CN105956198B - A kind of galactophore image searching system and method based on lesions position and content - Google Patents
A kind of galactophore image searching system and method based on lesions position and content Download PDFInfo
- Publication number
- CN105956198B CN105956198B CN201610445569.8A CN201610445569A CN105956198B CN 105956198 B CN105956198 B CN 105956198B CN 201610445569 A CN201610445569 A CN 201610445569A CN 105956198 B CN105956198 B CN 105956198B
- Authority
- CN
- China
- Prior art keywords
- image
- collection
- retrieved
- circle
- pretreated
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G06F19/321—
Landscapes
- Engineering & Computer Science (AREA)
- Library & Information Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Image Processing (AREA)
Abstract
The present invention provides a kind of galactophore image searching system and method based on lesions position and content, the system includes image pre-processing unit, image focus position similarity measurement unit, picture material similarity measurement unit and image synthesis similarity measurements measuring device, this method is the image to be retrieved for obtaining breast molybdenum target x-ray image, history image collection, selection standard image, image to be retrieved and history image collection are pre-processed, image focus position similarity measurement is carried out to pretreated image to be retrieved and pretreated history image collection, pretreated image to be retrieved and pretreated history image collection to breast molybdenum target x line image carry out picture material similarity measurement, obtain image synthesis similarity image serial number, obtain image searching result to be retrieved.Invention increases lesions position method for measuring similarity is based on, the retrieval performance of breast molybdenum target x line image is effectively improved, further assists diagnosis of the doctor to mammary gland disease.
Description
Technical field
The invention belongs to medical image post-procession technique fields, and in particular to a kind of mammary gland based on lesions position and content
Image indexing system and method.
Background technique
Currently, breast cancer screening is the important means realizing breast cancer early diagnosis and early controlling, 30% death rate can be reduced.Cream
Gland mammography image is breast cancer early detection, the important evidence of diagnosis, and the different manifestations of lesion in galactophore image become early
The sole criterion of phase Diagnosis of Breast cancer, but its diagnosis has biggish difficulty, it can be effectively by the retrieval of galactophore image
Assist diagnosis.
Content-based image retrieval technology starts from the initial stage nineties, its purposes medically is more next by now for development
It is more extensive, wherein the meaning in terms of galactophore image retrieval is extremely great, be according to being retrieved the characteristics of galactophore image,
Search result is considered as to the useful value of medical diagnosis.It is developed so far, retrieval technique has had more mature development, but still
So existing defects in terms of retrieval performance, be mainly manifested in retrieval performance it is lower on, main reason is that the letter that image contains
Breath amount and similarity measurement mode, that is, the method for the content and similarity retrieval of extracted image useful information.Therefore, such as
What improves the performance of retrieval, it is still necessary to further discuss and study.
Summary of the invention
In view of the deficiencies of the prior art, the present invention proposes a kind of galactophore image searching system based on lesions position and content
And method.
The technical scheme is that a kind of galactophore image searching system based on lesions position and content, including image
Pretreatment unit, image focus position similarity measurement unit, picture material similarity measurement unit and image synthesis similitude
Measure device;
Described image pretreatment unit, for obtaining the image F to be retrieved of breast molybdenum target radioscopic image0, history image collection
(F1,F2,…,Fn), selection standard image FC, to image F to be retrieved0With history image collection (F1,F2,…,Fn) pre-processed,
Obtain pretreated image set (I0,I1,I2,…,In), wherein including pretreated image I to be retrieved0After pretreatment
History image collection (I1,I2,…,In);The standard picture FCFor form is normal, the breast molybdenum target X ray picture that is of moderate size
Picture;
Described image lesions position similarity measurement unit, for the pretreated to be checked of mammograms
Rope image I0With pretreated history image collection (I1,I2,…,In) image focus position similarity measurement is carried out, obtain image
Lesions position similitude set (S1,S2,…,Sn);
Described image content similarities metric element, for the pretreated figure to be retrieved to mammograms
As I0With pretreated history image collection (I1,I2,…,In) picture material similarity measurement is carried out, it is similar to obtain picture material
Property set (E1,E2,…,En);
Described image integrates similarity measurements measuring device, is used for image focus position similitude set (S1,S2,…,Sn) according to
Image focus position similitude sorts from large to small, and marking serial numbers, by the image focus position similitude set after sequence
Serial number distributes the weight of A%, by picture material similitude set (E1,E2,…,En) according to picture material similitude from small to large
Sequence, and marking serial numbers, by the weight of serial number distribution (100-A) % of the picture material similitude set after sequence, complex chart
As lesions position similarity image serial number and picture material similarity image serial number obtain image synthesis similarity image serial number, obtain
To image searching result to be retrieved: i.e. image synthesis similarity image serial number is smaller, indicates that the image gets over phase with image to be retrieved
Seemingly.
Preferably, described image pretreatment unit includes: image denoising device and image intensifier;
Described image denoises device, for respectively to image F to be retrieved0With history image collection (F1,F2,…,Fn) carry out noise reduction
Processing, the image P to be retrieved after obtaining noise reduction0With the history image collection (P after noise reduction1,P2,…,Pn);
Described image booster, for respectively to the image P to be retrieved after noise reduction0With the history image collection (P after noise reduction1,
P2,…,Pn) image enhancement processing is carried out, obtain pretreated image I to be retrieved0With pretreated history image collection (I1,
I2,…,In);
Preferably, described image lesions position similarity measurement unit include: lesions position central point and radius determiner,
Image to be retrieved and image library and standard picture aligner, image focus position similitude determiner;
The lesions position central point and radius determiner, for determining pretreated image set (I0,I1,I2,…,
In) lesions position center point coordinate collection { (x0,y0),(x1,y1),(x2,y2)…,(xn,yn) and lesions position radius collection (R0,
R1,R2,…,Rn);
Preferably, the pretreated image set (I of the determination0,I1,I2,…,In) lesions position center point coordinate collection
{(x0,y0),(x1,y1),(x2,y2)…,(xn,yn) and lesions position radius collection (R0,R1,R2,…,Rn) method particularly includes:
Using classical Da-Jin algorithm Threshold Segmentation Algorithm to pretreated image set (I0,I1,I2,…,In) carry out two-value
Change is handled, highlight regions in the image after retaining binary conversion treatment, using the half of highlight regions X-direction maximum value as lesion
Place-centric point abscissa obtains disease using the half of highlight regions Y direction maximum value as lesions position central point ordinate
Stove place-centric point coordinate set { (x0,y0),(x1,y1),(x2,y2)…,(xn,yn), highlight regions all the points are traversed, from first
A point starts, and using right angled triangle Pythagorean theorem, finds out the distance for a little arriving central point, successively calculates all the points in highlight regions
At a distance from central point, using its maximum value as lesions position radius, lesions position radius collection (R is obtained0,R1,R2,…,Rn)。
The image to be retrieved and image library and standard picture aligner, for that will locate in advance using CPD method for registering is based on
Image set (I after reason0,I1,I2,…,In) and standard picture FCIt is matched, the lesion center point coordinate set after being converted
{(X0,Y0),(X1,Y1),(X2,Y2),…,(Xn,Yn), and by lesion center point the coordinate set { (X after conversion0,Y0),(X1,
Y1),(X2,Y2),…,(Xn,Yn) and lesions position radius collection (R0,R1,R2,…,Rn) it is shown in standard picture FCOn, using turn
Lesion center point coordinate set { (X after changing0,Y0),(X1,Y1),(X2,Y2),…,(Xn,Yn) and lesions position radius collection (R0,
R1,R2,…,Rn) determine focal area (Circle0,Circle1,Circle2,…,Circlen);
Described image lesions position similitude determiner, for calculating separately focal area (Circle0,Circle1,
Circle2,…,Circlen) in image to be retrieved focal area Circle0With the focal area (Circle of history image1,
Circle2,…,Circlen) intersection and union, enable
Obtain image focus position similitude set (S1,S2,S3,…,Sn);
Preferably, described image content similarities metric element includes: characteristics of image histogram extractor and picture material
Similitude determiner;
Described image feature histogram extractor, for extracting pretreated image set (I0,I1,I2,…,In) ash
Feature, shape feature and textural characteristics are spent, its image grey level histogram, straight based on edge orientation histogram, direction gradient is constructed
Side's figure and local binary patterns histogram, obtain gray feature vector (α0,α1,α2,…,αn), shape eigenvectors (β0,β1,
β2,…,βn) and texture feature vector (γ0,γ1,γ2,…,γn), merge gray feature vector (α0,α1,α2,…,αn), shape
Shape feature vector (β0,β1,β2,…,βn) and texture feature vector (γ0,γ1,γ2,…,γn), obtain multi-scale HoGC feature
Vector (ω0,ω1,ω2,…,ωn);
Described image content similarities determiner, for using EMD method by the multi-scale HoGC feature of image to be retrieved
Vector ω0With the multi-scale HoGC feature vector (ω of history image1,ω2,…,ωn) similarity measurement is carried out, it obtains in image
Hold similitude set (E1,E2,…,En)。
The method that image retrieval is carried out using the galactophore image searching system based on lesions position and content, including following step
It is rapid:
Step 1: obtaining the image F to be retrieved of breast molybdenum target radioscopic image0, history image collection (F1,F2,…,Fn), it chooses
Standard picture FC;
Step 2: to image F to be retrieved0With history image collection (F1,F2,…,Fn) pre-processed, it obtains pretreated
Image set (I0,I1,I2,…,In);
Step 2.1: respectively to image F to be retrieved0With history image collection (F1,F2,…,Fn) noise reduction process is carried out, it is dropped
Image P to be retrieved after making an uproar0With the history image collection (P after noise reduction1,P2,…,Pn);
Step 2.2: respectively to the image P to be retrieved after noise reduction0With the history image collection (P after noise reduction1,P2,…,Pn) into
Row image enhancement processing obtains pretreated image I to be retrieved0With pretreated history image collection (I1,I2,…,In);
Step 3: to the pretreated image I to be retrieved of mammograms0With pretreated history image collection
(I1,I2,…,In) carry out image focus position similarity measurement;
Step 3.1: determining pretreated image set (I0,I1,I2,…,In) lesions position center point coordinate collection { (x0,
y0),(x1,y1),(x2,y2)…,(xn,yn) and lesions position radius collection (R0,R1,R2,…,Rn);
Step 3.2: using CPD method for registering is based on, by pretreated image set (I0,I1,I2,…,In) and standard drawing
As FCIt is matched, lesion center point the coordinate set { (X after being converted0,Y0),(X1,Y1),(X2,Y2),…,(Xn,Yn), and
By lesion center point the coordinate set { (X after conversion0,Y0),(X1,Y1),(X2,Y2),…,(Xn,Yn) and lesions position radius collection
(R0,R1,R2,…,Rn) it is shown in standard picture FCOn, utilize lesion center point the coordinate set { (X after conversion0,Y0),(X1,Y1),
(X2,Y2),…,(Xn,Yn) and lesions position radius collection (R0,R1,R2,…,Rn) determine focal area (Circle0,Circle1,
Circle2,…,Circlen);
Step 3.3: calculating separately focal area (Circle0,Circle1,Circle2,…,Circlen) in figure to be retrieved
The focal area Circle of picture0With the focal area (Circle of history image1,Circle2,…,Circlen) intersection and simultaneously
Collection enables Obtain image focus position similitude
Gather (S1,S2,S3,…,Sn);
Step 4: to the pretreated image I to be retrieved of mammograms0With pretreated history image collection
(I1,I2,…,In) carry out picture material similarity measurement;
Step 4.1: extracting pretreated image set (I0,I1,I2,…,In) gray feature, shape feature and texture
Feature constructs its image grey level histogram, based on edge orientation histogram, histograms of oriented gradients and local binary patterns histogram
Figure, obtains gray feature vector (α0,α1,α2,…,αn), shape eigenvectors (β0,β1,β2,…,βn) and texture feature vector
(γ0,γ1,γ2,…,γn);
Step 4.2: merging gray feature vector (α0,α1,α2,…,αn), shape eigenvectors (β0,β1,β2,…,βn) and
Texture feature vector (γ0,γ1,γ2,…,γn), obtain multi-scale HoGC feature vector (ω0,ω1,ω2,…,ωn);
Step 4.3: using EMD method by the multi-scale HoGC feature vector ω of image to be retrieved0It is comprehensive with history image
Close histogram feature vector (ω1,ω2,…,ωn) similarity measurement is carried out, obtain picture material similitude set (E1,E2,…,
En);
Step 5: by image focus position similitude set (S1,S2,…,Sn) according to image focus position similitude from big
To small sequence, and marking serial numbers will scheme the weight of the serial number distribution A% of the image focus position similitude set after sequence
As content similarities set (E1,E2,…,En) sort from small to large according to picture material similitude, and marking serial numbers, it will sort
The weight of serial number distribution (100-A) % of picture material similitude set afterwards, synthetic image lesions position similarity image sequence
Number and picture material similarity image serial number obtain image synthesis similarity image serial number, obtain image searching result to be retrieved:
I.e. image synthesis similarity image serial number is smaller, indicates that the image is more similar to image to be retrieved.
Preferably, the step 3.2 the following steps are included:
Step 3.2.1: pretreated image set (I is extracted0,I1,I2,…,In) breast contours and standard drawing in image
As FCBreast contours;
Step 3.2.2: the affine transformation based on CPD is utilized, pretreated image set (I is registrated0,I1,I2,…,In) figure
Breast contours and standard picture F as inCBreast contours, obtain registration transformation matrix (T0,T1,T2,…,Tn);
Step 3.2.3: pass through registration transformation matrix (T0,T1,T2,…,Tn) to lesions position center point coordinate collection { (x0,
y0),(x1,y1),(x2,y2)…,(xn,yn) and lesions position radius collection (R0,R1,R2,…,Rn) converted, after obtaining conversion
Lesions position center point coordinate collection { (X0,Y0),(X1,Y1),(X2,Y2),…,(Xn,Yn)};
Step 3.2.4: by lesions position center point coordinate the collection { (X after conversion0,Y0),(X1,Y1),(X2,Y2),…,
(Xn,Yn) and lesions position radius collection (R0,R1,R2,…,Rn) it is shown in standard picture FCOn, it will be in the lesions position after conversion
Heart point coordinate set { (X0,Y0),(X1,Y1),(X2,Y2),…,(Xn,Yn) and lesions position radius collection (R0,R1,R2,…,Rn) really
Fixed circle (Circle0,Circle1,Circle2,…,Circlen) it is used as focal area.
Beneficial effects of the present invention:
The present invention proposes a kind of galactophore image searching system and method based on lesions position and content, it is traditional based on
On the basis of the image search method of content, increases based on lesions position method for measuring similarity, cream can be effectively improved
The retrieval performance of gland molybdenum target X-ray image, so as to further assist diagnosis of the doctor to mammary gland disease.
Detailed description of the invention
Fig. 1 is the structural frames of the galactophore image searching system based on lesions position and content in the specific embodiment of the invention
Figure;
Fig. 2 is the process of the galactophore image search method based on lesions position and content in the specific embodiment of the invention
Figure;
Fig. 3 is in the specific embodiment of the invention to the pretreated image to be retrieved of mammograms and pre-
Treated, and history image collection carries out the flow chart of image focus position similarity measurement;
Fig. 4 is to utilize the flow chart that focal area is determined based on CPD method for registering in the specific embodiment of the invention;
Fig. 5 is in the specific embodiment of the invention to the pretreated image to be retrieved of mammograms and pre-
Treated, and history image collection carries out the flow chart of picture material similarity measurement;
Fig. 6 is to be obtained at the image of image searching result to be retrieved in the specific embodiment of the invention according to history image collection
Manage process flow diagram flow chart.
Specific embodiment
The specific embodiment of the invention is described in detail with reference to the accompanying drawing.
A kind of galactophore image searching system based on lesions position and content, as shown in Figure 1, including image preprocessing list
Member, image focus position similarity measurement unit, picture material similarity measurement unit and image synthesis similarity measurements measuring device.
Image pre-processing unit, for obtaining the image F to be retrieved of breast molybdenum target radioscopic image0, history image collection (F1,
F2,…,Fn), selection standard image FC, to image F to be retrieved0With history image collection (F1,F2,…,Fn) pre-processed, it obtains
Pretreated image set (I0,I1,I2,…,In), wherein including pretreated image I to be retrieved0It is gone through with pretreated
History image set (I1,I2,…,In)。
Standard picture FCFor form is normal, the breast molybdenum target radioscopic image that is of moderate size.
Image pre-processing unit includes: image denoising device and image intensifier.
Image denoising device, for respectively to image F to be retrieved0With history image collection (F1,F2,…,Fn) carry out at noise reduction
Reason, the image P to be retrieved after obtaining noise reduction0With the history image collection (P after noise reduction1,P2,…,Pn)。
Image intensifier, for respectively to the image P to be retrieved after noise reduction0With the history image collection (P after noise reduction1,
P2,…,Pn) image enhancement processing is carried out, obtain pretreated image I to be retrieved0With pretreated history image collection (I1,
I2,…,In)。
Image focus position similarity measurement unit, for the pretreated figure to be retrieved to mammograms
As I0With pretreated history image collection (I1,I2,…,In) image focus position similarity measurement is carried out, obtain image focus
Position similitude set (S1,S2,…,Sn)。
Image focus position similarity measurement unit includes: lesions position central point and radius determiner, image to be retrieved
With image library and standard picture aligner, image focus position similitude determiner.
Lesions position central point and radius determiner, for determining pretreated image set (I0,I1,I2,…,In)
Lesions position center point coordinate collection { (x0,y0),(x1,y1),(x2,y2)…,(xn,yn) and lesions position radius collection (R0,R1,
R2,…,Rn)。
In present embodiment, pretreated image set (I is determined0,I1,I2,…,In) lesions position center point coordinate
Collect { (x0,y0),(x1,y1),(x2,y2)…,(xn,yn) and lesions position radius collection (R0,R1,R2,…,Rn) specific method
Are as follows:
Using classical Da-Jin algorithm Threshold Segmentation Algorithm to pretreated image set (I0,I1,I2,…,In) carry out two-value
Change is handled, highlight regions in the image after retaining binary conversion treatment, using the half of highlight regions X-direction maximum value as lesion
Place-centric point abscissa obtains disease using the half of highlight regions Y direction maximum value as lesions position central point ordinate
Stove place-centric point coordinate set { (x0,y0),(x1,y1),(x2,y2)…,(xn,yn), highlight regions all the points are traversed, from first
A point starts, and using right angled triangle Pythagorean theorem, finds out the distance for a little arriving central point, successively calculates all the points in highlight regions
At a distance from central point, using its maximum value as lesions position radius, lesions position radius collection (R is obtained0,R1,R2,…,Rn)。
Image to be retrieved and image library and standard picture aligner are based on CPD method for registering for utilizing, after pretreatment
Image set (I0,I1,I2,…,In) and standard picture FCIt carries out area to match, lesion center point the coordinate set { (X after being converted0,
Y0),(X1,Y1),(X2,Y2),…,(Xn,Yn), and by lesion center point the coordinate set { (X after conversion0,Y0),(X1,Y1),(X2,
Y2),…,(Xn,Yn) and lesions position radius collection (R0,R1,R2,…,Rn) it is shown in standard picture FCOn, utilize the disease after conversion
Stove center point coordinate collection { (X0,Y0),(X1,Y1),(X2,Y2),…,(Xn,Yn) and lesions position radius collection (R0,R1,R2,…,
Rn) determine focal area (Circle0,Circle1,Circle2,…,Circlen)。
Image focus position similitude determiner, for calculating separately focal area (Circle0,Circle1,
Circle2,…,Circlen) in image to be retrieved focal area Circle0With the focal area (Circle of history image1,
Circle2,…,Circlen) intersection and union, enable
Obtain image focus position similitude set (S1,S2,S3,…,Sn)。
Picture material similarity measurement unit, for the pretreated image I to be retrieved to mammograms0
With pretreated history image collection (I1,I2,…,In) picture material similarity measurement is carried out, obtain picture material similitude collection
Close (E1,E2,…,En)。
Picture material similarity measurement unit includes: that characteristics of image histogram extractor and picture material similitude determine
Device.
Characteristics of image histogram extractor, for extracting pretreated image set (I0,I1,I2,…,In) gray scale it is special
Sign, shape feature and textural characteristics construct its image grey level histogram, based on edge orientation histogram, histograms of oriented gradients
With local binary patterns histogram, gray feature vector (α is obtained0,α1,α2,…,αn), shape eigenvectors (β0,β1,β2,…,
βn) and texture feature vector (γ0,γ1,γ2,…,γn), merge gray feature vector (α0,α1,α2,…,αn), shape feature
Vector (β0,β1,β2,…,βn) and texture feature vector (γ0,γ1,γ2,…,γn), obtain multi-scale HoGC feature vector
(ω0,ω1,ω2,…,ωn)。
Picture material similitude determiner, for using EMD method by the multi-scale HoGC feature vector of image to be retrieved
ω0With the multi-scale HoGC feature vector (ω of history image1,ω2,…,ωn) similarity measurement is carried out, obtain picture material phase
Like property set (E1,E2,…,En)。
Image synthesis similarity measurements measuring device is used for image focus position similitude set (S1,S2,…,Sn) according to image
Lesions position similitude sorts from large to small, and marking serial numbers, by the serial number of the image focus position similitude set after sequence
The weight for distributing A%, by picture material similitude set (E1,E2,…,En) sort from small to large according to picture material similitude,
And marking serial numbers, by the weight of serial number distribution (100-A) % of the picture material similitude set after sequence, synthetic image lesion
Position similarity image serial number and picture material similarity image serial number obtain image synthesis similarity image serial number, obtain to be checked
Rope image searching result: i.e. image synthesis similarity image serial number is smaller, indicates that the image is more similar to image to be retrieved.
The method that image retrieval is carried out using the galactophore image searching system based on lesions position and content, as shown in Fig. 2,
The following steps are included:
Step 1: obtaining the image F to be retrieved of breast molybdenum target radioscopic image0, history image collection (F1,F2,…,Fn), it chooses
Standard picture FC, standard picture FCFor form is normal, the breast molybdenum target radioscopic image that is of moderate size.
Step 2: to image F to be retrieved0With history image collection (F1,F2,…,Fn) pre-processed, it obtains pretreated
Image set (I0,I1,I2,…,In), wherein including pretreated image I to be retrieved0With pretreated history image collection
(I1,I2,…,In)。
Step 2.1: respectively to image F to be retrieved0With history image collection (F1,F2,…,Fn) noise reduction process is carried out, it is dropped
Image P to be retrieved after making an uproar0With the history image collection (P after noise reduction1,P2,…,Pn)。
In present embodiment, selects median filter to be filtered using the transform method of spatial domain, realize to be retrieved
Image F0With history image collection (F1,F2,…,Fn) noise reduction process, reduce to image F to be retrieved0With history image collection (F1,
F2,…,Fn) in noise.
Step 2.2: respectively to the image P to be retrieved after noise reduction0With the history image collection (P after noise reduction1,P2,…,Pn) into
Row image enhancement processing obtains pretreated image I to be retrieved0With pretreated history image collection (I1,I2,…,In)。
In present embodiment, using the method for contrast enhancing to the image P to be retrieved after noise reduction0With the history after noise reduction
Image set (P1,P2,…,Pn) carry out image enhancement processing.Emphasize breast molybdenum target radioscopic image entirety or local characteristics, expanded view
Difference as between different objects feature inhibits uninterested feature, increases the contrast of suspected abnormality and surrounding tissue.
Step 3: to the pretreated image I to be retrieved of mammograms0With pretreated history image collection
(I1,I2,…,In) image focus position similarity measurement is carried out, as shown in Figure 3.
Step 3.1: determining pretreated image set (I0,I1,I2,…,In) lesions position center point coordinate collection { (x0,
y0),(x1,y1),(x2,y2)…,(xn,yn) and lesions position radius collection (R0,R1,R2,…,Rn)。
In present embodiment, using classical Da-Jin algorithm Threshold Segmentation Algorithm to pretreated image set (I0,I1,
I2,…,In) binary conversion treatment is carried out, highlight regions in the image after retaining binary conversion treatment are maximum by highlight regions X-direction
The half of value is as lesions position central point abscissa, using the half of highlight regions Y direction maximum value as in lesions position
Heart point ordinate obtains lesions position center point coordinate collection { (x0,y0),(x1,y1),(x2,y2)…,(xn,yn), traversal is highlighted
Region all the points, using right angled triangle Pythagorean theorem, find out the distance for a little arriving central point, successively count since first point
It calculates all the points in highlight regions and, using its maximum value as lesions position radius, obtains lesions position radius at a distance from central point
Collect (R0,R1,R2,…,Rn)。
Step 3.2: using CPD method for registering is based on, by pretreated image set (I0,I1,I2,…,In) and standard drawing
As FCIt carries out area to match, lesion center point the coordinate set { (X after being converted0,Y0),(X1,Y1),(X2,Y2),…,(Xn,Yn), and
By lesion center point the coordinate set { (X after conversion0,Y0),(X1,Y1),(X2,Y2),…,(Xn,Yn) and lesions position radius collection
(R0,R1,R2,…,Rn) it is shown in standard picture FCOn, utilize lesion center point the coordinate set { (X after conversion0,Y0),(X1,Y1),
(X2,Y2),…,(Xn,Yn) and lesions position radius collection (R0,R1,R2,…,Rn) determine focal area (Circle0,Circle1,
Circle2,…,Circlen), as shown in Figure 4.
It, can be by different size using CPD (Coherence Point Drift) method for registering is based in present embodiment
Mammary gland profile be standardized, reduce lesions position Similarity measures error.
Step 3.2.1: pretreated image set (I is extracted0,I1,I2,…,In) breast contours and standard drawing in image
As FCBreast contours.
Step 3.2.2: it using the affine transformation for being based on CPD (Coherence Point Drift), is registrated pretreated
Image set (I0,I1,I2,…,In) breast contours and standard picture F in imageCBreast contours, obtain registration transformation matrix
(T0,T1,T2,…,Tn)。
Step 3.2.3: pass through registration transformation matrix (T0,T1,T2,…,Tn) to lesions position center point coordinate collection { (x0,
y0),(x1,y1),(x2,y2)…,(xn,yn) and lesions position radius collection (R0,R1,R2,…,Rn) converted, after obtaining conversion
Lesions position center point coordinate collection { (X0,Y0),(X1,Y1),(X2,Y2),…,(Xn,Yn)}。
Step 3.2.4: by lesions position center point coordinate the collection { (X after conversion0,Y0),(X1,Y1),(X2,Y2),…,
(Xn,Yn) and lesions position radius collection (R0,R1,R2,…,Rn) it is shown in standard picture FCOn, it will be in the lesions position after conversion
Heart point coordinate set { (X0,Y0),(X1,Y1),(X2,Y2),…,(Xn,Yn) and lesions position radius collection (R0,R1,R2,…,Rn) really
Fixed circle (Circle0,Circle1,Circle2,…,Circlen) it is used as focal area.
Step 3.3: calculating separately focal area (Circle0,Circle1,Circle2,…,Circlen) in figure to be retrieved
The focal area Circle of picture0With the focal area (Circle of history image1,Circle2,…,Circlen) intersection and simultaneously
Collection enables Obtain image focus position phase
Like property set (S1,S2,S3,…,Sn)。
In present embodiment, The lesions position similitude ratio the big more phase
Seemingly.
Step 4: to the pretreated image I to be retrieved of mammograms0With pretreated history image collection
(I1,I2,…,In) picture material similarity measurement is carried out, as shown in Figure 5.
Step 4.1: extracting pretreated image set (I0,I1,I2,…,In) gray feature, shape feature and texture
Feature, building image grey level histogram are based on edge orientation histogram (Edge Direction Histogram, EDH), direction
Histogram of gradients (Histogram of oriented gradients, HOG) and local binary patterns histogram (Local
Binary Pattern, LBP), obtain gray feature vector (α0,α1,α2,…,αn), shape eigenvectors (β0,β1,β2,…,
βn) and texture feature vector (γ0,γ1,γ2,…,γn)。
Step 4.2: merging gray feature vector (α0,α1,α2,…,αn), shape eigenvectors (β0,β1,β2,…,βn) and
Texture feature vector (γ0,γ1,γ2,…,γn) obtain multi-scale HoGC feature vector (ω0,ω1,ω2,…,ωn)。
Step 4.3: using EMD (Earth Mover's Distance) method that the multi-scale HoGC of image to be retrieved is special
Levy vector ω0With the multi-scale HoGC feature vector (ω of history image1,ω2,…,ωn) similarity measurement is carried out, obtain image
Content similarities set (E1,E2,…,En)。
In present embodiment, (E1,E2,…,En)=(0.0223,0.0668,0.2641,0.3103,0.3179 ...,
0.6624).Numerical value is smaller, and feature histogram is more similar to get mammograms content similarities out.
Step 5: by image focus position similitude set (S1,S2,…,Sn) according to image focus position similitude from big
To small sequence, and marking serial numbers will scheme the weight of the serial number distribution 40% of the image focus position similitude set after sequence
As content similarities set (E1,E2,…,En) sort from small to large according to picture material similitude, and marking serial numbers, it will sort
The weight of the serial number distribution 60% of picture material similitude set afterwards, synthetic image lesions position similarity image serial number and figure
As content similarities picture numbers obtain image synthesis similarity image serial number, image searching result to be retrieved: i.e. image is obtained
Comprehensive similarity image serial number is smaller, indicates that the image is more similar to image to be retrieved.
In present embodiment, on the basis of traditional content-based image retrieval method, increase based on lesion position
Method for measuring similarity is set, the retrieval performance of mammograms can be effectively improved, so as to further assist
Diagnosis of the doctor to mammary gland disease obtains image processing process such as Fig. 6 of image searching result to be retrieved according to history image collection
It is shown.
Claims (6)
1. a kind of galactophore image searching system based on lesions position and content, which is characterized in that including image pre-processing unit,
Image focus position similarity measurement unit, picture material similarity measurement unit and image synthesis similarity measurements measuring device;
Described image pretreatment unit, for obtaining the image F to be retrieved of breast molybdenum target radioscopic image0, history image collection (F1,
F2..., Fn), selection standard image FC, to image F to be retrieved0With history image collection (F1, F2..., Fn) pre-processed, it obtains
Pretreated image set (I0, I1, I2..., In), wherein including pretreated image I to be retrieved0It is gone through with pretreated
History image set (I1, I2..., In);The standard picture FCFor form is normal, the breast molybdenum target radioscopic image that is of moderate size;
Described image lesions position similarity measurement unit, for the pretreated to be retrieved of breast molybdenum target radioscopic image
Image I0With pretreated history image collection (I1, I2..., In) image focus position similarity measurement is carried out, obtain image disease
Stove position similitude set (S1, S2..., Sn);Described image lesions position similarity measurement unit includes: lesions position center
Point and radius determiner, image to be retrieved and image library and standard picture aligner, image focus position similitude determiner;Institute
Lesions position central point and radius determiner are stated, for determining pretreated image set (I0, I1, I2..., In) lesion position
Set center point coordinate collection { (x0, y0), (x1, y1), (x2, y2) ..., (xn, yn) and lesions position radius collection (R0, R1, R2...,
Rn);The image to be retrieved and image library and standard picture aligner are based on CPD method for registering for utilizing, after pretreatment
Image set (I0, I1, I2..., In) and standard picture FCIt is matched, lesion center point the coordinate set { (X after being converted0,
Y0), (X1, Y1), (X2, Y2) ..., (Xn, Yn), and by lesion center point the coordinate set { (X after conversion0, Y0), (X1, Y1), (X2,
Y2) ..., (Xn, Yn) and lesions position radius collection (R0, R1, R2..., Rn) it is shown in standard picture FCOn, utilize the disease after conversion
Stove center point coordinate collection { (X0, Y0), (X1, Y1), (X2, Y2) ..., (Xn, Yn) and lesions position radius collection (R0, R1, R2...,
Rn) determine focal area (Circle0, Circle1, Circle2..., Circlen);Described image lesions position similitude determines
Device, for calculating separately focal area (Circle0, Circle1, Circle2..., Circlen) in image to be retrieved lesion
Region Circle0With the focal area (Circle of history image1, Circle2..., Circlen) intersection and union, enable Obtain image focus position similitude collection
Close (S1, S2, S3..., Sn);
Described image content similarities metric element, for the pretreated image I to be retrieved to breast molybdenum target radioscopic image0
With pretreated history image collection (I1, I2..., In) picture material similarity measurement is carried out, obtain picture material similitude collection
Close (E1, E2..., En);
Described image integrates similarity measurements measuring device, is used for image focus position similitude set (S1, S2..., Sn) according to image
Lesions position similitude sorts from large to small, and marking serial numbers, by the serial number of the image focus position similitude set after sequence
The weight for distributing A%, by picture material similitude set (E1, E2..., En) sort from small to large according to picture material similitude,
And marking serial numbers, by the weight of serial number distribution (100-A) % of the picture material similitude set after sequence, synthetic image lesion
Position similarity image serial number and picture material similarity image serial number obtain image synthesis similarity image serial number, obtain to be checked
Rope image searching result: i.e. image synthesis similarity image serial number is smaller, indicates that the image is more similar to image to be retrieved.
2. the galactophore image searching system according to claim 1 based on lesions position and content, which is characterized in that described
Image pre-processing unit includes: image denoising device and image intensifier;
Described image denoises device, for respectively to image F to be retrieved0With history image collection (F1, F2..., Fn) noise reduction process is carried out,
Image P to be retrieved after obtaining noise reduction0With the history image collection (P after noise reduction1, P2..., Pn);
Described image booster, for respectively to the image P to be retrieved after noise reduction0With the history image collection (P after noise reduction1,
P2..., Pn) image enhancement processing is carried out, obtain pretreated image I to be retrieved0With pretreated history image collection (I1,
I2..., In)。
3. the galactophore image searching system according to claim 1 based on lesions position and content, which is characterized in that described
Picture material similarity measurement unit includes: characteristics of image histogram extractor and picture material similitude determiner;
Described image feature histogram extractor, for extracting pretreated image set (I0, I1, I2..., In) gray scale it is special
Sign, shape feature and textural characteristics construct its image grey level histogram, based on edge orientation histogram, histograms of oriented gradients
With local binary patterns histogram, gray feature vector (α is obtained0, α1, α2..., αn), shape eigenvectors (β0, β1, β2...,
βn) and texture feature vector (γ0, γ1, γ2..., γn), merge gray feature vector (α0, α1, α2..., αn), shape feature
Vector (β0, β1, β2..., βn) and texture feature vector (γ0, γ1, γ2..., γn), obtain multi-scale HoGC feature vector
(ω0, ω1, ω2..., ωn);
Described image content similarities determiner, for using EMD method by the multi-scale HoGC feature vector of image to be retrieved
ω0With the multi-scale HoGC feature vector (ω of history image1, ω2..., ωn) similarity measurement is carried out, obtain picture material phase
Like property set (E1, E2..., En)。
4. the galactophore image searching system according to claim 1 based on lesions position and content, which is characterized in that described
Determine pretreated image set (I0, I1, I2..., In) lesions position center point coordinate collection { (x0, y0), (x1, y1), (x2,
y2) ..., (xn, yn) and lesions position radius collection (R0, R1, R2..., Rn) method particularly includes:
Using classical Da-Jin algorithm Threshold Segmentation Algorithm to pretreated image set (I0, I1, I2..., In) carry out at binaryzation
It manages, highlight regions in the image after retaining binary conversion treatment, using the half of highlight regions X-direction maximum value as lesions position
Central point abscissa obtains lesion position using the half of highlight regions Y direction maximum value as lesions position central point ordinate
Set center point coordinate collection { (x0, y0), (x1, y1), (x2, y2) ..., (xn, yn), highlight regions all the points are traversed, from first point
Start, using right angled triangle Pythagorean theorem, find out a little arrive central point distance, successively calculate highlight regions in all the points and in
The distance of heart point obtains lesions position radius collection (R using its maximum value as lesions position radius0, R1, R2..., Rn)。
5. using the side described in claim 1 for carrying out image retrieval based on the galactophore image searching system of lesions position and content
Method, which comprises the following steps:
Step 1: obtaining the image F to be retrieved of breast molybdenum target radioscopic image0, history image collection (F1, F2..., Fn), selection standard
Image FC;
Step 2: to image F to be retrieved0With history image collection (F1, F2..., Fn) pre-processed, obtain pretreated image
Collect (I0, I1, I2..., In);
Step 2.1: respectively to image F to be retrieved0With history image collection (F1, F2..., Fn) noise reduction process is carried out, after obtaining noise reduction
Image P to be retrieved0With the history image collection (P after noise reduction1, P2..., Pn);
Step 2.2: respectively to the image P to be retrieved after noise reduction0With the history image collection (P after noise reduction1, P2..., Pn) carry out image
Enhancing processing, obtains pretreated image I to be retrieved0With pretreated history image collection (I1, I2..., In);
Step 3: to the pretreated image I to be retrieved of breast molybdenum target radioscopic image0With pretreated history image collection
(I1, I2..., In) carry out image focus position similarity measurement;
Step 3.1: determining pretreated image set (I0, I1, I2..., In) lesions position center point coordinate collection { (x0, y0),
(x1, y1), (x2, y2) ..., (xn, yn) and lesions position radius collection (R0, R1, R2..., Rn);
Step 3.2: using CPD method for registering is based on, by pretreated image set (I0, I1, I2..., In) and standard picture FC
It is matched, lesion center point the coordinate set { (X after being converted0, Y0), (X1, Y1), (X2, Y2) ..., (Xn, Yn), and will turn
Lesion center point coordinate set { (X after changing0, Y0), (X1, Y1), (X2, Y2) ..., (Xn, Yn) and lesions position radius collection (R0,
R1, R2..., Rn) it is shown in standard picture FCOn, utilize lesion center point the coordinate set { (X after conversion0, Y0), (X1, Y1), (X2,
Y2) ..., (Xn, Yn) and lesions position radius collection (R0, R1, R2..., Rn) determine focal area (Circle0, Circle1,
Circle2..., Circlen);
Step 3.3: calculating separately focal area (Circle0, Circle1, Circle2..., Circlen) in image to be retrieved
Focal area Circle0With the focal area (Circle of history image1, Circle2..., Circlen) intersection and union, enable Obtain image focus position similitude
Gather (S1, S2, S3..., Sn);
Step 4: to the pretreated image I to be retrieved of breast molybdenum target radioscopic image0With pretreated history image collection
(I1, I2..., In) carry out picture material similarity measurement;
Step 4.1: extracting pretreated image set (I0, I1, I2..., In) gray feature, shape feature and textural characteristics,
It constructs its image grey level histogram, based on edge orientation histogram, histograms of oriented gradients and local binary patterns histogram, obtains
To gray feature vector (α0, α1, α2..., αn), shape eigenvectors (β0, β1, β2..., βn) and texture feature vector (γ0,
γ1, γ2..., γn);
Step 4.2: merging gray feature vector (α0, α1, α2..., αn), shape eigenvectors (β0, β1, β2..., βn) and texture
Feature vector (γ0, γ1, γ2..., γn), obtain multi-scale HoGC feature vector (ω0, ω1, ω2..., ωn);
Step 4.3: using EMD method by the multi-scale HoGC feature vector ω of image to be retrieved0With the synthesis histogram of history image
Figure feature vector (ω1, ω2..., ωn) similarity measurement is carried out, obtain picture material similitude set (E1, E2..., En);
Step 5: by image focus position similitude set (S1, S2..., Sn) according to image focus position similitude from big to small
Sequence, and marking serial numbers will be in images by the weight of the serial number distribution A% of the image focus position similitude set after sequence
Hold similitude set (E1, E2..., En) sort from small to large according to picture material similitude, and marking serial numbers, after sequence
Picture material similitude set serial number distribution (100-A) % weight, synthetic image lesions position similarity image serial number and
Picture material similarity image serial number obtains image synthesis similarity image serial number, obtains image searching result to be retrieved: scheming
The comprehensive similarity image serial number of picture is smaller, indicates that the image is more similar to image to be retrieved.
6. the side according to claim 5 for carrying out image retrieval based on the galactophore image searching system of lesions position and content
Method, which is characterized in that the step 3.2 the following steps are included:
Step 3.2.1: pretreated image set (I is extracted0, I1, I2..., In) breast contours and standard picture F in imageC
Breast contours;
Step 3.2.2: the affine transformation based on CPD is utilized, pretreated image set (I is registrated0, I1, I2..., In) in image
Breast contours and standard picture FCBreast contours, obtain registration transformation matrix (T0, T1, T2..., Yn);
Step 3.2.3: pass through registration transformation matrix (T0, T1, T2..., Tn) to lesions position center point coordinate collection { (x0, y0),
(x1, y1), (x2, y2) ..., (xn, yn) and lesions position radius collection (R0, R1, R2..., Rn) converted, after being converted
Lesions position center point coordinate collection { (X0, Y0), (X1, Y1), (X2, Y2) ..., (Xn, Yn)};
Step 3.2.4: by lesions position center point coordinate the collection { (X after conversion0, Y0), (X1, Y1), (X2, Y2) ..., (Xn, Yn)}
With lesions position radius collection (R0, R1, R2..., Rn) it is shown in standard picture FCOn, by the lesions position center point coordinate after conversion
Collect { (X0, Y0), (X1, Y1), (X2, Y2) ..., (Xn, Yn) and lesions position radius collection (R0, R1, R2..., Rn) determine circle
(Circle0, Circle1, Circle2..., Circlen) it is used as focal area.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610445569.8A CN105956198B (en) | 2016-06-20 | 2016-06-20 | A kind of galactophore image searching system and method based on lesions position and content |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610445569.8A CN105956198B (en) | 2016-06-20 | 2016-06-20 | A kind of galactophore image searching system and method based on lesions position and content |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105956198A CN105956198A (en) | 2016-09-21 |
CN105956198B true CN105956198B (en) | 2019-04-26 |
Family
ID=56907041
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610445569.8A Active CN105956198B (en) | 2016-06-20 | 2016-06-20 | A kind of galactophore image searching system and method based on lesions position and content |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105956198B (en) |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106846317B (en) * | 2017-02-27 | 2021-09-17 | 北京连心医疗科技有限公司 | Medical image retrieval method based on feature extraction and similarity matching |
CN108537893A (en) * | 2017-03-02 | 2018-09-14 | 南京同仁医院有限公司 | A kind of three-dimensional visualization model generation method of thyroid gland space occupying lesion |
CN107392204A (en) * | 2017-07-20 | 2017-11-24 | 东北大学 | A kind of galactophore image microcalcifications automatic checkout system and method |
CN107341265B (en) * | 2017-07-20 | 2020-08-14 | 东北大学 | Mammary gland image retrieval system and method fusing depth features |
CN107833631A (en) * | 2017-11-20 | 2018-03-23 | 新乡医学院 | A kind of medical image computer-aided analysis method |
CN111091906B (en) * | 2019-10-31 | 2023-06-20 | 中电药明数据科技(成都)有限公司 | Auxiliary medical diagnosis method and system based on real world data |
CN110837572B (en) * | 2019-11-15 | 2020-10-13 | 北京推想科技有限公司 | Image retrieval method and device, readable storage medium and electronic equipment |
CN111583320B (en) * | 2020-03-17 | 2023-04-07 | 哈尔滨医科大学 | Breast cancer ultrasonic image typing method and system fusing deep convolutional network and image omics characteristics and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101147159A (en) * | 2005-02-21 | 2008-03-19 | 三菱电机株式会社 | Fast method of object detection by statistical template matching |
CN101564323A (en) * | 2009-04-20 | 2009-10-28 | 华中科技大学 | Auxiliary equipment for diagnosing galactophore nidus based on galactophore X-ray photograph |
CN102306173A (en) * | 2011-08-25 | 2012-01-04 | 重庆理工大学 | Image similarity comparison method |
CN104915961A (en) * | 2015-06-08 | 2015-09-16 | 北京交通大学 | Lump image region display method and system based on mammary X-ray image |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2012154216A1 (en) * | 2011-05-06 | 2012-11-15 | Sti Medical Systems, Llc | Diagnosis support system providing guidance to a user by automated retrieval of similar cancer images with user feedback |
-
2016
- 2016-06-20 CN CN201610445569.8A patent/CN105956198B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101147159A (en) * | 2005-02-21 | 2008-03-19 | 三菱电机株式会社 | Fast method of object detection by statistical template matching |
CN101564323A (en) * | 2009-04-20 | 2009-10-28 | 华中科技大学 | Auxiliary equipment for diagnosing galactophore nidus based on galactophore X-ray photograph |
CN102306173A (en) * | 2011-08-25 | 2012-01-04 | 重庆理工大学 | Image similarity comparison method |
CN104915961A (en) * | 2015-06-08 | 2015-09-16 | 北京交通大学 | Lump image region display method and system based on mammary X-ray image |
Non-Patent Citations (1)
Title |
---|
"基于多特征融合图像检索系统设计与实现";邢春;《中国优秀硕士学位论文全文数据库信息科技辑》;20130715(第7 期);第I138-I263页 |
Also Published As
Publication number | Publication date |
---|---|
CN105956198A (en) | 2016-09-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105956198B (en) | A kind of galactophore image searching system and method based on lesions position and content | |
Chato et al. | Machine learning and deep learning techniques to predict overall survival of brain tumor patients using MRI images | |
CN108537773B (en) | Method for intelligently assisting in identifying pancreatic cancer and pancreatic inflammatory diseases | |
US7315639B2 (en) | Method of lung lobe segmentation and computer system | |
WO2021030629A1 (en) | Three dimensional object segmentation of medical images localized with object detection | |
CN113711271A (en) | Deep convolutional neural network for tumor segmentation by positron emission tomography | |
Tripathi et al. | A comparative analysis of segmentation techniques for lung cancer detection | |
CN110796672A (en) | Breast cancer MRI segmentation method based on hierarchical convolutional neural network | |
Haarburger et al. | Multi scale curriculum CNN for context-aware breast MRI malignancy classification | |
Maicas et al. | Pre and post-hoc diagnosis and interpretation of malignancy from breast DCE-MRI | |
Sammouda | Segmentation and analysis of CT chest images for early lung cancer detection | |
CN109191468A (en) | A kind of method, apparatus and storage medium of vessel extraction | |
Honghan et al. | Rms-se-unet: A segmentation method for tumors in breast ultrasound images | |
Teuwen et al. | Soft tissue lesion detection in mammography using deep neural networks for object detection | |
Dabass et al. | Effectiveness of region growing based segmentation technique for various medical images-a study | |
Durlak et al. | Growing a random forest with fuzzy spatial features for fully automatic artery-specific coronary calcium scoring | |
Ahmad et al. | Brain tumor detection & features extraction from MR images using segmentation, image optimization & classification techniques | |
EP1447772B1 (en) | A method of lung lobe segmentation and computer system | |
Huidrom et al. | Automated lung segmentation on computed tomography image for the diagnosis of lung cancer | |
CN109063208A (en) | A kind of medical image search method merging various features information | |
Tan et al. | A segmentation method of lung parenchyma from chest CT images based on dual U-Net | |
Devaki et al. | A novel approach to detect fissures in lung CT images using marker-based watershed transformation | |
Orban et al. | Lung nodule detection on digital tomosynthesis images: a preliminary study | |
Hammami et al. | Data augmentation for multi-organ detection in medical images | |
Mughal et al. | Early lung cancer detection by classifying chest CT images: a survey |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |