CN110276407A - A kind of Hepatic CT staging system and classification method - Google Patents
A kind of Hepatic CT staging system and classification method Download PDFInfo
- Publication number
- CN110276407A CN110276407A CN201910563651.4A CN201910563651A CN110276407A CN 110276407 A CN110276407 A CN 110276407A CN 201910563651 A CN201910563651 A CN 201910563651A CN 110276407 A CN110276407 A CN 110276407A
- Authority
- CN
- China
- Prior art keywords
- candidate
- liver
- liver neoplasm
- neoplasm
- feature
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration by the use of local operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30056—Liver; Hepatic
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30096—Tumor; Lesion
Abstract
The present invention discloses a kind of Hepatic CT staging system and classification method, including suspicious liver neoplasm detection module, candidate liver neoplasm segmentation module, candidate liver neoplasm categorization module;Suspicious liver neoplasm detection module is partitioned into liver parenchyma using the method that liver map is combined with manual correction, then suspicious liver neoplasm detection is carried out using variable annular filter, obtain candidate liver neoplasm, the center for determining variable annular filter is converted by gray scale weight distance, as candidate liver neoplasm seed point;Candidate liver neoplasm segmentation module completes candidate liver neoplasm segmentation using 3D region growing algorithm;Candidate liver neoplasm categorization module, which carries out coarse false positive to candidate liver neoplasm using empirical value method, to be reduced, then uses SVM algorithm as classifier, and candidate liver neoplasm is divided into tumour and other liver organizations;Present invention Hepatic CT lesion detection susceptibility with higher, and effectively increase the accuracy of classification.
Description
Technical field
The present invention relates to a kind of technical field of image processing more particularly to a kind of Hepatic CT staging system and classification sides
Method.
Background technique
Liver plays an important role in human metabolism as the maximum body of gland organ of human body.Liver is located at people
Body abdomen, and it is adjacent with multiple vitals, blood supply is abundant, it is easy to be transferred to peripheral organs once causing illness, while
It is easy to be influenced by other organs.In addition, liver cancer, as the second largest lethal illness, the every annual morbidity of China reaches every 100,000 26.39 people,
The deficiency of medicine classification level is to lead the high major reason of lethality.CT image, compared to the radiation techniques such as MR, X-ray, tool
Have imaging fast, contrast is high, the low advantage of cost, therefore, be widely used in liver diseases, the detection of special liver's tumour with point
Class, to push the development of the computer-aided detection system based on CT.
Since liver is adjacent with multiple organs, and the gray level of liver image is because of people, different because of equipment, meanwhile, close to liver
The damage on dirty boundary will lead to liver boundary and obscure, and causes the extraction of liver area extremely complex, accurately detects from liver area
Suspicious object is extremely difficult, and suspicious region here includes two crucial tissues: blood vessel and cancerous tissue;It is extracted in suspicious region
In the process, shape feature, image density, textural characteristics were widely used, however, the tumour of the liver of different times is close
Degree, texture information have different manifestations, therefore, rely on basic threshold information merely and are difficult to complete accurately mentioning for suspicious object
It takes;And the method based on shape information is difficult to realize since the position of tumour in liver can change with time change
The accurate detection of tumour;Last part is the classification of candidate tumor or disease, this stage generallys use traditional supporting vector
Candidate region is classified as liver neoplasm tissue and other liver organizations, traditional more bases of svm classifier algorithm by machine svm classifier algorithm
It is the kernel function for being inclined to locality in gaussian radial basis function core, abducent ability can be dropped with the increase of range parameter σ
It is low, it is difficult to classification is trained to large-scale sample, therefore not can guarantee the accuracy of liver neoplasm classification.
Summary of the invention
The object of the present invention is to provide a kind of Hepatic CT staging system and classification methods, to solve the above-mentioned prior art
There are the problem of, improve the susceptibility of Hepatic CT lesion detection, and improve the accuracy of categorizing system.
To achieve the above object, the present invention provides following schemes: the present invention provides a kind of Hepatic CT staging system,
Including suspicious liver neoplasm detection module: for detecting to the liver parenchyma being partitioned into, obtaining the kind of candidate liver neoplasm
Sub- point;Candidate liver neoplasm divides module: for carrying out region growth, dividing candidate to all candidate liver neoplasm seed points
Liver neoplasm obtains feature completely candidate liver neoplasm;Candidate liver neoplasm categorization module: for the feature being partitioned into is complete
Whole candidate liver neoplasm carries out feature extraction, classification, further determines that liver neoplasm region.
A kind of Hepatic CT staging method, comprising the following steps:
Step 1, suspicious liver neoplasm detection module are partitioned into liver parenchyma, carry out suspicious liver to the liver parenchyma being partitioned into
Dirty tumor target detection, obtains the seed point of candidate liver neoplasm;
Step 1-1, it is corrected based on graphical spectrum technology, in conjunction with manual segmentation, coarse segmentation goes out liver parenchyma;
Step 1-2, according to the morphological feature of liver neoplasm, using the detection method based on variable annular filter, to point
The liver parenchyma cut out is detected, and the size of variable annular filter and the size of candidate liver neoplasm is made to match;
Step 1-3, variable loop is improved using the method for gray scale weight distance conversion according to the gray feature of liver neoplasm
The difference effect of mode filter, obtains the center of variable annular filter, i.e., candidate liver neoplasm seed point;
Step 2, for the seed point of candidate's liver neoplasm obtained in step 1, candidate liver neoplasm segmentation module passes through
The 3D region growing algorithm of seed point conformity classification, seed point, is split candidate liver neoplasm, it is complete to obtain feature
Candidate liver neoplasm;
Step 3, candidate liver neoplasm categorization module are complete to the feature being partitioned into step 2 using empirical value method
Candidate liver neoplasm, which carries out coarse false positive, to be reduced, the candidate liver neoplasm after being screened;
The feature of step 4, candidate liver neoplasm categorization module selection characterization liver neoplasm, progress feature calculation, and according to
Liver neoplasm feature uses multinomial to classify as the kernel function of svm classifier algorithm to liver neoplasm.
Preferably, step 3 the following steps are included:
Step 3-1, the feature being partitioned into for step 2 complete candidate liver neoplasm will be in using empirical value
Between normal liver tissue in threshold range reject, retain the threshold region where tumour, blood vessel;
Step 3-2, to reject normal liver tissue candidate liver neoplasm, by false positive reduce rule R1, R2, R3 into
Row false positive is eliminated, and R1, R2, R3 are based respectively on volume characteristic, radial features and the sphericity feature of candidate liver neoplasm;
WithThe bottom threshold of volume, diameter and sphericity is respectively represented, candidate liver neoplasm
Characteristic value is less than the bottom threshold, then determines that the candidate liver neoplasm is candidate for false positive, and eliminated;
WithThe upper threshold of volume, diameter and sphericity is respectively represented, candidate liver neoplasm
Characteristic value is greater than the upper threshold, then determines that the candidate liver neoplasm is candidate for false positive, and eliminated.
Preferably, candidate liver neoplasm categorization module sets false sun according to the characteristic information of candidate liver neoplasm in step 3
Property reduce rule R1, R2, R3 threshold value, R1 upper threshold be 682.7cm3, R1 bottom threshold is 33.5cm3, R2 upper threshold is
8cm, R2 bottom threshold are 2cm, and R3 upper threshold is that 1.0, R3 bottom threshold is 0.45.
Preferably, the feature that 8 characterization liver neoplasms are chosen in step 4, carries out feature selecting and feature calculation, passes through
Svm classifier algorithm classifies to candidate liver neoplasm.
Preferably, cluster operation is carried out to training sample according to k means clustering algorithm in step 4, with the size of class cluster into
Row auxiliary estimation.
The invention discloses following technical effects:
1. candidate's liver neoplasm detection module of the invention, passes through variable annular filter detection method and gray scale weight distance
The method of conversion combines, and obtains candidate liver neoplasm, can not only detect solitary liver tumour, to the lesser frosted glass of gray scale
And nearby bleeding pipe liver neoplasm can be detected well, Hepatic CT lesion detection susceptibility with higher;
2. a pair candidate liver neoplasm divides the feature of module segmentation out completely candidate liver neoplasm, first carries out false positive and disappear
Except classifying again, the workload of subsequent classification is effectively reduced, improves candidate liver neoplasm classification effectiveness;
3. candidate liver neoplasm categorization module uses kernel function of the multinomial as svm classifier algorithm, have higher complete
Office's property, polynomial parameters, adjustment parameter, susceptibility and specificity are estimated based on svm classifier algorithm, improves the accurate of classification
Degree.
Detailed description of the invention
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is Hepatic CT lesion detection of the present invention and categorizing system overall flow;
Fig. 2 is variable annular filter construction schematic diagram of the present invention;
Fig. 3 is the present invention is based on the conversion of the distance of gray scale weight, and left-half is two-value apart from conversion effect, right half part
It is gray scale weight apart from conversion effect.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing and specific real
Applying mode, the present invention is described in further detail.
Referring to Fig.1-3, a kind of Hepatic CT staging system, including suspicious liver neoplasm detection module: for segmentation
Liver parenchyma out is detected, and the seed point of candidate liver neoplasm is obtained;Candidate liver neoplasm divides module: for all
Candidate liver neoplasm seed point carry out region growth, dividing candidate liver neoplasm obtains feature completely candidate liver neoplasm;
Candidate liver neoplasm categorization module: for by the feature being partitioned into, completely candidate liver neoplasm to carry out feature extraction, classification, into
One step determines liver neoplasm region.
A kind of Hepatic CT staging method, comprising the following steps:
Step 1, suspicious liver neoplasm detection module are partitioned into liver parenchyma, carry out suspicious liver to the liver parenchyma being partitioned into
Dirty tumor target detection, obtains the seed point of candidate liver neoplasm;
Step 1-1, it is corrected based on graphical spectrum technology, in conjunction with manual segmentation, coarse segmentation goes out liver parenchyma;
Step 1-2, according to the morphological feature of liver neoplasm, using the detection method based on variable annular filter, to point
The liver parenchyma cut out is detected, and the size of variable annular filter and the size of candidate liver neoplasm is made to match;
Original Liver CT image is f, passes through ring junction constitutive element SringExpansive working is carried out to Original Liver CT image, is obtained
CT image for liver f after expansionring, i.e., filtered CT image for liver, while after available Original Liver CT image and filtering
CT image for liver error image;
Variable annular filter is defined as follows:
X, y respectively corresponds the abscissa and ordinate of pixel, r1、r2Respectively indicate the inside radius and outer half of loop filter
Diameter (present invention in r1、r21) equal value is that ⊕ represents morphological dilation symbol.
Using the two-value switch technology of input picture, it is adaptively adjusted the radius size of variable annular filter, by two
Value converts the conversion value for determining current point, and the radius of the best variable annular filter as the point can closer to tumor center
The radius for becoming loop filter is bigger, when variable annular filter can just entangle tumour, the center of variable annular filter
It is just consistent with tumor center, and thus generate the maximum difference of Original Liver CT image and filtered CT image for liver;
Step 1-3, variable loop is improved using the method for gray scale weight distance conversion according to the gray feature of liver neoplasm
The difference effect of mode filter, obtains the center of variable annular filter, i.e., candidate liver neoplasm seed point;
Liver neoplasm wants small compared to the gray value of surrounding normal tissue, and contrast is lower, for the classification for improving liver neoplasm
Rate, in terms of strengthening contrast, based on gray scale weight apart from switch technology, generate the error image of more high contrast, at this time
The center of variable annular filter is the centre coordinate of corresponding tumor region, i.e., the seed point of candidate liver neoplasm.
Step 2, candidate liver neoplasm segmentation module carry out the seed point of candidate's liver neoplasm obtained in step 1 consistent
Property screening, if some seed point appears in isolated sectioning image, regarded as noise point, and delete, if certain
A seed point appears in 2 and the above serial section, then is left candidate seed point;The area 3D is passed through to candidate seed point
Domain growth algorithm is handled, if S is tumour set of voxels, L is other liver area set of voxels, if L is adjacent with S, and L
Gray scale difference value between candidate seed point is less than or equal to threshold value δ, then S is added in L, then the iteration above process, until S is not
Become, obtains feature completely candidate liver neoplasm, and completely candidate liver neoplasm is split by feature;
Step 3, candidate liver neoplasm categorization module are complete to the feature being partitioned into step 2 using empirical value method
Candidate liver neoplasm, which carries out coarse false positive, to be reduced, the candidate liver neoplasm after being screened;
Step 3-1, the feature being partitioned into for step 2 complete candidate liver neoplasm will be in using empirical value
Between normal liver tissue in threshold range reject, retain the threshold region where tumour, blood vessel;
Step 3-2, to reject normal liver tissue candidate liver neoplasm, by false positive reduce rule R1, R2, R3 into
Row false positive is eliminated, and R1, R2, R3 are based respectively on volume characteristic, radial features and the sphericity feature of candidate liver neoplasm;
WithThe bottom threshold of volume, diameter and sphericity is respectively represented, candidate liver neoplasm
Characteristic value is less than the bottom threshold, then determines that the candidate liver neoplasm is candidate for false positive, and eliminated;
WithThe upper threshold of volume, diameter and sphericity is respectively represented, candidate liver neoplasm
Characteristic value is greater than the upper threshold, then determines that the candidate liver neoplasm is candidate for false positive, and eliminated;
Candidate liver neoplasm categorization module according to the characteristic information of candidate liver neoplasm, setting false positive reduce rule R1,
The threshold value of R2, R3, R1 upper threshold are 682.7cm3, R1 bottom threshold is 33.5cm3, R2 upper threshold is 8cm, under R2 threshold value
It is limited to 2cm, R3 upper threshold is that 1.0, R3 bottom threshold is 0.45.
The feature of step 4, candidate liver neoplasm categorization module selection characterization liver neoplasm, progress feature calculation, and according to
Liver neoplasm feature uses multinomial to classify as the kernel function of svm classifier algorithm to liver neoplasm;
Step 4-1, the feature of 8 common characterization liver neoplasms is chosen as training sample;Characterize the institute of liver neoplasm
There is feature to be all based on form, shape, grayscale information to be calculated, the feature of 8 common characterization liver neoplasms includes 5
Gray feature: the maximum value of pixel gray value in candidate liver neoplasm region, pixel gray level in candidate liver neoplasm region
The average value of pixel gray value in the minimum value of value, candidate liver neoplasm region, pixel ash in candidate liver neoplasm region
The difference of pixel gray value and tumour gray average, further includes 3 forms in the standard deviation of angle value, candidate liver neoplasm region
Feature: circularity, area, the shortest distance;
Step 4-2, cluster operation is carried out to training sample according to k means clustering algorithm, is assisted with the size of class cluster
Estimation, the liver neoplasm feature after being trained;
Step 4-3, svm classifier algorithm is chosen, and using multinomial as the kernel function of svm classifier algorithm, according to svm classifier
Constraint condition of algorithm etc. carries out training sample to operate determining polynomial parameters, adjustment parameter, and then utilizes susceptibility and spy
Anisotropic formula is estimated to obtain susceptibility and specificity;Liver neoplasm feature after training that step 4-2 is calculated is input to
Svm classifier algorithm, according to the maximized characteristic of the distance between the two of the hyperplane of the parameter of Polynomial kernel function classifications, with
And polynomial parameters, adjustment parameter, susceptibility and the specificity of SVM classify to candidate liver neoplasm.
It should be appreciated that each section of the invention can be realized with hardware, software, firmware or their combination.
In the description of invention, it is to be understood that term " longitudinal direction ", " transverse direction ", "upper", "lower", "front", "rear",
The orientation or positional relationship of the instructions such as "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outside" is based on attached drawing institute
The orientation or positional relationship shown is merely for convenience of the description present invention, rather than the device or element of indication or suggestion meaning must
There must be specific orientation, be constructed and operated in a specific orientation, therefore be not considered as limiting the invention.
Embodiment described above is only that preferred embodiment of the invention is described, and is not carried out to the scope of the present invention
It limits, without departing from the spirit of the design of the present invention, those of ordinary skill in the art make technical solution of the present invention
Various changes and improvements, should all fall into claims of the present invention determine protection scope in.
Claims (6)
1. a kind of Hepatic CT staging system, it is characterised in that: including suspicious liver neoplasm detection module: for being partitioned into
Liver parenchyma detected, obtain the seed point of candidate liver neoplasm;Candidate liver neoplasm divides module: for all
Candidate liver neoplasm seed point carries out region growth, and dividing candidate liver neoplasm obtains feature completely candidate liver neoplasm;It waits
Select liver neoplasm categorization module: for completely candidate liver neoplasm to carry out feature extraction, classification by the feature being partitioned into, into one
It walks and determines liver neoplasm region.
2. a kind of Hepatic CT staging method, it is characterised in that: the following steps are included:
Step 1, suspicious liver neoplasm detection module are partitioned into liver parenchyma, and it is swollen to carry out suspicious liver to the liver parenchyma being partitioned into
Tumor target detection obtains the seed point of candidate liver neoplasm;
Step 1-1, it is corrected based on graphical spectrum technology, in conjunction with manual segmentation, coarse segmentation goes out liver parenchyma;
Step 1-2, according to the morphological feature of liver neoplasm, using the detection method based on variable annular filter, to being partitioned into
Liver parenchyma detected, so that the size of variable annular filter and the size of candidate liver neoplasm is matched;
Step 1-3, variable annular filter is improved using the method for gray scale weight distance conversion according to the gray feature of liver neoplasm
The difference effect of wave device, obtains the center of variable annular filter, i.e., candidate liver neoplasm seed point;
Step 2, for the seed point of candidate's liver neoplasm obtained in step 1, candidate liver neoplasm segmentation module passes through seed
The 3D region growing algorithm of point conformity classification, seed point, is split candidate liver neoplasm, it is completely candidate to obtain feature
Liver neoplasm;
Step 3, candidate liver neoplasm categorization module are completely candidate to the feature being partitioned into step 2 using empirical value method
Liver neoplasm, which carries out coarse false positive, to be reduced, the candidate liver neoplasm after being screened;
The feature of step 4, candidate liver neoplasm categorization module selection characterization liver neoplasm, carries out feature calculation, and according to liver
Tumoral character estimates polynomial parameters, adjustment parameter, susceptibility and specificity based on SVM algorithm, use multinomial as
The kernel function of SVM algorithm classifies to liver neoplasm.
3. a kind of Hepatic CT staging method according to claim 2, it is characterised in that: step 3 the following steps are included:
Step 3-1, completely candidate liver neoplasm using empirical value will be in intermediate threshold to the feature being partitioned into for step 2
The normal liver tissue being worth in range is rejected, and the threshold region where tumour, blood vessel is retained;
Step 3-2, to the candidate liver neoplasm for rejecting normal liver tissue, rule R1, R2, R3 is reduced by false positive and carry out vacation
The positive is eliminated, and R1, R2, R3 are based respectively on volume characteristic, radial features and the sphericity feature of candidate liver neoplasm;
WithRespectively represent the bottom threshold of volume, diameter and sphericity, the characteristic value of candidate liver neoplasm
Less than the bottom threshold, then determine that the candidate liver neoplasm is candidate for false positive, and eliminated;
WithRespectively represent the upper threshold of volume, diameter and sphericity, the feature of candidate liver neoplasm
Value is greater than the upper threshold, then determines that the candidate liver neoplasm is candidate for false positive, and eliminated.
4. a kind of Hepatic CT staging method according to claim 3, it is characterised in that: candidate liver is swollen in step 3
Characteristic information of the tumor categorization module according to candidate liver neoplasm, the threshold value of setting false positive reduction rule R1, R2, R3, R1 threshold value
The upper limit is 682.7cm3, R1 bottom threshold is 33.5cm3, R2 upper threshold is 8cm, and R2 bottom threshold is 2cm, R3 upper threshold
It is 0.45 for 1.0, R3 bottom threshold.
5. a kind of Hepatic CT staging method according to claim 2, it is characterised in that: choose 8 characterizations in step 4
The feature of liver neoplasm carries out feature selecting and feature calculation, is classified by SVM algorithm to candidate liver neoplasm.
6. a kind of Hepatic CT staging method according to claim 2, it is characterised in that: poly- according to k mean value in step 4
Class algorithm carries out cluster operation to training sample, carries out auxiliary estimation with the size of class cluster.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910563651.4A CN110276407A (en) | 2019-06-26 | 2019-06-26 | A kind of Hepatic CT staging system and classification method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910563651.4A CN110276407A (en) | 2019-06-26 | 2019-06-26 | A kind of Hepatic CT staging system and classification method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110276407A true CN110276407A (en) | 2019-09-24 |
Family
ID=67963465
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910563651.4A Pending CN110276407A (en) | 2019-06-26 | 2019-06-26 | A kind of Hepatic CT staging system and classification method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110276407A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110910371A (en) * | 2019-11-22 | 2020-03-24 | 北京理工大学 | Liver tumor automatic classification method and device based on physiological indexes and image fusion |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170148156A1 (en) * | 2015-11-25 | 2017-05-25 | Zebra Medical Vision Ltd. | Systems and methods for detecting a fatty liver from a computed tomography (ct) scan |
CN108596887A (en) * | 2018-04-17 | 2018-09-28 | 湖南科技大学 | A kind of abdominal CT sequence image liver neoplasm automatic division method |
CN109035227A (en) * | 2018-07-13 | 2018-12-18 | 哈尔滨理工大学 | The system that lung tumors detection and diagnosis is carried out to CT image |
-
2019
- 2019-06-26 CN CN201910563651.4A patent/CN110276407A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170148156A1 (en) * | 2015-11-25 | 2017-05-25 | Zebra Medical Vision Ltd. | Systems and methods for detecting a fatty liver from a computed tomography (ct) scan |
CN108596887A (en) * | 2018-04-17 | 2018-09-28 | 湖南科技大学 | A kind of abdominal CT sequence image liver neoplasm automatic division method |
CN109035227A (en) * | 2018-07-13 | 2018-12-18 | 哈尔滨理工大学 | The system that lung tumors detection and diagnosis is carried out to CT image |
Non-Patent Citations (1)
Title |
---|
王进科: "CT图像的肝脏及肺部肿瘤检测方法研究", 《中国博士学位论文全文数据库医药卫生科技辑(月刊)》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110910371A (en) * | 2019-11-22 | 2020-03-24 | 北京理工大学 | Liver tumor automatic classification method and device based on physiological indexes and image fusion |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Halder et al. | Lung nodule detection from feature engineering to deep learning in thoracic CT images: a comprehensive review | |
Zhang et al. | A deep learning framework for pancreas segmentation with multi-atlas registration and 3D level-set | |
Sheng et al. | Retinal vessel segmentation using minimum spanning superpixel tree detector | |
JP2020508143A (en) | Prostate Cancer Detection in Multiparametric MRI Using Random Forest with Instance Weighting and MR Prostate Segmentation by Deep Learning Using Holistic Nested Networks | |
Huang et al. | Automatic segmentation of breast lesions for interaction in ultrasonic computer-aided diagnosis | |
Huang et al. | Liver tumor detection and segmentation using kernel-based extreme learning machine | |
US8073226B2 (en) | Automatic detection and monitoring of nodules and shaped targets in image data | |
ES2711196T3 (en) | Systems and procedures for the segmentation and processing of tissue images and extraction of tissue characteristics to treat, diagnose or predict medical conditions | |
Lei et al. | Brain tumor segmentation in MR images using a sparse constrained level set algorithm | |
Singadkar et al. | Automatic lung segmentation for the inclusion of juxtapleural nodules and pulmonary vessels using curvature based border correction | |
Zabir et al. | Automatic brain tumor detection and segmentation from multi-modal MRI images based on region growing and level set evolution | |
WO2011008906A1 (en) | Computer-aided detection (cad) of intracranial aneurysms | |
Jony et al. | Detection of lung cancer from CT scan images using GLCM and SVM | |
CN109035227A (en) | The system that lung tumors detection and diagnosis is carried out to CT image | |
Hao et al. | Juxta-vascular pulmonary nodule segmentation in PET-CT imaging based on an LBF active contour model with information entropy and joint vector | |
Yan et al. | Automated breast cancer detection in mammography using ensemble classifier and feature weighting algorithms | |
Bai et al. | Cell segmentation based on FOPSO combined with shape information improved intuitionistic FCM | |
Li et al. | Detection of pulmonary nodules in CT images based on fuzzy integrated active contour model and hybrid parametric mixture model | |
Ananth et al. | Enhancing segmentation approaches from Gaussian mixture model and expected maximization to super pixel division algorithm | |
Sengodan et al. | Early detection and classification of malignant lung nodules from CT images: an optimal ensemble learning | |
CN110276407A (en) | A kind of Hepatic CT staging system and classification method | |
Aarthy et al. | Detection of lung nodule using multiscale wavelets and support vector machine | |
Fiori et al. | Automatic colon polyp flagging via geometric and texture features | |
Ma et al. | Automatic nasopharyngeal carcinoma segmentation in MR images with convolutional neural networks | |
Zhu et al. | A novel lung cancer detection algorithm for CADs based on SSP and Level Set |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190924 |