CN106600584A - Tsallis entropy selection-based suspected pulmonary nodule detection method - Google Patents

Tsallis entropy selection-based suspected pulmonary nodule detection method Download PDF

Info

Publication number
CN106600584A
CN106600584A CN201611115740.5A CN201611115740A CN106600584A CN 106600584 A CN106600584 A CN 106600584A CN 201611115740 A CN201611115740 A CN 201611115740A CN 106600584 A CN106600584 A CN 106600584A
Authority
CN
China
Prior art keywords
tsallis
tuberosity
entropys
feature
doubtful
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
Application number
CN201611115740.5A
Other languages
Chinese (zh)
Inventor
蓝天
丁熠
陈实
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Electronic Science and Technology of China
Original Assignee
University of Electronic Science and Technology of China
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by University of Electronic Science and Technology of China filed Critical University of Electronic Science and Technology of China
Priority to CN201611115740.5A priority Critical patent/CN106600584A/en
Publication of CN106600584A publication Critical patent/CN106600584A/en
Pending legal-status Critical Current

Links

Classifications

    • 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
    • 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
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
    • 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
    • G06T2207/30061—Lung
    • G06T2207/30064—Lung nodule
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/03—Recognition of patterns in medical or anatomical images
    • G06V2201/031—Recognition of patterns in medical or anatomical images of internal organs

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Multimedia (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

The invention relates to a Tsallis entropy selection-based suspected pulmonary nodule detection method. According to the method, suspected pulmonary nodules are screened out from segmented candidate nodules through the calculation of the Tsallis entropy, and a multi-scale feature extraction mask can be constructed; feature extraction is performed, and extraction features are introduced into a classifier, and a result can be finally obtained. Since the Tsallis entropy is introduced, the Tsallis entropy values of the candidate nodules are compared with a given empirical threshold value T, and therefore, the detection method is a novel method for screening the candidate nodules. With the method adopted, the false positive rate of nodule detection of a whole system can be decreased; and doctors can be assisted to conveniently and quickly diagnose pulmonary nodules.

Description

A kind of doubtful pulmonary nodule detection method selected based on Tsallis entropys
Technical field
The invention belongs to computer-aided diagnosises field, more specifically, it is related to a kind of select based on Tsallis entropys Doubtful pulmonary nodule detection method.
Background technology
Chinese annual cancer new cases are 3,120,000, every year because cancer mortality is more than 2,000,000, wherein dead most Cancer kind be pulmonary carcinoma.The cure rate of pulmonary carcinoma is closely related with clinical stagess during diagnosis, 5 years survival rates of the patients with lung cancer of early stage Survival rate for more than 90%, I phase patients with lung cancer is reduced to 60%, and the year survival rate of the patients with lung cancer of II to IV phase is dropped from 40% To 5%.Therefore, " early to find, early to diagnose, early treatment " is the key for improving patients with lung cancer survival rate.But it is not promising that Only 15% pulmonary carcinoma is discovered in an early phase.But in the ill symptomses that appearance cough and hemoptysis etc. are not healed for a long time, then go to cure Institute checked, is often advanced lung cancer with regard to own Jing.Therefore, how to accomplish that " early to find " and " early to diagnose " is the weight for needing research Want problem.
The CADe systems of detection Lung neoplasm are constituted generally by five subsystems：Collection, pretreatment, segmentation, nodule detection With reduction false positive.Collection mainly collection medical image.Main preconditioning technique has：Medium filtering, strengthens filtering, contrast The limited adaptive histogram equalization of degree, strengthens, Wiener filtering, fast Fourier transform, wavelet transformation automatically, and anti-geometry expands Dissipate, erosion filter, the method such as smothing filtering and noise amendment.
In recent decades, although various for the related algorithm species of image segmentation field, emerge in an endless stream, still cannot Fully meet the actual demand of people.Its reason is considerably complicated, including：People institute cannot be briefly described completely with mathematical model The practical problem for facing；Cutting object structural property varies；Image degradation and people are to segmentation result target It is different etc..These reasons are determined can not possibly realize a kind of pervasive, general dividing method, can only for particular problem and Specific demand gives reasonable selection, makes equilibrium or stress on the critical index such as precision, speed and robustness.Segmentation Two main methods of lung images are：The segmentation of segmentation and deformable model based on threshold value.In the segmentation based on threshold value On, the threshold value of a brightness implements lock out operation.Segmentation pulmonary's picture using major type of deformable model have：Actively Profile and the deformable model based on level set.The technology of other segmentation Lung neoplasms has：Cylindrical and spherical filter, form Student movement is calculated, thresholding, and multiple gray level thresholding and connection element is analyzed.The main feature for extracting has：Gray feature, form are special Levy, textural characteristics, contextual feature, surface.Main grader has：Linear discriminant analysiss, rule-based, cluster, horse Er Kefu random fields, artificial neural network (ANN), support vector machine (SVM), dual threshold cutting.
The content of the invention
It is an object of the invention to design a kind of method that utilization Tsallis entropys calculate to screen doubtful Lung neoplasm, dividing Candidate nodule is preselected before class, representative feature is put into into grader then, to improve arithmetic speed, subtracted Few false positive rate.
For achieving the above object, a kind of doubtful pulmonary nodule detection method selected based on Tsallis entropys of the present invention.It is main to wrap Include herein below：The theory of Tsallis entropys is introduced in the tuberosity mask of feature extraction, lung knot is distinguished by Tsallis entropy The tissue such as section and the blood vessel vertical with section, bronchus, effectively reduces the quantity of ROI, first carried out before detecting to Lung neoplasm The detection of doubtful Lung neoplasm, can effectively reduce the false positive of experiment, improve arithmetic speed.Know-why is as shown in figure 1, concrete skill Art flow process is as follows：
Step one：The acquisition of picture, it is intended to using public database LIDC, and pre- place is carried out to picture using medium filtering Reason；
Step 2：Primary segmentation is carried out to original image first by threshold value, secondly using the element of disc structure to thick The pulmonary of segmentation carries out Two-dimensional morphology opening operation, then the lung areas in image is carried out using 3D region growth algorithm Segmentation, it is exactly lung areas to split the part for obtaining, and is then used by Hole filling algorithms to repair hole, has finally used three-dimensional Closing operation of mathematical morphology comes the edge of refine pulmonary；
Step 3：Row threshold division is entered to original image with a series of threshold values first, then the result and lung of segmentation Portion's mask carries out logical AND operation, then carries out Two-dimensional morphology to the tuberosity split with a series of disc structure elements again and opens Computing, finally merges all of intermediate nodule mask and forms final nodule mask；
Step 4：Doubtful tuberosity is carried out.Selected threshold T, calculate tuberosity Tsallis entropy, when Tsallis entropy it is big It is obvious false positive when T, weeds out, the tuberosity that will be left behind is used as candidate nodule.
Step 5, feature extraction, it is intended to using traditional some features, and based on the three-dimensional block of different scale To extract.Feature mainly has three major types：Geometry, brightness, Gradient Features, all these feature contain two dimension and three Dimension.Finally again with SFS come preferably a series of feature；
Step 6：Tuberosity is classified with SVM classifier.
Description of the drawings
Fig. 1 is a kind of theory diagram of the doubtful pulmonary nodule detection method selected based on Tsallis entropys of the present invention.
Fig. 2 is a kind of technical scheme figure of the doubtful pulmonary nodule detection method selected based on Tsallis entropys of the present invention.
Specific implementation
Below in conjunction with the accompanying drawings the specific embodiment of the present invention is described, so as to those skilled in the art preferably Understand the present invention.Requiring particular attention is that, the known work(of main contents of the present invention in the following description, may be desalinated The detailed description that can and design will be ignored.
In the present embodiment, a kind of doubtful pulmonary nodule detection method selected based on Tsallis entropys of the present invention, main to wrap Include following link：1. picture obtain, 2. lung segmentation, 3. nodule segmentation, 4.Tsallis entropys select, 5. Multi resolution feature extraction, 6. pair doubtful tuberosity is classified.
The acquisition of picture, it is intended to using public database LIDC, and the pre- place of correlation is carried out using medium filtering to image Reason so as to can directly be processed by matlab.
The segmentation of pulmonary, the main content related to morphology using threshold method are split.First by two threshold values Primary segmentation is carried out to original image respectively, Two-dimensional morphology is carried out to the pulmonary of coarse segmentation using the element of disc structure secondly Then lung areas are split by opening operation using 3D region growth algorithm, and it is exactly lung areas to split the part for obtaining. Subsequently using Hole filling algorithms repairing these holes, three dimensional morphology closed operation has finally been used to come the side of refine pulmonary Edge.
The segmentation of Lung neoplasm, is split using morphologic method, first with a series of threshold values to original image Enter row threshold division, the result and pulmonary's mask of segmentation are carried out logical AND operation then, a series of disc structures are then used again Element opening operation is carried out to the tuberosity split, finally merge all of intermediate nodule mask and reconstruct nodule mask.
Selected threshold T, calculates the Tsallis entropy of candidate nodule, is significantly false sun when Tsallis entropy is more than T Property, weed out, tuberosity of the entropy more than T is remained as doubtful tuberosity.
Feature extraction, it is intended to which, using traditional some features, feature mainly has three major types：Geometry, brightness, Gradient Features.First to the nodular feature bounding box split, then carry out expanding 3 pixels on the basis of bounding box, make Not precisely the tuberosity of segmentation can major part include.As the accurate segmentation of tuberosity has certain limitations, so not dividing Feature extraction is carried out on the tuberosity mask for cutting, and feature extraction is carried out in the three-dimensional bits for expanding, enable tuberosity most of It is included.All these feature contain it is two dimension and three-dimensional, so that omnibearing tuberosity can be represented.It is last to use spy again Levy and select the algorithm of excellent namely dimensionality reduction to carry out preferably a series of feature, intend using advance algorithm SFS come preferred feature.
Grader is classified, and the feature that above-mentioned link is extracted is put into carries out two to all of tuberosity Classification, so as to obtain last result.
A kind of doubtful Lung neoplasm feature extracting method selected based on Tsallis entropys of the present invention is had the characteristics that：
The present invention proposes a kind of method of new screening candidate nodule, can preferably reduce the nodule detection of whole system False positive rate, new method can rapidly aid in diagnosis of the doctor to Lung neoplasm with convenient.
Although being described to illustrative specific embodiment of the invention above, in order to the technology of the art Personnel understand the present invention, it should be apparent that the invention is not restricted to the scope of specific embodiment, the common skill to the art For art personnel, as long as various change is in appended claim restriction and the spirit and scope of the present invention for determining, these Change is it will be apparent that all utilize the innovation and creation of present inventive concept in the row of protection.

Claims (2)

1. the present invention it is a kind of based on Tsallis entropys select doubtful pulmonary nodule detection method, mainly including herein below：In feature The theory of Tsallis entropys is introduced in the tuberosity mask of extraction, Lung neoplasm and the blood vertical with section are distinguished by Tsallis entropy Pipe, bronchus etc. are organized, and effectively reduce the quantity of ROI, the detection of doubtful Lung neoplasm, energy are first carried out before detecting to Lung neoplasm The false positive of experiment is effectively reduced, arithmetic speed is improved.Technical scheme is as follows：
Step one：The acquisition of picture, it is intended to using public database LIDC, and pretreatment is carried out to picture using medium filtering；
Step 2：Primary segmentation is carried out to original image first by threshold value, secondly using the element of disc structure to coarse segmentation Pulmonary carry out Two-dimensional morphology opening operation, then the lung areas in image are carried out point using 3D region growth algorithm Cut, it is exactly lung areas to split the part for obtaining, and is then used by Hole filling algorithms to repair hole, has finally used three-dimensional shaped State closed operation comes the edge of refine pulmonary；
Step 3：Row threshold division is entered to original image with a series of threshold values first, then the result and pulmonary split are covered Mould carries out logical AND operation, then carries out Two-dimensional morphology to the tuberosity split with a series of disc structure elements again and opens fortune Calculate, finally merge all of intermediate nodule mask and form final nodule mask；
Step 4：Selected threshold T, calculates the Tsallis entropy of candidate nodule, is significantly vacation when Tsallis entropy is more than T The positive, weeds out, and the tuberosity that will be left behind is used as doubtful tuberosity.
Step 5, feature extraction, it is intended to using traditional some features, and based on the three-dimensional block of different scale carrying Take.Feature mainly has three major types：Geometry, gray scale, the feature of gradient, all these feature all contain two dimension and three Dimension.Finally again with SFS come preferably a series of feature；
Step 6：Tuberosity is classified with SVM classifier.
2. a kind of doubtful pulmonary nodule detection method selected based on Tsallis entropys according to claim 1, its feature is The theory of Tsallis is introduced before tuberosity feature extraction, Lung neoplasm and the blood vertical with section are distinguished by Tsallis entropy Pipe, bronchus etc. are organized, so as to screen candidate nodule.
In the present invention, main characteristic is the introducing Tsallis entropy theories in feature extraction, due to Lung neoplasm with blood vessel in size With all very much like in shape, both no larger differences on gray average, and on Tsallis entropys, but have obvious Difference, so as to, before last the classification, first carry out primary election to doubtful tuberosity.Mainly include：(1) to the tuberosity split, lead to Cross Tsallis entropy to distinguish blood vessel and Lung neoplasm, so as to filter out candidate nodule, and then the multiple dimensioned feature extraction of construction is covered Film.(2) it is last, doubtful tuberosity is classified again with svm graders, so as to draw experimental result.
Tuberosity first to having split carries out the calculating of Tsallis entropy respectively.Tsallis entropys are the broad sense shapes of Shannon entropys Formula, Tsallis entropys are based on the mechanism by the theoretical Nonextensives for producing of Boltzmann-Gibbs.Select in image threshold In, there is Nonadditivity information in many images, therefore using the Shannon entropys with additivity as interpretational criteria, it is impossible to To preferable segmentation effect.Its formula is as follows：
S q ( p i ) = 1 q - 1 ( 1 - Σ i = 1 N ( p i ) q ) ( q ≠ 1 )
In formula：P={ p1,p2..., pNIt is arbitrary probability distribution, and meet pi≥0,Q is nonextensitivity coefficient, For describing the nonextensitivity of Tsallis entropys, i.e., to two subsystems A that can be decomposed into independent statistics and the system of B For, its Tsallis entropy can be expressed as：
Sq(A+B)=Sq(A)+Sq(B)+(1-q)Sq(A)Sq(B)
Tsallis entropys further contemplate the mutual relation between two subsystems, in image threshold is selected, in many images There is Nonadditivity information, thus using the Shannon entropys with additivity as interpretational criteria, it is impossible to access preferably sieve Select effect.
The respective Tsallis entropys of the candidate nodule for obtaining are compared with given empirical value T, is only retained Tsallis entropys and is less than The tuberosity of threshold value T carries out feature extraction as doubtful tuberosity, the then doubtful tuberosity to filtering out.
The feature that above-mentioned link is extracted is put into two classification are carried out to all of tuberosity, finally draw reality Test result.
CN201611115740.5A 2016-12-07 2016-12-07 Tsallis entropy selection-based suspected pulmonary nodule detection method Pending CN106600584A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201611115740.5A CN106600584A (en) 2016-12-07 2016-12-07 Tsallis entropy selection-based suspected pulmonary nodule detection method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201611115740.5A CN106600584A (en) 2016-12-07 2016-12-07 Tsallis entropy selection-based suspected pulmonary nodule detection method

Publications (1)

Publication Number Publication Date
CN106600584A true CN106600584A (en) 2017-04-26

Family

ID=58595288

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201611115740.5A Pending CN106600584A (en) 2016-12-07 2016-12-07 Tsallis entropy selection-based suspected pulmonary nodule detection method

Country Status (1)

Country Link
CN (1) CN106600584A (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107274402A (en) * 2017-06-27 2017-10-20 北京深睿博联科技有限责任公司 A kind of Lung neoplasm automatic testing method and system based on chest CT image
CN108154512A (en) * 2017-11-08 2018-06-12 东北大学 It is a kind of based on the multiple retinal images blood vessel segmentation system for going trend analysis
CN108470337A (en) * 2018-04-02 2018-08-31 江门市中心医院 A kind of sub- reality Lung neoplasm quantitative analysis method and system based on picture depth feature
CN108537784A (en) * 2018-03-30 2018-09-14 四川元匠科技有限公司 A kind of CT figure pulmonary nodule detection methods based on deep learning
CN110333078A (en) * 2019-08-21 2019-10-15 佛山科学技术学院 A kind of rolling bearing degenerate state stage determines method
CN111402270A (en) * 2020-03-17 2020-07-10 北京青燕祥云科技有限公司 Repeatable intra-pulmonary grinding glass and method for segmenting hypo-solid nodules
CN111401214A (en) * 2020-03-12 2020-07-10 四川大学华西医院 Multi-resolution integrated HER2 interpretation method based on deep learning
CN116452898A (en) * 2023-06-16 2023-07-18 中国人民大学 Lung adenocarcinoma subtype identification method and device based on image histology and deep learning

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107274402A (en) * 2017-06-27 2017-10-20 北京深睿博联科技有限责任公司 A kind of Lung neoplasm automatic testing method and system based on chest CT image
CN108154512A (en) * 2017-11-08 2018-06-12 东北大学 It is a kind of based on the multiple retinal images blood vessel segmentation system for going trend analysis
CN108537784A (en) * 2018-03-30 2018-09-14 四川元匠科技有限公司 A kind of CT figure pulmonary nodule detection methods based on deep learning
CN108537784B (en) * 2018-03-30 2021-08-24 四川元匠科技有限公司 CT image pulmonary nodule detection method based on deep learning
CN108470337A (en) * 2018-04-02 2018-08-31 江门市中心医院 A kind of sub- reality Lung neoplasm quantitative analysis method and system based on picture depth feature
CN110333078A (en) * 2019-08-21 2019-10-15 佛山科学技术学院 A kind of rolling bearing degenerate state stage determines method
CN110333078B (en) * 2019-08-21 2021-04-20 佛山科学技术学院 Rolling bearing degradation state stage determination method
CN111401214A (en) * 2020-03-12 2020-07-10 四川大学华西医院 Multi-resolution integrated HER2 interpretation method based on deep learning
CN111401214B (en) * 2020-03-12 2023-04-18 四川大学华西医院 Multi-resolution integrated HER2 interpretation method based on deep learning
CN111402270A (en) * 2020-03-17 2020-07-10 北京青燕祥云科技有限公司 Repeatable intra-pulmonary grinding glass and method for segmenting hypo-solid nodules
CN116452898A (en) * 2023-06-16 2023-07-18 中国人民大学 Lung adenocarcinoma subtype identification method and device based on image histology and deep learning
CN116452898B (en) * 2023-06-16 2023-10-17 中国人民大学 Lung adenocarcinoma subtype identification method and device based on image histology and deep learning

Similar Documents

Publication Publication Date Title
CN106600584A (en) Tsallis entropy selection-based suspected pulmonary nodule detection method
Alam et al. Multi-stage lung cancer detection and prediction using multi-class svm classifie
Wang et al. A multi-view deep convolutional neural networks for lung nodule segmentation
Vas et al. Lung cancer detection system using lung CT image processing
Beura et al. Mammogram classification using two dimensional discrete wavelet transform and gray-level co-occurrence matrix for detection of breast cancer
CN108986067A (en) Pulmonary nodule detection method based on cross-module state
CN104268873B (en) Breast tumor partition method based on nuclear magnetic resonance images
Gao et al. On combining morphological component analysis and concentric morphology model for mammographic mass detection
Han et al. Hybrid resampling and multi-feature fusion for automatic recognition of cavity imaging sign in lung CT
Palma et al. Detection of masses and architectural distortions in digital breast tomosynthesis images using fuzzy and a contrario approaches
CN109447998A (en) Based on the automatic division method under PCANet deep learning model
CN105740875A (en) Pulmonary nodule multi-round classification method based on multi-scale three-dimensional block feature extraction
CN104182755A (en) Mammary gland molybdenum target X-ray image block feature extraction method based on tower-shaped principal component analysis (PCA)
Bu et al. Lung nodule detection based on YOLOv3 deep learning with limited datasets
Pezeshki et al. Mass classification of mammograms using fractal dimensions and statistical features
De Nunzio et al. Approaches to juxta-pleural nodule detection in CT images within the MAGIC-5 Collaboration
Li et al. Breast cancer early diagnosis based on hybrid strategy
Ahmmed et al. Fuzzy logic based algorithm to classify tumor categories with position from brain MRI images
Rajendran et al. Novel fuzzy association rule image mining algorithm for medical decision support system
CN114373096A (en) Pulmonary nodule benign and malignant prediction system and method based on multi-feature fusion
Jia et al. Automated detection of pulmonary nodules in HRCT images
Nithila et al. Lung cancer diagnosis from CT images using CAD system: a review
Dolejší et al. Automatic two-step detection of pulmonary nodules
Jyothula et al. Lung Cancer Detection in CT Scans Employing Image Processing Techniques and Classification by Decision Tree (DT) and K-Nearest Neighbor (KNN)
Liu et al. Automatic Lung Parenchyma Segmentation of CT Images Based on Matrix Grey Incidence.

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
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20170426