WO2022237154A1 - 一种医学影像分割装置及方法 - Google Patents

一种医学影像分割装置及方法 Download PDF

Info

Publication number
WO2022237154A1
WO2022237154A1 PCT/CN2021/137328 CN2021137328W WO2022237154A1 WO 2022237154 A1 WO2022237154 A1 WO 2022237154A1 CN 2021137328 W CN2021137328 W CN 2021137328W WO 2022237154 A1 WO2022237154 A1 WO 2022237154A1
Authority
WO
WIPO (PCT)
Prior art keywords
medical image
trachea
segmentation
acquisition module
tracheal
Prior art date
Application number
PCT/CN2021/137328
Other languages
English (en)
French (fr)
Inventor
张凡
杨华
Original Assignee
上海杏脉信息科技有限公司
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 上海杏脉信息科技有限公司 filed Critical 上海杏脉信息科技有限公司
Publication of WO2022237154A1 publication Critical patent/WO2022237154A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30021—Catheter; Guide wire
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30061—Lung
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Definitions

  • the invention relates to an image processing device, in particular to a medical image segmentation device and method.
  • the diagnosis of diseases such as pulmonary bronchial stenosis, chronic obstructive pulmonary disease, and bronchiolitis obliterans depends on the quantitative analysis of the pulmonary trachea, and the construction of a tracheal tree model is helpful for the quantitative analysis of the morphological changes of the pulmonary trachea.
  • the tracheal tree model can also be applied to bronchial navigation in surgery.
  • the inventors have found in practical applications that in the prior art, medical personnel mainly rely on manual segmentation of medical images to obtain a tracheal tree model, which is inefficient.
  • the object of the present invention is to provide a medical image segmentation device and method for solving the problem of low efficiency of manual segmentation in the prior art.
  • the first aspect of the present invention provides a medical image segmentation device
  • the medical image segmentation device includes: a medical image acquisition module, used to acquire medical images, the medical images include at least Part of the lung trachea;
  • the first segmentation module is connected to the medical image acquisition module, and is used to segment the medical image to obtain the first segmentation result of the lung trachea;
  • the first partial image acquisition module is connected to the medical image acquisition module
  • the medical image acquisition module is connected to obtain at least one first-level partial image according to the medical image;
  • the second segmentation module is connected to the first partial image acquisition module, and is used to segment the first-level partial image to obtain Acquiring the second segmentation result of the pulmonary trachea;
  • a tracheal tree acquisition module connected to the first segmentation module and the second segmentation module, for obtaining a tracheal tree model, wherein the tracheal tree model is at least composed of the The first segmentation result and the second segmentation result of the above-mentioned lung
  • the medical image segmentation device further includes: a second partial image acquisition module, connected to the first partial image acquisition module, for acquiring at least A secondary partial image; a third segmentation module, connected to the second partial image acquisition module, for segmenting the secondary partial image to obtain a third segmentation result of the lung trachea; the tracheal tree
  • the acquisition module is also connected to the third segmentation module, and the trachea tree model is at least obtained by fusing the first segmentation result, the second segmentation result and the third segmentation result of the lung trachea.
  • the first partial image acquisition module includes: a tracheal branch point acquisition unit connected to the medical image acquisition module for acquiring the lung trachea in the medical image a tracheal branch point; a first partial image acquisition unit, connected to the tracheal branch point acquisition unit, and configured to acquire the primary partial image according to the tracheal branch point and the medical image.
  • the first partial image acquisition module includes: a tracheal terminal point acquisition unit connected to the medical image acquisition module for acquiring the lung trachea in the medical image the end point of the trachea; the second partial image acquisition unit, connected to the end point acquisition unit of the trachea, and configured to acquire the primary partial image according to the end point of the trachea and the medical image.
  • the tracheal terminal point obtaining unit is further configured to obtain the centerline of the pulmonary trachea, and obtain the tracheal terminal point according to the centerline of the pulmonary trachea.
  • the second partial image acquisition unit is further configured to cluster the end points of the trachea, and acquire the primary partial image based on the clustering result.
  • the medical image segmentation device further includes: a fracture area acquisition module, connected to the tracheal tree acquisition module, for acquiring the fracture area in the tracheal tree model; the fracture area A repair module, connected to the fractured area acquisition module, for repairing the fractured area.
  • the fracture area repair module includes: a seed point acquisition unit, configured to acquire a seed point, the seed point is located in the fracture area; a trachea repair unit, and the The seed point acquisition unit is connected to expand based on the seed point to acquire the trachea in the fractured area.
  • the tracheal tree acquisition module is further configured to prune the tracheal tree model.
  • a second aspect of the present invention provides a medical image segmentation method, the medical image segmentation method includes: acquiring a medical image, the medical image includes at least part of the patient's lung trachea; segmenting the medical image to obtain the A first segmentation result of the pulmonary trachea; acquiring at least one first-level partial image according to the medical image; segmenting the first-level partial image to obtain a second segmentation result of the pulmonary trachea; obtaining a tracheal tree model, wherein , the trachea tree model is at least obtained by fusing the first segmentation result and the second segmentation result of the lung trachea.
  • the medical image segmentation device can obtain the first segmentation result of the pulmonary trachea according to the medical image, and obtain the second segmentation result of the pulmonary trachea according to the first-level partial image. Based on this, the medical image segmentation device can at least obtain the pulmonary The first segmentation result of the trachea is fused with the second segmentation result to obtain a trachea tree model.
  • the above-mentioned process can be automatically completed by electronic equipment, basically without manual participation, and has high efficiency and accuracy.
  • FIG. 1A is a schematic structural diagram of a specific embodiment of the medical image segmentation device of the present invention.
  • Fig. 1B shows an example diagram of a medical image acquired by the medical image segmentation device of the present invention in a specific embodiment.
  • FIG. 1C is an example diagram of a first-level partial image acquired by the medical image segmentation device of the present invention in a specific embodiment.
  • FIG. 1D is a schematic structural diagram of a specific embodiment of the medical image segmentation device of the present invention.
  • FIG. 2 is a schematic structural diagram of a first partial image acquisition module in a specific embodiment of the medical image segmentation device of the present invention.
  • FIG. 3A is another structural schematic diagram of the first partial image acquisition module in a specific embodiment of the medical image segmentation device of the present invention.
  • FIG. 3B is a flow chart of repairing a broken trachea by the medical image segmentation device of the present invention in a specific embodiment.
  • FIG. 4A and FIG. 4B are diagrams showing examples of tracheal segmentation results acquired by the medical image acquisition device in a specific embodiment of the present invention.
  • FIG. 5 is a flow chart of a specific embodiment of the medical image segmentation method of the present invention.
  • the present invention provides a medical segmentation device.
  • the medical image segmentation device can obtain the first segmentation result of the lung trachea according to the medical image, and obtain the second segmentation result of the lung trachea according to the first-level partial image. Based on Here, the medical image segmentation device fuses at least the first segmentation result and the second segmentation result of the lung trachea to obtain a trachea tree model.
  • the above-mentioned process can be automatically completed by electronic equipment, basically without manual participation, and has high efficiency and accuracy.
  • the medical image segmentation device 1 includes a medical image acquisition module 11, a first segmentation module 12, a first partial image acquisition module 13, a second segmentation module 14 and a tracheal tree Get module 15.
  • the medical image acquisition module 11 is used to acquire a medical image, the medical image includes at least part of the patient's lung trachea, the medical image is, for example, a chest cavity image.
  • the medical image is, for example, a chest cavity image.
  • FIG. 1B an example diagram of the medical image is shown in FIG. 1B .
  • the first segmentation module 12 is connected to the medical image acquisition module 11 and configured to segment the medical image to obtain a first segmentation result of the lung trachea.
  • the first segmentation result is obtained by the first segmentation module 12 by segmenting the medical image itself, and belongs to the overall segmentation result of the lung and trachea.
  • the first segmentation results include, for example, segmentation results of grades 1-4 trachea.
  • the first partial image acquisition module 13 is connected to the medical image acquisition module 11, and is used to acquire at least one primary partial image according to the medical image, wherein the range of each primary partial image is smaller than that of the medical image range, and, compared with the medical image, the primary local image includes more details of the lungs and trachea.
  • an example diagram of the first-level partial image is shown in FIG. 1C .
  • the first partial image acquisition module 13 may obtain the first-level partial image, for example, by segmenting the medical image.
  • the second segmentation module 14 is connected to the first partial image acquisition module 13 and configured to segment the primary partial image to obtain a second segmentation result of the lung trachea.
  • the second segmentation result is obtained by segmenting the first-level partial image by the second segmentation module 14. Compared with the first segmentation result, the second segmentation result is a more detailed lung Segmentation results of the trachea.
  • the second segmentation results include, for example, segmentation results of grades 4-6 trachea.
  • the tracheal tree acquisition module 15 is connected to the first segmentation module 12 and the second segmentation module 14 for acquiring a tracheal tree model.
  • the trachea tree model is at least obtained by fusing the first segmentation result and the second segmentation result of the lung trachea.
  • the medical image segmentation apparatus 1 further includes a second partial image acquisition module 16 and a third segmentation module 17 .
  • the second partial image acquisition module 16 is connected to the first partial image acquisition module 13, and is used to acquire at least one secondary partial image according to the primary partial image, wherein the range of each secondary partial image is less than The range of the first-level partial image, and, compared with the first-level partial image, the second-level partial image includes further details of lung trachea.
  • the second partial image acquisition module 16 may obtain the secondary partial image by segmenting the primary partial image, for example.
  • the third segmentation module 17 is connected to the second partial image acquisition module 16 and configured to segment the secondary partial image to obtain a third segmentation result of the lung trachea.
  • the third segmentation result is obtained by segmenting the secondary partial image by the third segmentation module 17. Compared with the second segmentation result, the third segmentation result is more detailed Segmentation results of the trachea.
  • the third segmentation results include, for example, segmentation results of grades 6-8 trachea.
  • the trachea tree acquisition module 15 is also connected to the third segmentation module 17, at this time, the trachea tree model is at least obtained by fusing the first segmentation result, the second segmentation result and the third segmentation result of the pulmonary trachea .
  • the medical image segmentation device 1 is not limited to the above structure.
  • the medical image segmentation device may also include other modules to further segment the second partial shadow step by step, so as to obtain multi-level partial images containing more details of lung and trachea, and then obtain multi-level segmentation
  • the tracheal tree acquisition module can acquire a tracheal tree model including more branches based on the segmentation results of the plurality of levels.
  • the medical image segmentation device may further segment the second-level partial image to obtain a third-level partial image, and segment the third-level partial image to obtain a fourth-level partial image...; and, the medical image segmentation device
  • the fourth segmentation result can be obtained according to the third-level partial image
  • the fifth segmentation result can be obtained according to the fourth-level partial image...
  • the tracheal tree acquisition module can further fuse the fourth segmentation result, the fifth segmentation result... to form The tracheal tree model.
  • the medical image segmentation device 1 in this embodiment can obtain the first segmentation result of the pulmonary trachea according to the medical image, and obtain the second segmentation result of the pulmonary trachea according to the first-level partial image. Based on this, the medical The image segmentation device fuses at least the first segmentation result and the second segmentation result of the lung trachea to obtain a trachea tree model.
  • the above-mentioned process can be automatically completed by electronic equipment, basically without manual participation, and has high efficiency and accuracy.
  • the medical image segmentation device 1 in this embodiment can obtain the segmentation results of different levels of pulmonary trachea and fuse them into the trachea tree model.
  • the trachea tree model can include grades 1-8 trachea at the same time, but in related technologies, limited by the display range and resolution of medical images, it can only achieve automatic segmentation of trachea grades 4-6.
  • the trachea tree model acquired by the medical image segmentation device 1 in this embodiment can contain more levels of trachea, thereby providing medical personnel with more detailed trachea information.
  • the first partial image acquisition module 13 includes a tracheal branch point acquisition unit 131 and a first partial image acquisition unit 132 .
  • the tracheal branch point acquisition unit 131 is connected to the medical image acquisition module 11, and is used to obtain the tracheal branch point of the pulmonary trachea in the medical image, wherein the method of obtaining the tracheal branch point can be obtained through existing Realization of image recognition technology, etc.
  • the first partial image acquiring unit 132 is connected to the tracheal branch point acquiring unit 131, and is configured to acquire the primary partial image according to the tracheal branch point and the medical image.
  • the trachea branch point acquisition unit 131 can obtain the left and right pulmonary lobe branch points from the lung trachea mask through image recognition technology, and the first partial image acquisition unit 132 can convert the medical image Segmentation into a first-order partial image including the left pulmonary bronchus and a first-order partial image including the right pulmonary bronchus.
  • the corresponding modules in the medical image segmentation device can use the above-mentioned similar technology to segment the first-level partial image into second-level partial images according to the tracheal branch point, and divide the second-level partial image into three Level partial image..., the specific implementation will not be repeated here.
  • the first partial image acquisition module 13 includes a tracheal terminal point acquisition unit 133 and a second partial image acquisition unit 134 .
  • the tracheal end point acquisition unit 133 is connected to the medical image acquisition module 11, and is used to acquire the tracheal end point of the pulmonary trachea in the medical image, wherein the way of acquiring the tracheal end point can be through the existing Realization of image recognition technology, etc.
  • the tracheal end point acquiring unit 133 is further configured to acquire the centerline of the pulmonary trachea in the medical image, and acquire the tracheal end point according to the centerline of the pulmonary trachea.
  • the trachea terminal point obtaining unit 133 may obtain the terminal point of the central line of the pulmonary trachea as the trachea terminal point.
  • the second partial image acquisition unit 134 is connected to the tracheal end point acquisition unit 133 and configured to acquire the primary partial image according to the tracheal end point and the medical image. Specifically, the second partial image acquisition unit 134 may acquire an image within a certain range around each end point of the trachea as the first-level partial image, and the size and shape of this range may be set according to actual needs.
  • the second partial image acquisition unit 134 can also be used to cluster the end points of the trachea to obtain a clustering result, and obtain the primary partial image based on the clustering result. Specifically, after clustering, the second partial image acquisition unit 134 can acquire multiple tracheal end point sets, based on the multiple tracheal end point sets, multiple first-level partial images can be acquired, wherein each first-level local Each image contains all tracheal endpoints in a set of tracheal endpoints.
  • the corresponding modules in the medical image segmentation device can use the above-mentioned similar technology to segment the first-level partial image into a second-level partial image according to the end point of the trachea, and divide the second-level partial image into three Level partial image..., the specific implementation will not be repeated here.
  • the medical image segmentation device further includes a fracture region acquisition module and a fracture region repair module.
  • the fracture area acquisition module is connected to the trachea tree acquisition module, and is used to acquire the fracture area in the trachea tree model. Specifically, the fracture area acquisition module can obtain the starting point and end point of each bronchus by extracting the centerline of each bronchus in the lung trachea, and the tracheal tree can be obtained based on the starting point and end point of each bronchus Fractured regions in the model.
  • the fracture area repair module is connected to the fracture area repair module, and is used for repairing the fracture area.
  • the fracture area repair module includes a seed point acquisition unit and a trachea repair unit, wherein the seed point acquisition unit is configured to acquire one or more seed points in the fracture area.
  • the trachea repair unit is connected to the seed point acquisition unit, and is used for performing extended growth based on the seed point to acquire the trachea in the fractured area.
  • the specific implementation method of the trachea repair unit repairing the trachea includes:
  • the trachea tree acquisition module is also used to prune the trachea tree model .
  • the leak refers to the portion in the segmented trachea tree model that has a significantly larger radius difference from the actual trachea, such as the leaking part shown in FIG. 4A .
  • the tracheal tree acquisition module may prune the tracheal centerline based on the tracheal centerline, tracheal morphological connectivity and tracheal radius, and eliminate leaks in the tracheal tree model based on the pruning results.
  • the tracheal tree acquisition module calculates the radius of the tracheal branches based on the centerline of the trachea to obtain the minimum radius of each branch, and locates the part of the tracheal branch whose radius is greater than a times the minimum radius of the branch, that is, the part with leakage, wherein , a is a positive number greater than 1. Based on this, the tracheal tree acquisition module uses morphological expansion to generate a new trachea according to the centerline of the trachea and the radius of the upper and lower trachea of the leaking part, so as to realize the filtering of the leaking part of the trachea. For example, please refer to FIG. 4B , which shows an example diagram of a result obtained after the tracheal tree acquisition module in this embodiment prunes the tracheal tree model.
  • the present invention also provides a medical image segmentation method. Specifically, please refer to FIG. 5.
  • the medical image segmentation method can be implemented by the medical image segmentation device 1 shown in FIG. 1A or FIG. 1D, and specifically includes the following steps:
  • steps S11 to S15 correspond to the functions of the corresponding modules in the medical image segmentation device 1 shown in FIG. 1A or 1D . In order to save the length of the description, they will not be repeated here.
  • the present invention also provides a medical image segmentation device, which can realize the medical image segmentation method described in the present invention, but the realization device of the medical image segmentation method described in the present invention includes but is not limited to the following examples:
  • the structure of the medical image segmentation device, all structural deformations and replacements in the prior art made according to the principle of the present invention are included in the scope of protection of the present invention.
  • the medical image segmentation device of the present invention can obtain the first segmentation result of the pulmonary trachea according to the medical image, and obtain the second segmentation result of the pulmonary trachea according to the primary partial image. Based on this, the medical image segmentation device at least according to the The first segmentation result and the second segmentation result of the lung trachea are fused to obtain a trachea tree model.
  • the above process can be automatically completed by electronic equipment, basically without manual participation, and has high efficiency and accuracy.
  • the medical image segmentation device can be configured to extract the centerline of the trachea based on the trachea mask, and extract branch points and end points of the trachea based on the centerline of the trachea, so as to classify the lung trachea. Based on this, the medical image segmentation device can realize the main trachea model (corresponding to grade 1-4 trachea), branch trachea model (corresponding to grade 4-6 trachea) and local small trachea model (corresponding to 6 to 8 stages of trachea) segmentation.
  • the medical image segmentation device of the present invention adopts the idea of hierarchical segmentation described above, which can effectively improve the sensitivity of trachea segmentation below 2mm.
  • the present invention effectively overcomes various shortcomings in the prior art and has high industrial application value.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • Quality & Reliability (AREA)
  • Radiology & Medical Imaging (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

本发明提供一种医学影像分割装置及方法。所述医学影像分割装置包括：医学影像获取模块，用于获取医学影像，所述医学影像包括患者的至少部分肺部气管；第一分割模块，与所述医学影像获取模块相连，用于对所述医学影像进行分割以获取所述肺部气管的第一分割结果；第一局部影像获取模块，与所述医学影像获取模块相连，用于根据所述医学影像获取至少一个一级局部影像；第二分割模块，与所述第一局部影像获取模块相连，用于对所述一级局部影像进行分割以获取所述肺部气管的第二分割结果；气管树获取模块，与所述第一分割模块和所述第二分割模块相连，用于获取气管树模型。所述医学影像分割装置能够自动实现气管树模型的获取，效率和准确率较高。

Description

一种医学影像分割装置及方法 技术领域
本发明涉及一种图像处理装置，特别是涉及一种医学影像分割装置及方法。
背景技术
临床上肺支气管狭窄、慢性阻塞性肺病、闭塞性细支气管炎等疾病的诊断依赖于肺气管的定量化分析，构建气管树模型有助于定量分析肺气管形态变化。除此之外，气管树模型还能够应用于外科手术的支气管导航。然而，发明人在实际应用中发现，现有技术中主要依赖于医务人员通过人工方式对医学影像进行分割以得到气管树模型，此种方式效率较低。
发明内容
鉴于以上所述现有技术的缺点，本发明的目的在于提供一种医学影像分割装置及方法，用于解决现有技术中人工分割方式效率较低的问题。
为实现上述目的及其他相关目的，本发明的第一方面提供一种医学影像分割装置，所述医学影像分割装置包括：医学影像获取模块，用于获取医学影像，所述医学影像包括患者的至少部分肺部气管；第一分割模块，与所述医学影像获取模块相连，用于对所述医学影像进行分割以获取所述肺部气管的第一分割结果；第一局部影像获取模块，与所述医学影像获取模块相连，用于根据所述医学影像获取至少一个一级局部影像；第二分割模块，与所述第一局部影像获取模块相连，用于对所述一级局部影像进行分割以获取所述肺部气管的第二分割结果；气管树获取模块，与所述第一分割模块和所述第二分割模块相连，用于获取气管树模型，其中，所述气管树模型至少由所述肺部气管的第一分割结果和第二分割结果融合得到。
于所述第一方面的一实施例中，所述医学影像分割装置还包括：第二局部影像获取模块，与所述第一局部影像获取模块相连，用于根据所述一级局部影像获取至少一个二级局部影像；第三分割模块，与所述第二局部影像获取模块相连，用于对所述二级局部影像进行分割以获取所述肺部气管的第三分割结果；所述气管树获取模块还与所述第三分割模块相连，所述气管树模型至少由所述肺部气管的第一分割结果、第二分割结果和第三分割结果融合得到。
于所述第一方面的一实施例中，所述第一局部影像获取模块包括：气管分支点获取单元，与所述医学影像获取模块相连，用于获取所述医学影像中所述肺部气管的气管分支点；第一局部影像获取单元，与所述气管分支点获取单元相连，用于根据所述气管分支点和所述医学 影像获取所述一级局部影像。
于所述第一方面的一实施例中，所述第一局部影像获取模块包括：气管末端点获取单元，与所述医学影像获取模块相连，用于获取所述医学影像中所述肺部气管的气管末端点；第二局部影像获取单元，与所述气管末端点获取单元相连，用于根据所述气管末端点和所述医学影像获取所述一级局部影像。
于所述第一方面的一实施例中，所述气管末端点获取单元还用于获取所述肺部气管的中心线，并根据所述肺部气管的中心线获取所述气管末端点。
于所述第一方面的一实施例中，所述第二局部影像获取单元还用于对所述气管末端点进行聚类，并基于聚类结果获取所述一级局部影像。
于所述第一方面的一实施例中，所述医学影像分割装置还包括：断裂区域获取模块，与所述气管树获取模块相连，用于获取所述气管树模型中的断裂区域；断裂区域修复模块，与所述断裂区域获取模块相连，用于对所述断裂区域进行修复。
于所述第一方面的一实施例中，所述断裂区域修复模块包括：种子点获取单元，用于获取一种子点，所述种子点位于所述断裂区域内；气管修复单元，与所述种子点获取单元相连，用于基于所述种子点进行扩展以获取所述断裂区域的气管。
于所述第一方面的一实施例中，所述气管树获取模块还用于对所述气管树模型进行剪枝。
本发明的第二方面提供一种医学影像分割方法，所述医学影像分割方法包括：获取医学影像，所述医学影像包括患者的至少部分肺部气管；对所述医学影像进行分割以获取所述肺部气管的第一分割结果；根据所述医学影像获取至少一个一级局部影像；对所述一级局部影像进行分割以获取所述肺部气管的第二分割结果；获取气管树模型，其中，所述气管树模型至少由所述肺部气管的第一分割结果和第二分割结果融合得到。
如上所述，本发明所述医学影像分割装置及方法的一个技术方案具有以下有益效果：
所述医学影像分割装置能够根据医学影像获取肺部气管的第一分割结果，根据一级局部影像获取肺部气管的第二分割结果，基于此，所述医学影像分割装置至少根据所述肺部气管的第一分割结果和第二分割结果进行融合得到气管树模型。上述过程可以通过电子设备自动完成，基本无需人工参与，效率和准确率较高。
附图说明
图1A显示为本发明所述医学影像分割装置于一具体实施例中的结构示意图。
图1B显示为本发明所述医学影像分割装置于一具体实施例中获取的医学影像示例图。
图1C显示为本发明所述医学影像分割装置于一具体实施例中获取的一级局部影像示例图。
图1D显示为本发明所述医学影像分割装置于一具体实施例中的结构示意图。
图2显示为本发明所述医学影像分割装置于一具体实施例中第一局部影像获取模块的结构示意图。
图3A显示为本发明所述医学影像分割装置于一具体实施例中第一局部影像获取模块的另一结构示意图。
图3B显示为本发明所述医学影像分割装置于一具体实施例中对断裂气管进行修复的流程图。
图4A和图4B显示为本发明所述医学影像获取装置于一具体实施例中获取的气管分割结果示例图。
图5显示为本发明所述医学影像分割方法于一具体实施例中的流程图。
元件标号说明
1        医学影像分割装置
11       医学影像获取模块
12       第一分割模块
13       第一局部影像获取模块
131      气管分支点获取单元
132      第一局部影像获取单元
133      气管末端点获取单元
134      第二局部影像获取单元
14       第二分割模块
15       气管树获取模块
16       第二局部影像获取模块
17       第三分割模块
S11～S14 步骤
S21～S25 步骤
具体实施方式
以下通过特定的具体实例说明本发明的实施方式，本领域技术人员可由本说明书所揭露的内容轻易地了解本发明的其他优点与功效。本发明还可以通过另外不同的具体实施方式加以实施或应用，本说明书中的各项细节也可以基于不同观点与应用，在没有背离本发明的精神下进行各种修饰或改变。需说明的是，在不冲突的情况下，以下实施例及实施例中的特征可以相互组合。
需要说明的是，以下实施例中所提供的图示仅以示意方式说明本发明的基本构想，图示中仅显示与本发明中有关的组件而非按照实际实施时的组件数目、形状及尺寸绘制，其实际实施时各组件的型态、数量及比例可为一种随意的改变，且其组件布局型态也可能更为复杂。此外，在本文中，诸如“第一”、“第二”等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。
现有技术中主要依赖于医务人员通过人工方式对医学影像进行分割以得到气管树模型，此种方式效率较低。针对这一问题，本发明提供一种医学分割装置，所述医学影像分割装置能够根据医学影像获取肺部气管的第一分割结果，根据一级局部影像获取肺部气管的第二分割结果，基于此，所述医学影像分割装置至少根据所述肺部气管的第一分割结果和第二分割结果进行融合得到气管树模型。上述过程可以通过电子设备自动完成，基本无需人工参与，效率和准确率较高。
请参阅图1A，于本发明的一实施例中，所述医学影像分割装置1包括医学影像获取模块11、第一分割模块12、第一局部影像获取模块13、第二分割模块14和气管树获取模块15。
所述医学影像获取模块11用于获取医学影像，所述医学影像包括患者的至少部分肺部气管，所述医学影像例如为胸腔影像。本实施例中，所述医学影像的一个示例图如图1B所示。
所述第一分割模块12与所述医学影像获取模块11相连，用于对所述医学影像进行分割以获取所述肺部气管的第一分割结果。所述第一分割结果是由所述第一分割模块12通过对所述医学影像本身进行分割得到，属于肺部气管的整体分割结果。所述第一分割结果例如包括1～4级气管的分割结果。
所述第一局部影像获取模块13与所述医学影像获取模块11相连，用于根据所述医学影像获取至少一个一级局部影像，其中，各所述一级局部影像的范围小于所述医学影像的范围，并且，与所述医学影像相比，所述一级局部影像包括肺部气管更多的细节。本实施例中，所述一级局部影像的一个示例图如图1C所示。所述第一局部影像获取模块13例如可以通过对所述医学影像进行分割来得到所述一级局部影像。
所述第二分割模块14与所述第一局部影像获取模块13相连，用于对所述一级局部影像进行分割以获取所述肺部气管的第二分割结果。所述第二分割结果是由所述第二分割模块14对所述一级局部影像进行分割得到，与所述第一分割结果相比，所述第二分割结果是对更为细致的肺部气管的分割结果。所述第二分割结果例如包括4～6级气管的分割结果。
所述气管树获取模块15与所述第一分割模块12和所述第二分割模块14相连，用于获取气管树模型。其中，所述气管树模型至少由所述肺部气管的第一分割结果和第二分割结果融合得到。
可选地，请参阅图1D，所述医学影像分割装置1还包括第二局部影像获取模块16和第三分割模块17。
所述第二局部影像获取模块16与所述第一局部影像获取模块13相连，用于根据所述一级局部影像获取至少一个二级局部影像，其中，各所述二级局部影像的范围小于所述一级局部影像的范围，并且，与所述一级局部影像相比，所述二级局部影像包括肺部气管更进一步的细节。所述第二局部影像获取模块16例如可以通过对所述一级局部影像进行分割得到所述二级局部影像。
所述第三分割模块17与所述第二局部影像获取模块16相连，用于对所述二级局部影像进行分割以获取所述肺部气管的第三分割结果。所述第三分割结果是由所述第三分割模块17对所述二级局部影像进行分割得到，与所述第二分割结果相比，所述第三分割结果是对更为细致的肺部气管的分割结果。所述第三分割结果例如包括6～8级气管的分割结果。
所述气管树获取模块15还与所述第三分割模块17相连，此时，所述气管树模型至少由所述肺部气管的第一分割结果、第二分割结果和第三分割结果融合得到。
需要说明的是，本实施例中，所述医学影像分割装置1并不以上述结构为限。所述医学影像分割装置还可以包含其他模块以进一步对所述第二局部影进行逐级分割，从而得到包含更多肺部气管细节的、多个级别的局部影像，进而获取多个级别的分割结果，所述气管树获取模块基于所述多个级别的分割结果能够获取包含更多分支的气管树模型。例如，所述医学影像分割装置可以进一步对所述二级局部影像进行分割得到三级局部影像，对所述三级局部影像进行分割得到四级局部影像……；并且，所述医学影像分割装置可以根据三级局部影像获取第四分割结果，根据四级局部影像获取第五分割结果……；所述气管树获取模块可以进一步融合所述第四分割结果、所述第五分割结果……形成所述气管树模型。
根据以上描述可知，本实施例所述医学影像分割装置1能够根据医学影像获取肺部气管的第一分割结果，根据一级局部影像获取肺部气管的第二分割结果，基于此，所述医学影像 分割装置至少根据所述肺部气管的第一分割结果和第二分割结果进行融合得到气管树模型。上述过程可以通过电子设备自动完成，基本无需人工参与，效率和准确率较高。
此外，本实施例所述医学影像分割装置1能够获取不同等级肺部气管的分割结果，并将其融合至所述气管树模型中。例如，所述气管树模型可以同时包含1～8级气管，而相关技术中，受限于医学影像的显示范围和分辨率等，其只能实现4～6级气管的自动分割，因此，与相关技术相比，本实施例所述医学影像分割装置1获取的气管树模型能够包含更多级别的气管，从而为医务人员提供更加详细的气管信息。
请参阅图2，于本发明的一实施例中，所述第一局部影像获取模块13包括气管分支点获取单元131和第一局部影像获取单元132。所述气管分支点获取单元131与所述医学影像获取模块11相连，用于获取所述医学影像中所述肺部气管的气管分支点，其中，获取所述气管分支点的方式可以通过现有的图像识别技术等实现。所述第一局部影像获取单元132与所述气管分支点获取单元131相连，用于根据所述气管分支点和所述医学影像获取所述一级局部影像。
例如，所述气管分支点获取单元131可以通过图像识别技术从肺部气管掩膜中获取左右肺叶分支点，所述第一局部影像获取单元132可以根据所述左右肺叶分支点将所述医学影像分割为包括左肺支气管的一级局部影像和包括右肺支气管的一级局部影像。
可以理解的是，所述医学影像分割装置中的相应模块可以采用上述类似的技术，根据气管分支点将所述一级局部影像分割为二级局部影像，将所述二级局部影像分割为三级局部影像……，具体实现方式此处不作过多赘述。
请参阅图3A，于本发明的一实施例中，所述第一局部影像获取模块13包括气管末端点获取单元133和第二局部影像获取单元134。
所述气管末端点获取单元133与所述医学影像获取模块11相连，用于获取所述医学影像中所述肺部气管的气管末端点，其中，获取所述气管末端点的方式可以通过现有的图像识别技术等实现。
可选地，所述气管末端点获取单元133还用于获取所述医学影像中所述肺部气管的中心线，并根据所述肺部气管的中心线获取所述气管末端点。例如，所述气管末端点获取单元133可以获取所述肺部气管的中心线的端点作为所述气管末端点。
所述第二局部影像获取单元134与所述气管末端点获取单元133相连，用于根据所述气管末端点和所述医学影像获取所述一级局部影像。具体地，所述第二局部影像获取单元134可以获取各所述气管末端点周围某一范围内的影像作为所述一级局部影像，该范围的大小和 形状可以根据实际需求设置。
可选地，考虑到所述气管末端点获取单元133获取的气管末端点中可能存在距离较近的点，如果对各所述气管末端点均获取一个一级局部影像可能会增加不必要的运算量，针对这一问题，所述第二局部影像获取单元134还可以用于对所述气管末端点进行聚类以获取聚类结果，并基于所述聚类结果获取所述一级局部影像。具体地，经过聚类以后所述第二局部影像获取单元134能够获取多个气管末端点集合，基于所述多个气管末端点集合能够获取多个一级局部影像，其中，每个一级局部影像均包含一个气管末端点集合中的所有气管末端点。
可以理解的是，所述医学影像分割装置中的相应模块可以采用上述类似的技术，根据气管末端点将所述一级局部影像分割为二级局部影像，将所述二级局部影像分割为三级局部影像……，具体实现方式此处不作过多赘述。
考虑到放射影像的分辨率有限、小气管对比度较低、存在模糊等问题，气管分割容易出现断裂。针对这一问题，于本发明的一实施例中，所述医学影像分割装置还包括断裂区域获取模块和断裂区域修复模块。
所述断裂区域获取模块与所述气管树获取模块相连，用于获取所述气管树模型中的断裂区域。具体地，所述断裂区域获取模块可以通过提取所述肺部气管中各支气管的中心线进而获取各支气管的起始点和末端点，基于各支气管的起始点和末端点即可获取所述气管树模型中的断裂区域。
所述断裂区域修复模块与所述断裂区域修复模块相连，用于对所述断裂区域进行修复。
可选地，所述断裂区域修复模块包括种子点获取单元和气管修复单元，其中，所述种子点获取单元用于获取所述断裂区域内的一个或多个种子点。所述气管修复单元与所述种子点获取单元相连，用于基于所述种子点进行扩展生长以获取所述断裂区域的气管。
可选地，请参阅图3B，所述气管修复单元对气管进行修复的具体实现方法包括：
S11，获取所述气管树模型的输出概率图(probability map)和多尺度海森滤波图(multi-scale hessian-based filter map)。
S12，将所述概率图和所述多尺度海森滤波图进行叠加，以获取气管相似性概率图。
S13，采用所述气管树模型中的气管末端点作为种子点，基于所述气管相似性概率图，采用区域生长算法在所有种子点之间生成连接路径，并获取各连接路径的距离。
S14，对距离最小的连接路径所对应的种子点进行连接，以实现对所述断裂区域的修复。
此外，考虑到气管壁及其边界存在的模糊、断裂等情况容易导致气管分割出现泄漏，于本发明的一实施例中，所述气管树获取模块还用于对所述气管树模型进行剪枝。其中，泄漏 是指分割得到的气管树模型中存在的、与实际气管的半径差异明显较大的部分，例如图4A所示的泄漏部分。所述气管树获取模块可以基于气管中心线、气管的形态学连通性和气管半径对气管中心线进行剪枝，并基于剪枝结果消除所述气管树模型中的泄漏部分。
具体地，所述气管树获取模块基于气管中心线计算气管分支的半径进而得到各个分支的最小半径，并且定位气管分支中半径大于该分支最小半径a倍的部分，也即存在泄漏的部分，其中，a为大于1的正数。基于此，所述气管树获取模块根据气管中心线和泄漏部分上下段气管的半径、采用形态学膨胀生成新的气管，即可实现气管泄漏部分的滤除。例如，请参阅图4B，显示为本实施例中所述气管树获取模块对所述气管树模型进行剪枝后得到的一个结果示例图。
基于以上对所述医学影像分割装置的描述，本发明还提供一种医学影像分割方法。具体地，请参阅图5，于本发明的一实施例中，所述医学影像分割方法可以通过图1A或图1D所示的医学影像分割装置1实现，具体包括以下步骤：
S21，获取医学影像，所述医学影像包括患者的至少部分肺部气管。
S22，对所述医学影像进行分割以获取所述肺部气管的第一分割结果。
S23，根据所述医学影像获取至少一个一级局部影像。
S24，对所述一级局部影像进行分割以获取所述肺部气管的第二分割结果。
S25，获取气管树模型，其中，所述气管树模型至少由所述肺部气管的第一分割结果和第二分割结果融合得到。
上述步骤S11～S15与图1A或图1D所示医学影像分割装置1中的相应模块的功能一一对应，为节省说明书篇幅，此处不作过多赘述。
本发明所述的医学影像分割方法的保护范围不限于本实施例列举的步骤执行顺序，凡是根据本发明的原理所做的现有技术的步骤增减、步骤替换所实现的方案都包括在本发明的保护范围内。
本发明还提供一种医学影像分割装置，所述医学影像分割装置可以实现本发明所述的医学影像分割方法，但本发明所述的医学影像分割方法的实现装置包括但不限于本实施例列举的医学影像分割装置的结构，凡是根据本发明的原理所做的现有技术的结构变形和替换，都包括在本发明的保护范围内。
本发明所述医学影像分割装置能够根据医学影像获取肺部气管的第一分割结果，根据一级局部影像获取肺部气管的第二分割结果，基于此，所述医学影像分割装置至少根据所述肺部气管的第一分割结果和第二分割结果进行融合得到气管树模型。上述过程可以通过电子设 备自动完成，基本无需人工参与，效率和准确率较高。
此外，所述医学影像分割装置能够被配置为基于气管掩膜提取气管中心线，并基于气管中心线提取气管分支点和末端点，进而对肺部气管进行分级。基于此，所述医学影像分割装置能够通过对肺部气管进行分级分割来实现主气管模型(对应1～4级气管)、分支气管模型(对应4～6级气管)和局部小气管模型(对应6～8级气管)的分割。本发明所述医学影像分割装置采用上述分级分割的思路，能够有效提升2mm以下气管分割的敏感性。
综上所述，本发明有效克服了现有技术中的种种缺点而具高度产业利用价值。
上述实施例仅例示性说明本发明的原理及其功效，而非用于限制本发明。任何熟悉此技术的人士皆可在不违背本发明的精神及范畴下，对上述实施例进行修饰或改变。因此，举凡所属技术领域中具有通常知识者在未脱离本发明所揭示的精神与技术思想下所完成的一切等效修饰或改变，仍应由本发明的权利要求所涵盖。

Claims (10)

  1. 一种医学影像分割装置，其特征在于，所述医学影像分割装置包括：
    医学影像获取模块，用于获取医学影像，所述医学影像包括患者的至少部分肺部气管；
    第一分割模块，与所述医学影像获取模块相连，用于对所述医学影像进行分割以获取所述肺部气管的第一分割结果；
    第一局部影像获取模块，与所述医学影像获取模块相连，用于根据所述医学影像获取至少一个一级局部影像；
    第二分割模块，与所述第一局部影像获取模块相连，用于对所述一级局部影像进行分割以获取所述肺部气管的第二分割结果；
    气管树获取模块，与所述第一分割模块和所述第二分割模块相连，用于获取气管树模型，其中，所述气管树模型至少由所述肺部气管的第一分割结果和第二分割结果融合得到。
  2. 根据权利要求1所述的医学影像分割装置，其特征在于，所述医学影像分割装置还包括：
    第二局部影像获取模块，与所述第一局部影像获取模块相连，用于根据所述一级局部影像获取至少一个二级局部影像；
    第三分割模块，与所述第二局部影像获取模块相连，用于对所述二级局部影像进行分割以获取所述肺部气管的第三分割结果；
    所述气管树获取模块还与所述第三分割模块相连，所述气管树模型至少由所述肺部气管的第一分割结果、第二分割结果和第三分割结果融合得到。
  3. 根据权利要求1所述的医学影像分割装置，其特征在于，所述第一局部影像获取模块包括：
    气管分支点获取单元，与所述医学影像获取模块相连，用于获取所述医学影像中所述肺部气管的气管分支点；
    第一局部影像获取单元，与所述气管分支点获取单元相连，用于根据所述气管分支点和所述医学影像获取所述一级局部影像。
  4. 根据权利要求1所述的医学影像分割装置，其特征在于，所述第一局部影像获取模块包括：
    气管末端点获取单元，与所述医学影像获取模块相连，用于获取所述医学影像中所述肺部气管的气管末端点；
    第二局部影像获取单元，与所述气管末端点获取单元相连，用于根据所述气管末端点和所述医学影像获取所述一级局部影像。
  5. 根据权利要求4所述的医学影像分割装置，其特征在于：所述气管末端点获取单元还用于获取所述肺部气管的中心线，并根据所述肺部气管的中心线获取所述气管末端点。
  6. 根据权利要求4所述的医学影像分割装置，其特征在于：所述第二局部影像获取单元还用于对所述气管末端点进行聚类，并基于聚类结果获取所述一级局部影像。
  7. 根据权利要求1所述的医学影像分割装置，其特征在于，所述医学影像分割装置还包括：
    断裂区域获取模块，与所述气管树获取模块相连，用于获取所述气管树模型中的断裂区域；
    断裂区域修复模块，与所述断裂区域获取模块相连，用于对所述断裂区域进行修复。
  8. 根据权利要求7所述的医学影像分割装置，其特征在于，所述断裂区域修复模块包括：
    种子点获取单元，用于获取一种子点，所述种子点位于所述断裂区域内；
    气管修复单元，与所述种子点获取单元相连，用于基于所述种子点进行扩展以获取所述断裂区域的气管。
  9. 根据权利要求1所述的医学影像分割装置，其特征在于：所述气管树获取模块还用于对所述气管树模型进行剪枝。
  10. 一种医学影像分割方法，其特征在于，所述医学影像分割方法包括：
    获取医学影像，所述医学影像包括患者的至少部分肺部气管；
    对所述医学影像进行分割以获取所述肺部气管的第一分割结果；
    根据所述医学影像获取至少一个一级局部影像；
    对所述一级局部影像进行分割以获取所述肺部气管的第二分割结果；
    获取气管树模型，其中，所述气管树模型至少由所述肺部气管的第一分割结果和第二分割结果融合得到。
PCT/CN2021/137328 2021-05-11 2021-12-13 一种医学影像分割装置及方法 WO2022237154A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110512792.0 2021-05-11
CN202110512792.0A CN113139968B (zh) 2021-05-11 2021-05-11 一种医学影像分割装置及方法

Publications (1)

Publication Number Publication Date
WO2022237154A1 true WO2022237154A1 (zh) 2022-11-17

Family

ID=76816903

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/137328 WO2022237154A1 (zh) 2021-05-11 2021-12-13 一种医学影像分割装置及方法

Country Status (2)

Country Link
CN (1) CN113139968B (zh)
WO (1) WO2022237154A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113139968B (zh) * 2021-05-11 2023-08-22 上海杏脉信息科技有限公司 一种医学影像分割装置及方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080071160A1 (en) * 2004-06-22 2008-03-20 Koninklijke Philips Electronics N.V. Displaying A Tracheobronchial Tree
CN108171703A (zh) * 2018-01-18 2018-06-15 东北大学 一种从胸部ct图像中自动提取气管树的方法
CN110378923A (zh) * 2019-07-25 2019-10-25 杭州健培科技有限公司 一种智能肺部气管树分割提取和分级的方法与装置
CN113139968A (zh) * 2021-05-11 2021-07-20 上海杏脉信息科技有限公司 一种医学影像分割装置及方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006085254A1 (en) * 2005-02-11 2006-08-17 Koninklijke Philips Electronics N.V. Method of automatic extraction of the pulmonary artery tree from 3d medical images
WO2019000455A1 (zh) * 2017-06-30 2019-01-03 上海联影医疗科技有限公司 图像分割的方法及系统
CN108765445B (zh) * 2018-05-29 2021-08-20 上海联影医疗科技股份有限公司 一种肺气管分割方法及装置
CN111325729A (zh) * 2020-02-19 2020-06-23 青岛海信医疗设备股份有限公司 基于生物医学影像的生物组织的分割方法和通信终端
CN112651969B (zh) * 2021-02-08 2023-04-07 福州大学 结合多信息融合网络和区域增长的气管树分级提取方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080071160A1 (en) * 2004-06-22 2008-03-20 Koninklijke Philips Electronics N.V. Displaying A Tracheobronchial Tree
CN108171703A (zh) * 2018-01-18 2018-06-15 东北大学 一种从胸部ct图像中自动提取气管树的方法
CN110378923A (zh) * 2019-07-25 2019-10-25 杭州健培科技有限公司 一种智能肺部气管树分割提取和分级的方法与装置
CN113139968A (zh) * 2021-05-11 2021-07-20 上海杏脉信息科技有限公司 一种医学影像分割装置及方法

Also Published As

Publication number Publication date
CN113139968A (zh) 2021-07-20
CN113139968B (zh) 2023-08-22

Similar Documents

Publication Publication Date Title
CN105574859B (zh) 一种基于ct图像的肝脏肿瘤分割方法及装置
CN108171703B (zh) 一种从胸部ct图像中自动提取气管树的方法
CN108133476B (zh) 一种肺结节自动检测方法及系统
CN104992445A (zh) 一种ct图像肺实质的自动分割方法
CN103824295A (zh) 一种肺部ct图像中粘连血管型肺结节的分割方法
CN112651969B (zh) 结合多信息融合网络和区域增长的气管树分级提取方法
CN104504737B (zh) 一种从肺部ct图像获得三维气管树的方法
WO2022237154A1 (zh) 一种医学影像分割装置及方法
CN112006772B (zh) 完整人体体外呼吸道建立方法及系统
JP2008043736A (ja) 医用画像処理装置及び医用画像処理方法
CN107481251A (zh) 一种从肺部ct图像中提取末端支气管树的方法
CN104899851A (zh) 一种分割肺结节图像的方法
CN117373070B (zh) 血管分段标注的方法及装置、电子设备和存储介质
Bian et al. Small airway segmentation in thoracic computed tomography scans: A machine learning approach
CN110766667A (zh) 基于轮廓样本库匹配和模板引导的受损骨骼修复建模方法
CN104182456A (zh) 一种基于mrs-mm目标匹配模型的空间实体增量提取方法
CN115965641A (zh) 一种基于deeplabv3+网络的咽部图像分割及定位方法
CN111145226A (zh) 基于ct影像的三维肺部特征提取方法
Zhang et al. Difference-deformable convolution with pseudo scale instance map for cell localization
Huang et al. BronchusNet: Region and Structure Prior Embedded Representation Learning for Bronchus Segmentation and Classification
CN113052857A (zh) 一种基于CovSegNet的肺部病变图像分割方法
Shao et al. A segmentation method of airway from chest CT image based on VGG-Unet neural network
Bian et al. Accurate airway centerline extraction based on topological thinning using graph-theoretic analysis
CN112201329A (zh) 模板引导与数据驱动的三维碎骨分割与拼接方法
CN114511512A (zh) 一种基于交互指导的血管图像分割方法

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: 21941716

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: 21941716

Country of ref document: EP

Kind code of ref document: A1