in the preset color list, and the green color value is obtained as the outline color value.

[0072] If the obtained fracture length is 8mm, the number of fractures is 1, the fracture displacement angle is 5 degrees, and two of the three tissue feature parameters are less than or equal to their corresponding preset first feature thresholds, while one tissue feature parameter is greater than its corresponding preset first feature threshold, then the severity of the lesion is determined to be moderate. The second color value corresponding to moderate is then searched in the preset color list, and the outline color value is set to yellow. For example, if the obtained fracture length is 2mm, the number of fractures is 2, the fracture displacement angle is 15 degrees, and two of the three tissue feature parameters are less than or equal to their corresponding preset first feature thresholds, while one tissue feature parameter is greater than its corresponding preset first feature threshold, the severity of the lesion can also be determined to be moderate, and the outline color value is set to yellow.

[0073] Similarly, if the obtained fracture length is 4mm, the number of fractures is 5, the fracture displacement angle is 15 degrees, and one of the three tissue feature parameters is less than or equal to its corresponding preset first feature threshold, while the other two tissue feature parameters are greater than their corresponding preset first feature thresholds, then the severity of the lesion feature is determined to be severe, and the third color value corresponding to the moderate severity is searched in the preset color list, with orange as the outline color value.

[0074] For example, if the obtained fracture length is 10mm, the number of fractures is 5, and the fracture displacement angle is 15 degrees, and all three tissue feature parameters are greater than the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be urgent, and the fourth color value corresponding to the moderate severity is searched in the preset color list, with red as the outline color value.

[0075] After determining the contour color value corresponding to the severity of each lesion, in order to assign a specific brightness to each color value and further highlight the severity of the lesion through brightness, the step of determining the contour brightness value corresponding to the contour color value according to each lesion severity may include the following sub-steps: S1331. If the severity of the lesion is mild or moderate, then the contour brightness value corresponding to the contour color value is determined to be a preset first brightness value.

[0076] S1332. If the severity of the lesion is severe or urgent, and all three tissue feature parameters are less than or equal to the corresponding preset second feature threshold, then the contour brightness value corresponding to the contour color value is determined to be the preset second brightness value.

[0077] S1333. If the severity of the lesion is severe or urgent, and any one of the tissue feature parameters is greater than the corresponding preset second feature threshold, then the contour brightness value corresponding to the contour color value is determined to be a preset third brightness value.

[0078] S1334. If the severity of the lesion is severe or urgent, and any two of the tissue feature parameters are greater than the corresponding preset second feature threshold, then the contour brightness value corresponding to the contour color value is determined to be a preset fourth brightness value.

[0079] S1335. If the severity of the lesion is severe or urgent, and the three tissue feature parameters are greater than the corresponding preset second feature threshold, then the contour brightness value corresponding to the contour color value is determined to be the preset fifth brightness value.

[0080] Wherein, the preset first brightness value is less than the preset second brightness value, the preset second brightness value is less than the preset third brightness value, the preset third brightness value is less than the preset fourth brightness value, and the preset fourth brightness value is less than the preset fifth brightness value; the preset first feature threshold is less than the preset second feature threshold.

[0081] In practice, if the severity of the lesion is mild or moderate, it indicates that the patient may have a mild fracture. The fracture can be distinguished by color. Therefore, the outline brightness value corresponding to the green and yellow outline color values ​​can be set as the preset first brightness value.

[0082] If the severity of the lesion is indicated as severe or urgent, it suggests the patient may have a severe fracture, which is quite serious. In this case, each tissue characteristic parameter can be compared with its corresponding preset second characteristic threshold, where the preset second characteristic threshold is greater than the preset first characteristic threshold.

[0083] If all three tissue feature parameters are less than or equal to the corresponding preset second feature threshold, then the outline brightness values ​​of orange and red are determined to be preset second brightness values.

[0084] If any organizational feature parameter is greater than the corresponding preset second feature threshold, then the outline brightness value of orange and red is determined to be the preset third brightness value.

[0085] If any two tissue feature parameters are greater than the corresponding preset second feature threshold, then the outline brightness values ​​of orange and red are determined to be preset fourth brightness values.

[0086] Similarly, if all three tissue feature parameters are greater than the corresponding preset second feature threshold, then the red outline brightness value is determined to be the preset fifth brightness value.

[0087] For example, using the above example, the preset first feature threshold for fracture length is set to 5mm, the preset first feature threshold for the number of fractures is 3, and the preset first feature threshold for fracture displacement angle is 10 degrees. The preset second feature threshold for fracture length is set to 15mm, the preset first feature threshold for the number of fractures is 5, and the preset first feature threshold for fracture displacement angle is 30 degrees.

[0088] In practical applications, if the obtained fracture length is 3mm, the number of fractures is 2, the fracture displacement angle is 3 degrees, and all three tissue feature parameters are less than or equal to their corresponding preset first feature threshold, then the severity of the lesion feature is determined to be mild, and the green contour brightness value is determined to be the preset first brightness value.

[0089] If the obtained fracture length is 3mm, the number of fractures is 2, the fracture displacement angle is 35 degrees, and one tissue feature parameter is greater than the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be moderate, the yellow color value is used as the outline color value, and the yellow outline brightness value is determined to be the preset first brightness value.

[0090] If the obtained fracture length is 8mm, the number of fractures is 2, the fracture displacement angle is 15 degrees, two tissue feature parameters are greater than their corresponding preset first feature thresholds, and all three tissue feature parameters are less than their corresponding preset second feature thresholds, then the severity of the lesion is determined to be severe, with orange as the outline color value, and the orange outline brightness value is determined to be the preset second brightness value.

[0091] Similarly, if the obtained fracture length is 8mm, the number of fractures is 4, the fracture displacement angle is 15 degrees, and three tissue feature parameters are greater than their corresponding preset first feature thresholds, and all three tissue feature parameters are less than their corresponding preset second feature thresholds, then the severity of the lesion feature is determined to be urgent, the red color value is used as the outline color value, and the red outline brightness value is determined as the preset second brightness value.

[0092] If the obtained fracture length is 8mm, the number of fractures is 2, the fracture displacement angle is 35 degrees, two tissue feature parameters are greater than their corresponding preset first feature threshold, and one tissue feature parameter is greater than its corresponding preset second feature threshold, then the orange outline brightness value is determined to be the preset third brightness value.

[0093] Similarly, if the obtained fracture length is 8mm, the number of fractures is 6, the fracture displacement angle is 35 degrees, three tissue feature parameters are greater than their corresponding preset first feature threshold, and one tissue feature parameter is greater than its corresponding preset second feature threshold, then the red outline brightness value is determined to be the preset third brightness value.

[0094] For example, if the obtained fracture length is 20mm, the number of fractures is 2, the fracture displacement angle is 35 degrees, and two tissue feature parameters are greater than their corresponding preset first feature threshold, and two tissue feature parameters are greater than their corresponding preset second feature threshold, then the orange outline brightness value is determined to be the preset fourth brightness value.

[0095] Similarly, if the obtained fracture length is 20mm, the number of fractures is 4, the fracture displacement angle is 35 degrees, three tissue feature parameters are greater than their corresponding preset first feature threshold, and two tissue feature parameters are greater than their corresponding preset second feature threshold, then the red outline brightness value is determined to be the preset fourth brightness value.

[0096] For example, if the obtained fracture length is 20mm, the number of fractures is 6, the fracture displacement angle is 35 degrees, and all three tissue feature parameters are greater than their corresponding preset first feature threshold and preset second feature threshold, then the severity of the lesion feature is determined to be urgent, and the brightness value of the red outline is determined to be the preset fifth brightness value.

[0097] In one operating mode, the more severe the condition, the higher the brightness value; conversely, the milder the condition, the lower the brightness value.

[0098] In one embodiment, the first to fifth brightness values ​​can be preset or adjusted according to the doctor's needs. The doctor can add the required brightness values ​​to the diagnostic requirement information, and then read the diagnostic requirement information to adjust the corresponding brightness values ​​accordingly.

[0099] During rendering, the edges of the tissue contour image can be rendered directly based on the contour brightness and contour color values, which can highlight the skeletal tissue and lesion tissue after rendering.

[0100] For subtle bone spurs or fracture lines, as well as the shape and displacement angle of bones, existing technologies use a light box to illuminate a slide for doctors to view. This application can directly adjust the brightness of the lesion outline in the image, thereby making the lesion and bone (including fracture lines or bone spurs) stand out in the image, facilitating doctor's viewing. Furthermore, by adjusting the brightness based on information input by the doctor, it can meet the doctor's viewing requirements. Because for fracture-related diseases, each doctor needs to view the same bone content in the image. For the bone corresponding to the lesion tissue, doctors mainly look at the fracture line, bone spurs, and bone shape. Even adjusting the brightness will not change these details or cause any deviation. Therefore, the brightness of the bone and lesion can be directly adjusted. Moreover, adjusting the brightness makes the bone and lesion stand out in the image, allowing doctors to view them more intuitively.

[0101] S14. Obtain grayscale value parameters from the tissue rendering image, and perform pixel enhancement processing on the tissue of the tissue rendering image according to the grayscale value parameters to obtain an orthopedic reconstruction image, wherein the grayscale value parameters include the grayscale value of the lesion area in the treatment tissue and the grayscale value of the adjacent area of ​​the lesion area.

[0102] After rendering the outline of the tissue (including the outline of the bones and lesions), although the bones and lesion tissue can be highlighted, the pixels in the middle of the lesion may still be blurry. In order to highlight the lesion, pixel enhancement processing can be performed on it.

[0103] In one embodiment, the enhancement process is mainly aimed at distinguishing the tissue from the surrounding area. To this end, grayscale parameters can be obtained from the tissue rendering image. The grayscale parameters include the grayscale values ​​of the lesion area within the treatment tissue and the grayscale values ​​of the adjacent areas of the lesion area. The tissue in the tissue rendering image is then pixel-enhanced based on the grayscale parameters to obtain an orthopedic reconstruction image.

[0104] Before enhancement, the enhancement area of ​​the image can be determined by identifying the coordinates of the entire region enclosed by the boundary of the tissue's outer contour. By combining multiple coordinate points, a complete set of tissue contour coordinates can be obtained. Then, based on the coordinates of this set, the entire tissue region, including bone tissue and lesion tissue, can be determined. Finally, pixel enhancement is performed on the pixels of these tissues to highlight the entire tissue region.

[0105] It should be noted that for non-skeletal lesions, such as joint soft tissue or inflamed tissue, pixel-by-pixel enhancement may result in the enhanced inflammatory tissue (such as myositis) or fibrotic lesions appearing similar to tumor tissue or calcifications, leading to confusion with other diseases. Therefore, pixel-by-pixel enhancement is not feasible. (See reference...) Figure 1 and 2It is known that for bone lesions, the bone spurs and fracture lines protruding at the fracture site (especially at the edge of the bone) are relatively blurry in X-ray images. Pixel enhancement of the lesion tissue in this area can not only improve image quality and highlight key details, making blurry or low-resolution images clearer, thus helping doctors to identify fractures more accurately; but also, by rendering and pixel enhancement, only the originally blurry lesion tissue (including fracture lines and bone) in the image is restored, and the restored tissue will not be confused with other diseases.

[0106] In highlighting the entire tissue region, to distinguish the tissue region from its surrounding areas, in one embodiment, obtaining grayscale parameters from the tissue rendering image and performing pixel enhancement processing on the tissue in the tissue rendering image based on the grayscale parameters to obtain an orthopedic reconstruction image may include the following sub-steps: S141. Obtain the grayscale value of the lesion area and the grayscale value of the adjacent area from the tissue rendering image.

[0107] S142. Determine the target gray value based on the comparison result between the gray values ​​of the adjacent regions and the gray values ​​of the lesion region, and use the target gray value to render the pixels of the tissue rendering image to obtain a pixel rendering image.

[0108] S143. Adjust the size of the pixel-rendered image to obtain the orthopedic reconstruction image.

[0109] In one operation, the grayscale values ​​of the lesion area and adjacent areas can be obtained from the tissue rendering image. It should be noted that for fracture-related diseases, the fracture lesion may be on the bone, or on the edge or side of the bone, for example... Figure 1 and 2 In the example image, the fracture is located at the edge of the bone. Therefore, in the image cropped earlier, the image corresponding to the lesion tissue and the image corresponding to the bone tissue do not overlap in the original orthopedic case image. If the fracture is within the bone, the image corresponding to the lesion tissue can be a portion of the image corresponding to the bone tissue. For example, if there is a fracture in the middle of the radius, after cropping the bone tissue image of the radius, the fracture lesion is then cropped from the middle of the bone tissue image of the radius to obtain the bone tissue image of the lesion tissue.

[0110] When performing pixel enhancement, if the two overlap, pixel enhancement can be performed on the lesion in the tissue rendering image corresponding to the bone tissue, or pixel enhancement can be performed on the lesion in the tissue rendering image corresponding to the lesion tissue.

[0111] If the two do not overlap, pixel enhancement can be performed only on the lesion in the tissue rendering image corresponding to the lesion tissue.

[0112] For example, in the case of the radial fracture mentioned above, in practice, the grayscale value of the central lesion area in the tissue rendering image corresponding to the bone tissue can be obtained, and then the grayscale value of the surrounding area can be obtained. For the tissue rendering image of the lesion tissue, the grayscale value of the lesion area along the central fracture line can be obtained, and then the grayscale value of the surrounding area can be obtained.

[0113] To differentiate between the grayscale parameters of the rendered image for bone tissue and the grayscale parameters of the rendered image for lesion tissue, the grayscale value of the central lesion region in the rendered image for bone tissue can be the average grayscale value of the entire region, while the grayscale value of the surrounding region is the average grayscale value of the surrounding region. However, for the grayscale parameters of the rendered image for lesion tissue, the grayscale value of the central lesion region can be the average grayscale value of any pixel along the fracture line, while the grayscale value of the surrounding region is the average grayscale value of several arbitrary pixels surrounding the fracture line.

[0114] After determining the grayscale parameters for each tissue rendering image, a target grayscale value can be calculated based on the comparison results between the grayscale values ​​of adjacent areas and the grayscale values ​​of the lesion area in the grayscale parameters.

[0115] The target grayscale value is then used to render the pixels of the tissue rendering image to obtain a pixel-rendered image. After rendering, to further enhance the image, the size of the pixel-rendered image can be adjusted to obtain an orthopedic reconstruction image.

[0116] To improve the image quality and identifiability of the lesion area, in an optional embodiment, determining a target gray value based on a comparison between the gray values ​​of the adjacent areas and the gray value of the lesion area, and then using the target gray value to render the pixels of the tissue rendering image to obtain a pixel-rendered image, may include the following sub-steps: S1421. If the difference between the gray value of the adjacent region and the gray value of the lesion region is greater than the gray value difference threshold and the gray value of the lesion region is less than the gray value threshold, then the gray values ​​of the adjacent region and the gray value of the lesion region are replaced to obtain the target gray value, and the pixels of the lesion region are rendered according to the replaced gray value to obtain a grayscale rendering image.

[0117] S1422. If the difference between the gray values ​​of the adjacent regions and the gray values ​​of the lesion region is less than the gray value difference threshold, or the gray value of the lesion region is greater than the gray value threshold, then the sum of the gray values ​​of the adjacent regions and the gray values ​​of the lesion region is calculated to obtain the target gray value, and the pixels of the lesion region are rendered according to the target gray value to obtain a grayscale rendering image.

[0118] In practice, if the difference between the grayscale value of an adjacent region and the grayscale value of the lesion region is greater than a grayscale difference threshold, and the grayscale value of the lesion region is less than the grayscale threshold, it indicates that the grayscale value of the lesion region in the image is low, making it relatively blurry, while the grayscale values ​​of its surrounding areas are higher. To highlight the lesion region in the image, the grayscale values ​​of the adjacent regions and the lesion region can be interacted, and the resulting grayscale value can be used as the target grayscale value. After interaction, the pixels of the lesion region can be rendered using the grayscale values ​​of the adjacent regions, and vice versa. Through interactive rendering, the grayscale value of the lesion region can be increased, while the grayscale value of the surrounding areas can be decreased, thus highlighting the lesion region in the image and obtaining a grayscale rendered image.

[0119] Similarly, if the difference between the grayscale value of an adjacent region and the grayscale value of the lesion region is less than the grayscale difference threshold, it indicates that their grayscale values ​​are similar, meaning that the pixel grayscale values ​​of both regions may be relatively high or relatively low. In this case, to highlight the lesion region, the sum of the grayscale values ​​of the adjacent regions and the lesion region can be calculated to obtain the target grayscale value. Then, the pixels of the lesion region can be rendered based on the target grayscale value to obtain a grayscale rendered image.

[0120] After obtaining the orthopedic reconstruction image, the image size is adjusted. The adjustment can be done by enlarging or reducing the size, or it can be adjusted according to the size requirements of the diagnostic information.

[0121] After the adjustment, doctors can see the lesions more clearly, which will help them make a diagnosis and treatment.

[0122] As explained above, the generated orthopedic reconstruction image may be one or more. To facilitate doctors viewing multiple orthopedic reconstruction images simultaneously, in one embodiment, after obtaining grayscale value parameters from the tissue rendering image and performing pixel enhancement processing on the tissue in the tissue rendering image based on the grayscale value parameters to obtain the orthopedic reconstruction image, the method may further include the following steps: S15. Determine the number of orthopedic reconstruction images, and then create several layers based on the number of orthopedic reconstruction images.

[0123] S16. Add each of the orthopedic reconstruction images to the layer and overlay several of the layers to obtain an orthopedic overlay image.

[0124] First, determine the number of orthopedic reconstruction images, and then create several layers based on the number of orthopedic reconstruction images. Referring to the example above, assuming there are two orthopedic reconstruction images, one of bone tissue and the other of lesion tissue, we can construct two layers, adding each orthopedic reconstruction image to one layer, and then overlaying the two layers to obtain an overlay orthopedic image.

[0125] When doctors review the images later, they can view the superimposed reconstructed images of the lesions and bones, or switch layers to view the orthopedic reconstructed images of the bones or lesions separately, so that doctors can make a diagnosis and the image display effect can be improved.

[0126] In this embodiment, the present invention provides a method for reconstructing orthopedic case images. Its advantages are as follows: The present invention can acquire orthopedic case images and diagnostic requirement information. After determining that the skeleton in the orthopedic case image meets the reconstruction requirements based on the skeletal keywords in the diagnostic requirement information, it identifies the skeletal tissue contour and lesion tissue contour of the orthopedic case image according to the diagnostic requirement information, and extracts the image corresponding to the contour to obtain a tissue contour image. It extracts tissue feature parameters of the skeletal tissue from the tissue contour image, determines the severity of lesion features based on the values ​​of the tissue feature parameters, and renders the edges of the tissue contour image according to the severity of lesion features to obtain a tissue rendering image. It obtains grayscale value parameters from the tissue rendering image and performs pixel enhancement processing on the tissue in the tissue rendering image according to the grayscale value parameters to obtain the orthopedic reconstructed image. This invention can extract lesions and bone tissue from orthopedic case images, determine the severity of the lesions based on the size of the tissue's characteristic parameters, and then render the tissue outline according to the color and brightness corresponding to the severity of the lesions. The rendered image is then pixel-enhanced to highlight the lesions and bones, allowing doctors to quickly and intuitively determine the actual condition of the lesions based on the color and brightness of the reconstructed image. This eliminates the need for repeated image reviews and reliance on test reports for diagnosis, effectively improving the efficiency of doctors in diagnosing and studying lesions.

[0127] This invention also provides a device for reconstructing orthopedic case images, see [link to relevant documentation]. Figure 2 The diagram shows a schematic structural diagram of an orthopedic case image reconstruction device provided in an embodiment of the present invention.

[0128] As an example, the orthopedic case image reconstruction device may include: The acquisition module 201 is used to acquire orthopedic case images and diagnostic requirement information respectively, and after determining that the skeleton of the orthopedic case image meets the reconstruction requirements based on the skeletal keywords in the diagnostic requirement information, the orthopedic case image is formatted according to the diagnostic requirement information to obtain a formatted image. The recognition module 202 is used to call the edge detection algorithm to recognize the bone tissue contour and lesion tissue contour of the format-adjusted image according to the bone keywords of the diagnostic requirement information, and to extract the image corresponding to the contour to obtain the tissue contour image. The rendering module 203 is used to extract tissue feature parameters of bone tissue from the tissue contour image, determine the severity of lesion tissue based on the value of the tissue feature parameters, determine contour color value and contour brightness value according to the severity of lesion tissue, and render the edge of the tissue contour image using the contour color value and the contour brightness value to obtain a tissue rendering image. The tissue feature parameters include the fracture length, number of fractures and fracture displacement angle of bone tissue. The enhancement module 204 is used to obtain grayscale value parameters from the tissue rendering image, and perform pixel enhancement processing on the tissue of the tissue rendering image according to the grayscale value parameters to obtain an orthopedic reconstruction image. The grayscale value parameters include the grayscale value of the lesion area in the treatment tissue and the grayscale value of the adjacent area of ​​the lesion area.

[0129] Optionally, the rendering module is further configured to: Determine the comparison result of each tissue feature parameter with the corresponding preset first feature threshold; The severity of the corresponding lesion tissue is determined based on the three comparison results, and the corresponding color value is found based on the severity of the lesion tissue to obtain the outline color value; The contour brightness value corresponding to the contour color value is determined based on the severity of the lesion tissue. The edges of the tissue contour image are rendered based on the contour color value and the contour brightness value to obtain a tissue rendering image.

[0130] Optionally, the rendering module is further configured to: If each comparison result is that the tissue feature parameter is less than or equal to the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be mild, and the first color value corresponding to the mild severity of the lesion feature is searched in the preset color list to obtain the outline color value; If any of the comparison results is greater than the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be moderate, and the second color value corresponding to the moderate severity of the lesion feature is searched in the preset color list to obtain the outline color value; If any two of the comparison results are greater than the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be severe, and the third color value corresponding to the severity of the lesion feature is searched in the preset color list to obtain the outline color value; If each comparison result indicates that the tissue feature parameter is greater than the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be urgent, and the fourth color value corresponding to the severity of the lesion feature being urgent is searched in the preset color list to obtain the outline color value.

[0131] Optionally, the rendering module is further configured to: If the severity of the lesion is mild or moderate, then the contour brightness value corresponding to the contour color value is determined to be a preset first brightness value; If the severity of the lesion is severe or urgent, and all three tissue feature parameters are less than or equal to the corresponding preset second feature threshold, then the contour brightness value corresponding to the contour color value is determined to be the preset second brightness value. If the severity of the lesion is severe or urgent, and any one of the tissue feature parameters is greater than the corresponding preset second feature threshold, then the contour brightness value corresponding to the contour color value is determined to be a preset third brightness value. If the severity of the lesion is severe or urgent, and any two of the tissue feature parameters are greater than the corresponding preset second feature threshold, then the contour brightness value corresponding to the contour color value is determined to be a preset fourth brightness value. If the severity of the lesion is severe or urgent, and the three tissue feature parameters are greater than the corresponding preset second feature threshold, then the contour brightness value corresponding to the contour color value is determined to be the preset fifth brightness value. Wherein, the preset first brightness value is less than the preset second brightness value, the preset second brightness value is less than the preset third brightness value, the preset third brightness value is less than the preset fourth brightness value, and the preset fourth brightness value is less than the preset fifth brightness value; The preset first feature threshold is less than the preset second feature threshold.

[0132] Optionally, the enhancement module is further configured to: The grayscale values ​​of the lesion region and the grayscale values ​​of adjacent regions are obtained from the rendered image of the tissue. The target gray value is determined based on the comparison between the gray values ​​of the adjacent regions and the gray value of the lesion region, and the target gray value is used to render the pixels of the tissue rendering image to obtain a pixel rendering image. Adjusting the size of the pixel-rendered image yields the orthopedic reconstruction image.

[0133] Optionally, the enhancement module is further configured to: If the difference between the gray value of the adjacent region and the gray value of the lesion region is greater than the gray value difference threshold and the gray value of the lesion region is less than the gray value threshold, then the gray values ​​of the adjacent region and the gray value of the lesion region are replaced to obtain the target gray value, and the pixels of the lesion region are rendered according to the replaced gray value to obtain a pixel rendering image. If the difference between the gray values ​​of the adjacent regions and the gray values ​​of the lesion region is less than the gray value difference threshold, or the gray value of the lesion region is greater than the gray value threshold, then the sum of the gray values ​​of the adjacent regions and the gray values ​​of the lesion region is calculated to obtain the target gray value, and the pixels of the lesion region are rendered according to the target gray value to obtain a pixel-rendered image.

[0134] Reference Figure 3 The diagram shows a structural schematic of an orthopedic case image reconstruction device according to another embodiment of the present invention.

[0135] Optionally, the device further includes: The determining module 205 is used to determine the number of orthopedic reconstruction images after the steps of obtaining grayscale value parameters from the tissue rendering image and performing pixel enhancement processing on the tissue of the tissue rendering image according to the grayscale value parameters to obtain orthopedic reconstruction images, and then creating several layers according to the number of orthopedic reconstruction images. The overlay module 206 is used to add each of the orthopedic reconstruction images into the layer and overlay several of the layers to obtain an orthopedic overlay image.

[0136] Those skilled in the art will understand that, for ease of description and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0137] Furthermore, this application also 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 executes the program to implement the orthopedic case image reconstruction method as described in the above embodiments.

[0138] Furthermore, embodiments of this application also provide a computer-readable storage medium storing a computer-executable program for causing a computer to perform the orthopedic case image reconstruction method as described in the above embodiments.

[0139] In the description of the embodiments of the present invention, it should be noted that the terms "above," "below," etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. When an element such as a layer, region, or substrate is referred to as being "above" or "on top of" another element, it may be directly on the other element, or there may be an intermediate element. Conversely, when an element is referred to as being "directly on" or "above" another element, there is no intermediate element. It should also be understood that when an element is referred to as being "below" or "under" another element, it may be directly below or under the other element, or there may be an intermediate element. Conversely, when an element is referred to as being "directly below" or "under" another element, there is no intermediate element. Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0140] Those skilled in the art will understand that embodiments of this application may also include computer program products. Therefore, this application may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0141] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), devices, and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0142] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0143] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0144] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for reconstructing orthopedic case images, characterized in that, The method includes: The orthopedic case images and diagnostic requirement information are acquired separately. After determining that the skeleton of the orthopedic case image meets the reconstruction requirements based on the skeletal keywords in the diagnostic requirement information, the format of the orthopedic case image is adjusted according to the diagnostic requirement information to obtain a format-adjusted image. Based on the bone keywords in the diagnostic requirements information, the edge detection algorithm is invoked to identify the bone tissue contour and lesion tissue contour of the format-adjusted image, and the image corresponding to the contour is cropped to obtain the tissue contour image; Tissue feature parameters of bone tissue are extracted from the tissue contour image. The severity of the lesion is determined based on the values ​​of the tissue feature parameters. The contour color value and contour brightness value are determined according to the severity of the lesion. The edges of the tissue contour image are rendered using the contour color value and the contour brightness value to obtain a tissue rendering image. The tissue feature parameters include the fracture length, number of fractures, and fracture displacement angle of the bone tissue. Grayscale parameters are obtained from the tissue rendering image, and pixel enhancement processing is performed on the tissue in the tissue rendering image according to the grayscale parameters to obtain an orthopedic reconstruction image. The grayscale parameters include the grayscale values ​​of the lesion area within the treatment tissue and the grayscale values ​​of the adjacent areas of the lesion area.

2. The method for reconstructing orthopedic case images according to claim 1, characterized in that, The process includes extracting tissue feature parameters of bone tissue from the tissue contour image, determining the severity of lesions based on the values ​​of the tissue feature parameters, determining contour color values ​​and contour brightness values ​​based on the severity of lesions, and rendering the edges of the tissue contour image using the contour color values ​​and contour brightness values ​​to obtain a tissue rendering image, including: Determine the comparison result of each tissue feature parameter with the corresponding preset first feature threshold; The severity of the corresponding lesion tissue is determined based on the three comparison results, and the corresponding color value is found based on the severity of the lesion tissue to obtain the outline color value; The contour brightness value corresponding to the contour color value is determined based on the severity of the lesion tissue. The edges of the tissue contour image are rendered based on the contour color value and the contour brightness value to obtain a tissue rendering image.

3. The method for reconstructing orthopedic case images according to claim 2, characterized in that, The step of determining the severity of the corresponding lesion tissue based on the three comparison results, and finding the corresponding color value based on the severity of the lesion tissue to obtain the contour color value includes: If each comparison result is that the tissue feature parameter is less than or equal to the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be mild, and the first color value corresponding to the mild severity of the lesion feature is searched in the preset color list to obtain the outline color value; If any one of the comparison results is that the tissue feature parameter is greater than the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be moderate, and the second color value corresponding to the moderate severity of the lesion feature is searched in the preset color list to obtain the outline color value; If only any two of the comparison results are that the tissue feature parameter is greater than the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be severe, and the third color value corresponding to the severity of the lesion feature is searched in the preset color list to obtain the outline color value; If each comparison result indicates that the tissue feature parameter is greater than the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be urgent, and the fourth color value corresponding to the severity of the lesion feature being urgent is searched in the preset color list to obtain the outline color value.

4. The method for reconstructing orthopedic case images according to claim 3, characterized in that, The step of determining the contour brightness value corresponding to the contour color value based on the severity of each lesion tissue includes: If the severity of the lesion is mild or moderate, then the contour brightness value corresponding to the contour color value is determined to be a preset first brightness value; If the severity of the lesion is severe or urgent, and all three tissue feature parameters are less than or equal to the corresponding preset second feature threshold, then the contour brightness value corresponding to the contour color value is determined to be the preset second brightness value. If the severity of the lesion is severe or urgent, and only one of the tissue feature parameters is greater than the corresponding preset second feature threshold, then the contour brightness value corresponding to the contour color value is determined to be a preset third brightness value. If the severity of the lesion is severe or urgent, and only any two of the tissue feature parameters are greater than the corresponding preset second feature threshold, then the contour brightness value corresponding to the contour color value is determined to be a preset fourth brightness value. If the severity of the lesion is severe or urgent, and the three tissue feature parameters are greater than the corresponding preset second feature threshold, then the contour brightness value corresponding to the contour color value is determined to be the preset fifth brightness value. Wherein, the preset first brightness value is less than the preset second brightness value, the preset second brightness value is less than the preset third brightness value, the preset third brightness value is less than the preset fourth brightness value, and the preset fourth brightness value is less than the preset fifth brightness value; The preset first feature threshold is less than the preset second feature threshold.

5. The method for reconstructing orthopedic case images according to claim 1, characterized in that, The step of obtaining grayscale value parameters from the tissue rendering image and performing pixel enhancement processing on the tissue in the tissue rendering image based on the grayscale value parameters to obtain an orthopedic reconstruction image includes: The grayscale values ​​of the lesion region and the grayscale values ​​of adjacent regions are obtained from the rendered image of the tissue. The target gray value is determined based on the comparison between the gray values ​​of the adjacent regions and the gray value of the lesion region, and the target gray value is used to render the pixels of the tissue rendering image to obtain a pixel rendering image. Adjusting the size of the pixel-rendered image yields the orthopedic reconstruction image.

6. The method for reconstructing orthopedic case images according to claim 5, characterized in that, The step of determining a target gray value based on a comparison between the gray values ​​of the adjacent regions and the gray value of the lesion region, and then using the target gray value to render the pixels of the tissue rendering image to obtain a pixel rendering image, includes: If the difference between the gray value of the adjacent region and the gray value of the lesion region is greater than the gray value difference threshold and the gray value of the lesion region is less than the gray value threshold, then the gray values ​​of the adjacent region and the gray value of the lesion region are replaced to obtain the target gray value, and the pixels of the lesion region are rendered according to the replaced gray value to obtain a pixel rendering image. If the difference between the gray values ​​of the adjacent regions and the gray values ​​of the lesion region is less than the gray value difference threshold, or the gray value of the lesion region is greater than the gray value threshold, then the sum of the gray values ​​of the adjacent regions and the gray values ​​of the lesion region is calculated to obtain the target gray value, and the pixels of the lesion region are rendered according to the target gray value to obtain a pixel-rendered image.

7. The method for reconstructing orthopedic case images according to any one of claims 1-6, characterized in that, After the steps of obtaining grayscale value parameters from the tissue rendering image and performing pixel enhancement processing on the tissue in the tissue rendering image based on the grayscale value parameters to obtain an orthopedic reconstruction image, the method further includes: Once the number of orthopedic reconstruction images is determined, several layers are created based on the number of orthopedic reconstruction images; Each of the orthopedic reconstruction images is added to the layer and several layers are superimposed to obtain an orthopedic overlay image.

8. A device for reconstructing orthopedic case images, characterized in that, The device includes: The acquisition module is used to acquire orthopedic case images and diagnostic requirement information respectively, and after determining that the skeleton of the orthopedic case image meets the reconstruction requirements based on the skeletal keywords in the diagnostic requirement information, the orthopedic case image is formatted according to the diagnostic requirement information to obtain a formatted image. The recognition module is used to call the edge detection algorithm to recognize the bone tissue contour and lesion tissue contour of the format-adjusted image based on the bone keywords of the diagnostic requirement information, and to extract the image corresponding to the contour to obtain the tissue contour image. The rendering module is used to extract tissue feature parameters of bone tissue from the tissue contour image, determine the severity of lesion tissue based on the value of the tissue feature parameters, determine contour color value and contour brightness value according to the severity of lesion tissue, and render the edge of the tissue contour image using the contour color value and contour brightness value to obtain a tissue rendering image. The tissue feature parameters include fracture length, number of fractures and fracture displacement angle of bone tissue. An enhancement module is used to obtain grayscale value parameters from the tissue rendering image, and perform pixel enhancement processing on the tissue in the tissue rendering image according to the grayscale value parameters to obtain an orthopedic reconstruction image. The grayscale value parameters include the grayscale value of the lesion area within the treatment tissue and the grayscale value of the adjacent area of ​​the lesion area.

9. The orthopedic case image reconstruction apparatus according to claim 8, characterized in that, The rendering module is also used for: Determine the comparison result of each tissue feature parameter with the corresponding preset first feature threshold; The severity of the corresponding lesion tissue is determined based on the three comparison results, and the corresponding color value is found based on the severity of the lesion tissue to obtain the outline color value; The contour brightness value corresponding to the contour color value is determined based on the severity of the lesion tissue. The edges of the tissue contour image are rendered based on the contour color value and the contour brightness value to obtain a tissue rendering image.

10. The orthopedic case image reconstruction apparatus according to claim 9, characterized in that, The rendering module is also used for: If each comparison result is that the tissue feature parameter is less than or equal to the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be mild, and the first color value corresponding to the mild severity of the lesion feature is searched in the preset color list to obtain the outline color value; If any one of the comparison results is that the tissue feature parameter is greater than the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be moderate, and the second color value corresponding to the moderate severity of the lesion feature is searched in the preset color list to obtain the outline color value; If only any two of the comparison results are that the tissue feature parameter is greater than the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be severe, and the third color value corresponding to the severity of the lesion feature is searched in the preset color list to obtain the outline color value; If each comparison result indicates that the tissue feature parameter is greater than the corresponding preset first feature threshold, then the severity of the lesion feature is determined to be urgent, and the fourth color value corresponding to the severity of the lesion feature being urgent is searched in the preset color list to obtain the outline color value.

Images