WO2026055893A1 - Human body positioning method for x-ray transmission examination - Google Patents

Abstract

A human body positioning method for X-ray transmission examination. The method comprises: acquiring a visible light image of a patient on a hospital bed, and a preoperative medical image of the patient obtained by using a fluoroscopic examination device (S101); on the basis of the visible light image, obtaining a corresponding parameterized patient model, and on the basis of the parameterized patient model, performing fitting to obtain a standard skeletal model (S102); performing segmentation and reconstruction on the preoperative medical image, so as to obtain a personalized skeletal model (S103); and on the basis of a target bony landmark, fusing the personalized skeletal model and the standard skeletal model to obtain a target fused image, so as to position a target organ of the patient (S104). By using the method in the present disclosure, a target organ can be quickly positioned preoperatively on the basis of a target bony landmark.

Description

A method for human body positioning in X-ray transmission examination Technical Field

This disclosure relates to the interdisciplinary field of medicine and engineering, and in particular to a method for locating the human body in X-ray transmission examination.

Background Technology

During surgery, commonly used X-ray transmission examination methods include: 1) Fluoroscopy: Fluoroscopy equipment is used during surgery to guide the surgeon through real-time imaging with penetrating X-rays. This method is typically used in orthopedic surgery and the placement of internal fixation devices. 2) Digital subtraction angiography (DSA): In interventional procedures, surgeons can introduce contrast agents into the patient's blood vessels and, combined with X-ray imaging technology, monitor the intravascular conditions in real time to guide the interventional procedure. These methods provide real-time X-ray transmission imaging information during surgery, assisting surgeons in accurate positioning and manipulation, and playing an important auxiliary role in some surgeries requiring precise positioning and manipulation.

Fluoroscopy X-ray equipment (also known as X-ray fluoroscopy examination equipment) is widely used for real-time dynamic imaging during surgery, mainly including C-arms, O-arms, and G-arms. A C-arm X-ray device typically consists of a fixed articulated arm and a movable X-ray detector, shaped like the letter "C". This device is widely used in orthopedic surgery, trauma surgery, cardiac catheterization, and other surgical and treatment procedures to provide real-time X-ray imaging. An O-arm X-ray device is an X-ray device that rotates around the patient and is typically used for imaging and guiding surgery within the operating room. O-arm devices can provide high-resolution 3D imaging, aiding surgeons in accurate positioning and surgical planning during complex procedures. G-arm X-ray devices are similar to C-arms but slightly different in construction, and are also commonly used for X-ray imaging and guiding surgery within the operating room. They provide multi-angle X-ray imaging, helping surgeons to observe and guide the procedure in real time.

X-ray transmission aims to obtain prior information about the location and shape of the organ being examined. Due to the radiation hazards associated with X-ray transmission, the maximum number of scans a patient can safely undergo within a given timeframe is limited. Because of this limitation, physicians must carefully plan the imaging protocol before surgery. Any solution that facilitates rapid and accurate localization of the target organ without excessive beam out-of-focus imaging will greatly assist clinicians and reduce the risk of radiation damage to the patient. For example, in radiation therapy (such as cancer treatments that use high-dose radiation to kill cancer cells and shrink tumors), it is common practice to use a patient-specific mold (usually made of plastic or plaster) to keep the patient in the exact same posture during preoperative organ scans and subsequent treatments. This customized mold helps to accurately target the organ region without rescanning the patient and repositioning the tumor before each treatment. However, this method limits the clinician's operational space and incurs considerable time and financial costs. Existing technologies also propose in vivo organ deformation models based on different subject postures, which can extract organ shape representations for specific patients and predict their deformable shapes based on different postural parameters. However, this method uses FEM (Finite Element Method) to simulate posture-related organ deformation, and its accuracy and generalization ability for different organs need to be verified.

Summary of the Invention

This disclosure presents a method, system, electronic device, and computer-readable storage medium for human body localization during X-ray transmission examination to rapidly locate target organs preoperatively based on target bone landmarks.

To achieve the above objectives, the first aspect of this disclosure provides a method for human body localization in X-ray transmission examination, comprising:

Acquire visible light images of the patient in bed and preoperative medical images of the patient obtained using fluoroscopic examination equipment;

A corresponding patient parametric model is obtained based on the visible light image, and a standard skeletal model is obtained by fitting the patient parametric model.

A personalized skeletal model is obtained by segmenting and reconstructing the preoperative medical images.

Based on the target bone landmarks, the personalized bone model is fused with the standard bone model to obtain a target fusion image, thereby achieving target organ localization for the patient.

In the method of the first aspect of this disclosure, the step of fitting a standard bone model based on the patient parametric model includes: generating bone wrapping surface data and obtaining skin surface data based on the patient parametric model; and fitting a standard bone model based on the bone wrapping surface data and the skin surface data.

In the method of the first aspect of this disclosure, the standard bone model is obtained by fitting the bone wrapping surface data and the skin surface data using a nonlinear least squares method.

In the method of the first aspect of this disclosure, the step of fitting the standard bone model based on the bone wrapping surface data and the skin surface data using a nonlinear least squares method includes: setting deformable surface parameters, obtaining energy based on the deformable surface parameters, the bone wrapping surface data and the skin surface data, and minimizing the energy using a nonlinear least squares method to obtain the standard bone model.

In the method of the first aspect of this disclosure, the step of fusing the personalized skeletal model and the standard skeletal model based on target bone markers to obtain a target fused image includes: selecting a first set of target bone markers of the personalized skeletal model in a preset order and a preset number; selecting a second set of target bone markers of the standard skeletal model in a preset order and a preset number; and fusing the first set of target bone markers and the second set of target bone markers in a corresponding order to obtain the target fused image.

In the method of the first aspect of this disclosure, fusing the first target bone marker set and the second target bone marker set in a corresponding order to obtain the target fused image includes: matching each first target bone marker in the first target bone marker set and the second target bone marker set in a corresponding order; obtaining a spatial transformation combination; and adjusting the personalized skeleton model based on the spatial transformation combination so that the adjusted personalized skeleton model is consistent with the coordinate system of the standard skeleton model, thereby obtaining the target fused image.

To achieve the above objectives, a second aspect of this disclosure provides a human body positioning system for X-ray transmission examination, comprising:

The image acquisition module is used to acquire visible light images of the patient in the hospital bed and preoperative medical images of the patient obtained using fluoroscopic examination equipment.

The skeletal modeling module is used to obtain a corresponding patient parametric model based on the visible light image, and to fit a standard skeletal model based on the patient parametric model.

The reconstruction module is used to segment and reconstruct the preoperative medical images to obtain a personalized skeletal model.

The fusion module is used to fuse the personalized bone model with the standard bone model based on the target bone landmarks to obtain a target fusion image, so as to realize the target organ localization of the patient.

In the system of the second aspect of this disclosure, the bone modeling module, when used to fit a standard bone model based on the patient parametric model, is specifically used to: generate bone wrapping surface data and obtain skin surface data based on the patient parametric model; and fit a standard bone model based on the bone wrapping surface data and the skin surface data.

To achieve the above objectives, a third aspect of this disclosure provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; and the processor executes the computer-executable instructions stored in the memory to implement the method proposed in the first aspect of this disclosure.

To achieve the above objectives, a fourth aspect of this disclosure provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, are used to implement the method proposed in the first aspect of this disclosure.

The present disclosure provides a method, system, electronic device, and storage medium for human body localization in X-ray transmission examination. This involves acquiring visible light images of a patient in bed and preoperative medical images of the same patient obtained using fluoroscopy equipment; obtaining a corresponding parametric model of the patient based on the visible light images; fitting a standard skeletal model based on the patient parametric model; segmenting and reconstructing the preoperative medical images to obtain a personalized skeletal model; and fusing the personalized skeletal model with the standard skeletal model based on target bone landmarks to obtain a target fused image, thereby achieving target organ localization. In this scenario, by combining the standard skeletal model obtained from the visible light images and the personalized skeletal model obtained from the preoperative medical images, and fusing the personalized skeletal model with the standard skeletal model based on target bone landmarks to obtain a target fused image, the position and shape of the target organ in the patient's posture corresponding to the visible light images can be obtained from the bone positions in the target fused image. This enables rapid preoperative localization of the target organ based on target bone landmarks, thereby better assisting physicians in developing preoperative imaging plans and reducing the scanning radiation dose during intraoperative X-ray transmission examination.

It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description.

Attached Figure Description

The above and/or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

Figure 1 is a schematic flowchart of a human body positioning method for X-ray transmission examination provided in an embodiment of this disclosure;

Figure 2 is a schematic diagram of the shooting scene provided in the embodiment of this disclosure;

Figure 3 is a schematic diagram of a visible light image and a three-dimensional model of an SMPL patient provided in an embodiment of this disclosure;

Figure 4 is a schematic diagram of the skeletal landmarks and three-dimensional body surface model provided in the embodiments of this disclosure;

Figure 5 is a schematic diagram of fitting a standard skeletal model based on a three-dimensional model of an SMPL patient provided in an embodiment of this disclosure;

Figure 6 is a schematic diagram of preoperative medical images and segmentation and reconstruction results provided in an embodiment of this disclosure;

Figure 7 is a schematic diagram of the selection of bone landmarks provided in an embodiment of this disclosure;

Figure 8 is a schematic diagram of the effect after fusing the reconstructed spinal model and the standard skeletal model provided in the embodiments of this disclosure;

Figure 9 is a block diagram of a human body positioning system for X-ray transmission examination provided in an embodiment of this disclosure.

Detailed Implementation

Embodiments of this disclosure are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this disclosure, and should not be construed as limiting this disclosure.

The following describes a human body positioning method and system for X-ray transmission examination according to embodiments of the present disclosure with reference to the accompanying drawings.

This disclosure provides a method for human body localization during X-ray transmission examination to quickly locate target organs preoperatively based on target bone landmarks.

Figure 1 is a schematic flowchart of a human body positioning method for X-ray transmission examination provided in an embodiment of this disclosure.

As shown in Figure 1, this method for locating the human body during X-ray transmission examination includes the following steps:

Step S101: Acquire visible light images of the patient in the hospital bed and preoperative medical images of the patient obtained using fluoroscopic examination equipment.

In step S101, the visible light image can be captured by a visible light camera arranged on the X-ray fluoroscopy equipment. Specifically, a visible light camera is installed near the detector plane of the X-ray fluoroscopy equipment, and a visible light image of the patient on the hospital bed is captured using the visible light camera.

In step S101, preoperative medical images can be obtained by scanning with fluoroscopy equipment. The main scanning methods for preoperative medical images include: Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Positron Emission Tomography-CT (PET-CT), and ultrasound imaging.

Figure 2 is a schematic diagram of the imaging scenario provided in an embodiment of this disclosure. As shown in Figure 2, the fluoroscopic examination device is an X-ray fluoroscopic examination device. The X-ray fluoroscopic examination device obtains preoperative medical images of the patient on the hospital bed through an X-ray source and a detector. A visible light camera is installed near the detector plane of the X-ray fluoroscopic examination device to capture visible light images of the patient on the hospital bed.

Step S102: Obtain the corresponding patient parameterized model based on the visible light image, and fit the standard bone model based on the patient parameterized model.

In simple terms, 3D human modeling has several parametric representations. SMPL (Skinned Multi-Person Linear) is a parametric 3D human model that efficiently describes and generates 3D human morphology using a small number of parameters. SMPL can convert input pose and shape parameters into the pose and shape of a 3D human model. Specifically, SMPL is defined as a parametric human model M(β,θ), mapping the shape parameter vector β and pose parameter vector θ to the vertices of the human body, where β controls the shape of the human body and θ controls the pose. SMPL-H is a variant of the SMPL model that adds details to the hand bones, making the generated human model more realistic. SMPL-X is an extended version of the SMPL model, suitable for people of various body types, muscle mass, and body shape characteristics, and can simulate more realistic human shapes and movements. STAR (Sparse Trained Articulated Human Body Regressor) is a framework for body pose reconstruction and animation generation. It combines the techniques of SMPL and BlendSCAPE, transforming pose-related deformation factors into a set of sparse spatial local pose correction hybrid shape functions. Each joint only affects a sparse subset of mesh vertices, enabling the reconstruction of 3D human poses from a single image or video, and the generation and editing of animations.

In this embodiment of the disclosure, taking the SMPL model used when performing three-dimensional modeling of the patient as an example, the patient parameterized model obtained based on the visible light image in step S102 is the SMPL patient three-dimensional model.

The specific steps for obtaining the corresponding 3D model of the SMPL patient based on the visible light image include: (1) Pose estimation: The joint position and pose of the patient in the visible light image are estimated by computer vision and deep learning technology to obtain the joint position and pose estimation results; (2) Shape modeling: The 3D shape of the patient is estimated by using the joint position and pose estimation results and adopting the existing human shape modeling method to obtain the patient's shape parameter vector β and pose parameter vector θ; (3) Alignment and matching: The learned SMPL is obtained, and the obtained patient's shape parameter vector β and pose parameter vector are used as input data to the learned SMPL output patient vertices. All the vertices form the 3D model of the SMPL patient, thereby realizing the alignment and matching of the estimated patient's 3D shape with the SMPL model.

Figure 3 is a schematic diagram of a visible light image and a three-dimensional model of an SMPL patient provided in an embodiment of this disclosure. In Figure 3, (a) is a visible light image of the patient, and (b) and (c) are the front view and side view of the three-dimensional model of the SMPL patient obtained by the method of this disclosure, respectively.

In step S102, a standard bone model is obtained by fitting a patient parametric model, including: generating bone wrapping surface data and obtaining skin surface data based on the patient parametric model; and fitting a standard bone model based on the bone wrapping surface data and skin surface data. Specifically, the standard bone model is obtained by fitting the bone wrapping surface data and skin surface data using a nonlinear least squares method.

A standard skeletal model is obtained by fitting nonlinear least squares data based on skeletal surface data and skin surface data. The process includes: setting deformable surface parameters, obtaining energy based on deformable surface parameters, skeletal surface data and skin surface data, and minimizing energy using nonlinear least squares to obtain the standard skeletal model.

Specifically, taking the 3D model of an SMPL patient as an example, the process of fitting a standard skeletal model based on the 3D model of an SMPL patient is as follows:

Figure 4 is a schematic diagram of the skeletal landmarks and three-dimensional body surface model provided in the embodiments of this disclosure; Figure 5 is a schematic diagram of fitting a standard skeletal model based on a three-dimensional model of an SMPL patient provided in the embodiments of this disclosure.

Using the generated 3D model of the SMPL patient and the corresponding camera parameters of the visible light camera, the 3D pose is projected onto a 2D image plane. From the projected 2D image, specific joint points or skeletal landmarks (referred to as bone landmarks) are extracted, such as 2D skeletal landmarks in the head, shoulder, elbow, and wrist, as shown in Figure 4(a), which includes multiple bone landmarks in the head, shoulder, elbow, wrist, hip, knee, and ankle. These bone landmarks, when connected, form the initial spatial configuration of the standard skeletal model. For the initial spatial configuration, a watertight Genus-0 surface, referred to as the bone wrapping surface W (i.e., bone wrapping surface data), is first generated to enclose this initial spatial configuration. This bone wrapping surface is a smooth, impermeable 2D manifold surface that does not include areas such as the inside of the ribs, nor does it include the pelvis or the small opening between the ulna and radius.

Based on the generated SMPL patient 3D model, a body surface 3D model is obtained (as shown in Figure 4(b)). This body surface 3D model is the skin surface data, also known as the skin surface M. A standard skeletal model B is generated based on the skin surface M. First, the initial vertex configuration of the deformable surface X (i.e., the deformable surface parameters) is set. Then, a nonlinear least squares method is used to minimize the energy, which consists of a fitting term and a regularization term. The fitting term is used to attract the deformable surface X to the skeletonized surface W; the regularization term is used to prevent the deformable surface X from deviating from its initial vertex configuration. Deformation in a physically unreasonable way:

In the formula, B is the standard skeleton model (also known as energy), _ωfit is the weight of the fitting term, and _ωreg is the weight of the regularization term.

Regularization is expressed as a discrete bending energy that penalizes changes in the mean curvature:

In the formula, _{x_i} and Represent the deformed surface X and the initial vertex configuration, respectively. The vertex R_{_i} ∈ SO(3) represents the Laplace Δx_{_i} of the vertex and The optimal rotation matrix for alignment, where A_{_i} represents the Voronoi area.

The fitting term penalizes the squared distance between vertex _{x_i} ∈ X and target position _{t_i} ∈ W:

In the formula, the target position t _{_i} is a point on the skeleton wrapping surface W, which has three types: nearest point correspondence, fixed correspondence, or collision target. The weight ω_{_i} is determined by the type of the target position t_{_i} (0.1 for nearest point correspondence, 1 for fixed correspondence, and 100 for collision target). After minimizing formula (1), the final standard skeleton model B is obtained. This standard skeleton model is a three-dimensional skeleton model. Figure 5(a) shows the three-dimensional model of the SMPL patient with skeleton markers, and Figure 5(b) shows the obtained standard skeleton model B.

Step S103: Segment and reconstruct the preoperative medical images to obtain a personalized skeletal model.

In step S103, the preoperative medical images can target the human spine, thus obtaining a personalized skeletal model that is a reconstructed spinal model. It should be noted that the preoperative medical images can also target different parts such as the arm, head, and pelvis as needed, thereby obtaining a reconstructed model of the corresponding bones.

In this embodiment of the disclosure, taking a CT scan image of the human spine as an example of preoperative medical imaging, Figure 6 is a schematic diagram of the preoperative medical image and segmentation and reconstruction results provided in this embodiment of the disclosure. As shown in Figure 6(a), a CT scan image of the human spine taken before surgery is obtained. The CT scan image is segmented and reconstructed to obtain the reconstructed spinal model shown in Figure 6(b). The reconstructed spinal model includes the lumbar vertebrae, thoracic vertebrae, and cervical vertebrae.

Step S104: Based on the target bone landmarks, the personalized bone model is fused with the standard bone model to obtain a target fusion image, so as to realize the localization of the patient's target organ.

In step S104, based on the target bone markers, the personalized bone model and the standard bone model are fused to obtain a target fused image. This includes: selecting a first set of target bone markers from the personalized bone model according to a preset order and a preset number; selecting a second set of target bone markers from the standard bone model according to a preset order and a preset number; and fusing the first and second target bone marker sets in a corresponding order to obtain the target fused image. Specifically, fusing the first and second target bone marker sets in a corresponding order to obtain the target fused image includes: matching each first target bone marker in the first and second target bone marker sets one-to-one according to the corresponding order; obtaining a spatial transformation combination; and adjusting the personalized bone model based on the spatial transformation combination to make the coordinate system of the adjusted personalized bone model consistent with that of the standard bone model, thereby obtaining the target fused image.

In step S104, taking the reconstruction of the spinal model as an example, the target bone landmarks are selected from the lumbar, thoracic, and cervical vertebrae. The preset number is, for example, 5, and the preset order is, for example, the first cervical vertebra, the seventh cervical vertebra, the seventh thoracic vertebra, the fourth lumbar vertebra, and the fifth lumbar vertebra.

Figure 7 is a schematic diagram of the selection of bone landmarks provided in the embodiments of this disclosure. As shown in Figure 7, the reconstructed spinal model includes lumbar, thoracic, cervical, and sacral vertebrae, wherein the cervical vertebrae include the first cervical vertebra C1, the second cervical vertebra C2, the third cervical vertebra C3, ..., the seventh cervical vertebra C7. The thoracic vertebrae include the first thoracic vertebra T1, the second thoracic vertebra T2, ..., the seventh thoracic vertebra T7, ..., the eleventh thoracic vertebra T11, and the twelfth thoracic vertebra T12. The lumbar vertebrae include the first lumbar vertebra L1, the second lumbar vertebra L2, ..., the fourth lumbar vertebra L4, and the fifth lumbar vertebra L5. The sacral vertebrae include the first sacral vertebra S1, etc. The five primary target bone landmarks are selected from the reconstructed spinal model: _m1 (center of the first cervical vertebra C1), _m2 (center of the seventh cervical vertebra C7), _m3 (center of the seventh thoracic vertebra T7), _m4 (center of the fourth lumbar vertebra L4), and _m5 (center of the fifth lumbar vertebra L5). This results in a set of primary target bone landmarks, denoted as LM1 = { _m1 , _m2 , _m3 , _m4 , _m5 }. These primary target bone landmarks are chosen primarily because of their anatomical specificity and identifiability. For example, the seventh cervical vertebra C7 is the lowest cervical vertebra, characterized by a prominent spinous process, known as a prominent vertebra.

For the standard skeletal model, a second set of target bone landmarks is obtained according to a preset order and a preset number. The second set of target bone landmarks LM2 = { _n1 , _n2 , _n3 , _n4 , _n5 }, where the five second target bone landmarks _n1 , _n2 , _n3 , _n4 , and _n5 are the center points of the first cervical vertebra _C′1 , the seventh cervical vertebra _C′7 , the seventh thoracic vertebra _T′7 , the fourth lumbar vertebra L′ ₄ , and the fifth lumbar vertebra L′ ₅ in the standard skeletal model, respectively.

The target bone landmarks in the two sets of target bone landmark sets are arranged in a one-to-one correspondence: i = 1, 2, 3, 4, 5, _mi ∈ LM1, n _i ∈ LM2. Then n _i = _lRmi + t + ε _i , i = 1, 2, 3, 4, 5. Where l > 0 is the scaling factor; R ∈ SO (3) is the three-dimensional rotation matrix; It is a three-dimensional translation vector; We model the unknown additive noise, assuming it follows a zero-mean isotropic Gaussian distribution with a standard deviation of σ_{_i} . Under the condition of maximum likelihood estimation, we optimize the scaling factor l ^* , rotation matrix R ^* , and translation vector t ^* :

In the formula, l ^* , R ^* , and t ^* constitute a spatial transformation combination. The spatial transformation combination obtained by optimization is applied to the reconstructed spinal model to adjust the coordinate system of the reconstructed spinal model, thereby making the coordinate system of the adjusted reconstructed spinal model consistent with that of the standard skeletal model.

Figure 8 is a schematic diagram of the effect after fusing the reconstructed spinal model and the standard skeletal model provided in the embodiments of this disclosure. Figure 8(a) shows the fused image obtained by fusing the reconstructed spinal model and the standard skeletal model. Figures 8(b) and (c) show the side view and front view of the fused image of the SMPL patient 3D model corresponding to the reconstructed spinal model and the standard skeletal model.

In step S104, the target fusion image may include a first fusion image obtained by fusing the personalized bone model and the standard bone model. The target fusion image may also include a second fusion image of the patient parameterized model corresponding to the personalized bone model and the standard bone model, a third fusion image of the preoperative medical image corresponding to the personalized bone model and the standard bone model, etc.

In step S104, since the relative positions of bones and target organs are fixed in the personalized skeletal model, fusing the personalized skeletal model with the standard skeletal model maps the relative positions of bones and target organs onto the standard skeletal model. This allows the target fusion image to include the position and shape information of the target organ in the patient's posture in the visible light image. Therefore, based on the target fusion image, the position and shape of the human body, especially the target organ, can be quickly located during X-ray transmission examination. This assists doctors in developing imaging plans before surgery and reduces scanning radiation dose.

To achieve the above embodiments, this disclosure also proposes a human body positioning system for X-ray transmission examination.

Figure 9 is a block diagram of a human body positioning system for X-ray transmission examination provided in an embodiment of this disclosure.

As shown in Figure 9, this human positioning system for X-ray transmission examination includes an image acquisition module, a skeleton modeling module, a reconstruction module, and a fusion module, wherein:

The image acquisition module is used to acquire visible light images of the patient in the hospital bed and preoperative medical images of the patient obtained using fluoroscopic examination equipment.

The skeletal modeling module is used to obtain the corresponding patient parametric model based on the visible light image, and to fit the standard skeletal model based on the patient parametric model.

The reconstruction module is used to segment and reconstruct preoperative medical images to obtain personalized bone models;

The fusion module is used to fuse personalized bone models with standard bone models based on target bone landmarks to obtain a target fusion image, thereby enabling the localization of the patient's target organs.

Furthermore, in one possible implementation of this disclosure embodiment, the bone modeling module obtains a standard bone model based on the patient parametric model by: generating bone wrapping surface data and obtaining skin surface data based on the patient parametric model; and obtaining a standard bone model based on the bone wrapping surface data and the skin surface data.

Furthermore, in one possible implementation of this disclosure embodiment, in the bone modeling module, a standard bone model is obtained by fitting the bone wrapping surface data and skin surface data using a nonlinear least squares method.

Furthermore, in one possible implementation of this disclosure embodiment, the bone modeling module uses nonlinear least squares to fit a standard bone model based on bone wrapping surface data and skin surface data, including: setting deformable surface parameters, obtaining energy based on deformable surface parameters, bone wrapping surface data and skin surface data, and minimizing energy using nonlinear least squares to obtain a standard bone model.

Furthermore, in one possible implementation of this disclosure embodiment, the target bone landmarks in the fusion module are selected from the lumbar, thoracic, and cervical vertebrae.

Furthermore, in one possible implementation of this disclosure embodiment, the fusion module is specifically used to: select a first target bone marker set of a personalized skeletal model according to a preset order and a preset number; select a second target bone marker set of a standard skeletal model according to a preset order and a preset number; and fuse the first target bone marker set and the second target bone marker set in a corresponding order to obtain a target fused image.

Furthermore, in one possible implementation of this embodiment, the fusion module fuses the first target bone landmark set and the second target bone landmark set in a corresponding order to obtain a target fused image, including: matching each first target bone landmark in the first target bone landmark set and the second target bone landmark set in a corresponding order; obtaining a spatial transformation combination; and adjusting the personalized skeleton model based on the spatial transformation combination so that the adjusted personalized skeleton model is consistent with the coordinate system of the standard skeleton model, thereby obtaining the target fused image.

It should be noted that the foregoing explanation of the human body positioning method embodiment for X-ray transmission examination also applies to the human body positioning system for X-ray transmission examination in this embodiment, and will not be repeated here.

In this embodiment, a visible light image of the patient in bed and a preoperative medical image of the patient obtained using a fluoroscopic examination device are acquired. A corresponding parametric model of the patient is obtained based on the visible light image, and a standard skeletal model is fitted based on the patient parametric model. A personalized skeletal model is obtained by segmenting and reconstructing the preoperative medical image. Based on target bone landmarks, the personalized skeletal model and the standard skeletal model are fused to obtain a target fused image, thereby achieving target organ localization for the patient. In this case, by combining the standard skeletal model obtained from the visible light image and the personalized skeletal model obtained from the preoperative medical image, and fusing the personalized skeletal model with the standard skeletal model based on target bone landmarks to obtain a target fused image, the position and shape of the target organ in the patient's posture corresponding to the visible light image can be obtained based on the bone position in the target fused image. This enables rapid preoperative localization of the target organ based on target bone landmarks, thereby better assisting doctors in developing preoperative imaging plans and reducing the scanning radiation dose during subsequent intraoperative X-ray transmission examinations.

To implement the above embodiments, this disclosure also proposes an electronic device, including: a processor and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method provided in the foregoing embodiments.

To implement the above embodiments, this disclosure also proposes a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the methods provided in the foregoing embodiments.

To implement the above embodiments, this disclosure also proposes a computer program product, including a computer program that, when executed by a processor, implements the methods provided in the foregoing embodiments.

In the foregoing descriptions of the embodiments, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this disclosure, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of preferred embodiments of this disclosure includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of this disclosure pertain.

The logic and/or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

It should be understood that various parts of this disclosure can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

Furthermore, the functional units in the various embodiments of this disclosure can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of the present disclosure have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present disclosure. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present disclosure.

Claims (10)

A method for locating the human body during X-ray transmission examination, characterized by comprising: Acquire visible light images of the patient in bed and preoperative medical images of the patient obtained using fluoroscopic examination equipment; A corresponding patient parametric model is obtained based on the visible light image, and a standard skeletal model is obtained by fitting the patient parametric model. A personalized skeletal model is obtained by segmenting and reconstructing the preoperative medical images. Based on the target bone landmarks, the personalized bone model is fused with the standard bone model to obtain a target fusion image, thereby achieving target organ localization for the patient. The human body positioning method for X-ray transmission examination according to claim 1 is characterized in that, the step of obtaining a standard skeletal model based on the patient parametric model includes: Generate bone-wrapped surface data and obtain skin surface data based on the patient parameterization model; A standard bone model is obtained by fitting the bone wrapping surface data and the skin surface data. The human body positioning method for X-ray transmission examination according to claim 2 is characterized in that the standard bone model is obtained by fitting the bone wrapping surface data and the skin surface data using a nonlinear least squares method. The human body positioning method for X-ray transmission examination according to claim 3 is characterized in that, the step of fitting the standard bone model using nonlinear least squares method based on the bone wrapping surface data and the skin surface data includes: The deformable surface parameters are set, and the energy is obtained based on the deformable surface parameters, the bone wrapping surface data, and the skin surface data. The energy is then minimized using a nonlinear least squares method to obtain the standard bone model. The human body localization method for X-ray transmission examination according to claim 1 is characterized in that, the step of fusing the personalized skeletal model with the standard skeletal model based on target bone landmarks to obtain a target fused image includes: Select the first set of target bone landmarks of the personalized skeletal model according to a preset order and a preset number; Select the second set of target bone landmarks of the standard skeletal model according to a preset order and a preset number; The first target bone landmark set and the second target bone landmark set are fused in a corresponding order to obtain the target fused image. The method for human body localization in X-ray transmission examination according to claim 5, characterized in that, the step of fusing the first target bone landmark set and the second target bone landmark set in a corresponding order to obtain the target fused image includes: The first target bone markers in the first target bone marker set and the second target bone marker set are matched one-to-one in the corresponding order; A spatial transformation combination is obtained, and the personalized skeleton model is adjusted based on the spatial transformation combination so that the coordinate system of the adjusted personalized skeleton model is consistent with that of the standard skeleton model, thereby obtaining the target fused image. A human body positioning system for X-ray transmission examination, characterized in that it comprises: The image acquisition module is used to acquire visible light images of the patient in the hospital bed and preoperative medical images of the patient obtained using fluoroscopic examination equipment. The skeletal modeling module is used to obtain a corresponding patient parametric model based on the visible light image, and to fit a standard skeletal model based on the patient parametric model. The reconstruction module is used to segment and reconstruct the preoperative medical images to obtain a personalized skeletal model. The fusion module is used to fuse the personalized bone model with the standard bone model based on the target bone landmarks to obtain a target fusion image, so as to realize the target organ localization of the patient. The human body positioning system for X-ray transmission examination according to claim 7 is characterized in that, when the bone modeling module is used to fit a standard bone model based on the patient parametric model, it is specifically used to: generate bone wrapping surface data and obtain skin surface data based on the patient parametric model; and fit a standard bone model based on the bone wrapping surface data and the skin surface data. An electronic device, characterized in that it comprises: a processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-6. A computer-readable storage medium, characterized in that the computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-6.