Diagnostic method and system of a kind of medical image under VR environment
Technical field
The present invention relates to technical field of medical instruments, particularly relate to a kind of diagnostic method of medical image under VR environment and
System.
Background technology
Doctor diagnoses according to CT scan (Computed Tomography, CT) to conditions of patients
When, it is necessary to the skilled grasp of human anatomic structure, to normal physiological mechanism, Histological Study, and to various diseases caused by
Symptom, sign, auxiliary examination result (clinical manifestation), development lapses to and prognosis will have certain understanding.Doctor is intuitively
The image seen from CT pieces, it is likely that simply a jiao of iceberg, the experience of doctor's long term accumulation can form one in brain
The network system, same CT image may have diversified result.
CT is, using the penetrability and fluorescence of X-ray, those who are investigated to be placed in into fluorescent screen (or image amplifier) and X
Between spool, x-ray on fluorescent screen (or image amplifier) through forming visible image after human body and regard the inspection read.
X-ray beam is scanned to the certain thickness aspect in human body a part, and the x-ray information through the aspect, transformation are collected by detector
It is visible ray, electric signal is changed into by optical-electrical converter, then digital information is converted to through analog/digital converter, input is calculated
Machine treatment, and computer is calculated the x-ray attenuation coefficient of each voxel by the data for obtaining, and is arranged in matrix, reconstruction image;Again
Through digital/analog converter, the blockage that each numeral is converted to black and white GTG turns into pixel (pixel), is arranged in a matrix into
CT images.
But, existing CT images are all bidimensional images, and institutional framework overlaps each other, it is easy to the situation failed to pinpoint a disease in diagnosis,
Especially in location overlap and the display at embedded position.
Virtual reality (Virtual Reality, VR) is provided and gives people one as one of in recent years most hot technology, VR
Brand-new steric environment, allows people really to observe a stereopsis for specific things, and VR technologies are applied to medical science
In the diagnosis of image, accuracy of the doctor to difficult diagnostic imaging can be greatly increased.
The content of the invention
The technical problem to be solved in the present invention is to provide diagnostic method and system of a kind of medical image under VR environment, with
Solve the problems, such as existing for prior art easy fails to pinpoint a disease in diagnosis.
In order to solve the above technical problems, the embodiment of the present invention provides a kind of diagnostic method of medical image under VR environment,
Including:
Obtain two-dimensional ct image;
The two-dimensional ct image to obtaining carries out three-dimensional reconstruction operation, obtains the corresponding three-dimensional mould of the two-dimensional ct image
Type;
In default VR display modules, the threedimensional model is cut off using section and section is shown.
Further, the two-dimensional ct image of described pair of acquisition carries out three-dimensional reconstruction operation, obtains the two-dimensional ct shadow
As corresponding threedimensional model includes:
By the different region of the two-dimensional ct video generation;
After Area generation, the corresponding threedimensional model of the two-dimensional ct image is rebuild according to iso-surface patch algorithm.
Further, the region that the two-dimensional ct video generation is different includes:
S1, the two-dimensional ct image order to obtaining is scanned, and acquisition first is not belonged to also in region
Pixel, and described first pixel not belonged to also in region that will be obtained is used as sub-pixel (x0, y0);
S2, centered on sub-pixel (x0, y0), detect around sub-pixel (x0, y0) close to pixel (x, y)
Whether pixel value is less than default sub-pixel value, if so, then by less than the pixel (x, y) and seed of default sub-pixel value
Pixel (x0, y0) merges in same region, and will be pressed into storehouse less than the pixel (x, y) of default sub-pixel value;
S3, ejects a pixel from storehouse, using the pixel for ejecting as new sub-pixel (x0, y0), returns to S2
Continue executing with;
S4, when storehouse is space-time, returns to S1 and continues executing with;
S5, repeats S1-S4, until each pixel in the two-dimensional ct image has ownership.
Further, after the Area generation, the corresponding three-dimensional of the two-dimensional ct image is rebuild according to iso-surface patch algorithm
Model includes:
After Area generation, the pixel value in region is set as a constant a, the pixel value outside region is set as
0, it is the face of a to obtain contour surface according to iso-surface patch algorithm, wherein, the contour surface for getting is the two-dimensional ct image for the face of a
Corresponding threedimensional model.
Further, it is described using section the threedimensional model is cut off and section show including:
WillVector withVector makees dot product, obtains the cosine value COS (α) of ∠ α;
Judge COS (α) whether more than 0;
If more than 0, A points above cutting planes are along normal direction, then retaining A points and rendering A points;
Otherwise, then A points are rejected;
Wherein, ∠ α=∠ OAP, O point is threedimensional model central point, and A points are any point on threedimensional model, and P points are to cut
A coordinate points in cutting plane,It is the normal line vector of cutting planes.
Further, methods described also includes:
If | OA |>Width, then be rendered to default color by A points;
Wherein, | OA | is the distance between O points to A points, and O points are threedimensional model central point, and A points are appointing on threedimensional model
A bit, width is cross-section image display width to meaning.
The embodiment of the present invention also provides a kind of diagnostic system of medical image under VR environment, including:
Acquisition module, for obtaining two-dimensional ct image;
Module is rebuild, for carrying out three-dimensional reconstruction operation to the two-dimensional ct image for obtaining, the two-dimensional ct shadow is obtained
As corresponding threedimensional model;
Display module, in default VR display modules, the threedimensional model is cut off using section and
Section shows.
Further, the reconstruction module includes:
Area generation unit, for by the different region of the two-dimensional ct video generation;
Model Reconstruction unit, for Area generation after, rebuild the two-dimensional ct image according to iso-surface patch algorithm corresponding
Threedimensional model;
Wherein, the Area generation unit includes:
Acquisition submodule, is scanned for the two-dimensional ct image order to obtaining, and obtains first and does not return also
Belong to the pixel in region, and described first pixel not belonged to also in region that will be obtained as sub-pixel (x0,
y0);
Merge submodule, for centered on sub-pixel (x0, y0), detect around sub-pixel (x0, y0) close to
Pixel (x, y) pixel value whether be less than default sub-pixel value, if so, then will be less than default sub-pixel value picture
Plain (x, y) merges in same region with sub-pixel (x0, y0), and will be less than default sub-pixel value pixel (x,
Y) it is pressed into storehouse;
Ejection submodule, for ejecting a pixel from storehouse, using the pixel of ejection as new sub-pixel (x0,
Y0), the merging submodule is returned to continue executing with;
Judging submodule, for being space-time when storehouse, returns to the acquisition submodule and continues executing with;
Submodule is called, for repeating S1-S4, until each pixel in the two-dimensional ct image has ownership;
The Model Reconstruction unit, specifically for Area generation after, the pixel value in region is set as a constant
A, the pixel value outside region is set as 0, and it is the face of a to obtain contour surface according to iso-surface patch algorithm, wherein, the equivalence for getting
Face is the corresponding threedimensional model of the two-dimensional ct image for the face of a.
Further, the display module includes:
Dot product unit, for inciting somebody to actionVector withVector makees dot product, obtains the cosine value COS (α) of ∠ α;
Judging unit, for judging COS (α) whether more than 0;
Rendering unit, if being more than 0, A points above cutting planes are along normal direction for COS (α), then retains A points simultaneously
Render A points;
Culling unit, if being not more than 0 for COS (α), rejects A points;
Wherein, ∠ α=∠ OAP, O point is threedimensional model central point, and A points are any point on threedimensional model, and P points are to cut
A coordinate points in cutting plane,It is the normal line vector of cutting planes.
Further, the rendering unit, if specifically for | OA |>Width, then be rendered to default color by A points;
Wherein, | OA | is the distance between O points to A points, and O points are threedimensional model central point, and A points are appointing on threedimensional model
A bit, width is cross-section image display width to meaning.
Above-mentioned technical proposal of the invention has the beneficial effect that:
In such scheme, three-dimensional reconstruction operation is carried out by the two-dimensional ct image for obtaining, obtain the two-dimensional ct image
Corresponding threedimensional model;In default VR display modules, the threedimensional model is cut off using section and section is aobvious
Show, so, with 3D form simulation two-dimensional ct images, and the corresponding threedimensional model of two-dimensional ct image is relied under VR environment, can
The hidden danger situation of overlapping region and embedded region between true display tissue under VR environment, is easy to doctor to observe respectively
The minor detail at position, each hidden danger that may be present is not let off, so as to aid in the doctor quickly to carry out overlapping region, embedded area
The accurate judgement in domain and diagnostic imaging, it is possible to increase diagnosis accuracy of the doctor to conditions of patients.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of diagnostic method of the medical image provided in an embodiment of the present invention under VR environment;
Fig. 2 be around sub-pixel provided in an embodiment of the present invention close to pixel schematic diagram;
Fig. 3 is threedimensional model cutting planes sectional schematic diagram provided in an embodiment of the present invention;
Fig. 4 is the display schematic diagram under VR environment provided in an embodiment of the present invention;
Fig. 5 is the structural representation of diagnostic system of the medical image provided in an embodiment of the present invention under VR environment.
Specific embodiment
To make the technical problem to be solved in the present invention, technical scheme and advantage clearer, below in conjunction with accompanying drawing and tool
Body embodiment is described in detail.
The present invention is directed to the existing easy problem failed to pinpoint a disease in diagnosis, there is provided a kind of diagnostic method of medical image under VR environment
And system.
Embodiment one
Referring to shown in Fig. 1, diagnostic method of the medical image provided in an embodiment of the present invention under VR environment, including:
S101, obtains two-dimensional ct image;
S102, the two-dimensional ct image to obtaining carries out three-dimensional reconstruction operation, obtains the two-dimensional ct image corresponding
Threedimensional model;
S103, in default VR display modules, is cut off to the threedimensional model using section and section is shown.
The diagnostic method of medical image described in the embodiment of the present invention under VR environment, by the two-dimensional ct image to obtaining
Three-dimensional reconstruction operation is carried out, the corresponding threedimensional model of the two-dimensional ct image is obtained;In default VR display modules, using cutting
The threedimensional model is cut off piece and section shows, so, with 3D form simulation two-dimensional ct images, and under VR environment
The corresponding threedimensional model of two-dimensional ct image is relied on, overlapping region that can be under VR environment between true display tissue and interior
The hidden danger situation in embedding region, is easy to doctor it was observed that the minor detail at each position, each hidden danger that may be present is not let off, from
And aiding in the doctor quickly to carry out overlapping region, the accurate judgement in embedded region and diagnostic imaging, it is possible to increase doctor is to patient
The diagnosis accuracy of the state of an illness.
In the specific embodiment of diagnostic method of the foregoing medical image under VR environment, further, described pair is obtained
The two-dimensional ct image for taking carries out three-dimensional reconstruction operation, and obtaining the corresponding threedimensional model of the two-dimensional ct image includes:
By the different region of the two-dimensional ct video generation;
After Area generation, the corresponding threedimensional model of the two-dimensional ct image is rebuild according to iso-surface patch algorithm.
In the present embodiment, the two-dimensional ct image is medical science two-dimensional CT image.
It is further, described by institute in the specific embodiment of diagnostic method of the foregoing medical image under VR environment
Stating the different region of two-dimensional ct video generation includes:
S1, the two-dimensional ct image order to obtaining is scanned, and acquisition first is not belonged to also in region
Pixel, and described first pixel not belonged to also in region that will be obtained is used as sub-pixel (x0, y0);
S2, centered on sub-pixel (x0, y0), detect around sub-pixel (x0, y0) close to pixel (x, y)
Whether pixel value is less than default sub-pixel value, if so, then by less than the pixel (x, y) and seed of default sub-pixel value
Pixel (x0, y0) merges in same region, and will be pressed into storehouse less than the pixel (x, y) of default sub-pixel value;
S3, ejects a pixel from storehouse, using the pixel for ejecting as new sub-pixel (x0, y0), returns to S2
Continue executing with;
S4, when storehouse is space-time, returns to S1 and continues executing with;
S5, repeats S1-S4, until each pixel in the two-dimensional ct image has ownership.
In the present embodiment, after S5 has been performed, region growing terminates, and the basic thought of region growing is by with similitude
Pixel set get up to constitute region, specifically first to each needs segmentation region look for a sub-pixel as growth rise
Point, then by around sub-pixel close to face in domain there is the potting gum of same or similar property to seed with sub-pixel
In region where pixel, and by the pixel with sub-pixel with same or similar property as new sub-pixel continue into
The process of row S2, S3, S4, can be included until not meeting pre-conditioned pixel again, and so, an Area generation is just
Complete, wherein, it is pre-conditioned to be:Around sub-pixel close to pixel whether be less than default sub-pixel value.
In the present embodiment, as shown in Fig. 2 around sub-pixel (x0, y0) close to pixel (x, y) in Fig. 2
(x1, y1), (x2, y2), (x3, y3), (x4, y4), (x5, y5), (x6, y6), (x7, y7) and (x8, y8).
In the specific embodiment of diagnostic method of the foregoing medical image under VR environment, further, the region
After generation, rebuilding the corresponding threedimensional model of the two-dimensional ct image according to iso-surface patch algorithm includes:
After Area generation, the pixel value in region is set as a constant a, the pixel value outside region is set as
0, it is the face of a to obtain contour surface according to iso-surface patch algorithm, wherein, the contour surface for getting is the two-dimensional ct image for the face of a
Corresponding threedimensional model.
In the present embodiment, the basic thought of iso-surface patch algorithm is cube (voxel) one by one in processing data, classification
Go out the cube intersected with contour surface, the intersection point of contour surface and cube side is gone out using interpolation calculation.According to cube each
Summit and the relative position of contour surface, generation contour surface is connected by contour surface with the intersection point on cube side by certain way, as
One in the cube of contour surface approaches expression.
In the present embodiment, after Area generation, the pixel value in region is set as a constant a, it is preferable that a=1;
Pixel value outside region is set as 0, then looks for the face that contour surface is 1 according to iso-surface patch algorithm, and the contour surface found out is exactly
Color value (the RGB of the corresponding threedimensional model of two-dimensional ct image of reconstruction, the threedimensional model that will have been rebuild and the threedimensional model
Value) store in folder path set in advance.
In the present embodiment, by storage to the threedimensional model rebuild in folder path set in advance and described three
The color value of dimension module is imported in previously fabricated zSpace stereoscopic devices, the VR in previously fabricated zSpace stereoscopic devices
In display module using section can the threedimensional model built up of counterweight cut off and section shows, to aid in doctor to carry out image
Diagnosis.
In the present embodiment, principle combination Fig. 3 that serial section is carried out under VR environment is stated, and in figure 3, O points are
Threedimensional model central point, P points are a certain coordinate points of cutting planes,It is the normal line vector of cutting planes, A points are three-dimensional mould
Any point in type, width is cross-section image display width, ∠ α=∠ OAP;
In the present embodiment,Vector withVector makees dot product can obtain the value of COS (α), COS (α)=Dot
((A.WorldPos-P.WorldPos),), wherein, Dot represents dot product, A.WorldPos, P.WorldPos difference table
Show world coordinates, the world coordinates of P points of A points;When the value of COS (α) is more than 0, the A points are in cutting planes in normal direction
Side, then retain A points and render A points.Conversely, then rejecting A points.Again according to Pythagorean theorem COS (α)=| OA |/| PA |, i.e. | OA |
=COS (α) * | PA |, if | OA |>| OB | (i.e. width), then A points be rendered into default color, wherein, it is described default
Color can be red, and final effect is as shown in Figure 4.
It is described that the threedimensional model is cut off and cut using section as an alternative embodiment in the present embodiment
Face show including:
WillVector withVector makees dot product, obtains the cosine value COS (α) of ∠ α;
Judge COS (α) whether more than 0;
If more than 0, A points above cutting planes are along normal direction, then retaining A points and rendering A points;
Otherwise, then A points are rejected;
Wherein, ∠ α=∠ OAP, O point is threedimensional model central point, and A points are any point on threedimensional model, and P points are to cut
A coordinate points in cutting plane,It is the normal line vector of cutting planes.
In the present embodiment, used as another alternative embodiment, methods described also includes:
If | OA |>Width, then be rendered to default color by A points;
Wherein, | OA | is the distance between O points to A points, and O points are threedimensional model central point, and A points are appointing on threedimensional model
A bit, width is cross-section image display width to meaning.
Embodiment two
The present invention also provides a kind of specific embodiment of diagnostic system of medical image under VR environment, due to the present invention
The specific reality of diagnostic method of diagnostic system of the medical image of offer under VR environment with foregoing medical image under VR environment
Apply that mode is corresponding, during diagnostic system of the medical image under VR environment can be by performing above method specific embodiment
Process step realize the purpose of the present invention, therefore diagnostic method specific embodiment of the above-mentioned medical image under VR environment
In explanation, the specific embodiment of the diagnostic system of the medical image under VR environment of present invention offer is provided,
Be will not be described in great detail in specific embodiment below the present invention.
Referring to shown in Fig. 5, the embodiment of the present invention also provides a kind of diagnostic system of medical image under VR environment, including:
Acquisition module 11, for obtaining two-dimensional ct image;
Module 12 is rebuild, for carrying out three-dimensional reconstruction operation to the two-dimensional ct image for obtaining, the two-dimensional ct is obtained
The corresponding threedimensional model of image;
Display module 13, in default VR display modules, using section the threedimensional model is cut off with
And section shows.
The diagnostic system of medical image described in the embodiment of the present invention under VR environment, by the two-dimensional ct image to obtaining
Three-dimensional reconstruction operation is carried out, the corresponding threedimensional model of the two-dimensional ct image is obtained;In default VR display modules, using cutting
The threedimensional model is cut off piece and section shows, so, with 3D form simulation two-dimensional ct images, and under VR environment
The corresponding threedimensional model of two-dimensional ct image is relied on, overlapping region that can be under VR environment between true display tissue and interior
The hidden danger situation in embedding region, is easy to doctor it was observed that the minor detail at each position, each hidden danger that may be present is not let off, from
And aiding in the doctor quickly to carry out overlapping region, the accurate judgement in embedded region and diagnostic imaging, it is possible to increase doctor is to patient
The diagnosis accuracy of the state of an illness.
In the specific embodiment of diagnostic system of the foregoing medical image under VR environment, further, the reconstruction
Module includes:
Area generation unit, for by the different region of the two-dimensional ct video generation;
Model Reconstruction unit, for Area generation after, rebuild the two-dimensional ct image according to iso-surface patch algorithm corresponding
Threedimensional model;
Wherein, the Area generation unit includes:
Acquisition submodule, is scanned for the two-dimensional ct image order to obtaining, and obtains first and does not return also
Belong to the pixel in region, and described first pixel not belonged to also in region that will be obtained as sub-pixel (x0,
y0);
Merge submodule, for centered on sub-pixel (x0, y0), detect around sub-pixel (x0, y0) close to
Pixel (x, y) pixel value whether be less than default sub-pixel value, if so, then will be less than default sub-pixel value picture
Plain (x, y) merges in same region with sub-pixel (x0, y0), and will be less than default sub-pixel value pixel (x,
Y) it is pressed into storehouse;
Ejection submodule, for ejecting a pixel from storehouse, using the pixel of ejection as new sub-pixel (x0,
Y0), the merging submodule is returned to continue executing with;
Judging submodule, for being space-time when storehouse, returns to the acquisition submodule and continues executing with;
Submodule is called, for repeating S1-S4, until each pixel in the two-dimensional ct image has ownership;
The Model Reconstruction unit, specifically for Area generation after, the pixel value in region is set as a constant
A, the pixel value outside region is set as 0, and it is the face of a to obtain contour surface according to iso-surface patch algorithm, wherein, the equivalence for getting
Face is the corresponding threedimensional model of the two-dimensional ct image for the face of a.
In the specific embodiment of diagnostic system of the foregoing medical image under VR environment, further, the display
Module includes:
Dot product unit, for inciting somebody to actionVector withVector makees dot product, obtains the cosine value COS (α) of ∠ α;
Judging unit, for judging COS (α) whether more than 0;
Rendering unit, if being more than 0, A points above cutting planes are along normal direction for COS (α), then retains A points simultaneously
Render A points;
Culling unit, if being not more than 0 for COS (α), rejects A points;
Wherein, ∠ α=∠ OAP, O point is threedimensional model central point, and A points are any point on threedimensional model, and P points are to cut
A coordinate points in cutting plane,It is the normal line vector of cutting planes.
It is further, described to render in the specific embodiment of diagnostic system of the foregoing medical image under VR environment
Unit, if specifically for | OA |>Width, then be rendered to default color by A points;
Wherein, | OA | is the distance between O points to A points, and O points are threedimensional model central point, and A points are appointing on threedimensional model
A bit, width is cross-section image display width to meaning.
The above is the preferred embodiment of the present invention, it is noted that for those skilled in the art
For, on the premise of principle of the present invention is not departed from, some improvements and modifications can also be made, these improvements and modifications
Should be regarded as protection scope of the present invention.