Guide system for correcting rotational alignment of femur using artificial intelligence image processing and method thereof

Abstract

An embodiment of the present invention provides a guide system for correcting the rotational alignment of a femur using artificial intelligence image processing and a method therefor, wherein the guide system comprises: an imaging device for capturing images of affected and the unaffected sides of a patient; an image display device for displaying the affected side image containing the femur coupled with an implant and the unaffected side image containing the normal femur captured by the imaging device; and a user device that acquires an affected side image and an unaffected side image from the affected side and unaffected side images, respectively, and aligns the affected side image with the unaffected side image on the basis of the femur to calculate the rotational angle difference between the fractured femur in the affected side image and the normal femur in the unaffected side image using artificial intelligence image processing, thereby guiding the rotational alignment correction of the affected side.

Description

Guide system and method for correcting rotational alignment of the femur using artificial intelligence image processing

[0001] The present invention relates to a guide system and method for correcting the rotational alignment of a femur using artificial intelligence image processing, and more particularly, to a guide system and method for correcting the rotational alignment of a femur using artificial intelligence image processing for medical images used in an operating room.

[0002]

[0003] Femur fractures caused by factors such as traffic accidents or industrial accidents generally account for a significant proportion of all fracture patients. However, many patients suffer from problems such as nonunion and delayed union due to misalignment. Therefore, a precise femoral rotational alignment correction system is needed to address these issues.

[0004] Figure 1 is a diagram simulating the images of the healthy and affected sides during conventional femoral fracture surgery. Conventional femoral fracture surgery involves taking multiple mobile fluoroscopy images (hereinafter referred to as C-arm images). Currently, C-arm images are used to determine proper rotational alignment. However, because C-arm images only provide two-dimensional images, it is difficult to discern three-dimensional rotational images. This presents significant challenges in achieving accurate anatomical reduction.

[0005] Furthermore, radiation exposure for both patients and medical staff is a serious concern due to the multiple imaging procedures. Furthermore, postoperative CT scans are essential to confirm accurate alignment, and the increased likelihood of reoperation increases discomfort and costs.

[0006]

[0007] The technical task to be achieved by the present invention is to provide a guide system and method for correcting rotational alignment of the femur, which improves the quality of C-arm images used in an operating room and guides the correction of rotational alignment through comparison of the affected side and the healthy side using artificial intelligence image processing.

[0008] The technical problems to be solved by the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned can be clearly understood by a person having ordinary skill in the technical field to which the present invention belongs from the description below.

[0009]

[0010] In order to achieve the above technical task, one embodiment of the present invention provides a guide system for correcting rotational alignment of a femur using artificial intelligence image processing, characterized by including: an image capturing device for capturing the affected side and the healthy side of a patient; an image display device for displaying an image of the affected side including a femur with an implant combined therewith and an image of the healthy side including a normal femur captured by the image capturing device; and a user device for obtaining an image of the affected side and an image of the healthy side from the image of the affected side and the image of the healthy side, and aligning the image of the affected side with the image of the healthy side based on the femur through artificial intelligence image processing to calculate a difference in rotational angle between a fractured femur in the image of the affected side and a normal femur in the image of the healthy side, thereby guiding the correction of rotational alignment of the affected side.

[0011] In an embodiment of the present invention, the image capturing device is a C-arm using X-ray, the user device captures the affected-side image and the healthy-side image displayed on the image display device or obtains the affected-side image and the healthy-side image through communication, and the user device may include: an image acquisition unit that obtains the affected-side image and the healthy-side image from the affected-side image and the healthy-side image, respectively; an image quality enhancement unit that removes an implant combined with a femur from the affected-side image obtained through the artificial intelligence image processing and improves the quality of the image including clarity; an image matching unit that overlays the healthy-side image and the affected-side image, matches the size and position of the femur, and calculates an angular difference in rotational alignment; and a result display unit (340) that displays the result of the matching by the image matching unit so that a user can recognize it.

[0012] In an embodiment of the present invention, the user device is an MR-based (AR / VR) device, and extracts a femur from the affected-side image and the healthy-side image displayed on the image display device using artificial intelligence image processing, constructs a 3D model, and then performs 3D model alignment using the 3D model.

[0013] In order to achieve the above technical problem, another embodiment of the present invention provides a guide method for correcting the rotational alignment of a femur using artificial intelligence image processing, characterized in that it comprises: a step in which an image acquisition unit of a user device acquires an affected-side image and a healthy-side image from an affected-side image and a healthy-side image by an image photographing device; a step in which an image quality enhancement unit removes an implant in the affected-side image through artificial intelligence image processing; a step in which the image quality enhancement unit improves the quality by performing imaging processing on the affected-side image from which the implant has been removed; a step in which an image matching unit matches the healthy-side image and the image-processed affected-side image; and a step in which a result display unit displays a result of the matching including an angular difference in the rotational alignment of the affected-side femur and the healthy-side femur calculated as a result of the matching.

[0014] In an embodiment of the present invention, the step of removing the implant in the affected-side image may include a step of restoring image distortion and noise that occur when the image quality enhancement unit acquires the affected-side image and the healthy-side image through artificial intelligence image processing; a step of distinguishing the femur and the implant in the affected-side image through artificial intelligence image processing and removing the implant; and a step of restoring the original femur image in the area where the implant was removed by the image quality enhancement unit.

[0015] In an embodiment of the present invention, the step of improving the quality of the affected side image by imaging processing may include the step of performing high-resolution processing to improve the resolution of the affected side image from which the implant has been removed; the step of performing high-definition processing to improve the clarity of the affected side image from which the implant has been removed; and the step of performing transparency processing for comparison on the affected side or healthy side image on which the high-resolution processing and the high-definition processing have been performed.

[0016] In an embodiment of the present invention, the step of aligning the healthy side image and the image-processed affected side image may include the steps of: finding the femoral head and the femoral shaft in the healthy side image and the image-processed affected side image, respectively; matching the image-processed affected side image to the healthy side image by overlapping it with the scale of the femoral head and the position of the femoral shaft; and calculating the angular difference in rotational alignment between the femoral head of the healthy side and the femoral head of the affected side.

[0017]

[0018] According to an embodiment of the present invention, a guide system and method for correcting rotational alignment of a femur can be provided, which guides the correction of rotational alignment by converting a photographed C-arm monitor image into its original form without distortion or noise through artificial intelligence image processing technology that improves readability by extracting only the femur from a C-arm image.

[0019] In addition, by using an artificial intelligence image processing system to guide the correction of the rotational alignment of the femur, which is difficult to correct in the operating room, we can provide a precision medical service that improves the surgical results of femur fracture surgery and enables rapid recovery.

[0020] The effects of the present invention are not limited to the effects described above, and should be understood to include all effects that can be inferred from the detailed description of the present invention or the composition of the invention described in the claims.

[0021]

[0022] Figure 1 is a drawing simulating images of the healthy side and the affected side in a conventional femoral fracture treatment surgery.

[0023] FIG. 2 is a drawing for explaining a guide system for correcting rotational alignment of a femur using artificial intelligence image processing according to one embodiment of the present invention.

[0024] FIG. 3 is a block diagram showing a guide system for correcting rotational alignment of a femur using artificial intelligence image processing according to one embodiment of the present invention.

[0025] Figure 4 illustrates an example of a user device that displays an image captured by a video displayed on a monitor.

[0026] Figure 5 is a diagram showing a user device displaying the results of alignment.

[0027] Figure 6 illustrates an AR or VR device as another example of a user device.

[0028] Figure 7 is a flowchart showing a guide method for correcting rotational alignment of a femur using artificial intelligence image processing according to one embodiment of the present invention.

[0029] Figure 8 shows an example of a monitor displaying a C-arm image.

[0030] Figure 9 shows a side image obtained by capturing a side image displayed on a monitor using a user device.

[0031] Figure 10 shows a side view image acquired by the user device.

[0032] Figure 11 shows an image of the affected side from which the implant has been removed through artificial intelligence image processing for the affected side image of Figure 10.

[0033] Figure 12 shows an image with improved resolution and quality of the image on the back.

[0034] Figure 13 is an image of the image processed in Figure 12 with its color changed and horizontally flipped.

[0035] Figure 14 is an image with the transparency changed so that the image processed in Figure 13 becomes transparent.

[0036] Figure 15 additionally illustrates a process for improving image quality and visibility.

[0037] Figure 16 is an image of finding the femoral head and femoral shaft in the image-processed healthy side image.

[0038] Figure 17 is an image of finding the femoral head and femoral shaft in the image of the affected side that has been image-processed.

[0039] Figure 18 is an image that overlays the healthy image of Figure 16 and the affected image of Figure 17.

[0040] Figure 19 shows an image of a screw inserted after correction of rotational alignment.

[0041]

[0042] <Explanation of symbols>

[0043] 10: Guide system to correct rotational alignment of the femur

[0044] 100: Video recording device

[0045] 200: Video display device

[0046] 300: User device

[0047] 310: Image acquisition unit

[0048] 320: Video Quality Improvement Department

[0049] 330: Image alignment unit

[0050] 340: Results display section

[0051]

[0052] The present invention is susceptible to various modifications and variations, and specific embodiments are illustrated in the drawings and described in detail in the text. However, this is not intended to limit the present invention to a specific disclosed form, but rather to encompass all modifications, equivalents, and alternatives falling within the spirit and technical scope of the present invention. Throughout the description of each drawing, similar reference numerals have been used to designate similar components.

[0053] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and will not be interpreted in an idealized or overly formal sense unless explicitly defined herein.

[0054] Hereinafter, with reference to the attached drawings, a preferred embodiment of the present invention will be described in more detail.

[0055] FIG. 2 is a drawing for explaining a guide system (10) for correcting the rotational alignment of the femur using artificial intelligence image processing according to one embodiment of the present invention. FIG. 3 is a block diagram showing a guide system (10) for correcting the rotational alignment of the femur using artificial intelligence image processing according to one embodiment of the present invention.

[0056] Referring to FIGS. 2 and 3, a guide system (10) for correcting rotational alignment of a femur using artificial intelligence image processing may include an image capturing device (100), an image display device (200), and a user device (300).

[0057] The imaging device (100) may be a C-arm. The imaging display device (200) may be a monitor that displays images captured by the C-arm. The C-arm can capture images of the fractured femur (affected side) and the normal part on the healthy side of a patient positioned on a table using X-rays. Images of the affected side and the healthy side may be displayed on the monitor. The image of the affected side (e.g., left femur image) may show the femur and the implant that connects the fractured femur. The image of the healthy side may show an image of the unfractured femur (e.g., right femur image).

[0058] The user device (300) can guide rotational alignment correction by comparing the healthy-side image and the affected-side image displayed on a monitor in an operating room. To this end, the user device (300) can photograph the affected-side image and the healthy-side image displayed on the monitor or obtain the affected-side image and the healthy-side image through communication. The user device (300) may be a device capable of processing images, such as a smartphone, tablet, virtual reality device (VR), or augmented reality device (AR). The user device (300) may include an image acquisition unit (310), an image quality enhancement unit (320), an image alignment unit (330), and a result display unit (340).

[0059] Figure 4 shows an example of a user device (300) that displays an image captured by a video displayed on a monitor.

[0060] Referring to FIG. 4, the image acquisition unit (310) can capture the affected-side image and the healthy-side image displayed on the monitor to acquire the affected-side image and the healthy-side image, respectively. Of course, alternatively, the image acquisition unit (310) can also acquire the affected-side image and the healthy-side image, respectively, from the affected-side image and the healthy-side image through communication with the monitor or the image capture device. The user device (300) can capture the affected-side image and the healthy-side image displayed on the monitor by fixing it at a preset position with respect to the monitor or, depending on the situation, holding it in the user's hand. For example, a smartphone can be used as the user device (300), and the camera module of the smartphone can capture the affected-side image and the healthy-side image displayed on the monitor as the image acquisition unit (310) to acquire the affected-side image and the healthy-side image, respectively. The acquired affected-side image and the healthy-side image can be used by the user in a timely manner in the operating room to guide femoral rotational alignment.

[0061] The image quality enhancement unit (320) can remove implants attached to the femur from the acquired affected-side image and improve the image quality, such as clarity. The image quality enhancement unit (320) can use artificial intelligence image processing technology to extract only the femur from the C-arm image (affected-side image, healthy-side image) to improve readability. For example, artificial intelligence software can be used to remove implants (e.g., nails, plates, etc.) that are unnecessary for image comparison from the affected-side image displayed on the smartphone screen. In this case, when the user selects an implant in the affected-side image using a finger, touch pen, cursor, etc., the artificial intelligence software provided in the image quality enhancement unit (320) can recognize the implant, separate it from the femur, and then delete the implant. Alternatively, the artificial intelligence software can be trained to distinguish between bone and implant and automatically separate and delete them. That is, the image quality enhancement unit (320) can extract only the femur from the affected-side image, remove the implant, and then perform high-resolution, high-definition, and low-noise processing to improve the readability of the affected-side image. The process of removing an implant from an image of the affected side, restoring an image of the bone, and improving image quality is described below as an example.

[0062] The image quality enhancement unit (320) may first perform image preprocessing to remove specific portions of the image, such as implants, from the affected image. Before initiating the implant removal process, preprocessing can be performed to make the element being processed as clear as possible. Next, the image is converted to grayscale. Most medical images are grayscale, so this may not be necessary. However, if the image is in color, converting to grayscale makes imaging easier. Next, noise reduction is performed. Medical images can be noisy, so applying a noise reduction filter, such as a Gaussian filter, can smooth the image and better distinguish between bone and implants. This distinction can be achieved using thresholding, which is a method for dividing an image into multiple segments based on pixel intensities. Because metal implants typically appear brighter (higher intensity) than bone, thresholding techniques can be helpful in separating these segments. For example, adaptive thresholding can help distinguish implants from bone by considering the pixel's neighborhood and then finding the optimal threshold for each neighborhood. Region-based segmentation can then be performed. Once a thresholded image is available, implants can be identified using region-based segmentation. Contours are then found. This allows the boundaries of the implant to be identified. This allows for the implant region to be identified. Implant regions can be manually selected or identified using AI software that utilizes machine learning models based on characteristics such as area, shape, and intensity. The implant image, once identified and distinguished through this process, can be removed. The removed implant region can then be filled in with adjacent pixels. One example of this filling method is to fill in the identified region with the values ​​of adjacent pixels, such as by smoothing the region using interpolation. In some cases, this can even reconstruct the underlying bone structure.This can be accomplished using advanced methods such as morphological techniques or inpainting. After implant removal and restoration, image post-processing can be performed. After implant removal, additional post-processing steps, such as contrast adjustment or additional noise reduction, may be required to improve image quality. Finally, the images can be reviewed and adjusted using image editing software to correct any inaccuracies.

[0063] Figure 5 is a diagram showing a user device (300) displaying the result of alignment.

[0064] The image alignment unit (330) can guide rotational alignment correction using the healthy side image and the affected side image. For example, the image alignment unit (330) can overlay the affected side image on the healthy side image, match the size and position of the femur, and calculate the difference in angle. This alignment process is particularly important in medical imaging, especially in surgeries requiring alignment of various anatomical structures. In the case of femur fracture surgery, aligning the fractured side with the healthy side to accurately adjust the angle or position is essential for a successful surgery. A specific operation process of the image alignment unit (330) is as follows. The image alignment unit (330) can first perform image preprocessing to make the image suitable for alignment. In addition, the image alignment unit (330) can perform image enhancement by applying the aforementioned techniques, such as edge enhancement, noise reduction, and contrast adjustment. Thereafter, if the resolutions of the affected side and the healthy side images differ, they can be resampled to the same resolution. Afterwards, the image can be cropped or masked to focus only on the region of interest (e.g., the femur). The image registration unit (330) can also detect key features of the two images to be used to align the healthy and affected images. Specifically, methods such as Scale-Invariant Feature Transform (SIFT) or Speeded-Up Robust Features (SURF) can be used to detect key features in the two images. In addition, the image registration unit (330) can match these features between the two images using methods such as Approximate Nearest Neighbors (FLANN). After this process, the image registration unit (330) can determine a transformation to align the features of the two images. First, depending on the misalignment, a fixed (rotation and translation), an affine (including scaling and shearing), or a more complex nonlinear transformation can be selected. The transformation parameters are reliably estimated using a method such as the RANSAC algorithm.Afterwards, a transformation can be applied to align the images. Specifically, the fractured lateral image can be resampled to align with the normal lateral image using an expected transformation. Furthermore, methods such as bicubic or trilinear interpolation can make the transformation smooth and continuous.

[0065] The result display unit (340) can display the result of such alignment on the screen of the smartphone so that the user can recognize it. The result of the alignment can include the angular difference of the femur in the healthy side image and the affected side image. The user can correct this angle by correcting the rotational angle of the affected side. That is, the correction guide function of the user device as described above can be utilized to correct the rotational angle before the final process such as fixing the implant. Accordingly, the rotational alignment of the femur can be corrected accurately and quickly, which is good for both the patient and the doctor, and the X-ray exposure dose can be significantly reduced.

[0066] Figure 6 illustrates an AR or VR device as another example of a user device.

[0067] In the guide system (10) for correcting the rotational alignment of the femur using artificial intelligence image processing of the present embodiment, in addition to the aforementioned smartphone, an MR-based (AR / VR) surgical guide device as exemplified in FIG. 6 may be adopted as a user device. For example, it is possible to improve visibility and surgical convenience and maximize alignment accuracy by implementing it in real time with AR in the operating room. In conjunction with such an MR-based (AR / VR) surgical guide device, a femur can be extracted from a C-arm image using a bone alignment guide marker in real time, a 3D model can be constructed, and then the 3D model can be aligned using this. As a result, X-ray-free (radiation-free) surgery can be pursued.

[0068] Figure 7 is a flowchart illustrating a guide method for correcting rotational alignment of the femur using artificial intelligence image processing according to one embodiment of the present invention. Figure 8 illustrates an example of a monitor displaying a C-arm image. Figure 9 illustrates an image of the affected side obtained by photographing the image of the affected side displayed on the monitor using a user device (300).

[0069] In a guide method for correcting femoral rotational alignment using artificial intelligence image processing, the user device (300) first acquires C-arm images of the healthy and affected sides (S10). For example, prior to surgery, the aforementioned imaging device (100) takes C-arm lateral images of the healthy and affected sides. Depending on the surgical procedure, the procedure may proceed up to the final fixation stage after nail insertion.

[0070] Referring to Fig. 8, in this state, the affected side image and the healthy side image can be displayed on the monitor (image display device (200)). In this state, the user device (300) can capture C-ARM images of the healthy side and the affected side (the affected side image and the healthy side image) to obtain the healthy side image and the affected side image. That is, as illustrated in Fig. 9, if the user device (300) is a smartphone, the smartphone can capture the monitor to obtain the affected side image and the healthy side image.

[0071] Fig. 10 shows a side image obtained by a user device (300). Fig. 11 shows a side image from which an implant has been removed through artificial intelligence image processing for the side image of Fig. 10.

[0072] Thereafter, as illustrated in FIGS. 10 and 11, the implant is removed from the affected image (S20). Artificial intelligence software may be applied to restore image distortion that occurs when a C-arm monitor image (affected-side image, healthy-side image) is photographed with a user device (300). This restoration may include both restoration of distortion, such as removing noise that occurs when photographing with the user device (300), and restoration of the original image in the area where the implant was removed. For example, for effective image comparison, implants such as plates or nails are removed from the affected-side image. The operation method of the artificial intelligence software for such imaging processing is as described above. For example, for the affected-side image illustrated in FIG. 10, the image quality enhancement unit (320) may use a program such as a magic eraser to delete the implant, and restore the shape of the femur in the area where the implant was removed, as illustrated in FIG. 11.

[0073] Figure 12 shows an image with improved resolution and quality of the image on the affected side. Figure 13 shows the image processed in Figure 12 with its colors changed and horizontally flipped. Figure 14 shows the image processed in Figure 13 with its transparency changed to make it transparent.

[0074] Next, the image of the affected side from which the implant has been removed is image-processed to improve its quality (S30). As described above, even after removing the implant from the affected side image, the image may still not be clear. Therefore, an imaging processing process to improve the image quality, such as resolution and clarity, of the image of the affected side from which the implant has been removed may be further performed by the image quality enhancement unit (320). That is, for comparison with the image of the healthy side in the subsequent alignment process, the image processing may be performed using software that can at least produce a quality comparable to that of the image of the healthy side.

[0075] For example, as illustrated in FIGS. 12 to 14, high-resolution, high-definition, low-noise processing can be performed on the image of the affected side from which the implant has been removed. That is, as in FIG. 12, high-resolution processing can be performed on the image of the affected side from which the implant has been removed, as in FIG. 13, high-definition processing can be performed, and as in FIG. 14, low-noise processing can be performed through transparency processing. These processes can also be performed by the image processing artificial intelligence software applied during the implant removal described above.

[0076] Figure 15 additionally illustrates a process for improving image quality and visibility.

[0077] After the imaging processing described above, as shown in FIG. 15, additional modifications can be made to improve the image quality and visibility so that only the bones are visible.

[0078] Fig. 16 is an image of finding the femoral head and femoral shaft in the image-processed healthy side image. Fig. 17 is an image of finding the femoral head and femoral shaft in the image-processed affected side image. Fig. 18 is an image of overlaying the healthy side image of Fig. 16 and the affected side image of Fig. 17.

[0079] As described above, after improving the image quality, the image alignment unit (330) can align the healthy image and the image-processed affected image (S40). After removing the implant from the affected image and improving the image quality, it can be overlaid on the healthy image and compared by matching the scale, angle, and position. For example, the femoral head and femoral shaft are found in the healthy image and the image-processed affected image. Thereafter, the image-processed affected image can be overlaid on the healthy image by matching the position and scale and compared. As described above, the affected image is transparentized, so that when it is overlaid on the healthy image, the femoral head and femoral shaft can be recognized in the healthy image, and this can be visually adjusted to match the femoral shaft of the affected side. In the operating room, a C-arm image should be taken with an imaging device (100) to compare the healthy and affected sides, and it should be confirmed that the rotation angles of both sides are the same when looking at the shape of the foot. However, in the past, it was difficult to confirm whether the rotation angles were exactly the same simply by looking at the affected and healthy images displayed on the monitor. In this embodiment, the image matching unit (330) of the user device overlays the affected and healthy images to match them. That is, the position and size of the femoral shaft on the healthy side and the femoral shaft on the affected side can be matched, and the difference in the rotation angle can be extracted. The difference in the rotation angles extracted in this way can be used as a guide to correct the rotation angles of the affected and healthy sides to match.

[0080] Next, the result of the alignment can be displayed through the user device (300) (S50).

[0081] Using images of the affected and healthy sides captured using a smartphone, the femur can be extracted and a 3D model constructed. Analysis of the alignment results by the image alignment unit (330) on the smartphone can display the current anteversion angle and the angular difference between the healthy side and the affected side (see Fig. 5). A post-operative CT scan showed an anteversion of less than 3 degrees, demonstrating excellent correction.

[0082] Figure 19 shows an image of a screw inserted after correction of rotational alignment.

[0083] As previously mentioned, the rotational angle is corrected through rotational alignment correction, and then the implant is secured with screws to complete the correction surgery. The alignment is finally confirmed before the surgery is completed.

[0084] The guide system and method for correcting femoral rotational alignment using artificial intelligence image processing according to the present invention can be easily implemented and used via a smartphone app. This eliminates the need for a separate, complex system and allows for a simple and fast use of the torsion angle calculation app, enabling highly convenient and accurate rotational angle correction guidance.

[0085] According to an embodiment of the present invention, a system and method for photographing a C-arm monitor image to restore distortion and reduce noise, extracting a femur and improving the image to improve readability of the image, and guiding position and size matching and correction for comparing images of the healthy side and the affected side can be provided.

[0086] The foregoing description of the present invention is for illustrative purposes only, and those skilled in the art will readily appreciate that the present invention can be readily modified into other specific forms without altering the technical spirit or essential characteristics of the present invention. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. For example, each component described as a single entity may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined manner.

[0087] The scope of the present invention is indicated by the claims described below, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention.

[0088]

[0089] According to the present invention, a guide system and method for correcting rotational alignment of a femur that guides the correction of rotational alignment by converting a photographed C-arm monitor image into its original form without distortion or noise through artificial intelligence image processing technology that improves readability by extracting only the femur from a C-arm image, thereby being more effective in the medical field.

Claims

1. An imaging device that photographs the affected and healthy sides of the patient; An image display device that displays an image of the affected side including the femur with the implant attached and an image of the healthy side including the normal femur, which are captured by the above image capturing device; and A guide system for correcting rotational alignment of a femur using artificial intelligence image processing, characterized in that it includes a user device that obtains a diseased-side image and a healthy-side image from the diseased-side image and the healthy-side image, and guides the rotational alignment correction of the diseased-side by aligning the diseased-side image with the healthy-side image based on the femur through artificial intelligence image processing and calculating the difference in rotational angle between the fractured femur in the diseased-side image and the normal femur in the healthy-side image.

2. In claim 1, The above imaging device is a C-arm that uses X-rays, The user device captures the affected image and the healthy image displayed on the image display device or obtains the affected image and the healthy image through communication, The above user device, An image acquisition unit that acquires the affected-side image and the healthy-side image from the affected-side image and the healthy-side image, respectively; An image quality enhancement unit that removes an implant attached to the femur from the image of the affected side obtained through the above artificial intelligence image processing and improves the quality of the image, including clarity; An image matching unit that overlays the healthy image and the affected image, matches the size and position of the femur, and calculates the angular difference of rotational alignment; and A guide system for correcting rotational alignment of a femur using artificial intelligence image processing, characterized in that it includes a result display unit that displays the result of alignment by the image alignment unit so that a user can recognize it.

3. In claim 1, A guide system for correcting rotational alignment of the femur using artificial intelligence image processing, characterized in that the user device is an MR-based (AR / VR) device, extracts the femur using artificial intelligence image processing from the affected-side image and the healthy-side image displayed on the image display device, constructs a 3D model, and then performs 3D model alignment using the 3D model.

4. In a guide method for correcting rotational alignment of the femur using artificial intelligence image processing, A step of acquiring a diseased image and a healthy image from the diseased image and the healthy image by the image acquisition unit of the user device using an image capturing device; A step of removing an implant from the image of the affected side through artificial intelligence image processing by the image quality improvement unit; A step of improving the image quality by processing the image of the affected side from which the implant has been removed; A step of aligning the image alignment unit to the healthy image and the image-processed affected image; A guide method for correcting rotational alignment of a femur using artificial intelligence image processing, characterized in that the result display unit includes a step of displaying the result of alignment including the angular difference in rotational alignment of the affected femur and the healthy femur as a result of the alignment.

5. In claim 4, The step of removing the implant within the above image is as follows: A step of restoring image distortion that occurs when the image quality improvement unit acquires the affected-side image and the healthy-side image through artificial intelligence image processing; The step of the image quality improvement unit distinguishing the femur and the implant in the image of the affected side through artificial intelligence image processing and removing the implant; and A guide method for correcting rotational alignment of a femur using artificial intelligence image processing, characterized in that the image quality improvement unit includes a step of restoring the original femur image to the area where the implant was removed.

6. In claim 5, The step of improving the quality by imaging the above-mentioned image is as follows: A step of performing high-resolution processing to improve the resolution of the image of the affected side from which the implant has been removed; A step of performing a high-definition operation to improve clarity on the image of the affected side from which the implant was removed; and A guide method for correcting rotational alignment of a femur using artificial intelligence image processing, characterized in that it includes a step of performing low-noise processing through transparency on the image of the affected side on which the high-resolution processing and the high-definition work have been performed.

7. In claim 6, The step of aligning the above healthy image and the image-processed affected image is as follows: A step of finding the femoral head and femoral shaft in the above healthy side image and the image-processed affected side image, respectively; A step of superimposing the image-processed image of the affected side on the image of the healthy side and aligning the position of the femoral shaft by matching the scale of the femoral head; and A guide method for correcting rotational alignment of a femur using artificial intelligence image processing, characterized by including a step of calculating an angular difference in rotational alignment between the femoral head on the healthy side and the femoral head on the affected side.

Images