[go: up one dir, main page]

WO2022183719A1 - Procédé et dispositif de planification préopératoire à base d'apprentissage profond pour chirurgie de révision de remplacement total de la hanche - Google Patents

Procédé et dispositif de planification préopératoire à base d'apprentissage profond pour chirurgie de révision de remplacement total de la hanche Download PDF

Info

Publication number
WO2022183719A1
WO2022183719A1 PCT/CN2021/120275 CN2021120275W WO2022183719A1 WO 2022183719 A1 WO2022183719 A1 WO 2022183719A1 CN 2021120275 W CN2021120275 W CN 2021120275W WO 2022183719 A1 WO2022183719 A1 WO 2022183719A1
Authority
WO
WIPO (PCT)
Prior art keywords
dimensional
deep learning
prosthesis
pelvis
image
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2021/120275
Other languages
English (en)
Chinese (zh)
Inventor
张逸凌
刘星宇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Longwood Valley Medtech Co Ltd
Original Assignee
Longwood Valley Medtech Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Longwood Valley Medtech Co Ltd filed Critical Longwood Valley Medtech Co Ltd
Publication of WO2022183719A1 publication Critical patent/WO2022183719A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10—Computer-aided planning, simulation or modelling of surgical operations
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
    • A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
    • A61F2/02—Prostheses implantable into the body
    • A61F2/30—Joints
    • A61F2/46—Special tools for implanting artificial joints
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
    • A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
    • A61F2/02—Prostheses implantable into the body
    • A61F2/30—Joints
    • A61F2/46—Special tools for implanting artificial joints
    • A61F2/4603—Special tools for implanting artificial joints for insertion or extraction of endoprosthetic joints or of accessories thereof
    • A61F2/4607—Special tools for implanting artificial joints for insertion or extraction of endoprosthetic joints or of accessories thereof of hip femoral endoprostheses
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
    • A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
    • A61F2/02—Prostheses implantable into the body
    • A61F2/30—Joints
    • A61F2/46—Special tools for implanting artificial joints
    • A61F2/4603—Special tools for implanting artificial joints for insertion or extraction of endoprosthetic joints or of accessories thereof
    • A61F2/4609—Special tools for implanting artificial joints for insertion or extraction of endoprosthetic joints or of accessories thereof of acetabular cups
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10—Computer-aided planning, simulation or modelling of surgical operations
    • A61B2034/101—Computer-aided simulation of surgical operations
    • A61B2034/102—Modelling of surgical devices, implants or prosthesis
    • A61B2034/104—Modelling the effect of the tool, e.g. the effect of an implanted prosthesis or for predicting the effect of ablation or burring
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10—Computer-aided planning, simulation or modelling of surgical operations
    • A61B2034/101—Computer-aided simulation of surgical operations
    • A61B2034/105—Modelling of the patient, e.g. for ligaments or bones
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10—Computer-aided planning, simulation or modelling of surgical operations
    • A61B2034/107—Visualisation of planned trajectories or target regions
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10—Computer-aided planning, simulation or modelling of surgical operations
    • A61B2034/108—Computer aided selection or customisation of medical implants or cutting guides
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
    • A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
    • A61F2/02—Prostheses implantable into the body
    • A61F2/30—Joints
    • A61F2/46—Special tools for implanting artificial joints
    • A61F2002/4632—Special tools for implanting artificial joints using computer-controlled surgery, e.g. robotic surgery
    • A61F2002/4633—Special tools for implanting artificial joints using computer-controlled surgery, e.g. robotic surgery for selection of endoprosthetic joints or for pre-operative planning

Definitions

  • the invention relates to the technical field of artificial intelligence, in particular to a preoperative planning method and device for revision of total hip replacement based on deep learning.
  • Revision of artificial hip replacement refers to restoring joint function by re-implanting a new prosthesis that can be firmly fixed and restoring (or basically restoring) the anatomy of the joint.
  • X-rays are two-dimensional images, which are easily limited by the angle and direction, and are blocked by the previous replacement prosthesis. , it is impossible to make a reasonable judgment on the patient's real bone defect and bone residual volume, resulting in incomplete consideration of revision surgery; for bone and joint CT images, doctors can perform relatively three-dimensional segmentation and observation, but most segmentation methods need to be performed in each Manual positioning or manual segmentation in a CT image is time-consuming, labor-intensive, and inefficient.
  • the existing technology is difficult to accurately extract the original implant in the patient and the remaining bone image after the implant is separated, so that it is difficult to accurately judge the remaining bone volume.
  • the present invention provides a preoperative planning method and equipment for total hip replacement revision based on deep learning, which is used to solve the defects of artificial hip joint revision in the prior art, so that doctors can use artificial intelligence technology to separate the original artificial hip joint.
  • the residual bone volume of the patient who removed the original prosthesis was observed, and the visual matching of the newly inserted prosthesis model was performed.
  • the preoperative planning method for total hip replacement revision based on deep learning of the present invention includes: acquiring image data of the patient's pelvis and femur, obtaining a three-dimensional bone structure based on the pelvis and femur image data, and extracting the original three-dimensional bone structure.
  • Implants wherein, the three-dimensional bone structure includes a three-dimensional pelvis image, a three-dimensional left femur image, and a three-dimensional right femur image; after removing the original implant from the three-dimensional bone structure, the three-dimensional bone structure is calculated Bone defect amount of the structure; identify key points in the three-dimensional bone structure; Select the appropriate prosthesis model in the simulation to visually simulate the matching of the prosthesis.
  • the artificial neural network model is segmented and extracted.
  • obtaining a three-dimensional bone structure based on the pelvis and femur image data, and extracting the original implant in the three-dimensional bone structure includes: : perform image segmentation on the pelvis and femur image data based on a deep learning algorithm; perform 3D reconstruction based on the segmented image data to obtain the 3D pelvis image, 3D left femur image, and 3D right femur image, from the 3D image
  • the original implants are extracted from the skeletal structure and displayed visually.
  • the method further includes: segmenting the image data of the pelvis and femur.
  • an input segmentation adjustment instruction is received; the segmentation of the pelvis and femur image data is adjusted according to the segmentation adjustment instruction.
  • the image segmentation of the pelvis and femur image data based on a deep learning algorithm is: based on a segmentation neural network model Image data for image segmentation; wherein, the segmentation neural network model includes a cascaded first segmentation neural network and a second segmentation neural network; the associated parameters of the first segmentation neural network and the second segmentation neural network are based on Image data in pre-stored medical image databases are determined for training and testing.
  • the first segmentation neural network is FCN, Seg Net, Unet, 3D-Unet, Mask-RCNN, hollow convolution, ENet, CRFasRNN , at least one of PSPNet, ParseNet, RefineNet, ReSeg, LSTM-CF, DeepMask, DeepLabV1, DeepLabV2, DeepLabV3;
  • the second segmentation neural network is at least one of EEfficientDet, SimCLR, and PointRend.
  • the key points include the anterior superior iliac spine, the pubic symphysis, the lesser trochanter, the center of the femoral head, and the medullary canal axis;
  • the key points Recognition is achieved by at least one neural network model among MTCNN, locnet, Pyramid Residual Module, Densenet, hourglass, resnet, SegNet, Unet, R-CNN, Fast R-CNN, Faster R-CNN, R-FCN, and SSD.
  • selecting an appropriate prosthesis model in a database of pre-stored prosthesis models, and performing a visual simulation of the matching of the prosthesis includes the following steps: based on the following steps: For the key points in the three-dimensional bone structure, according to preset rules, calculate the difference in leg length and offset difference before revision; in the three-dimensional pelvis image with the original implant removed, visually place the matching acetabular cup false body model; wherein, the acetabular cup prosthesis model is selected and determined according to the key points, the shape and size of the original implant in the database of the pre-stored prosthesis model; and is determined according to the amount of the bone defect adding a spacer or superimposing an acetabular cup prosthesis model; in at least one of the three-dimensional left femoral image and the three-dimensional right femoral image from which the original implant is removed, according to the amount of bone defect, visually Placement of the revision femoral stem prosthesis model.
  • the deep learning-based preoperative planning method for total hip replacement revision further includes: determining the inappropriate placement positions and placement angles of the acetabular cup prosthesis model and the femoral stem prosthesis model. receiving at least one instruction of adjusting the position and adjusting the angle; according to the instruction of at least one of adjusting the position and adjusting the angle, placing the acetabular cup prosthesis model and the femoral stem prosthesis model The position and placement angle can be adjusted.
  • the deep learning-based preoperative planning method for revision of total hip replacement further includes: simulating an osteotomy operation based on a matched prosthesis model; measuring the vertical distance from the highest point of the femur to the femoral prosthesis stem and the osteotomy height, and calculate the postoperative leg length difference and eccentricity; and simulate the postoperative preview.
  • the artificial neural network model includes: a unet network module, which is used as a backbone network for roughing the original implants. segmentation; pointrend module for precise segmentation based on the coarse segmentation.
  • the original implant is segmented and extracted by using an artificial neural network model, including: obtaining the original implant area marked with the original implant.
  • an image data set of implants and the image data set is divided into a training set and a test set according to a preset ratio; the two-dimensional cross-sectional DICOM data in the image data set is converted into a picture in JPG format, and the annotation file is converted image in png format.
  • the present invention also provides a deep learning-based preoperative planning device for total hip replacement revision, comprising: an acquisition module for acquiring image data of a patient's pelvis and femur, obtaining a three-dimensional bone structure based on the pelvis and femur image data, and extracting The original implant in the three-dimensional bone structure; wherein, the three-dimensional bone structure includes a three-dimensional pelvis image, a three-dimensional left femur image and a three-dimensional right femur image; the original implant is removed from the three-dimensional bone structure After the implant is inserted, the bone defect amount of the three-dimensional bone structure is calculated; the identification module is used to identify the key points in the three-dimensional bone structure; the matching module is used to identify the key points and the original implant based on the key points.
  • the shape and size, as well as the amount of the bone defect select an appropriate prosthesis model from a database of pre-stored prosthesis models, and perform a visualization to simulate the matching of the prosthesis.
  • the present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, when the processor executes the program, the processor implements any one of the above-mentioned based Steps of a deep learning approach to preoperative planning for revision total hip arthroplasty.
  • the present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, implements any one of the above-mentioned deep learning-based total hip replacement revision surgery The steps of the pre-planning method.
  • the preoperative planning method, device, electronic device and storage medium for total hip replacement revision based on deep learning provided by the present invention have the following technical effects:
  • a three-dimensional bone structure is constructed, and the original implant of the patient's total hip joint is extracted at the same time, and then the original implant is removed through reverse extraction to determine the amount of bone defect.
  • the key points in the three-dimensional bone structure based on the key points, the shape and size of the original implant, and the bone defect, select a suitable prosthesis model in the database of pre-stored prosthesis models, and The remaining three-dimensional bone structure is visualized to simulate the matching of the prosthesis.
  • the doctor can know the condition of the bone defect and the shape and size of the original implant before the operation, and on the basis of separating the original implant, select the model and size of the newly implanted implant, and visualize it.
  • the matching of the prosthesis is simulated until the model of the prosthesis to be retrofitted meets the performance requirements.
  • the invention provides technical support for doctors to perform total hip replacement revision, makes the surgical operation more accurate and safer, and promotes the development of the surgical operation in the direction of intelligence, precision and minimal invasion.
  • Fig. 1a is one of the schematic flow charts of the preoperative planning method for total hip replacement revision based on deep learning of the present invention
  • Fig. 1b is a structural block diagram of an artificial neural network model adopted for segmenting the original implant in the preoperative planning method for total hip replacement revision based on deep learning of the present invention
  • 1c is a schematic diagram of the working principle of the original implant based on an artificial neural network model in the preoperative planning method for total hip replacement revision based on deep learning of the present invention
  • Fig. 2 is in the present invention's deep learning-based total hip replacement revision preoperative planning method, the schematic flow chart of obtaining three-dimensional bone image based on pelvis and femur image data;
  • FIG. 3 is a schematic structural diagram of a segmentation neural network model adopted for image segmentation in the preoperative planning method for total hip replacement revision based on deep learning of the present invention
  • FIG. 4 is a schematic diagram of a three-dimensional skeletal structure generated based on three-dimensional reconstruction in the preoperative planning method for total hip replacement revision based on deep learning of the present invention
  • FIG. 5 is a schematic diagram of identifying key points in the preoperative planning method for total hip replacement revision based on deep learning of the present invention
  • Figure 6 is a schematic diagram of a patient's skeletal defect
  • Fig. 7 is a flowchart of the steps of visually simulating the matching of prostheses in the preoperative planning method for revision total hip replacement based on deep learning;
  • FIG. 8 is a front view of placing an acetabular cup prosthesis in the preoperative planning method for revision total hip replacement based on deep learning provided by the present invention
  • Figure 9 is another view of placing the acetabular cup prosthesis in the preoperative planning method for revision total hip replacement based on deep learning provided by the present invention.
  • 10 is one of the effect diagrams of placing a revision femoral stem prosthesis in the preoperative planning method for revision total hip replacement based on deep learning provided by the present invention
  • 11 is the second schematic flowchart of the preoperative planning method for revision of total hip replacement based on deep learning provided by the present invention.
  • FIG. 12 is a schematic diagram after performing an osteotomy in the deep learning-based preoperative planning method for total hip replacement revision provided by the present invention.
  • 13 is the third schematic flowchart of the preoperative planning method for revision total hip replacement based on deep learning provided by the present invention.
  • FIG. 14 is a schematic structural diagram of a preoperative planning device for total hip replacement revision based on deep learning provided by the present invention.
  • FIG. 15 is a schematic structural diagram of an electronic device provided by the present invention.
  • Fig. 1a is a schematic flowchart of a preoperative planning method for total hip replacement revision based on deep learning provided by the present invention, including the following steps:
  • Step S110 acquiring image data of the patient's pelvis and femur, obtaining a three-dimensional bone structure based on the pelvis and femur image data, and extracting the original implant in the three-dimensional bone structure; removing the original implant in the three-dimensional bone structure Then, the bone defect amount of the three-dimensional bone structure is calculated.
  • the three-dimensional bone structure includes a three-dimensional pelvis image, a three-dimensional left femur image, and a three-dimensional right femur image.
  • Step S120 identifying key points in the three-dimensional bone structure.
  • Step S130 based on the key points, the shape and size of the original implant, select an appropriate prosthesis model in the database of pre-stored prosthesis models, and perform the matching of the visually simulated prosthesis.
  • the original implant may include at least one of an original prosthesis and other metal implants, the prosthesis is generally made of metal, and the original prosthesis is a metal prosthesis.
  • the pelvis and femur image data can be CT (Computed Tomography, electronic computed tomography) image data in the format of digital imaging and communications (DICOM, Digital Imaging and Communications in Medicine), or magnetic resonance imaging (MRI, Magnetic Resonance Imaging) image data, but the present invention is not limited to this, and other pelvis and femur image data can also be used in the present invention.
  • CT Computer Tomography, electronic computed tomography
  • DICOM Digital Imaging and Communications in Medicine
  • MRI Magnetic Resonance Imaging
  • Step S110 the original implant is extracted by the following methods:
  • an image dataset with the original implant that annotates the original implant area that is, obtain an image dataset with the original implant, manually label the original implant area, and as an image dataset.
  • the image dataset is divided into a training set and a test set according to a preset ratio.
  • the artificial neural network model includes: a unet network module and a pointrend module.
  • the artificial neural network used in this embodiment will be further described below.
  • the first stage uses 4 downsampling to learn the deep features of the image, followed by 4 upsampling to re-store the feature map into the image, where each downsampling layer includes 2 convolutional layers and 1 pooling layer, The size of the convolution kernel is 3*3, the size of the convolution kernel in the pooling layer is 2*2, and the number of convolution kernels in each convolution layer is 128, 256, 512; each upsampling layer includes 1 upsampling layer and 2 convolutional layers, where the convolution kernel size of the convolutional layer is 3*2, the size of the convolution kernel in the upsampling layer is 2*2, and the convolution kernel in each upsampling layer The number is 512, 256, 128. There is a dropout layer after the last upsampling, and the dropout rate is set to 0.7. All convolutional layers are followed by activation functions as relu functions.
  • the background pixel value of the data label is set to 0, the femur is 1, the tibia is 2, the fibula is 3, the patella is 4, the training batch_size is 6, the learning rate is set to 1e-4, and the optimizer Using the Adam optimizer, the loss function used is DICE loss, and all the training sets are sent to the network for training. According to the change of the loss function during the training process, the size of the training batch is adjusted, and the rough segmentation results of each part are finally obtained. After entering the pointrend module, bilinear interpolation is used to upsample the prediction results of the previous step, and then the N most uncertain points are selected in this denser feature map, such as points with a probability close to 0.5.
  • the output of the artificial neural network model is the metal prosthesis part as the original implant. Referring to Fig. 1c, a schematic diagram of the working principle of the original implant based on the artificial neural network model is shown.
  • the three-dimensional bone structure is constructed based on deep learning through the image data of the patient's pelvis and femur, and the original implant of the patient's total hip joint is extracted at the same time, and then the original implant is removed by reverse extraction, Determine the amount of bone defect.
  • the appropriate prosthesis model in the database of pre-stored prosthesis models, and the remaining three-dimensional bone structure Perform a visual simulation of the fitting of the prosthesis.
  • the doctor can know the condition of the bone defect and the shape and size of the original implant before the operation, and select the model and size of the newly implanted prosthesis on the basis of separating the original implant. Visually simulate the fit of the prosthesis until the model of the prosthesis to be retrofitted meets the performance requirements.
  • the invention provides technical support for doctors to perform total hip replacement revision, makes the surgical operation more accurate and safer, and promotes the development of the surgical operation in the direction of intelligence, precision and minimal invasion.
  • Step S110 acquiring image data of the patient's pelvis and femur, obtaining a three-dimensional bone structure based on the pelvis and femur image data, and extracting the original implant in the three-dimensional bone structure.
  • obtaining the three-dimensional skeletal structure based on the pelvis and femur image data may be performed by means of a deep learning algorithm in artificial intelligence.
  • 3D reconstruction is performed based on the segmented image data, a 3D pelvis image, a 3D left femur image, a 3D right femur image, and the original implant are obtained and visualized.
  • Fig. 2 is a schematic flowchart of obtaining a three-dimensional skeletal image based on pelvis and femur image data in the preoperative planning method for total hip replacement revision based on deep learning provided by the present invention, including the following steps:
  • Step S1101 acquiring image data of the pelvis and the femur.
  • Step S1102 Perform image segmentation on the pelvis and femur image data based on the deep learning algorithm in artificial intelligence.
  • Artificial Intelligence is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. Artificial intelligence is a branch of computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can respond in a similar way to human intelligence. Research in this field includes robotics, language recognition, image recognition, Natural language processing and expert systems, etc. Artificial intelligence can simulate the information process of human consciousness and thinking.
  • Deep Learning is a new research direction in the field of Machine Learning (ML), which is introduced into Machine Learning to make it closer to its original goal - Artificial Intelligence. Deep learning is to learn the inherent laws and representation levels of sample data, and the information obtained during these learning processes is of great help to the interpretation of data such as text, images, and sounds. Its ultimate goal is to enable machines to have the ability to analyze and learn like humans, and to recognize data such as words, images, and sounds.
  • ML Machine Learning
  • the deep learning algorithm is a segmentation neural network model, that is, image segmentation is performed on image data based on the segmentation neural network model.
  • the associated parameters of the segmentation neural network model are determined by training and testing based on image datasets in a medical image database (eg, a lower extremity medical image database).
  • a medical image database eg, a lower extremity medical image database
  • the image dataset in the lower extremity medical image database is an image dataset that annotates the pelvis, left femur, right femur and related parts, and the image dataset is divided into training set and test set; based on the training set and test set, Train a segmentation neural network model.
  • FIG. 3 there is shown a schematic structural diagram of a segmentation neural network model used for image segmentation in the deep learning-based preoperative planning method for total hip replacement revision according to the present invention.
  • the segmentation neural network model includes a cascaded first segmentation neural network 1 and a second segmentation neural network 2 .
  • the input information of the segmentation neural network model is the pelvis and femur image data, for example, the pelvis and femur image data A1 shown in FIG. 3, the pelvis and femur image data A2, the pelvis and femur image data A3... Image data An-1, and, pelvis and femur image data An.
  • the output end of the segmentation neural network is connected to the input end of the 3D reconstruction module 3, and through 3D reconstruction, a 3D bone structure is generated, including a 3D pelvis image, a 3D left femur image, a 3D right femur image, and the original implant.
  • the first segmentation neural network 1 may include FCN, SegNet, Unet, 3D-Unet, Mask-RCNN, Atrous Convolution, ENet, CRFasRNN, PSPNet, ParseNet, RefineNet, ReSeg, LSTM-CF , at least one of DeepMask, DeepLabV1, DeepLabV2, and DeepLabV3.
  • the second segmentation neural network 2 may include at least one of EEfficientDet, SimCLR, and PointRend.
  • the associated parameters of the first segmentation neural network 1 and the second segmentation neural network 2 are determined by training and testing based on image data in a pre-stored medical image database.
  • Step S1103 Perform three-dimensional reconstruction based on the segmented image data to obtain a three-dimensional pelvis image, a three-dimensional left femur image, a three-dimensional right femur image, and the original implant.
  • 3D reconstruction refers to the establishment of a mathematical model suitable for computer representation and processing of three-dimensional objects. It is the basis for processing, operating and analyzing its properties in a computer environment. Technology.
  • Step S1104 Visually display the 3D reconstructed 3D pelvis image, the 3D left femur image, and the 3D right femur image, and extract the original implant therefrom.
  • Extract the original implant mainly including the original prosthesis, and test the original prosthesis to determine its shape and size.
  • FIG. 4 shows the three-dimensional bone structure generated based on the three-dimensional reconstruction in the deep learning-based preoperative planning method for total hip replacement revision according to the present invention. It can be clearly seen from FIG. The right femur 4a was reconstructed, the left femur 4b was 3D reconstructed, and the original metal implant 4d was 3D reconstructed. Based on this, it can be seen that the segmentation method of the present application greatly reduces the influence of artifacts.
  • Step S1105 determine whether the image segmentation that is the basis of the 3D pelvis image, the 3D left femur image, and the 3D right femur image needs to be optimized, if it needs to be optimized, go to step S1106; .
  • step S1102 it is determined whether the segmentation of the pelvis and femur image data in step S1102 is reasonable. Whether it is reasonable or not can be determined by manual inspection or by equipment detection.
  • cross-sectional CT, sagittal CT, and coronal CT images located on the left.
  • Cross-sectional CT, sagittal CT, coronal CT images and 3D bone images can realize three-axis linkage, and can be observed simultaneously through 2D and 3D views. You can also adjust the transparency or opacity of the 3D reconstructed bones, and adjust the display or hide state of each segmented bone.
  • Step S1106 receiving the input division adjustment instruction, and returning to step S1102. Until the pelvis, left femur, right femur and original implant can be displayed independently and completely.
  • Step S1107 the 3D bone structure generation operation is ended.
  • step S120 key points in the three-dimensional skeletal structure are identified.
  • key points are identified from a three-dimensional skeletal structure such as FIG. 4, and the key point identification can be achieved by an artificial neural network model.
  • an artificial neural network model For example, it can be implemented for at least one neural network model among MTCNN, locnet, Pyramid Residual Module, Densenet, hourglass, resnet, SegNet, Unet, R-CNN, Fast R-CNN, Faster R-CNN, R-FCN, SSD .
  • the identified key points may include: the anterior superior iliac spine, the symphysis pubis, the lesser trochanter, the center of the femoral head, and the medullary canal axis.
  • the anterior superior iliac spine, the pubic symphysis, the lesser trochanter, the center of the femoral head, and the axis of the medullary cavity are used as bony landmarks, which play an important role in clinical reference in terms of position and angle in total hip surgery, and provide clinical data measurement .
  • FIG. 5 is a schematic diagram of identifying key points in the preoperative planning method for revision total hip replacement based on deep learning of the present invention.
  • the viewing angle of the pelvis and/or femur can be adjusted.
  • human observation can be used to detect whether the identification is correct, and the parts that need to be adjusted are manually marked.
  • step S130 will be described.
  • step S130 based on the key points obtained in step S120 and the shape and size of the original implant, a suitable hip joint prosthesis model is selected from the database of pre-stored prosthesis models, and the matching of the visual simulation prosthesis is performed.
  • the database storing the prosthesis model is data pre-stored in the system. It mainly stores the prosthesis models related to the total hip joint for total hip replacement surgery. Models vary in size and size.
  • Design of prosthetic models related to the total hip joint In one embodiment, a normal human hip joint can be scanned by CT, and digital technology can be used to measure the joint shape and the shape after osteotomy, and then a digital joint model database can be established to provide morphology for the design of the total hip prosthesis model. data.
  • step S120 the shape and size of the original implant based on the three-dimensional bone image of the patient are determined.
  • the system searches for matching objects in the database of pre-stored prosthetic models and makes intelligent recommendations.
  • the model, placement position and placement angle of the hip joint prosthesis model are given.
  • the intelligently recommended prosthesis model is matched and displayed on the 3D bone structure.
  • an indicator of the amount of bone defect can also be added, here The amount of bone defect obtained by removing the original implant in the three-dimensional bone structure.
  • Step a reverse extraction, remove the original implant in the three-dimensional bone structure, and obtain the remaining three-dimensional bone structure;
  • step b) by comparing the three-dimensional femoral structure with the remaining three-dimensional bone structure, the condition of the bone defect is determined, and the amount of the bone defect is determined.
  • step 130 can be improved to select an appropriate prosthesis model in a database of pre-stored prosthesis models based on the key points, the shape and size of the original implant, and the amount of bone defect.
  • Figure 6 shows a patient with a bone defect. Specifically, the three-dimensional reconstruction model 6a of the pelvis with the original implant removed, and the acetabular bone defect 6b are shown. After the original prosthesis was extracted from the acetabulum, the defect of the acetabular floor was observed. That is to say, after extracting and hiding the original prosthesis, the bone defect can be clearly observed.
  • the surgical strategy can be preliminarily determined according to the skeletal defect.
  • FIG. 7 is a flowchart of the steps of visually simulating the matching of the prosthesis in the preoperative planning method for revision total hip replacement based on deep learning, including the following steps:
  • Step S1301 based on the key points in the three-dimensional skeletal structure, according to preset rules, calculate the difference in leg length and offset difference before revision;
  • Step S1302 in the three-dimensional pelvis image with the original implant removed, visually place the matched acetabular cup prosthesis model.
  • the acetabular cup prosthesis model is selected and determined according to the key points, the shape and size of the original implant in the database of the pre-stored prosthesis model; and the block or superposition operation is added according to the amount of bone defect;
  • Step S1303 in the 3D left femur image and/or the 3D right femur image with the original implant removed, visually place the revised femoral stem prosthesis model according to the amount of bone defect.
  • Step S1304 determine whether the placement position and placement angle of the acetabular cup prosthesis model and the femoral stem prosthesis model are appropriate, if the placement angle is not appropriate, execute step S1305, if the placement angle is appropriate, execute step S1306.
  • Step S1305 receive an instruction to adjust the position and/or angle, and adjust;
  • step S1306 the matching of the visual simulation prosthesis ends.
  • the spacers can be added or the acetabular cups can be superimposed to form a double-cup and three-cup structure to form a cup on cup scheme, and the simulation matching effect can be observed in real time;
  • the selection of 3D acetabular and 3D femoral prosthesis includes selection Prosthesis type and/or prosthesis size and/or three-dimensional space position; prostheses can be displayed and/or not displayed in any combination.
  • FIG. 8 is a front view of the placement of the acetabular cup prosthesis, from FIG. 8 can be seen three-dimensional reconstruction of the pelvis 8a and the acetabular cup prosthesis 8b.
  • Figure 9 is another view of the placement of the acetabular cup prosthesis, from Figure 8 a three-dimensional reconstruction of the pelvis 9a and the acetabular cup prosthesis 9b can be seen.
  • FIG. 9 is an effect diagram of placing an acetabular cup provided by the present invention, as shown in the figure.
  • the figure shows that the acetabular cup implanted in the pelvis can choose the appropriate model, and the acetabular cup can be moved back and forth and rotated, and the anteversion angle of 20°, the abduction angle of 40 degrees, and the coverage can be shown in the lower corner of the figure.
  • the rate is 97%.
  • Fig. 10 is one of the effect diagrams of placing a revision femoral stem prosthesis provided by the present invention, as shown in the figure.
  • the figure shows the implantation of the pelvis and the right leg. From Figure 10, it can be seen that the three-dimensional reconstruction of the right femur 10a, the three-dimensional reconstruction of the left femur 10b, the three-dimensional reconstruction of the pelvis 10c, and the femoral stem prosthesis 10d .
  • FIG. 11 is the second schematic flowchart of the preoperative planning method for total hip replacement revision based on deep learning provided by the present invention, comprising the following steps:
  • Step S1110 acquiring image data of the patient's pelvis and femur, obtaining a three-dimensional bone structure based on the pelvis and femur image data, and extracting the original implant in the three-dimensional bone structure;
  • Step S1120 reversely extracting and removing the remaining three-dimensional bone structure of the original implant, and determining the amount of bone defect
  • Step S1130 identifying key points in the three-dimensional bone structure
  • Step S1140 selecting an appropriate prosthesis model from a database of pre-stored prosthetic models based on key points, the shape and size of the original implant, and the amount of bone defect;
  • Step S1150 simulating an osteotomy operation based on the matched prosthesis model
  • Step S1160 measuring the vertical distance from the highest point of the femur to the femoral prosthesis stem and the osteotomy height, and calculating the postoperative leg length difference and eccentricity;
  • Step S1170 simulate postoperative preview.
  • the simulated matching effect can be observed in one or more states, including osteotomy or non-osteotomy, and bone transparent or opaque.
  • the actual osteotomy position is determined, the femoral stem and the acetabular cup are matched, the vertical distance from the highest point of the femur to the femoral prosthesis stem and the osteotomy height are measured, and the postoperative leg length difference and eccentricity are calculated. distance.
  • FIG. 12 is a schematic diagram after performing an osteotomy operation in the deep learning-based preoperative planning method for total hip replacement revision provided by the present invention.
  • Fig. 13 shows a flow chart of steps of an embodiment of the preoperative planning method for revision total hip replacement based on deep learning of the present invention.
  • the preoperative planning method and system provided by the present invention realize the automatic segmentation of the pelvis, left femur, right femur and metal implant based on deep learning, which improves segmentation efficiency and accuracy.
  • the system can fill the bone defect by adding spacers or superimposing the acetabular cup according to the acetabular bone defect after separating the original implant, that is, the cup on cup scheme.
  • the size and position of the implanted prosthesis can be determined before the operation, and whether the prosthesis meets the performance requirements can be virtually tested on the basis of separating the original metal implant, so as to Optimize the reconstruction of the articular surface and the determination of the position of the prosthesis; provide technical support for doctors to make surgery more accurate and safer; and promote the development of surgery in the direction of intelligence, precision, and minimal invasion.
  • FIG. 14 is a schematic structural diagram of a revision device for total hip replacement provided by the present invention, including an acquisition module 1401 , an identification module 1402 and a matching module 1403 .
  • the acquisition module 1401 is configured to acquire image data of the patient's pelvis and femur, obtain a three-dimensional bone structure based on the pelvis and femur image data, and extract the original implants in the three-dimensional bone structure; wherein, the three-dimensional bone structure includes three-dimensional pelvic images, 3D left femur image and 3D right femur image;
  • the identification module 1402 is configured to identify key points in the three-dimensional skeletal structure
  • the matching module 1403 is configured to select an appropriate prosthesis model in the database of pre-stored prosthesis models based on the key points, the shape and size of the original implant, and perform the matching of the visually simulated prosthesis.
  • the three-dimensional bone structure is constructed based on deep learning through the image data of the patient's pelvis and femur, and the original implant of the patient's total hip joint is extracted at the same time, and then the original implant is removed by reverse extraction, Determine the amount of bone defect.
  • the appropriate prosthesis model in the database of pre-stored prosthesis models, and the remaining three-dimensional bone structure Perform a visual simulation of the fitting of the prosthesis.
  • the doctor can know the condition of the bone defect and the shape and size of the original implant before the operation, and select the model and size of the newly implanted prosthesis on the basis of separating the original implant. Visually simulate the fit of the prosthesis until the model of the prosthesis to be retrofitted meets the performance requirements.
  • the invention provides technical support for doctors to perform total hip replacement revision, makes the surgical operation more accurate and safer, and promotes the development of the surgical operation in the direction of intelligence, precision and minimal invasion.
  • the matching module 1403 further selects an appropriate prosthesis model from a database of pre-stored prosthesis models based on key points, the shape and size of the original implant, and the amount of bone defect; wherein, The amount of bone defect was calculated by removing the original implant from the three-dimensional bone structure.
  • the acquisition module 1401 further includes: a segmentation unit 1401A and a reconstruction unit 1401B, wherein:
  • the segmentation unit 1401A is used to perform image segmentation on the pelvis and femur image data based on the deep learning algorithm
  • the reconstruction unit 1401B is configured to perform 3D reconstruction based on the segmented image data, obtain a 3D pelvis image, a 3D left femur image, a 3D right femur image, and the original implant, and visualize them.
  • an adjustment unit 1401C is also provided after the reconstruction unit, and the adjustment unit 1401C is used to determine whether the segmentation of the pelvis and femur image data needs to be optimized. split to adjust.
  • the segmentation unit 1401A is further configured to: perform image segmentation on the pelvis and femur image data based on the segmentation neural network model; wherein, the segmentation neural network model includes a cascaded first segmentation neural network and a second segmentation neural network; the first segmentation neural network The parameters associated with the second segmentation neural network are determined by training and testing based on pre-stored image data in a medical image database.
  • the first segmentation neural network is any of FCN, SegNet, Unet, 3D-Unet, Mask-RCNN, Hollow Convolution, ENet, CRFasRNN, PSPNet, ParseNet, RefineNet, ReSeg, LSTM-CF, DeepMask, DeepLabV1, DeepLabV2, DeepLabV3. one or more; and/or, the second segmentation neural network is any one or more of EEfficientDet, SimCLR, and PointRend.
  • the key points include the anterior superior iliac spine, the pubic symphysis, the lesser trochanter, the center of the femoral head, and the axis of the medullary cavity; the key points are identified through MTCNN, locnet, Pyramid Residual Module, Densenet, hourglass, resnet, SegNet, Unet , R-CNN, Fast R-CNN, Faster R-CNN, R-FCN, SSD any one or more neural network model implementation.
  • the matching module 1403 is also configured to perform the following operations:
  • the matched acetabular cup prosthesis model is placed visually; wherein, the acetabular cup prosthesis model is based on key points and original implant models in the database of pre-stored prosthesis models.
  • the shape and size of the input object are selected and determined; and according to the amount of bone defect, add spacers or superimpose operations;
  • the revised femoral stem prosthesis model is visually placed according to the amount of bone defect.
  • the matching module 1403 is further configured to perform: judging whether the placement position and placement angle of the acetabular cup prosthesis model and the femoral stem prosthesis model are appropriate, and if not, receiving an instruction to adjust the position and/or angle to adjust.
  • the apparatus also includes a preview module 1404 configured to perform: simulating an osteotomy operation based on the matched prosthesis model; measuring the vertical distance from the highest point of the femur to the femoral prosthesis stem and the osteotomy height, and calculating the postoperative leg length difference and eccentricity; and, simulated postoperative preview.
  • FIG. 15 illustrates a schematic diagram of the physical structure of an electronic device.
  • the electronic device may include: a processor (processor) 1510, a communication interface (Communications Interface) 1520, a memory (memory) 1530 and a communication bus 1540,
  • the processor 1510 , the communication interface 1520 , and the memory 1530 communicate with each other through the communication bus 1540 .
  • the processor 1510 can invoke the logic instructions in the memory 1530 to execute a deep learning-based preoperative planning method for total hip replacement revision, the method comprising: acquiring image data of the patient's pelvis and femur, and obtaining a three-dimensional bone structure based on the pelvis and femur image data , and extract the original implants in the three-dimensional bone structure; wherein, the three-dimensional bone structure includes the three-dimensional pelvis image, the three-dimensional left femur image and the three-dimensional right femur image; identify the key points in the three-dimensional bone structure; Depending on the shape and size of the implant, a suitable prosthesis model is selected from the database of pre-stored prosthesis models to visually simulate the matching of the prosthesis.
  • the above-mentioned logic instructions in the memory 1530 may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as an independent product.
  • the technical solution of the present invention can be embodied in the form of a software product in essence, or the part that contributes to the prior art or the part of the technical solution.
  • the computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present invention.
  • the aforementioned storage medium includes: U disk, mobile hard disk, Read-Only Memory (ROM, Read-Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes .
  • the present invention also provides a computer program product
  • the computer program product includes a computer program stored on a non-transitory computer-readable storage medium
  • the computer program includes program instructions
  • the program instructions when executed by a computer, the computer can execute
  • the deep learning-based preoperative planning method for total hip replacement revision includes: acquiring image data of the patient's pelvis and femur, obtaining a three-dimensional bone structure based on the pelvic and femoral image data, and extracting the three-dimensional bone structure.
  • the 3D bone structure includes 3D pelvis image, 3D left femur image and 3D right femur image; identify key points in the 3D bone structure; based on the key points, the shape and size of the original implant , select an appropriate prosthesis model from the database of pre-stored prosthesis models, and perform a visual simulation of the matching of the prosthesis.
  • the present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the computer program is implemented to perform the deep learning-based total hip replacement revision provided by each of the above.
  • Preoperative planning methods that include:
  • the patient's pelvis and femur image data obtain the three-dimensional bone structure based on the pelvis and femur image data, and extract the original implants in the three-dimensional bone structure;
  • the three-dimensional bone structure includes the three-dimensional pelvis image, the three-dimensional left femur image and the three-dimensional right femur image.
  • Lateral femur image identify key points in the 3D bone structure; based on the key points, the shape and size of the original implant, select the appropriate prosthesis model in the database of pre-stored prosthesis models, and visually simulate the matching of the prosthesis .
  • the device embodiments described above are only illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place , or distributed to multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.
  • each embodiment can be implemented by means of software plus a necessary general hardware platform, and certainly can also be implemented by hardware.
  • the above-mentioned technical solutions can be embodied in the form of software products in essence or the parts that make contributions to the prior art, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic Disks, optical discs, etc., include instructions for causing a computer device (which may be a personal computer, server, or network device, etc.) to perform the methods of various embodiments or portions of embodiments.

Landscapes

  • Health & Medical Sciences (AREA)
  • Transplantation (AREA)
  • Orthopedic Medicine & Surgery (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Biomedical Technology (AREA)
  • Veterinary Medicine (AREA)
  • Public Health (AREA)
  • Physical Education & Sports Medicine (AREA)
  • Cardiology (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Vascular Medicine (AREA)
  • Surgery (AREA)
  • Molecular Biology (AREA)
  • Robotics (AREA)
  • Medical Informatics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Image Analysis (AREA)

Abstract

La présente invention concerne un procédé et un dispositif de planification préopératoire à base d'apprentissage profond pour une chirurgie de révision de remplacement total de la hanche. Le procédé comprend : l'acquisition de données d'image du bassin et du fémur d'un patient, l'obtention d'une structure osseuse tridimensionnelle sur la base des données d'image du bassin et du fémur, et l'extraction d'un implant d'origine dans la structure osseuse tridimensionnelle ; l'identification d'un point clé dans la structure osseuse tridimensionnelle ; et la sélection, sur la base du point clé et de la forme et de la taille de l'implant original, un modèle de prothèse adapté dans une base de données pour préstocker des modèles de prothèse, de façon à effectuer un alignement de prothèse visualisé et simulé. L'invention fournit une assistance technique pour les médecins pour effectuer une révision de remplacement total de la hanche, de sorte que les opérations chirurgicales sont plus précises et plus sûres, de façon à favoriser le développement intelligent, précis et mini-invasif d'opérations chirurgicales.
PCT/CN2021/120275 2021-03-02 2021-09-24 Procédé et dispositif de planification préopératoire à base d'apprentissage profond pour chirurgie de révision de remplacement total de la hanche Ceased WO2022183719A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110231578.8A CN112971981B (zh) 2021-03-02 2021-03-02 基于深度学习的全髋关节图像处理方法和设备
CN202110231578.8 2021-03-02

Publications (1)

Publication Number Publication Date
WO2022183719A1 true WO2022183719A1 (fr) 2022-09-09

Family

ID=76352114

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/120275 Ceased WO2022183719A1 (fr) 2021-03-02 2021-09-24 Procédé et dispositif de planification préopératoire à base d'apprentissage profond pour chirurgie de révision de remplacement total de la hanche

Country Status (2)

Country Link
CN (1) CN112971981B (fr)
WO (1) WO2022183719A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11944392B2 (en) 2016-07-15 2024-04-02 Mako Surgical Corp. Systems and methods for guiding a revision procedure
US12370061B2 (en) 2019-08-29 2025-07-29 Mako Surgical Corp. Robotic surgery system for augmented arthroplasty procedures

Families Citing this family (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112971981B (zh) * 2021-03-02 2022-02-08 北京长木谷医疗科技有限公司 基于深度学习的全髋关节图像处理方法和设备
CN113744214B (zh) * 2021-08-24 2022-05-13 北京长木谷医疗科技有限公司 基于深度强化学习的股骨柄放置装置及电子设备
CN113962927B (zh) * 2021-09-01 2022-07-12 北京长木谷医疗科技有限公司 基于强化学习的髋臼杯位置调整方法、装置及存储介质
CN113870229B (zh) * 2021-09-29 2022-03-11 北京长木谷医疗科技有限公司 股骨近端髓腔图像数据处理方法、装置、设备和存储介质
CN114092466B (zh) * 2021-12-01 2024-08-23 杭州柳叶刀机器人有限公司 使用深度学习辅助计算腿长的方法与系统、可读存储介质
CN113870261B (zh) * 2021-12-01 2022-05-13 杭州柳叶刀机器人有限公司 用神经网络识别力线的方法与系统、存储介质及电子设备
CN114419618B (zh) * 2022-01-27 2024-02-02 北京长木谷医疗科技股份有限公司 基于深度学习的全髋关节置换术前规划系统
CN114663363B (zh) * 2022-03-03 2023-11-17 四川大学 一种基于深度学习的髋关节医学图像处理方法和装置
CN114431957B (zh) * 2022-04-12 2022-07-29 北京长木谷医疗科技有限公司 基于深度学习的全膝关节置换术后翻修术前规划系统
CN115393272B (zh) * 2022-07-15 2023-04-18 北京长木谷医疗科技有限公司 基于深度学习的膝关节髌骨置换三维术前规划系统及方法
CN115252233B (zh) * 2022-07-28 2024-04-26 大连理工大学 基于深度学习的全髋关节置换术中髋臼杯的自动化规划方法
CN114983566B (zh) * 2022-08-02 2022-12-13 北京壹点灵动科技有限公司 假体翻修的配准方法及装置、存储介质和处理器
CN115381553B (zh) * 2022-09-21 2023-04-07 北京长木谷医疗科技有限公司 复杂性骨性融合膝关节的智能定位装置设计方法及系统
CN116211458B (zh) * 2022-12-12 2023-10-03 高峰医疗器械(无锡)有限公司 种植体规划方法、装置、设备及存储介质
CN116172763A (zh) * 2022-12-27 2023-05-30 杭州键嘉医疗科技股份有限公司 一种端到端的髋关节置换手术规划方案智能化生成装置和方法
CN116350349B (zh) * 2023-05-31 2023-07-28 中日友好医院(中日友好临床医学研究所) 一种基于cjfh分型的保髋手术治疗系统及装置
CN116650110B (zh) * 2023-06-12 2024-05-07 北京长木谷医疗科技股份有限公司 基于深度强化学习的膝关节假体自动放置方法及装置
CN119488389A (zh) * 2023-08-16 2025-02-21 武汉联影智融医疗科技有限公司 髋关节假体确定方法、装置和计算机设备
CN117009826B (zh) * 2023-08-30 2024-04-02 北京长木谷医疗科技股份有限公司 一种基于深度学习的髋臼杯角度自适应确定方法及装置
CN117853458B (zh) * 2024-01-10 2024-08-16 北京长木谷医疗科技股份有限公司 翻修手术机器人植入物智能识别评估方法、装置及设备
CN118983058B (zh) * 2024-07-24 2025-08-15 中日友好医院(中日友好临床医学研究所) 一种股骨头坏死临床决策支持系统及装置
CN118806432B (zh) * 2024-07-31 2025-08-22 北京长木谷医疗科技股份有限公司 一种基于强化学习的髋臼周围截骨规划方法及装置
CN119423978B (zh) * 2024-10-09 2025-10-17 北京长木谷医疗科技股份有限公司 一种基于人工智能的保髋手术术前规划方法及装置

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106999245A (zh) * 2014-12-08 2017-08-01 思想外科有限公司 用于全关节成形术中的基于计划、数字化、以及配准的植入物
CN110648337A (zh) * 2019-09-23 2020-01-03 武汉联影医疗科技有限公司 髋关节分割方法、装置、电子设备和存储介质
CN111179350A (zh) * 2020-02-13 2020-05-19 张逸凌 基于深度学习的髋关节图像处理方法及计算设备
CN111938813A (zh) * 2020-07-21 2020-11-17 南京市第一医院 一种髋关节翻修手术的术前规划方法
WO2021034706A1 (fr) * 2019-08-16 2021-02-25 Tornier, Inc. Planification préopératoire de procédures de révision chirurgicale pour des articulations orthopédiques
CN112971981A (zh) * 2021-03-02 2021-06-18 北京长木谷医疗科技有限公司 基于深度学习的全髋关节置换翻修术前规划方法和设备

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5318571A (en) * 1992-05-14 1994-06-07 Benson Gail M Method and apparatus for simplifying total hip arthroplasty
US9023112B2 (en) * 2011-02-24 2015-05-05 Depuy (Ireland) Maintaining proper mechanics THA
US9131945B2 (en) * 2013-03-11 2015-09-15 DePuy Synthes Products, Inc. Customized patient-specific revision surgical instruments and method
KR102275444B1 (ko) * 2013-03-15 2021-07-08 씽크 써지컬, 인크. 전 관절 재성형술을 위한 시스템 및 프로세스

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106999245A (zh) * 2014-12-08 2017-08-01 思想外科有限公司 用于全关节成形术中的基于计划、数字化、以及配准的植入物
WO2021034706A1 (fr) * 2019-08-16 2021-02-25 Tornier, Inc. Planification préopératoire de procédures de révision chirurgicale pour des articulations orthopédiques
CN110648337A (zh) * 2019-09-23 2020-01-03 武汉联影医疗科技有限公司 髋关节分割方法、装置、电子设备和存储介质
CN111179350A (zh) * 2020-02-13 2020-05-19 张逸凌 基于深度学习的髋关节图像处理方法及计算设备
CN111938813A (zh) * 2020-07-21 2020-11-17 南京市第一医院 一种髋关节翻修手术的术前规划方法
CN112971981A (zh) * 2021-03-02 2021-06-18 北京长木谷医疗科技有限公司 基于深度学习的全髋关节置换翻修术前规划方法和设备

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11944392B2 (en) 2016-07-15 2024-04-02 Mako Surgical Corp. Systems and methods for guiding a revision procedure
US12343093B2 (en) 2016-07-15 2025-07-01 Mako Surgical Corp. Systems and methods for guiding a revision procedure
US12370061B2 (en) 2019-08-29 2025-07-29 Mako Surgical Corp. Robotic surgery system for augmented arthroplasty procedures

Also Published As

Publication number Publication date
CN112971981B (zh) 2022-02-08
CN112971981A (zh) 2021-06-18

Similar Documents

Publication Publication Date Title
WO2022183719A1 (fr) Procédé et dispositif de planification préopératoire à base d'apprentissage profond pour chirurgie de révision de remplacement total de la hanche
US12412666B2 (en) Systems and methods for simulating spine and skeletal system pathologies
US12089900B2 (en) Systems and methods for modeling spines and treating spines based on spine models
US11869670B2 (en) Systems and methods for automated segmentation of patient specific anatomies for pathology specific measurements
CN114170128B (zh) 基于深度学习的骨骼分割方法和系统
AU2020101836A4 (en) A method for generating femoral x-ray films based on deep learning and digital reconstruction of radiological image
WO2022170768A1 (fr) Procédé et appareil de traitement d'image d'articulation unicondylienne, dispositif et support de stockage
US10874460B2 (en) Systems and methods for modeling spines and treating spines based on spine models
WO2022042459A1 (fr) Procédé et système de planification préopératoire de remplacement total du genou basés sur l'apprentissage profond, et support associé
WO2024001140A1 (fr) Procédé et appareil de segmentation de sous-région de corps vertébral, et support de stockage
JP2021176573A (ja) 骨再建及び整形外科インプラント
US8644568B1 (en) Automated patient-specific bone-implant biomechanical analysis
CN100576210C (zh) 用于一组骨关节的生物力学模拟方法
CN115252233B (zh) 基于深度学习的全髋关节置换术中髋臼杯的自动化规划方法
KR20210019136A (ko) 수술실에서 관성 측정 장치를 등록하는 방법
AU2022217138B2 (en) Computer-assisted surgical planning
US20230085093A1 (en) Computerized prediction of humeral prosthesis for shoulder surgery
KR102877777B1 (ko) 수술 전 수술 플래닝을 용이하게 하기 위해 생리학적으로 건강하고 그리고 생리학적으로 결함이 있는 해부학적 구조의 재구성 및 특성화를 위한 시스템 및 방법
CN118902705A (zh) 一种人工髋关节假体型号匹配及安全位置选择方法和系统
WO2024059693A2 (fr) Génération d'images médicales synthétiques, données de caractéristiques pour la segmentation d'images d'apprentissage, et images médicales peintes à l'aide de modèles génératifs
CN110811829B (zh) 一种基于股骨旋转轴与内翻轴分析模型的构建方法及系统
CN114947905B (zh) 一种生成髋臼月状面结构模型的系统及方法
Zhou et al. Improving inter-fragmentary alignment for virtual 3D reconstruction of highly fragmented bone fractures
Zhang et al. Development and model form assessment of an automatic subject-specific vertebra reconstruction method
Semenchenia et al. Construction of individual geometric 3D model of the lumbar spine of a person based on the analysis of medical images

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21928794

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21928794

Country of ref document: EP

Kind code of ref document: A1