CN115713590A - Three-dimensional reconstruction image processing method and system based on CT - Google Patents

Abstract

The invention relates to the technical field of image processing, in particular to a three-dimensional reconstruction image processing method and system based on CT. Establishing an initial three-dimensional space according to the CT image data of the patient, segmenting the CT image data of the patient according to the marking points to obtain each tissue part, and finally generating a three-dimensional display model of the lesion organ according to each tissue part in the initial three-dimensional space. Therefore, the two-dimensional CT image of the patient is converted into the three-dimensional CT image to be displayed, so that the doctor can recognize each organ, the three-dimensional display model of the lesion organ is generated by each tissue part divided by the CT image, the lesion organ is directly displayed in the three-dimensional space, doctors and students and other inexperienced doctors are helped to recognize the focus, the accuracy of disease diagnosis is further improved, even non-professionals can recognize the lesion organ in the CT image through the scheme, the recognition difficulty of the CT image is greatly reduced, and the working efficiency of the doctor is improved.

Description

A CT-based three-dimensional reconstruction image processing method and system

technical field

The invention relates to the technical field of image processing, in particular to a CT-based three-dimensional reconstruction image processing method and system.

Background technique

CT scanning scans a layer of a certain thickness of the human body inspection site through the X-ray beam, and the detector receives the X-rays that pass through this layer, converts the received X-rays into visible light, and converts them into electrical signals by a photoelectric converter, and then simulates them. The analog/digital converter (Analog/Digital converter) converts it into a digital signal, and finally inputs the digital signal into a computer for processing to obtain a two-dimensional CT image.

CT scanning is a commonly used medical auxiliary diagnosis and treatment method. However, due to the abstraction of two-dimensional CT images and the complexity of organ structures, it is difficult to identify them. Only professional doctors such as radiologists or experienced doctors can quickly see them. Understand. For new medical students or inexperienced doctors, it takes a long time to see abnormal lesions in patient CT images.

Contents of the invention

Aiming at the deficiencies in the prior art, the present invention proposes a CT-based three-dimensional reconstruction image processing method and system to reduce the recognition difficulty of CT images.

In a first aspect, the present invention provides a CT-based three-dimensional reconstruction image processing method.

In the first practicable manner, a CT-based three-dimensional reconstruction image processing method includes:

Obtain patient CT image data;

Obtain the initial three-dimensional space according to the patient's CT image data;

Obtain the marker points, and segment and process the patient's CT image data according to the marker points to obtain each tissue part;

In the initial three-dimensional space, a three-dimensional display model of the diseased organ is generated according to each tissue location.

In combination with the first achievable way, in the second achievable way, the initial three-dimensional space is obtained according to the patient's CT image data, including:

Determine the initial position according to the image size, number of pixels and image layer thickness in the patient's CT image data;

Obtain three-dimensional coordinates according to the coordinate information in the patient's CT image data;

An initial three-dimensional space is established according to the initial position and three-dimensional coordinates.

Combined with the first possible way, in the third possible way, the CT image data of the patient is segmented and processed according to the marker points to obtain various tissue parts, including:

Each marked point is taken as the initial origin, and each initial origin is expanded according to a preset growth rule to obtain several sets of pixel points, and each set of pixel points is each tissue part.

Combined with the third possible way, in the fourth possible way, the growth rule is: include the pixels adjacent to and similar to the initial origin into the initial origin, form a new origin, and repeat the new origin as the initial origin Do the above.

Combining with the first achievable way, in the fifth achievable way, a three-dimensional display model of the diseased organ is generated according to each tissue part in the initial three-dimensional space, including:

Obtain the label of each tissue part; the label is used to represent the name and lesion condition of each tissue part;

Generate a three-dimensional model according to each tissue part in the initial three-dimensional space;

The label of each tissue part is used as the first label of each tissue part corresponding to the organ in the three-dimensional model;

Judging whether each organ in the three-dimensional model is abnormal, and obtaining the judgment result; and using the judgment result as the second label of each organ in the three-dimensional model;

According to the first label and the second label, an organ with pathological changes in the three-dimensional model is marked to obtain a three-dimensional display model of the diseased organ.

Combined with the fifth possible way, in the sixth possible way, the labels of various tissue parts are obtained, including:

Perform matching operations on each tissue part in the preset organ database to obtain tissue images similar to each tissue part;

The name and lesion corresponding to the tissue image with the highest similarity to each tissue part are used as labels for each tissue part.

Combined with the fifth possible way, in the seventh possible way, it is judged whether there is any abnormality in each organ in the three-dimensional model, and the judgment result is obtained, including:

Compare the 3D model with the preset standard 3D model, and segment the parts that do not meet the parameter range of the standard 3D model;

Scale up the segmented parts and obtain texture information;

Analyze according to the texture information, judge whether there is an abnormality, and obtain the judgment result.

In combination with the fifth possible way, in the eighth possible way, according to the first label and the second label, the organs with pathological conditions in the three-dimensional model are marked, and the three-dimensional display model of the diseased organ is obtained, including:

displaying only the organs with the first label in the three-dimensional model with the first color;

Display the organs with only the second label in the 3D model with the second color;

The organ with both the first label and the second label in the three-dimensional model is displayed with the third color to obtain a three-dimensional display model of the diseased organ.

In a second aspect, the present invention provides a CT-based three-dimensional reconstruction image processing system.

In a ninth practicable manner, a CT-based three-dimensional reconstruction image processing system includes:

A patient CT image data acquisition module configured to acquire patient CT image data;

an initial three-dimensional space establishment module configured to establish an initial three-dimensional space according to the patient's CT image data;

The tissue site acquisition module is configured to acquire marker points, and segment and process the patient's CT image data according to the marker points to obtain each tissue site;

The three-dimensional display model generation module of the diseased organ is configured to generate a three-dimensional display model of the diseased organ according to various tissue parts in the initial three-dimensional space.

As can be seen from the above technical solutions, the beneficial technical effects of the present invention are as follows:

1. Establish an initial three-dimensional space based on the patient's CT image data, segment and process the patient's CT image data according to the marker points, and obtain various tissue parts. Finally, generate a three-dimensional display model of the diseased organ according to each tissue part in the initial three-dimensional space. In this way, converting the two-dimensional CT images of patients into three-dimensional displays is beneficial for doctors to identify various organs, and the three-dimensional display models of diseased organs are generated through the tissue parts divided into CT images, so that the diseased organs can be directly displayed in three-dimensional space Come out, help medical students and other inexperienced doctors to identify lesions, and further improve the accuracy of disease diagnosis. Even non-professionals can identify diseased organs in CT images through this solution, which greatly reduces the difficulty of identifying CT images. , Improve the doctor's work efficiency.

2. Identify the lesion on the tissue part of the two-dimensional CT image, and judge the organ abnormality of the three-dimensional model, so as to identify the diseased organ in the patient's CT image in different dimensions, which is conducive to improving the accuracy of disease diagnosis. It is more conducive to deepening doctors' understanding of the pathological changes of organs in different dimensions of diseases.

3. Use different colors to mark the identification of different dimensions, further reducing the difficulty of identifying diseased organs in the 3D model, and helping to distinguish the appearance of diseased organs in different dimensions.

Description of drawings

In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the accompanying drawings that need to be used in the description of the specific embodiments or the prior art. Throughout the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, elements or parts are not necessarily drawn in actual scale.

Fig. 1 is a schematic diagram of a CT-based three-dimensional reconstruction image processing method provided by the present invention;

FIG. 2 is a schematic structural diagram of a CT-based three-dimensional reconstruction image processing system provided by the present invention.

Detailed ways

Embodiments of the technical solutions of the present invention will be described in detail below in conjunction with the accompanying drawings. The following examples are only used to illustrate the technical solutions of the present invention more clearly, and therefore are only examples, rather than limiting the protection scope of the present invention.

It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application shall have the usual meanings understood by those skilled in the art to which the present invention belongs. The terms "first", "second" and the like in the description and claims of the embodiments of the present disclosure and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific sequence or sequence. It should be understood that the data so used may be interchanged under appropriate circumstances so as to facilitate the embodiments of the disclosed embodiments described herein. Furthermore, the terms "comprising" and "having", as well as any variations thereof, are intended to cover a non-exclusive inclusion. Unless stated otherwise, the term "plurality" means two or more. In the embodiments of the present disclosure, the character "/" indicates that the preceding and following objects are an "or" relationship. For example, A/B means: A or B. The term "and/or" is an associative relationship describing objects, indicating that there can be three relationships. For example, A and/or B means: A or B, or, A and B, these three relationships. The term "correspondence" may refer to an association relationship or a binding relationship, and the correspondence between A and B means that there is an association relationship or a binding relationship between A and B.

As shown in FIG. 1, this embodiment provides a CT-based three-dimensional reconstruction image processing method, including:

Step S01, acquiring patient CT image data;

Step S02, acquiring an initial three-dimensional space according to the patient's CT image data;

Step S03, obtaining the marker points, and performing segmentation processing on the CT image data of the patient according to the marker points, to obtain each tissue part;

Step S04, generating a three-dimensional display model of the diseased organ according to each tissue location in the initial three-dimensional space.

Optionally, obtaining the initial three-dimensional space according to the patient's CT image data includes: determining the initial position according to the image size, the number of pixels and the image layer thickness in the patient's CT image data; obtaining the three-dimensional coordinates according to the coordinate information in the patient's CT image data; An initial three-dimensional space is established according to the initial position and three-dimensional coordinates.

Optionally, a table lookup operation is performed on the image size, number of pixels and image layer thickness in the preset human body position table to obtain the initial position. The preset human body position table stores the corresponding relationship between image size, pixel point and image layer thickness and the initial position.

In some embodiments, the part of the human body scanned by the CT image is determined according to the size of the image, the number of pixels, and the thickness of the image layer, and the position of the part of the human body in the three-dimensional human body model is determined as the initial position. The approximate position, range, and direction of the human body part scanned by the image in three-dimensional space.

Optionally, the coordinate information in the patient's CT image data includes voxel coordinates. When the CT image is formed, the selected slice is divided into several cuboids with the same volume, and each cuboid is a voxel.

Optionally, obtain the three-dimensional coordinates according to the coordinate information through the following formula:

Among them, (X, Y) is the coordinates of the voxel of point a in the CT image, (x`, y`, z`) is the three-dimensional coordinates of point a corresponding to the real world, and the coordinates of the patient are (x, y, z); X _x , X _y , and X _z are the angles between the X-axis direction of the CT image and the x, y, and z-axis directions of the patient, respectively; Y _x , Y y , _and Y _z are the angles between the Y-axis direction of the CT image and the x, y, and The angle relationship in the z-axis direction, △x is the pixel interval in the x-direction, △y is the pixel interval in the y-direction, (S _x , S _y , S _z ) is the three-dimensional corresponding to the pixel point in the upper left corner of the CT image coordinate. In some embodiments, the cosine of the included angle represents the relationship of the included angle, and the cosine of the included angle is the projection value of each direction of the CT image on the three directions of the patient coordinate system.

Optionally, establishing an initial three-dimensional space based on the initial position and three-dimensional coordinates includes: initializing a three-dimensional space through a computer according to the initial position, the length and width of the CT image, the number of pixels, and the three-dimensional coordinates, referring to the prior art.

Optionally, obtaining the marker points includes: selecting the coordinates of a certain point in the CT image as the marker point, or the doctor inputs the coordinates of any point in the CT image as the marker point.

Optionally, the CT image data of the patient is segmented according to the marker points to obtain various tissue parts, including: taking each marker point as the initial origin, expanding each initial origin according to a preset growth rule, and obtaining several sets of pixel points , each set of pixel points is each tissue site.

Optionally, the growth rule is: incorporate pixels adjacent to and similar to the initial origin into the initial origin to form a new origin, and repeat the above operations with the new origin as the initial origin.

Optionally, according to the gray level and texture of the initial origin and its adjacent pixels, it is judged whether the two are similar.

Optionally, when the difference between the initial origin and its adjacent pixel points is within a preset range, it is determined that the two are similar.

In some embodiments, segmenting the CT image data of the patient according to the marker points includes the following steps:

Step 21, using the marked point as the initial origin;

Step 22. Determine whether the initial origin has adjacent and similar pixels, and if so, include the adjacent and similar pixels into the initial origin set to form a new initial origin;

Step 23, repeat step 22 until the initial origin has no similar and adjacent pixel points, the set formed by the initial origin is a segmented area, that is, the tissue site;

Step 24, if there are pixels in the patient's CT image that have not been selected or judged similarity, select one of the pixels that have not been selected or judged similarity as a marker point, and return to step 21, step 22, step twenty three;

Step 25. If there is no pixel in the patient's CT image that has not been selected or whose similarity has been judged, it is determined that the segmentation of the patient's CT image is completed.

Optionally, generating a three-dimensional display model of the diseased organ according to each tissue part in the initial three-dimensional space, including: obtaining the label of each tissue part; the label is used to represent the name and lesion condition of each tissue part; The three-dimensional model is generated by the part; the label of each tissue part is used as the first label of the corresponding organ of each tissue part in the three-dimensional model; whether there is abnormality in each organ in the three-dimensional model is judged, and the judgment result is obtained; The second label of the organ; according to the first label and the second label, the organ with pathological changes in the three-dimensional model is marked to obtain a three-dimensional display model of the diseased organ.

Optionally, acquiring the labels of each tissue part includes: performing a matching operation on each tissue part in a preset organ database to obtain tissue images similar to each tissue part; The name and lesion status of each tissue site were used as labels.

In some embodiments, reference may be made to prior art for generating a three-dimensional model according to each tissue site in the initial three-dimensional space.

Optionally, judging whether each organ in the three-dimensional model is abnormal, and obtaining the judgment result includes: comparing the three-dimensional model with a preset standard three-dimensional model, and segmenting parts that do not meet the parameter range of the standard three-dimensional model; Segmented parts are scaled up in equal proportions, and texture information is obtained; analysis is performed based on the texture information to determine whether there is any abnormality, and obtain the judgment result.

Optionally, according to the first label and the second label, the organs with pathological conditions in the three-dimensional model are marked to obtain the three-dimensional display model of the diseased organ, including: displaying the organs with only the first label in the three-dimensional model with the first color; Displaying the organs with only the second label in the 3D model with the second color; displaying the organs with both the first label and the second label in the 3D model with the third color to obtain a three-dimensional display model of the diseased organ. In this way, not only the lesions are identified on the tissue parts of the two-dimensional CT images, but also the organ abnormalities of the three-dimensional model are judged, so that the abnormalities in the patient's CT images can be identified in two dimensions, which is conducive to providing a basis for disease diagnosis. Accuracy is more conducive to deepening doctors' understanding of the pathological changes of organs in different dimensions of diseases. The identification of different dimensions is marked with different colors, which further reduces the difficulty of identifying diseased organs in the 3D model, and is conducive to distinguishing the appearance of diseased organs in different dimensions.

Optionally, input the lesion conditions of various tissue parts and the abnormal conditions of organs in the three-dimensional model into the disease case storage database for analysis and search, obtain similar cases, and display the similar cases as reference materials for the three-dimensional display model of diseased organs to the doctor . In this way, according to the lesions of various tissue parts and the abnormality of organs in the three-dimensional model, similar cases can be found in the disease case storage, providing more reference materials for doctors, avoiding doctors to search for similar cases again, and improving work efficiency. Moreover, it provides learning and accumulation opportunities for medical students or doctors with insufficient experience, which is practical and convenient.

Optionally, when the similarity of the similar case is greater than a preset threshold, the disease diagnosis result of the similar case is directly displayed to the doctor as a reference disease result of the three-dimensional display model of the diseased organ.

As shown in FIG. 2, this embodiment provides a CT-based three-dimensional reconstruction image processing system, including: patient CT image data acquisition module 101, initial three-dimensional space establishment module 102, tissue site acquisition module 103, three-dimensional model display model of diseased organs Generate module 104 . The patient CT image data acquisition module 101 is configured to acquire patient CT image data; the initial three-dimensional space establishment module 102 is configured to establish an initial three-dimensional space according to the patient CT image data; the tissue site acquisition module 103 is configured to acquire marker points, and according to the marker Carry out segmentation processing on the patient's CT image data to obtain various tissue parts; the three-dimensional model display model generation module 104 of the diseased organ is configured to generate a three-dimensional display model of the diseased organ according to each tissue part in the initial three-dimensional space.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it still The technical solutions described in the foregoing embodiments may be modified, or some or all of the technical features may be equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention. , which should be included within the scope of the claims and description of the present invention.

Claims (9)

1. A three-dimensional reconstruction image processing method based on CT is characterized by comprising the following steps:

acquiring CT image data of a patient;

acquiring an initial three-dimensional space according to CT image data of a patient;

acquiring a mark point, and segmenting the CT image data of the patient according to the mark point to obtain each tissue part;

and generating a three-dimensional display model of the lesion organ according to each tissue part in the initial three-dimensional space.

2. The method of claim 1, wherein acquiring an initial three-dimensional space from CT image data of a patient comprises:

determining an initial position according to the image size, the number of pixel points and the image layer thickness in the CT image data of the patient;

acquiring three-dimensional coordinates according to coordinate information in CT image data of a patient;

and establishing an initial three-dimensional space according to the initial position and the three-dimensional coordinates.

3. The method of claim 1, wherein segmenting the CT image data of the patient based on the marker points to obtain tissue regions comprises:

and expanding each initial origin point by taking each mark point as the initial origin point according to a preset growth rule to obtain a plurality of pixel point sets, wherein each pixel point set is each tissue part.

4. The method of claim 3, wherein the growth rule is: and bringing the pixel points which are adjacent and similar to the initial origin into the initial origin to form a new origin, and repeatedly executing the operation by taking the new origin as the initial origin.

5. The method of claim 1, wherein generating a three-dimensional representation of a diseased organ from each tissue site within the initial three-dimensional space comprises:

obtaining labels of each tissue site; the label is used for representing the name and the pathological change condition of each tissue part;

generating a three-dimensional model from each tissue site within the initial three-dimensional space;

taking the label of each tissue part as a first label of the corresponding organ of each tissue part in the three-dimensional model;

judging whether each organ in the three-dimensional model has abnormal conditions or not, and obtaining a judgment result; taking the judgment result as a second label of each organ in the three-dimensional model;

and labeling the organ with the pathological change condition in the three-dimensional model according to the first label and the second label to obtain a three-dimensional display model of the pathological change organ.

6. The method of claim 5, wherein obtaining a label for each tissue site comprises:

matching each tissue part in a preset organ database to obtain a tissue image similar to each tissue part;

the name and lesion state corresponding to the tissue image having the highest similarity for each tissue region are used as the label for each tissue region.

7. The method of claim 5, wherein determining whether each organ in the three-dimensional model is abnormal and obtaining the determination result comprises:

comparing the three-dimensional model with a preset standard three-dimensional model, and segmenting parts which do not accord with the parameter range of the standard three-dimensional model;

amplifying the segmented parts in equal proportion, and acquiring texture information;

and analyzing according to the texture information, judging whether an abnormal condition exists or not, and obtaining a judgment result.

8. The method of claim 5, wherein labeling an organ with a pathological condition in the three-dimensional model according to the first label and the second label to obtain a three-dimensional display model of the pathological organ comprises:

displaying only the first labeled organ in the three-dimensional model through a first color;

displaying only the second labeled organ in the three-dimensional model by a second color;

and displaying the organ with the first label and the second label in the three-dimensional model through a third color to obtain a three-dimensional display model of the lesion organ.

9. A CT-based three-dimensional reconstructed image processing system, comprising:

a patient CT image data acquisition module configured to acquire patient CT image data;

an initial three-dimensional space establishing module configured to establish an initial three-dimensional space according to the patient CT image data;

the tissue part acquisition module is configured to acquire the mark points and perform segmentation processing on the CT image data of the patient according to the mark points to obtain tissue parts;

and the lesion organ three-dimensional model display model generation module is configured to generate a lesion organ three-dimensional display model according to each tissue part in the initial three-dimensional space.

Images