CN104217213A - Medical image multi-stage classification method based on symmetry theory - Google Patents
Medical image multi-stage classification method based on symmetry theory Download PDFInfo
- Publication number
- CN104217213A CN104217213A CN201410409810.2A CN201410409810A CN104217213A CN 104217213 A CN104217213 A CN 104217213A CN 201410409810 A CN201410409810 A CN 201410409810A CN 104217213 A CN104217213 A CN 104217213A
- Authority
- CN
- China
- Prior art keywords
- image
- classification
- medical image
- symmetry
- symmetric
- 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.)
- Granted
Links
Landscapes
- Apparatus For Radiation Diagnosis (AREA)
- Image Analysis (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention belongs to the technical field of medical information, and particularly relates to a medical image multi-stage classification method based on a symmetry theory. The method comprises the following steps that: images to be classified provide classification requests, wherein the images to be classified need to be original medical image data; and an image preprocessing process comprises the steps of image-based modeling, multi-stage classification and result display. The concepts of weak symmetry and high symmetry provided by the invention belong to redefinition on the medical images. A weak symmetry judging algorithm and a high symmetry judging algorithm are provided for realizing the multi-stage classification of the medical images. The classification accuracy of the multi-stage classification is very high, and the direct connection and layer-by-layer deep classification is realized in each stage, so the diagnosis precision of a doctor is improved, and the diagnosis time of the doctor is shortened. The medical image classification is realized by adopting the symmetry theory, so that the medical image multi-stage classification method based on the symmetry theory has higher accuracy.
Description
Technical field
The invention belongs to medical information technical field, be specifically related to a kind of medical image multistage sorting technique based on symmetric theory.
Background technology
Owing to containing abundant image and medical information in medical image, become the focus of medical science and the research of computing machine cross discipline in recent years towards the data mining technology of medical image.Along with the fast development of medical digital equipment, medical information database is widely used.The structured text information of patient, and a large amount of destructuring medical images, the data mining for medical image provides abundant data resource.Medical image effectively assist physicians can detect, locates and judge the good pernicious of it to pathological change region in diagnostic procedure, is therefore widely used in clinical diagnostic process.But, even if different judgements may be there is to same medical image in the doctor with different knowledge background, so, maintenance data method for digging Research of Medical image classification method, assist physician is diagnosed according to medical image, improve its efficiency and precision, there is higher learning value and actual application prospect.
At present, both at home and abroad in Medical Images Classification research, the main sorting technique adopted comprises statistical method, neural net method, Fuzzy Pattern Recognition Method, machine learning method etc.Existing sorting technique is positive anomaly or the position of locating exception by Medical Images Classification, and do not realize temporarily the multistage sorting technique that medical image is successively deep.The perpendicular bisector both sides that the imaging results of medical image shows about image are near symmetrical, and healthy image all presents approximate symmetrical structure in image gray levels distribution and picture shape position.And the image that pathology occurs will destroy the structure of this near symmetrical.Propose a kind of symmetric knowledge that can make full use of medical image itself to realize multistage sorting technique be a problem demanding prompt solution for this reason.
Summary of the invention
The object of the invention is to propose a kind of classification method of medical image improving medical image multistage classification accuracy based on symmetric theory.
The object of the present invention is achieved like this:
The present invention includes following steps:
(1) image to be classified proposes classification request: image to be classified should be original medical image data;
(2) Image semantic classification process: area-of-interest (ROI is extracted to original medical image, Region Of Interest), the grey level histogram of computed image ROI region, obtain the trough list of the grey level histogram of image ROI region, according to trough list to image grading texture feature extraction, according to actual needs the classification texture image obtained is standardized to unified size; ROI region is divided into the left and right sides by the perpendicular bisector about image ROI region, and the perpendicular bisector about classification texture image is divided into the left and right sides;
(3) image modeling: according to strong symmetry and the weak symmetric concept of medical image, sets up the graph model of medical image multistage classification;
(4) multistage classification: based on the weak symmetry decision method of grey level histogram Intersection, to medical image compared with coarseness having been carried out the classification of first stage; Based on point-symmetric strong symmetry decision method, in conjunction with weak symmetry decision method, be that abnormal image has carried out more fine-grained subordinate phase classification to first stage classification results, located the position of lesion region; Finally, utilize the feature that lesion region is extracted, the classification of phase III has been carried out to lesion region;
(5) result is shown: the image in raw image data storehouse is realized classification by the multistage sorting technique based on symmetric theory.
Image modeling is: set up medical image multistage disaggregated model according to weak symmetry and strong symmetric concept, weak symmetry is: for a medical image G, weak symmetry refers to the number of pixels that D in G (L) and D (R) has in K at each group, weighs weak symmetry by the distance that they intersect
Wherein, group is the width that histogram is often organized apart from K.
Strong symmetry is: strong symmetry refer to the number of the point in TI in circle or on circle number, weigh strong symmetry rad (v (i))=mov (v (i)) with their radius.
Multistage is categorized as: first to medical image again compared with the classification of coarseness being applied weak symmetry decision method and carry out the first stage; Then apply strong symmetry decision method and weak symmetry decision method to combine first stage classification results be the classification that abnormal image carries out more fine-grained subordinate phase, the position of location lesion region; Finally, utilize the feature that lesion region is extracted, the classification of phase III has been carried out to lesion region.
Beneficial effect of the present invention is:
The weak symmetry that the present invention proposes and strong symmetric concept redefine one of medical image.The multistage classification that weak symmetry decision method and strong symmetry decision method realize medical image is proposed.The classification accuracy of this multistage classification is very high, and each stage is directly connected successively deeply, thus improves the diagnostic accuracy of doctor and shorten Diagnostic Time.Adopt symmetric theory to realize the classification of medical image, make the medical image multistage sorting technique based on symmetric theory have higher accuracy.
Accompanying drawing explanation
Fig. 1 is the process about perpendicular bisector symmetry division;
Fig. 2 is MI iconic model;
Fig. 3 is weak symmetry example;
Fig. 4 is the example of initiatively change and passive change
Fig. 5 is strong symmetry example;
Fig. 6 is the medical image multistage sorting technique process flow diagram based on symmetric theory;
Embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is further illustrated:
First pre-service is carried out to medical image:
1. each the original brain CT image zooming-out ROI region in pair original image storehouse;
2. intercept ROI region and correct;
3. the trough distribution situation of computed image ROI region grey level histogram, obtains the trough table of grey level histogram;
4. threshold value is set according to trough table and repeatedly texture is extracted to image, thus obtain Mipmap image;
5. the image finally sized by Mipmap image normalization will being COLUMN × ROW;
6. image ROI region is divided into left and right two parts about perpendicular bisector;
7. the texture image after standardization size is divided into left and right two parts about perpendicular bisector;
Be kept in corresponding database to Image Segmentation Using process, through above process, all corresponding four the equal-sized images of each original image.Calculate d (L, the R) value in grey image R OI region and rad (v (i))=mov (v (the i)) value of texture image; Calculate the value of gray level image d (L, R), realize the classification of the first stage of medical image; Then the classification of combining the subordinate phase realizing medical image of the weak symmetry d (L, R) based on gray level image and the strong symmetry rad (v (i)) based on Texture Points on texture image is utilized, the position of location lesion region; Finally feature is extracted to lesion region, utilize these features to realize the classification of the phase III of medical image;
A concrete Images Classification process is as follows:
1. couple medical image MI to be sorted proposes classification request;
2. couple medical image MI to be sorted carries out pre-service, obtains corresponding four images (about gray level image L, R and the texture image L of perpendicular bisector symmetry
h, R
h);
3. obtain d (L, the R) value of gray level image L and R according to weak symmetric definition, apply weak symmetry decision method realizes first stage classification to medical image, be namely divided into normal medical image and abnormal medical image;
4. the relation of initiatively change and passive change in medical image is defined according to the medical knowledge of medical image, namely initiatively change is less than passive change, initiatively the relation of change and passive change is present in normal medical image and abnormal medical image, by val=d (L, L ')-d (R, R ') calculate, wherein the value of val is the numerical value of initiatively magnitude relationship between change and passive change, d (L, L ') be on the left of normal medical image and abnormal medical image on the left of weak symmetric value, d (R, R ') be on the right side of normal medical image and abnormal medical image on the right side of weak symmetric value,
5. the abnormal medical image in pair first stage carries out the classification of subordinate phase, the active calculating abnormal medical image and normal medical image changes and the relation value val=d (L of passive change, L ')-d (R, R '), if val<0, then the middle generation of the left side L ' of abnormal medical image is initiatively change, and what right side R ' occurred is passive change.So abnormal area T is present in the left side L ' of abnormal medical image.Otherwise work as val>0, abnormal area is present in the right side R ' of abnormal medical image.The classification of subordinate phase realizes the location to abnormal area T;
6. extract the edge contour in T region and calculate following characteristics value:
Wherein C is compactness, davg is Average normalized radius, Ro is roughness, δ
2be variance, Sk is measure of skewness, Pe is peak value, A is the area of tumor's profiles, and p is the girth of tumor's profiles.N is the number that edge is put, if the center of the minimum enclosed rectangle of ROI region is (x
0, y
0), radius d (i) is each marginal point (x
i, y
i) to picture centre (x
0, y
0) distance, K is the quantity that edge is put, L
iit is the Euclidean distance on edge between point and point, μ is average, adopt iteration stochastic sampling method, estimate by spectral clustering the class label not marking sample in each sampling, SVM is used to carry out model learning, successive optimization model, to improve classify accuracy, realize the classification of phase III, be categorized as optimum and pernicious two classes by abnormal area T;
Above example shows, the medical image multistage sorting technique based on symmetric theory that the present invention proposes has actual using value.
The present invention also has some technical characteristics like this:
1. multistage classification medical image model:
A medical image (Medical Image, MI) is set G={ (L, R, T) | L={v
1, v
2..., v
n, n is the number of MI left pixel point, v
i∈ [0,255].R={u
1, u
2..., u
m, m is the number of MI right pixel point, u
i∈ [0,255], T={w
1, w
2..., w
p, 0<p<n or 0<p<m, w
i∈ [0,255] }.
Wherein L and R is that MI is divided equally the set of rear left side and right pixel by median perpendicular respectively.T is the pixel set of abnormal area in MI,
or
.If MI is normal, then T set is for empty, otherwise T set is not empty.
In G, because some gene in organ occurs abnormal, when Normocellular chromosome has run into this gene just energy entice cell generation distorted proliferation, hinder Normocellular growth, thus the side cell ramp of exception, opposite side then hinders Normocellular growth due to reasons such as extruding coverings, makes its passive generation deformation.Existence due to T region makes L and R set change.It is that larger change occurs in abnormal side pixel set that this medical knowledge is embodied on image, but not the change that the side pixel set of exception occurs is little compared with abnormal side.Fig. 2 (a) divides equally for left and right two parts about brain median perpendicular for normal medical science brain image, being evenly distributed of pixel in L and R set in G.Fig. 2 (b) is abnormal brain image example, and we can find that in G, there is an abnormal area T gathers.Due to the existence of T set, in G, in L and R set there is very large difference in the distribution of pixel.
In abnormal image (Abnormal Image, AMI), if
, then the change that the left side of AMI occurs is called initiatively change, and the change that right side occurs is called passive change; If
, then the change that the right side of AMI occurs is called initiatively change, and the change that left side occurs is called passive change.
In Fig. 3 (b), L and R pixel set difference is very large, causes the existence in T region.
, namely extremely appear at left side, so the change that in Fig. 3 (b), left side occurs is exactly active change, right side is passive change.Do you how to determine the difference of L and R pixel set in G? if D (X) is the grey level histogram containing K group distance of a pixel set X, provide weak symmetric definition below.
For a medical image G, weak symmetry refers to the number of pixels that D in G (L) and D (R) has in K at each group, weighs weak symmetry by the distance that they intersect
Wherein, group is the width that histogram is often organized apart from K.
Fig. 3 is the grey level histogram of left, right parts of images in Fig. 2 (a).Horizontal ordinate represents the distribution of gray-scale value, and span [0 255], ordinate represents the number K of pixel on a certain gray-scale value.The distance of the pixel intersection of sets of L and R can be learnt according to above formula.The part that the larger explanation of weak symmetry d (L, R) L and R intersects is more, and difference is less.The part that the less explanation of weak symmetry d (L, R) L and R intersects is fewer, and difference is larger.
Normal medical image (Normal Medical Image is comprised at G, and abnormal medical image (Abnormal Medical Image NMI), AMI), how they are distinguished, provide the relation of initiatively change and passive change in NMI and AMI below.Initiatively the relation of change and passive change is present in the weak symmetric contrast of NMI and AMI, and initiatively change is less than passive change.
Initiatively the relation of change and passive change as shown in Figure 4.Fig. 4 (a) is NMI, Fig. 4 (b) is AMI.Fig. 4 (c) (d) is the histogram on Fig. 4 (a) left side and right side, and Fig. 4 (e) (f) is the histogram on Fig. 4 (b) left side and right side.If the left side of known NMI and right pixel set are the left side of L and R, AMI and right pixel set is L ' and R '.Location abnormal area
or
apply following formula.
val＝d(L,L')-d(R,R')
If val<0, then the middle generation of the L ' of AMI is initiatively change, and what R ' occurred is passive change.So T region is present in the L ' of AMI.Otherwise work as val>0, T region is present in R '.The position of abnormal area T in gray level medical image can be located according to the relation of initiatively change and passive change.According to formula, the val>0 of (a) and (b) in Fig. 4, namely T region appears at the right side of Fig. 4 (b).
The texture image of medical image is made up of many textures, and every texture is made up of several points.The texture image that NMI is corresponding divide equally about brain median perpendicular after left and right sides near symmetrical, give strong symmetric definition according to this principle.In the texture image of AMI, the texture number that the side that initiatively change occurs presents is more than the side texture number of passive change.
A texture image (Texture Image, TI) is a set F={ (L
h, R
h) | L
h={ L
h1, L
h2..., L
hn, L
hil
hin the i-th texture, Lhi={v (1), v (2) ..., v (n), v (i) represent composition L
hithe point of texture.R
h={ R
h1, R
h2..., R
hn, R
hir
hin the i-th texture, R
hj=v (1), v (2) ..., and v (m) }, v (i) represents composition R
hithe point of texture }, wherein L
hand R
hit is the set of left side and the right side texture divided equally about median perpendicular.
In texture image TI, to form some v (i) of texture for the center of circle, the mobility mov (v (i)) of some v (i) does circle for radius.Point in circle or on circle is all thought with this point-symmetric.
Strong symmetry to refer in TI the number of the point in circle or on circle number, weigh strong symmetry rad (v (i))=mov (v (i)) with their radius.
The partial enlarged drawing of Fig. 5 (a) to be a texture image TI, Fig. 5 (b) and Fig. 5 (c) be Fig. 5 (a) and symmetrical about center line.Select the side i.e. left side that in TI, Texture Points is many as template, to mark the rad (v (i)) of each point according to strong symmetric concept, as shown in Fig. 5 (d), in figure, represent that significance index is a little different with the point of different colours.Fig. 5 (f) is the symmetric graph of Fig. 5 (d) about center line.Fig. 5 (g) is the exemplary plot of strong symmetry rad (v (i)).
2. the concrete steps of weak symmetry decision method:
For G, the distribution of pixel is very important feature.Judge whether T region exists by weak symmetry d (L, R), can judge that whether medical image is normal.Image applications character 1 for the existence of T region compares the relation of initiatively change and passive change in NMI and AMI.Known L and R is the set of left side in NMI and right pixel, and L ' and R ' is the set of left side and right pixel in AMI.Calculate the weak symmetry d (L of L and L ', L ') with the weak symmetry d (R of R and R ', R '), if d is (L, L ') <d (R, R '), then the change that pixel distribution occurs in L and L ' is greater than the change that in R and R ', pixel distribution occurs.Show that T region appears in the L ' of AMI, namely
.Otherwise appear in the R ' of AMI, namely
3. the concrete steps of strong symmetry decision method:
Because the change such as direction, size, position of weak symmetric characteristics to image or image-region is insensitive, so weak symmetric characteristics can not catch the local attribute of objects in images well, the strong symmetry decision method proposing image herein addresses this problem.Different from grey level histogram feature, textural characteristics is not the feature based on pixel, and it needs to carry out statistical computation in the region comprising multiple pixel.The position of gray-level pixels point in the positioning image of textural characteristics energy local.Each texture is all made up of the point of gray scale acute variation, is judged the global symmetry of texture image by strong symmetric definition.On texture, each some v (i) divides some v ' (i) being mapped to opposite side equally about median perpendicular, can draw the mobile range of a v ' (i) according to strong symmetric definition.Point so in mobile range is all thought with origin symmetry.
Texture image set F={L
h, R
hin, L in the texture image of normal medical image
hand R
hit is near symmetrical.If there is lesion region, so the side texture number of lesion region is by more, namely forms counting of texture and increases.Scanning L
hand R
hthe number of the point of middle composition texture, the many side of the number supposing to put is as primary template.The all points in mobile range of elimination method cancellation of opposite side application Texture Points are mapped to about axis of symmetry.Remaining some markf=suml-sumr after first time cancellation.Then L is calculated
hand R
hin the number of remaining point, using the side of non-template as the point of secondary template cancellation opposite side.Remaining some markl=suml-sumr after second time cancellation.The relatively number sum=markf+markl of both sides left point, if sum >=0 illustrates that left side exists T region, sum<0 illustrates that right side exists T region.The basic thought of Here it is the strong symmetry decision method of image.
4. the concrete steps of medical image multistage sorting technique:
By learning the investigation of doctor, it utilizes medical image to carry out diagnostic procedure to be: first judge that whether this medical image MI is normal according to the medical knowledge of doctor.Secondly, research is continued to AMI, appear in L or R according to the clinical diagnosis micro-judgment of doctor is abnormal.Finally, further research is done to the region T in AMI, judge the good pernicious of T region according to doctor's diagnostic experiences in the past.The diagnostic procedure of simulating doctor herein proposes the sorting technique of MI multistage sorter (multi-stage classification, MSC), and its implementation procedure as shown in Figure 6.
MSC-1: application WSDA has carried out first stage classification to medical image MI, is namely categorized as NMI and AMI to MI.
MSC-2: process further obtain corresponding texture image F to being labeled as abnormal image AMI, applies WSDA and SSDA simultaneously and judges the left side of abnormal area T in medical image or right side.
MSC-3: herein according to the edge contour of abnormal area T, the multiple features extracting T realize carrying out good pernicious classification to AMI.
Key of the present invention is to use medical image symmetric theory to analyze medical image, thus utilizes weak symmetry decision method and strong symmetry decision method to carry out multistage classification to image.
The present invention takes into full account the structural and regular of medical image itself.By being that namely two parts of left and right near symmetrical can realize multistage classification fast by Iamge Segmentation, also disclosed the feature of the medical image of pathology, thus make this multistage sorting technique itself closer to the right mind process of the mankind.
Claims (6)
1., based on a medical image multistage sorting technique for symmetric theory, it is characterized in that, comprise the steps:
(1) image to be classified proposes classification request, and image to be classified should be original medical image data;
(2) Image semantic classification process: region of interest ROI is extracted to original medical image, the grey level histogram of computed image ROI region, obtain the trough list of the grey level histogram of image ROI region, according to trough list to image grading texture feature extraction, according to actual needs the classification texture image obtained is standardized to unified size; ROI region is divided into the left and right sides by the perpendicular bisector about image ROI region, and the perpendicular bisector about classification texture image is divided into the left and right sides;
(3) image modeling: according to strong symmetry and the weak symmetric concept of medical image, sets up the graph model of medical image multistage classification;
(4) multistage classification: based on the weak symmetry decision method of grey level histogram Intersection, to medical image compared with coarseness having been carried out the classification of first stage; Based on point-symmetric strong symmetry decision method, in conjunction with weak symmetry decision method, be that abnormal image has carried out more fine-grained subordinate phase classification to first stage classification results, located the position of lesion region; Finally, utilize the feature that lesion region is extracted, the classification of phase III has been carried out to lesion region;
(5) result is shown: the image in raw image data storehouse is realized classification by the multistage sorting technique based on symmetric theory.
2. a kind of medical image multistage sorting technique based on symmetric theory according to claim 1, it is characterized in that, described image modeling process is: the graph model setting up the classification of medical image multistage according to the strong symmetric concept of medical image and weak symmetric concept.
3. a kind of medical image multistage sorting technique based on symmetric theory according to claim 1, it is characterized in that, described weak symmetry is: for a medical image G, weak symmetry refers to the number of pixels that D in G (L) and D (R) has in K at each group, weighs weak symmetry by the distance that they intersect
Wherein, group is the width that histogram is often organized apart from K.
4. a kind of medical image multistage sorting technique based on symmetric theory according to claim 1, it is characterized in that, described strong symmetry is: strong symmetry refer to the number of the point in TI in circle or on circle number, weigh strong symmetry rad (v (i))=mov (v (i)) with their radius.
5. a kind of medical image multistage sorting technique based on symmetric theory according to claim 1, it is characterized in that, described multistage assorting process is: based on the weak symmetry decision method of grey level histogram Intersection, to medical image compared with coarseness having been carried out the classification of first stage; Based on point-symmetric strong symmetry decision method, in conjunction with weak symmetry decision method, be that abnormal image has carried out more fine-grained subordinate phase classification to first stage classification results, located the position of lesion region; Finally, utilize the feature that lesion region is extracted, the classification of phase III has been carried out to lesion region.
6. a kind of medical image multistage sorting technique based on symmetric theory according to claim 1, it is characterized in that, the pretreated database of described medical image is: by carrying out pre-service to the image of often opening in existing image library, then carries out multistage classified image modeling to pretreated image set and obtains a multistage classification atlas D={G
1, G
2..., G
n.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410409810.2A CN104217213B (en) | 2014-08-20 | 2014-08-20 | A kind of medical image multistage sorting technique based on symmetric theory |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410409810.2A CN104217213B (en) | 2014-08-20 | 2014-08-20 | A kind of medical image multistage sorting technique based on symmetric theory |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104217213A true CN104217213A (en) | 2014-12-17 |
| CN104217213B CN104217213B (en) | 2018-04-27 |
Family
ID=52098676
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410409810.2A Active CN104217213B (en) | 2014-08-20 | 2014-08-20 | A kind of medical image multistage sorting technique based on symmetric theory |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104217213B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573742A (en) * | 2014-12-30 | 2015-04-29 | 中国科学院深圳先进技术研究院 | Medical image classification method and system |
| CN104933712A (en) * | 2015-06-12 | 2015-09-23 | 哈尔滨工程大学 | Graph model displaying method based on cerebral CT image |
| CN106446923A (en) * | 2016-05-25 | 2017-02-22 | 哈尔滨工程大学 | Medical image classification method based on corner matching |
| CN109272508A (en) * | 2018-08-02 | 2019-01-25 | 哈尔滨工程大学 | A kind of petri net image partition method based on rough set and rough entropy |
| CN110210543A (en) * | 2019-05-24 | 2019-09-06 | 上海联影智能医疗科技有限公司 | Image classification system, method, equipment and storage medium |
| CN111242144A (en) * | 2020-04-26 | 2020-06-05 | 北京邮电大学 | Method and device for detecting abnormality of power grid equipment |
| CN111488933A (en) * | 2020-04-13 | 2020-08-04 | 上海联影智能医疗科技有限公司 | Image classification method, network, computer device and storage medium |
| CN113313203A (en) * | 2021-06-22 | 2021-08-27 | 哈尔滨工程大学 | Medical image classification method based on extension theory and deep learning |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102467667A (en) * | 2010-11-11 | 2012-05-23 | 江苏大学 | Classification method of medical image |
| CN102509109A (en) * | 2011-10-20 | 2012-06-20 | 西北民族大学 | Method for distinguishing Thangka image from non-Thangka image |
| CN103226582A (en) * | 2013-04-08 | 2013-07-31 | 哈尔滨工程大学 | Medical image retrieving method on basis of uncertain fixed point image |
| WO2014080305A2 (en) * | 2012-11-20 | 2014-05-30 | Koninklijke Philips N.V. | Integrated phenotyping employing image texture features. |
-
2014
- 2014-08-20 CN CN201410409810.2A patent/CN104217213B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102467667A (en) * | 2010-11-11 | 2012-05-23 | 江苏大学 | Classification method of medical image |
| CN102509109A (en) * | 2011-10-20 | 2012-06-20 | 西北民族大学 | Method for distinguishing Thangka image from non-Thangka image |
| WO2014080305A2 (en) * | 2012-11-20 | 2014-05-30 | Koninklijke Philips N.V. | Integrated phenotyping employing image texture features. |
| CN103226582A (en) * | 2013-04-08 | 2013-07-31 | 哈尔滨工程大学 | Medical image retrieving method on basis of uncertain fixed point image |
Non-Patent Citations (2)
| Title |
|---|
| 林伟曦: "人脑CT图像特征提取及其分类", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 黎维娟: "多模态影像脑部疾病检索研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573742B (en) * | 2014-12-30 | 2018-03-23 | 中国科学院深圳先进技术研究院 | Classification method of medical image and system |
| CN104573742A (en) * | 2014-12-30 | 2015-04-29 | 中国科学院深圳先进技术研究院 | Medical image classification method and system |
| CN104933712A (en) * | 2015-06-12 | 2015-09-23 | 哈尔滨工程大学 | Graph model displaying method based on cerebral CT image |
| CN104933712B (en) * | 2015-06-12 | 2017-12-01 | 哈尔滨工程大学 | Graph model methods of exhibiting based on brain CT images |
| CN106446923B (en) * | 2016-05-25 | 2019-08-06 | 哈尔滨工程大学 | Classification method of medical image based on corners Matching |
| CN106446923A (en) * | 2016-05-25 | 2017-02-22 | 哈尔滨工程大学 | Medical image classification method based on corner matching |
| CN109272508A (en) * | 2018-08-02 | 2019-01-25 | 哈尔滨工程大学 | A kind of petri net image partition method based on rough set and rough entropy |
| CN109272508B (en) * | 2018-08-02 | 2021-10-26 | 哈尔滨工程大学 | Petri network image segmentation method based on rough set and rough entropy |
| CN110210543A (en) * | 2019-05-24 | 2019-09-06 | 上海联影智能医疗科技有限公司 | Image classification system, method, equipment and storage medium |
| CN110210543B (en) * | 2019-05-24 | 2021-08-17 | 上海联影智能医疗科技有限公司 | Image classification system, method, apparatus and storage medium |
| CN111488933A (en) * | 2020-04-13 | 2020-08-04 | 上海联影智能医疗科技有限公司 | Image classification method, network, computer device and storage medium |
| CN111488933B (en) * | 2020-04-13 | 2024-02-27 | 上海联影智能医疗科技有限公司 | Image classification method, network, computer device, and storage medium |
| CN111242144A (en) * | 2020-04-26 | 2020-06-05 | 北京邮电大学 | Method and device for detecting abnormality of power grid equipment |
| CN113313203A (en) * | 2021-06-22 | 2021-08-27 | 哈尔滨工程大学 | Medical image classification method based on extension theory and deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104217213B (en) | 2018-04-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104217213A (en) | Medical image multi-stage classification method based on symmetry theory | |
| CN103098090B (en) | Multiparameter three-dimensional magnetic resonance imaging brain tumor partition method | |
| CN103996018B (en) | Face identification method based on 4DLBP | |
| CN102999886B (en) | Image Edge Detector and scale grating grid precision detection system | |
| CN109378052B (en) | The preprocess method and system of image labeling | |
| CN111931811B (en) | Calculation method based on super-pixel image similarity | |
| CN107256558A (en) | The cervical cell image automatic segmentation method and system of a kind of unsupervised formula | |
| CN104715251B (en) | A kind of well-marked target detection method based on histogram linear fit | |
| CN107730515A (en) | Panoramic picture conspicuousness detection method with eye movement model is increased based on region | |
| CN104282008B (en) | The method and apparatus that Texture Segmentation is carried out to image | |
| CN101551855A (en) | Auxiliary diagnostic system for tracing self-adaptive kernel matching and auxiliary diagnostic method thereof | |
| CN104573742A (en) | Medical image classification method and system | |
| CN107437068A (en) | Pig individual discrimination method based on Gabor direction histograms and pig chaeta hair pattern | |
| CN106548478B (en) | Active contour image partition method based on local fit image | |
| CN104281856A (en) | Image preprocessing method and system for brain medical image classification | |
| CN106909881A (en) | The method and system of corn breeding base ridge number are extracted based on unmanned aerial vehicle remote sensing images | |
| US20240304007A1 (en) | Method and system for the segmentation and clustering of nuclei based on single-cell pathological images | |
| Di Cataldo et al. | ANAlyte: A modular image analysis tool for ANA testing with indirect immunofluorescence | |
| CN104732230A (en) | Pathology image local-feature extracting method based on cell nucleus statistical information | |
| CN104751461A (en) | White cell nucleus segmentation method based on histogram threshold and low rank representation | |
| Hao et al. | VP-Detector: A 3D multi-scale dense convolutional neural network for macromolecule localization and classification in cryo-electron tomograms | |
| CN107423771A (en) | A kind of two phase method for detecting change of remote sensing image | |
| CN104835155A (en) | Fractal-based early-stage breast cancer calcification point computer auxiliary detection method | |
| CN102609721B (en) | Remote sensing image clustering method | |
| CN113240620B (en) | Highly-adhesive and multi-size brain neuron automatic segmentation method based on point marking |
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 |