CN113139968B - Medical image segmentation device and method - Google Patents
Medical image segmentation device and method Download PDFInfo
- Publication number
- CN113139968B CN113139968B CN202110512792.0A CN202110512792A CN113139968B CN 113139968 B CN113139968 B CN 113139968B CN 202110512792 A CN202110512792 A CN 202110512792A CN 113139968 B CN113139968 B CN 113139968B
- Authority
- CN
- China
- Prior art keywords
- tracheal
- medical image
- segmentation
- trachea
- local 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.)
- Active
Links
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
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
The invention provides a medical image segmentation device and a medical image segmentation method. The medical image segmentation apparatus includes: a medical image acquisition module for acquiring a medical image, the medical image comprising at least part of a lung trachea of a patient; the first segmentation module is connected with the medical image acquisition module and is used for segmenting the medical image to acquire a first segmentation result of the lung trachea; the first local image acquisition module is connected with the medical image acquisition module and is used for acquiring at least one primary local image according to the medical image; the second segmentation module is connected with the first local image acquisition module and is used for segmenting the primary local image to acquire a second segmentation result of the lung trachea; and the tracheal tree acquisition module is connected with the first segmentation module and the second segmentation module and is used for acquiring a tracheal tree model. The medical image segmentation device can automatically acquire the tracheal tree model, and has higher efficiency and accuracy.
Description
Technical Field
The present invention relates to image processing apparatuses, and more particularly, to a medical image segmentation apparatus and method.
Background
Diagnosis of diseases such as pulmonary bronchoconstriction, chronic obstructive pulmonary disease, bronchiolitis obliterans and the like in clinic depends on quantitative analysis of pulmonary bronchi, and the establishment of a tracheal tree model is helpful for quantitative analysis of morphological changes of the pulmonary bronchi. In addition, the tracheal tree model can be applied to bronchial navigation for surgical procedures. However, the inventor finds that in practical application, the prior art mainly relies on medical personnel to segment a medical image manually to obtain a model of a tracheal tree, and this mode is inefficient.
Disclosure of Invention
In view of the above-mentioned drawbacks of the prior art, an object of the present invention is to provide a medical image segmentation apparatus and method for solving the problem of low efficiency of manual segmentation in the prior art.
To achieve the above and other related objects, a first aspect of the present invention provides a medical image segmentation apparatus comprising: a medical image acquisition module for acquiring a medical image, the medical image comprising at least part of a lung trachea of a patient; the first segmentation module is connected with the medical image acquisition module and is used for segmenting the medical image to acquire a first segmentation result of the lung trachea; the first local image acquisition module is connected with the medical image acquisition module and is used for acquiring at least one primary local image according to the medical image; the second segmentation module is connected with the first local image acquisition module and is used for segmenting the primary local image to acquire a second segmentation result of the lung trachea; and the tracheal tree acquisition module is connected with the first segmentation module and the second segmentation module and is used for acquiring a tracheal tree model, wherein the tracheal tree model is at least obtained by fusion of a first segmentation result and a second segmentation result of the lung trachea.
In an embodiment of the first aspect, the medical image segmentation apparatus further includes: the second local image acquisition module is connected with the first local image acquisition module and is used for acquiring at least one second local image according to the first local image; the third segmentation module is connected with the second local image acquisition module and is used for segmenting the second local image to acquire a third segmentation result of the lung trachea; the tracheal tree acquisition module is further connected with the third segmentation module, and the tracheal tree model is at least obtained by fusing a first segmentation result, a second segmentation result and a third segmentation result of the pulmonary tracheal.
In an embodiment of the first aspect, the first local image acquisition module includes: the tracheal branch point acquisition unit is connected with the medical image acquisition module and is used for acquiring the tracheal branch point of the lung trachea in the medical image; the first local image acquisition unit is connected with the tracheal branch point acquisition unit and is used for acquiring the first-stage local image according to the tracheal branch point and the medical image.
In an embodiment of the first aspect, the first local image acquisition module includes: the tracheal end point acquisition unit is connected with the medical image acquisition module and is used for acquiring a tracheal end point of the lung trachea in the medical image; the second local image acquisition unit is connected with the tracheal end point acquisition unit and is used for acquiring the first-level local image according to the tracheal end point and the medical image.
In an embodiment of the first aspect, the tracheal endpoint obtaining unit is further configured to obtain a centerline of the pulmonary trachea, and obtain the tracheal endpoint according to the centerline of the pulmonary trachea.
In an embodiment of the first aspect, the second local image obtaining unit is further configured to cluster the tracheal endpoint, and obtain the primary local image based on a clustering result.
In an embodiment of the first aspect, the medical image segmentation apparatus further includes: the fracture area acquisition module is connected with the tracheal tree acquisition module and is used for acquiring fracture areas in the tracheal tree model; and the fracture area repairing module is connected with the fracture area acquisition module and is used for repairing the fracture area.
In an embodiment of the first aspect, the fracture area repair module includes: a seed point obtaining unit, configured to obtain a seed point, where the seed point is located in the fracture area; and the tracheal repair unit is connected with the seed point acquisition unit and is used for expanding based on the seed point to acquire the tracheal of the fracture area.
In an embodiment of the first aspect, the tracheal tree obtaining 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 comprising: acquiring a medical image, the medical image including at least a portion of a lung airway of a patient; segmenting the medical image to obtain a first segmentation result of the lung trachea; acquiring at least one primary local image according to the medical image; segmenting the primary local image to obtain a second segmentation result of the lung trachea; and obtaining a tracheal tree model, wherein the tracheal tree model is at least obtained by fusing a first segmentation result and a second segmentation result of the pulmonary trachea.
As described above, one technical solution of the medical image segmentation apparatus and method of the present invention has the following beneficial effects:
the medical image segmentation device can acquire a first segmentation result of the lung trachea according to the medical image, acquire a second segmentation result of the lung trachea according to the first-level local image, and based on the first segmentation result and the second segmentation result of the lung trachea, the medical image segmentation device fuses at least to acquire a trachea tree model. The process can be automatically completed through the electronic equipment, manual participation is basically not needed, and the efficiency and the accuracy are high.
Drawings
Fig. 1A is a schematic structural diagram of a medical image segmentation apparatus according to an embodiment of the invention.
FIG. 1B is a diagram illustrating an example of a medical image acquired by a medical image segmentation apparatus according to an embodiment of the present invention.
FIG. 1C is a diagram illustrating an exemplary primary local image acquired by a medical image segmentation apparatus according to an embodiment of the present invention.
Fig. 1D is a schematic structural diagram of a medical image segmentation apparatus according to an embodiment of the invention.
Fig. 2 is a schematic structural diagram of a first local image acquisition module of the medical image segmentation apparatus according to an embodiment of the invention.
Fig. 3A is a schematic diagram of another structure of a first local image capturing module of the medical image segmentation apparatus according to an embodiment of the invention.
Fig. 3B is a flowchart illustrating the repair of a fractured trachea according to an embodiment of the medical image segmentation apparatus according to the present invention.
Fig. 4A and 4B are diagrams illustrating an exemplary result of tracheal segmentation obtained by the medical image acquisition device according to an embodiment of the present invention.
Fig. 5 is a flowchart of a medical image segmentation method according to an embodiment of the invention.
Description of element reference numerals
1. Medical image segmentation device
11. Medical image acquisition module
12. First dividing module
13. First local image acquisition module
131. Tracheal branch point acquisition unit
132. First local image acquisition unit
133. Tracheal end point acquisition unit
134. Second local image acquisition unit
14. Second dividing module
15. Tracheal tree acquisition module
16. Second local image acquisition module
17. Third dividing module
S11 to S14 steps
S21 to S25 steps
Detailed Description
Other advantages and effects of the present invention will become apparent to those skilled in the art from the following disclosure, which describes the embodiments of the present invention with reference to specific examples. The invention may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present invention. It should be noted that the following embodiments and features in the embodiments may be combined with each other without conflict.
It should be noted that the illustrations provided in the following embodiments merely illustrate the basic concept of the present invention by way of illustration, and only the components related to the present invention are shown in the illustrations, not according to the number, shape and size of the components in actual implementation, and the form, number and proportion of each component in actual implementation may be arbitrarily changed, and the layout of the components may be more complex. Moreover, relational terms such as "first," "second," and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.
In the prior art, medical staff is mainly relied on to divide medical images in a manual mode to obtain a tracheal tree model, and the mode is low in efficiency. In order to solve the problems, the invention provides a medical image segmentation device, which can acquire a first segmentation result of a lung trachea according to a medical image, acquire a second segmentation result of the lung trachea according to a first-level local image, and at least fuse the first segmentation result and the second segmentation result of the lung trachea to obtain a trachea tree model. The process can be automatically completed through the electronic equipment, manual participation is basically not needed, and the efficiency and the accuracy are high.
Referring to fig. 1A, in an embodiment of the invention, the medical image segmentation apparatus 1 includes a medical image acquisition module 11, a first segmentation module 12, a first local image acquisition module 13, a second segmentation module 14, and a tracheal tree acquisition module 15.
The medical image acquisition module 11 is configured to acquire a medical image, including at least a portion of a lung trachea of a patient, such as a chest image. In this embodiment, an exemplary view of the medical image is shown in fig. 1B.
The first segmentation module 12 is connected to the medical image acquisition module 11, and is configured to segment the medical image to acquire a first segmentation result of the pulmonary trachea. The first segmentation result is obtained by the first segmentation module 12 through segmenting the medical image, and belongs to the overall segmentation result of the lung trachea. The first segmentation result includes, for example, a segmentation result of a 1-4-stage trachea.
The first local image acquisition module 13 is connected to the medical image acquisition module 11, and is configured to acquire at least one primary local image according to the medical image, where a range of each primary local image is smaller than a range of the medical image, and the primary local image includes more details of a pulmonary trachea than the medical image. In this embodiment, an exemplary diagram of the primary local image is shown in fig. 1C. The first local image acquisition module 13 may obtain the first local image by, for example, dividing the medical image.
The second segmentation module 14 is connected to the first local image acquisition module 13, and is configured to segment the first local image to acquire a second segmentation result of the pulmonary trachea. The second segmentation result is obtained by segmenting the primary local image by the second segmentation module 14, and compared with the first segmentation result, the second segmentation result is obtained by segmenting a finer lung trachea. The second division result includes, for example, a division result of 4 to 6-stage 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 tracheal tree model is at least obtained by fusing a first segmentation result and a second segmentation result of the lung trachea.
Optionally, referring to fig. 1D, the medical image segmentation apparatus 1 further includes a second local image acquisition module 16 and a third segmentation module 17.
The second local image acquisition module 16 is connected to the first local image acquisition module 13, and is configured to acquire at least one secondary local image according to the primary local image, where a range of each secondary local image is smaller than a range of the primary local image, and the secondary local image includes further details of a pulmonary trachea compared to the primary local image. The second local image acquisition module 16 may obtain the second local image by, for example, dividing the first local image.
The third segmentation module 17 is connected to the second local image acquisition module 16, and is configured to segment the second local image to acquire a third segmentation result of the pulmonary trachea. The third segmentation result is obtained by segmenting the secondary local image by the third segmentation module 17, and compared with the second segmentation result, the third segmentation result is obtained by segmenting a finer lung trachea. The third segmentation result includes, for example, a segmentation result of 6-8-stage trachea.
The tracheal tree obtaining module 15 is further connected to the third segmentation module 17, where the tracheal tree model is obtained by fusing at least the first segmentation result, the second segmentation result and the third segmentation result of the pulmonary tracheal.
In the present embodiment, the medical image segmentation apparatus 1 is not limited to the above configuration. The medical image segmentation device may further include other modules to further segment the second partial image step by step, so as to obtain a plurality of levels of partial images including more lung tracheal details, and further obtain a plurality of levels of segmentation results, where the tracheal tree acquisition module may acquire a tracheal tree model including more branches based on the plurality of levels of segmentation results. For example, the medical image segmentation apparatus may further segment the secondary local image to obtain a tertiary local image, and segment the tertiary local image to obtain a quaternary local image … …; in addition, the medical image segmentation device may obtain a fourth segmentation result according to the third-level local image, and obtain a fifth segmentation result … … according to the fourth-level local image; the tracheal tree acquisition module may further fuse the fourth segmentation result and the fifth segmentation result … … to form the tracheal tree model.
As can be seen from the above description, the medical image segmentation apparatus 1 according to the present embodiment is capable of obtaining a first segmentation result of a lung and a trachea according to a medical image, obtaining a second segmentation result of the lung and the trachea according to a first-level local image, and based on this, the medical image segmentation apparatus performs fusion according to at least the first segmentation result and the second segmentation result of the lung and the trachea to obtain a tracheal tree model. The process can be automatically completed through the electronic equipment, manual participation is basically not needed, and the efficiency and the accuracy are high.
In addition, the medical image segmentation apparatus 1 according to the present embodiment can acquire segmentation results of lung airways of different levels and fuse the segmentation results into the airway tree model. For example, the tracheal tree model may include 1-8 level tracheal at the same time, but in the related art, the display range and resolution of the medical image are limited, and only the automatic segmentation of 4-6 level tracheal can be achieved, so compared with the related art, the tracheal tree model obtained by the medical image segmentation device 1 in this embodiment can include more levels of tracheal, thereby providing more detailed tracheal information for medical staff.
Referring to fig. 2, in an embodiment of the invention, the first local image acquisition module 13 includes a tracheal branch point acquisition unit 131 and a first local image acquisition unit 132. The tracheal branch point obtaining unit 131 is connected to the medical image obtaining module 11, and is configured to obtain a tracheal branch point of the pulmonary tracheal in the medical image, where a manner of obtaining the tracheal branch point may be implemented by using an existing image recognition technology or the like. The first local image acquisition unit 132 is connected to the tracheal branch point acquisition unit 131, and is configured to acquire the first local image according to the tracheal branch point and the medical image.
For example, the tracheal branch point obtaining unit 131 may obtain left and right pulmonary lobe branch points from a pulmonary tracheal mask by an image recognition technique, and the first partial image obtaining unit 132 may divide the medical image into a first partial image including a left pulmonary bronchus and a first partial image including a right pulmonary bronchus according to the left and right pulmonary lobe branch points.
It can be appreciated that the corresponding modules in the medical image segmentation apparatus may adopt the similar techniques described above, and the primary local image is segmented into the secondary local image according to the tracheal branch point, and the secondary local image is segmented into the tertiary local image … …, and detailed description thereof will be omitted herein.
Referring to fig. 3A, in an embodiment of the invention, the first local image acquisition module 13 includes a tracheal endpoint acquisition unit 133 and a second local image acquisition unit 134.
The tracheal endpoint obtaining unit 133 is connected to the medical image obtaining module 11, and is configured to obtain a tracheal endpoint of the pulmonary tracheal in the medical image, where a manner of obtaining the tracheal endpoint may be implemented by using an existing image recognition technology or the like.
Optionally, the tracheal end point obtaining unit 133 is further configured to obtain a centerline of the pulmonary trachea in the medical image, and obtain the tracheal end point according to the centerline of the pulmonary trachea. For example, the tracheal end point acquisition unit 133 may acquire an end point of the center line of the pulmonary trachea as the tracheal end point.
The second local image acquisition unit 134 is connected to the tracheal end point acquisition unit 133, and is configured to acquire the first local image according to the tracheal end point and the medical image. Specifically, the second local image obtaining unit 134 may obtain, as the first local image, an image in a certain range around each tracheal endpoint, where the size and shape of the range may be set according to actual requirements.
Alternatively, in view of the problem that there may be a point closer to the tracheal end point obtained by the tracheal end point obtaining unit 133, if one primary local image is obtained for each tracheal end point, an unnecessary amount of computation may be increased, for this problem, the second local image obtaining unit 134 may be further configured to cluster the tracheal end points to obtain a clustering result, and obtain the primary local image based on the clustering result. Specifically, the second local image obtaining unit 134 may obtain a plurality of sets of tracheal end points after clustering, and may obtain a plurality of primary local images based on the plurality of sets of tracheal end points, where each primary local image includes all tracheal end points in one set of tracheal end points.
It can be appreciated that the corresponding modules in the medical image segmentation apparatus may adopt the similar techniques described above, and the primary local image is segmented into the secondary local image according to the tracheal end point, and the secondary local image is segmented into the tertiary local image … …, and detailed description thereof will be omitted herein.
Considering the problems of limited resolution of the radiological image, low contrast of the small trachea, blurring and the like, the segmentation of the trachea is easy to break. In view of this problem, in an embodiment of the present invention, the medical image segmentation apparatus further includes a fracture region acquisition module and a fracture region restoration module.
The fracture area acquisition module is connected with the tracheal tree acquisition module and is used for acquiring the fracture area in the tracheal tree model. Specifically, the fracture region acquisition module may acquire a starting point and an ending point of each bronchus by extracting a center line of each bronchus in the pulmonary bronchus, and may acquire a fracture region in the tracheal tree model based on the starting point and the ending point of each bronchus.
The fracture area repair module is connected with the fracture area repair module and is used for repairing the fracture area.
Optionally, the fracture area repairing module comprises a seed point acquiring unit and an air pipe repairing unit, wherein the seed point acquiring unit is used for acquiring one or more seed points in the fracture area. The tracheal repair unit is connected with the seed point acquisition unit and is used for performing expansion growth based on the seed point so as to acquire the tracheal of the fracture area.
Optionally, referring to fig. 3B, a specific implementation method of repairing a trachea by the trachea repairing unit includes:
s11, obtaining an output probability map (probability map) and a multi-scale hessian filter map (multi-scale hessian-based filter map) of the tracheal tree model.
S12, superposing the probability map and the multi-scale hessian filter map to obtain a tracheal similarity probability map.
S13, adopting a tracheal end point in the tracheal tree model as a seed point, generating connection paths among all the seed points by adopting a region growing algorithm based on the tracheal similarity probability map, and acquiring the distance of each connection path.
And S14, connecting seed points corresponding to the connecting paths with the smallest distance so as to repair the fracture area.
In addition, in consideration of the fact that the air pipe wall and the boundary thereof have the conditions of blurring, fracture and the like, the air pipe segmentation is easy to leak, and in an embodiment of the invention, the air pipe tree acquisition module is further used for pruning the air pipe tree model. Here, the leakage refers to a portion existing in the segmented tracheal tree model and having a significantly large difference from the radius of the actual trachea, for example, a leakage portion shown in fig. 4A. The tracheal tree acquisition module can prune the tracheal centerline based on the tracheal centerline, morphological connectivity of the trachea and the tracheal radius, and eliminate leakage in the tracheal tree model based on pruning results.
Specifically, the tracheal tree acquisition module calculates the radius of the tracheal branch based on the tracheal center line to obtain the minimum radius of each branch, and positions the part of the tracheal branch with the radius being greater than a times the minimum radius of the branch, namely the part with leakage, wherein a is a positive number greater than 1. Based on the method, the tracheal tree acquisition module generates a new tracheal by morphological expansion according to the tracheal center line and the radiuses of the upper and lower sections of the tracheal of the leakage part, and can realize the filtering of the leakage part of the tracheal. For example, referring to fig. 4B, a diagram of an example of a result obtained by pruning the tracheal tree model by the tracheal tree obtaining module in this embodiment is shown.
Based on the description of the medical image segmentation device, the invention further provides a medical image segmentation method. Specifically, referring to fig. 5, in an embodiment of the present invention, the medical image segmentation method may be implemented by the medical image segmentation apparatus 1 shown in fig. 1A or 1D, and specifically includes the following steps:
s21, acquiring a medical image, wherein the medical image comprises at least part of lung trachea of a patient.
S22, segmenting the medical image to obtain a first segmentation result of the lung trachea.
S23, at least one primary local image is obtained according to the medical image.
S24, segmenting the primary local image to obtain a second segmentation result of the lung trachea.
S25, acquiring a tracheal tree model, wherein the tracheal tree model is at least obtained by fusion of a first segmentation result and a second segmentation result of the pulmonary trachea.
The steps S11 to S15 correspond to the functions of the corresponding modules in the medical image segmentation apparatus 1 shown in fig. 1A or fig. 1D one by one, and are not repeated here for saving the description space.
The protection scope of the medical image segmentation method of the present invention is not limited to the execution sequence of the steps listed in the present embodiment, and all the schemes implemented by the steps of increasing or decreasing and step replacing in the prior art according to the principles of the present invention are included in the protection scope of the present invention.
The invention also provides a medical image segmentation device, which can realize the medical image segmentation method of the invention, but the device for realizing the medical image segmentation method of the invention comprises but is not limited to the structure of the medical image segmentation device listed in the embodiment, and all structural modifications and substitutions of the prior art according to the principles of the invention are included in the protection scope of the invention.
The medical image segmentation device can acquire a first segmentation result of the lung trachea according to the medical image, acquire a second segmentation result of the lung trachea according to the first-level local image, and fuse the first segmentation result and the second segmentation result of the lung trachea at least to acquire a trachea tree model based on the first segmentation result and the second segmentation result of the lung trachea. The process can be automatically completed through the electronic equipment, manual participation is basically not needed, and the efficiency and the accuracy are high.
Furthermore, the medical image segmentation apparatus can be configured to extract a tracheal centerline based on the tracheal mask and to extract tracheal branch points and end points based on the tracheal centerline, thereby classifying the pulmonary trachea. Based on this, the medical image segmentation apparatus can achieve segmentation of the main tracheal model (corresponding to 1-4 level tracheal), the branch tracheal model (corresponding to 4-6 level tracheal), and the local small tracheal model (corresponding to 6-8 level tracheal) by performing hierarchical segmentation on the pulmonary tracheal. The medical image segmentation device adopts the thought of grading segmentation, and can effectively improve the sensitivity of tracheal segmentation below 2 mm.
In summary, the present invention effectively overcomes the disadvantages of the prior art and has high industrial utility value.
The above embodiments are merely illustrative of the principles of the present invention and its effectiveness, and are not intended to limit the invention. Modifications and variations may be made to the above-described embodiments by those skilled in the art without departing from the spirit and scope of the invention. Accordingly, it is intended that all equivalent modifications and variations of the invention be covered by the claims, which are within the ordinary skill of the art, be within the spirit and scope of the present disclosure.
Claims (6)
1. A medical image segmentation apparatus, the medical image segmentation apparatus comprising:
a medical image acquisition module for acquiring a medical image, the medical image comprising at least part of a lung trachea of a patient;
the first segmentation module is connected with the medical image acquisition module and is used for segmenting the medical image to acquire a first segmentation result of the lung trachea;
the first local image acquisition module is connected with the medical image acquisition module and is used for acquiring at least one primary local image according to the medical image, the range of the primary local image is smaller than that of the medical image, and the primary local image contains more details of a lung trachea compared with the medical image;
the second segmentation module is connected with the first local image acquisition module and is used for segmenting the primary local image to acquire a second segmentation result of the lung trachea;
the tracheal tree acquisition module is connected with the first segmentation module and the second segmentation module and is used for acquiring a tracheal tree model, wherein the tracheal tree model is at least obtained by fusing a first segmentation result and a second segmentation result of the pulmonary tracheal;
the fracture area acquisition module is connected with the tracheal tree acquisition module and is used for acquiring fracture areas in the tracheal tree model;
the fracture area repairing module is connected with the fracture area acquiring module and is used for repairing the fracture area;
the first local image acquisition module comprises an end point acquisition unit of the trachea and a second local image acquisition unit, wherein the end point acquisition unit of the trachea is used for acquiring end points of the trachea of the lung in the medical image, and the second local image acquisition unit is used for clustering the end points of the trachea and acquiring the first-level local image based on a clustering result;
the fracture area repairing module comprises a seed point obtaining unit and a trachea repairing unit, wherein the seed point obtaining unit is used for obtaining a seed point which is positioned in the fracture area, and the trachea repairing unit is used for expanding based on the seed point to obtain a trachea of the fracture area; the method for repairing the trachea by the trachea repairing unit comprises the following steps:
s11, obtaining an output probability map and a multi-scale Hemson filter map of the tracheal tree model; s12, superposing the probability map and the multi-scale hessian filter map to obtain a tracheal similarity probability map; s13, adopting a tracheal end point in the tracheal tree model as a seed point, generating connection paths among all seed points by adopting a region growing algorithm based on the tracheal similarity probability map, and acquiring the distance of each connection path; and S14, connecting seed points corresponding to the connecting paths with the smallest distance so as to repair the fracture area.
2. The medical image segmentation apparatus as set forth in claim 1, further comprising:
the second local image acquisition module is connected with the first local image acquisition module and is used for acquiring at least one second local image according to the first local image;
the third segmentation module is connected with the second local image acquisition module and is used for segmenting the second local image to acquire a third segmentation result of the lung trachea;
the tracheal tree acquisition module is further connected with the third segmentation module, and the tracheal tree model is at least obtained by fusing a first segmentation result, a second segmentation result and a third segmentation result of the pulmonary tracheal.
3. The medical image segmentation device of claim 1, wherein the first local image acquisition module comprises:
the tracheal branch point acquisition unit is connected with the medical image acquisition module and is used for acquiring the tracheal branch point of the lung trachea in the medical image;
the first local image acquisition unit is connected with the tracheal branch point acquisition unit and is used for acquiring the first-stage local image according to the tracheal branch point and the medical image.
4. The medical image segmentation apparatus as set forth in claim 1, wherein: the tracheal endpoint acquisition unit is also used for acquiring the central line of the pulmonary trachea and acquiring the tracheal endpoint according to the central line of the pulmonary trachea.
5. The medical image segmentation apparatus as set forth in claim 1, wherein: the tracheal tree acquisition module is also used for pruning the tracheal tree model.
6. A medical image segmentation method, characterized in that the medical image segmentation method comprises:
acquiring a medical image, the medical image including at least a portion of a lung airway of a patient;
segmenting the medical image to obtain a first segmentation result of the lung trachea;
acquiring at least one primary local image according to the medical image, wherein the range of the primary local image is smaller than that of the medical image, and the primary local image contains more details of a lung trachea compared with the medical image;
segmenting the primary local image to obtain a second segmentation result of the lung trachea;
obtaining a tracheal tree model, wherein the tracheal tree model is at least obtained by fusing a first segmentation result and a second segmentation result of the pulmonary trachea;
obtaining a fracture area in the tracheal tree model, and repairing the fracture area;
wherein, obtaining at least one first-level local image according to the medical image comprises: acquiring a tracheal end point of the pulmonary trachea in the medical image; clustering the tracheal end points, and acquiring the primary local image based on a clustering result;
repairing the fracture region includes: obtaining a seed point, wherein the seed point is positioned in the fracture area; obtaining an output probability map and a multi-scale hessian filter map of the tracheal tree model; superposing the probability map and the multi-scale hessian filter map to obtain a tracheal similarity probability map; adopting a tracheal end point in the tracheal tree model as a seed point, generating connection paths among all seed points by adopting a region growing algorithm based on the tracheal similarity probability map, and acquiring the distance of each connection path; and connecting seed points corresponding to the connecting paths with the smallest distance so as to repair the fracture area.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110512792.0A CN113139968B (en) | 2021-05-11 | 2021-05-11 | Medical image segmentation device and method |
| PCT/CN2021/137328 WO2022237154A1 (en) | 2021-05-11 | 2021-12-13 | Medical image segmentation apparatus and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110512792.0A CN113139968B (en) | 2021-05-11 | 2021-05-11 | Medical image segmentation device and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113139968A CN113139968A (en) | 2021-07-20 |
| CN113139968B true CN113139968B (en) | 2023-08-22 |
Family
ID=76816903
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110512792.0A Active CN113139968B (en) | 2021-05-11 | 2021-05-11 | Medical image segmentation device and method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113139968B (en) |
| WO (1) | WO2022237154A1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113139968B (en) * | 2021-05-11 | 2023-08-22 | 上海杏脉信息科技有限公司 | Medical image segmentation device and method |
Citations (5)
| 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 |
| CN108765445A (en) * | 2018-05-29 | 2018-11-06 | 上海联影医疗科技有限公司 | A kind of lung qi pipe dividing method and device |
| CN109215032A (en) * | 2017-06-30 | 2019-01-15 | 上海联影医疗科技有限公司 | The method and system of image segmentation |
| CN111325729A (en) * | 2020-02-19 | 2020-06-23 | 青岛海信医疗设备股份有限公司 | Biological tissue segmentation method based on biomedical images and communication terminal |
| CN112651969A (en) * | 2021-02-08 | 2021-04-13 | 福州大学 | Trachea tree hierarchical extraction method combining multi-information fusion network and regional growth |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006000953A1 (en) * | 2004-06-22 | 2006-01-05 | Koninklijke Philips Electronics N.V. | Displaying a tracheobronchial tree |
| CN108171703B (en) * | 2018-01-18 | 2020-09-15 | 东北大学 | Method for automatically extracting trachea tree from chest CT image |
| CN110378923A (en) * | 2019-07-25 | 2019-10-25 | 杭州健培科技有限公司 | A kind of method and apparatus that the segmentation of intelligence intratracheal tree is extracted and is classified |
| CN113139968B (en) * | 2021-05-11 | 2023-08-22 | 上海杏脉信息科技有限公司 | Medical image segmentation device and method |
-
2021
- 2021-05-11 CN CN202110512792.0A patent/CN113139968B/en active Active
- 2021-12-13 WO PCT/CN2021/137328 patent/WO2022237154A1/en active Application Filing
Patent Citations (5)
| 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 |
| CN109215032A (en) * | 2017-06-30 | 2019-01-15 | 上海联影医疗科技有限公司 | The method and system of image segmentation |
| CN108765445A (en) * | 2018-05-29 | 2018-11-06 | 上海联影医疗科技有限公司 | A kind of lung qi pipe dividing method and device |
| CN111325729A (en) * | 2020-02-19 | 2020-06-23 | 青岛海信医疗设备股份有限公司 | Biological tissue segmentation method based on biomedical images and communication terminal |
| CN112651969A (en) * | 2021-02-08 | 2021-04-13 | 福州大学 | Trachea tree hierarchical extraction method combining multi-information fusion network and regional growth |
Non-Patent Citations (1)
| Title |
|---|
| 基于CT影像的肺部气管树分割算法的研究;朱辰坤;《中国优秀博硕士学位论文全文数据库(硕士)医药卫生科技辑》;20150515;摘要、第3-4章 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113139968A (en) | 2021-07-20 |
| WO2022237154A1 (en) | 2022-11-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104992445B (en) | A kind of automatic division method of CT images pulmonary parenchyma | |
| CN110232383A (en) | A kind of lesion image recognition methods and lesion image identifying system based on deep learning model | |
| CN109166105A (en) | The malignancy of tumor risk stratification assistant diagnosis system of artificial intelligence medical image | |
| CN104504737B (en) | A kind of method that three-dimensional tracheae tree is obtained from lung CT image | |
| CN104867147A (en) | SYNTAX automatic scoring method based on coronary angiogram image segmentation | |
| CN101116103A (en) | Method of automatic extraction of the pulmonary artery tree from 3d medical images | |
| CN102693540A (en) | Liver segmentation method and system thereof | |
| CN112651969B (en) | Trachea tree hierarchical extraction method combining multi-information fusion network and regional growth | |
| JP2008043736A (en) | Medical image processing apparatus and method for processing medical image | |
| CN113139968B (en) | Medical image segmentation device and method | |
| CN109727260A (en) | A kind of three-dimensional lobe of the lung dividing method based on CT images | |
| CN109410166A (en) | Full-automatic partition method for pulmonary parenchyma CT image | |
| Bian et al. | Small airway segmentation in thoracic computed tomography scans: A machine learning approach | |
| CN111145226B (en) | Three-dimensional lung feature extraction method based on CT image | |
| CN103985122B (en) | The dirtiest extracting method of based on heart CT image | |
| CN111260669A (en) | Lung lobe segmentation method and device based on CT image | |
| Li et al. | Renal cortex segmentation using optimal surface search with novel graph construction | |
| CN111080556A (en) | Method, system, equipment and medium for strengthening trachea wall of CT image | |
| CN117373070B (en) | Method and device for labeling blood vessel segments, electronic equipment and storage medium | |
| CN110246126A (en) | A method of extracting terminal bronchi tree from lung CT image | |
| Seibold et al. | Accurate fine-grained segmentation of human anatomy in radiographs via volumetric pseudo-labeling | |
| CN111415341A (en) | Pneumonia stage evaluation method, pneumonia stage evaluation device, pneumonia stage evaluation medium and electronic equipment | |
| Perchet et al. | Advanced navigation tools for virtual bronchoscopy | |
| CN114693622B (en) | Plaque erosion automatic detection system based on artificial intelligence | |
| US20230410291A1 (en) | Method and system for segmenting and identifying at least one tubular structure in medical images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |