CN108985345A - A kind of detection device based on the classification of lung's Medical image fusion - Google Patents
A kind of detection device based on the classification of lung's Medical image fusion Download PDFInfo
- Publication number
- CN108985345A CN108985345A CN201810662571.XA CN201810662571A CN108985345A CN 108985345 A CN108985345 A CN 108985345A CN 201810662571 A CN201810662571 A CN 201810662571A CN 108985345 A CN108985345 A CN 108985345A
- Authority
- CN
- China
- Prior art keywords
- image
- lung
- classification
- pixel
- fusion
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
- G06F18/2148—Generating training patterns; Bootstrap methods, e.g. bagging or boosting characterised by the process organisation or structure, e.g. boosting cascade
-
- 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
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Medical Informatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Probability & Statistics with Applications (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
A kind of detection device based on the classification of lung's Medical image fusion is claimed in the present invention, comprising: acquisition unit obtains the CT lung scans image of the patient；Pretreatment portion is pre-processed to obtain binaryzation gray level image；Morphological erosion and bulge carry out the image after morphological erosion operation is corroded, and carry out the image after dilation operation is expanded to the binaryzation gray level image；Operational part carries out opening operation to the image after the corrosion and obtains opening operation figure, carries out closed operation to the image after the expansion and obtains closed operation figure；Fourier transformation portion obtains Fourier transformation value；Lung's Medical image fusion categorization module, for carrying out integrated classification differentiation to CT lung scans image；Lesion test section is judged to obtain the suspected abnormality region of the patient for distinguishing result to the Fourier transformation value combination row integrated classification.The identifying and diagnosing accuracy that the present invention can reduce complexity, improve pulmonary lesions.
Description
Technical field
The invention belongs to medical image processing technology field, in particular to it is a kind of based on lung's Medical image fusion classification
Detection device.
Background technique
Existing lung's Medical Image Fusion is mainly to the medical image of two kinds of different modalities, according to medical image mode
Difference, Medical image fusion system can be divided into three types: anatomic medicine image and anatomic medicine image co-registration, dissection are cured
Learn image and functional medicine image co-registration and functional medicine image and functional medicine image co-registration.MRI-PET and MRI-SPECT
Medical image fusion system belongs to anatomic medicine image and functional medicine image co-registration, and the input picture of the system is gray scale and puppet
It is colored.The MRI-PET combined integratedization external member that PHILIPS Co. releases is by commercial MRI image scanning instrument and has special shielding
PET combine, obtained image is to the diagnosis of cancer metastasis and preoperative has important clinical value by stages.
Since Pixel-level Multiscale Fusion method is directly handled the pixel value of image in different scale images, energy
Retain the Pixel Information of input picture to the maximum extent and improves the quality of blending image.Therefore, Pixel-level multi-scale method exists
Medical image fusion field becomes a research hotspot.Currently, mainly melting both at home and abroad from picture breakdown and reconstructing method and image
Normally two aspects of method are set out, and propose new Pixel-level Multiscale Fusion method.In terms of picture breakdown and reconstructing method,
Different scale is carried out to picture signal using Fourier transformation and inverse Fourier transform based on the image interfusion method of frequency domain to divide
Solution and reconstruct, but such method has the characteristics that time complexity height and long operational time, this is auxiliary with the medical treatment of high real-time
Diagnosis contradiction is helped, and the experiment porch hardware facility and software facility of processing are required very high.Later, researchers proposed
Image is handled using spatial filter to carry out multi-resolution decomposition and reconstruct to input picture, although such method can be fast
Picture breakdown and reconstruct are carried out fastly, but the image interfusion method noise immunity based on airspace is poor.But existing method is not for
Medical image with mode is merged using same feature.
Summary of the invention
Present invention seek to address that the above problem of the prior art.It proposes a kind of reduction complexity, improve blending image
Quality, the detection device based on the classification of lung's Medical image fusion for improving classification accuracy.
Technical scheme is as follows:
A kind of detection device based on the classification of lung's Medical image fusion, for handling the CT lung scans image of patient
Obtain the suspected abnormality region of the patient comprising: acquisition unit obtains the CT lung scans image of the patient；Pretreatment
Portion is pre-processed to obtain binaryzation gray level image to the CT lung scans image after acquisition；Morphological erosion and expansion
Portion carries out the image after morphological erosion operation is corroded to the binaryzation gray level image, to the binaryzation grayscale image
As carrying out the image after dilation operation is expanded；Operational part carries out opening operation to the image after the corrosion and obtains opening operation
Figure carries out closed operation to the image after the expansion and obtains closed operation figure；Fourier transformation portion, to the opening operation figure and described
Closed operation figure carries out Fourier transformation and obtains Fourier transformation value；Lung's Medical image fusion categorization module, for lung CT
Scan image carries out integrated classification differentiation, and differentiation result is transferred to lesion test section and is judged；Lesion test section, is used for
Result is distinguished to the Fourier transformation value combination row integrated classification to be judged to obtain the suspected abnormality region of the patient.Institute
Lung's Medical image fusion categorization module is stated to comprise the following modules:
Data preprocessing module: obtaining the CT lung scans image of the patient, carries out including that the removal lung CT is swept
Pre-treatment step including baseline and high-frequency noise of the tracing as in, and it is classified as CT lung scans gray scale anatomic image and puppet
Color function CT lung scans image, is respectively labeled asWith
Multi-resolution decomposition module: during multi-resolution decomposition, by change propagate filter in Gaussian convolution core it is big
It is small, to the two width source images to step 1WithIt is smoothed, using smoothing processing to signalWithCarry out multiple dimensioned point
Solution obtains smoothed image and detail pictures；Fusion is weighted using comentropy for smoothed image S to obtain merging smooth figure
Picture, and detail pictures are merged to obtain fusion detail pictures F using multiple featuresS；Propagate the prototype of filter are as follows: Sp=
1/Zp∑t∈N(p)wp,t×It, wherein SpFor smoothed out image, ZpFor normalization factor, N (p) is using p as center pixel
Neighborhood, i indicate pixel sequence number, ItFor in the image pixel value of coordinate position t, wp,tTo propagate filter weights；Wherein, the first item of filter weights is propagated
It indicates to utilize the pixel the distance between adjacent with pixel p and pixel tWith Gaussian convolution core σaGaussian function；It passes
Broadcast the Section 2 of filter weightsIt indicates to roll up using all pixels distance and Gauss adjacent with pixel p
Product core σrGaussian function；
Reconstruct Fusion Module: using the simple operation in Pixel-level to fused smoothed image FDWith detail pictures FSInto
Row reconstructs to obtain blending image F based on ADMM；Export final blending image F；
Categorization module: and classified using improved Least risk Bayes Multiple Classifier Fusion image F, wherein
AdaBoost method is as follows:
Network data is inputted with a matrix type, initializes weight i=1,2 ... ..., n, is executed and is recycled m=1,2 ... ...,
M, by ωiValue substitute into AdaBoost frame in, be trained by Least risk Bayes classifier；Assuming that P:X ∈ y i,
Classifier traverses entire data set, and P is marked to classify the sample of correct sample and classification error, according to overall sample
This quantity calculates the classification error rate α of P come the sample number that judges incorrectlym, by classification error rate αmIt updates, obtains training sample
Weight beContinue the circulation of beginning next round, until M circulation knot
Beam；By repeatedly recycling, the Least risk Bayes sorting algorithm based on AdaBoost can sum up M classifier Pm out, pass through
Algorithm obtainsFinal P (x) is namely based in the in-depth filtration algorithm of content by M times
Obtained final classification device after study, classification obtains the classification of lung images.
Further, the multi-resolution decomposition module is merged smoothed image S using comentropy and is smoothly schemed
As FDWeight computing is carried out using the comentropy that Shannon proposes, wherein information entropy functionThe sequence of weight is mainly carried out using the probability distribution of pixel value, the same value
Sum of all pixels more multilist shows that the pixel is more important and assigns the pixel biggish weight, on the contrary then assign the pixel lesser power
Weight.
Further, the pretreatment portion includes image smoothing and de-noising, image sharpening, image background segmentation and image two
Value step.
Beneficial effects of the present invention
The present invention carries out the Medical image fusion of two kinds of different modalities of MRI and SPECT using filter and multiple features method,
Input picture is subjected to multi-resolution decomposition using the propagation filter of different Gaussian convolution cores, this method smoothly locates image
Reason improves smoothed image to the robustness of noise.Relative to traditional fusion method, brightness, direction and phase property are utilized
The weight of blending image is constructed, the more important informations of input picture can be obtained, and then provide more accurate auxiliary for doctor
Medical information.Adaboost classifier training method provided by the invention, optimized AdaBoost algorithm can exclude
Unnecessary training data feature, focuses on crucial training data, and according to different topic distillation strategies to data into
Row filtering, all taking into account a possibility that all classification errors, largely reduces the risk of erroneous judgement.
Detailed description of the invention
Fig. 1 is that the present invention provides preferred embodiment lung Medical image fusion classification method schematic diagram；
Fig. 2 is the detection device schematic diagram based on the classification of lung's Medical image fusion.
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, detailed
Carefully describe.Described embodiment is only a part of the embodiments of the present invention.
The technical solution that the present invention solves above-mentioned technical problem is:
It is illustrated in figure 2 a kind of detection device based on the classification of lung's Medical image fusion, for handling the lung of patient
CT scan image and obtain the suspected abnormality region of the patient comprising: acquisition unit obtains the CT lung scans of the patient
Image；Pretreatment portion is pre-processed to obtain binaryzation gray level image to the CT lung scans image after acquisition；Morphology
Corrosion and bulge carry out the image after morphological erosion operation is corroded to the binaryzation gray level image, to described two
Value gray level image carries out the image after dilation operation is expanded；Operational part carries out opening operation to the image after the corrosion
Opening operation figure is obtained, closed operation is carried out to the image after the expansion and obtains closed operation figure；Fourier transformation portion opens fortune to described
Nomogram and the closed operation figure carry out Fourier transformation and obtain Fourier transformation value；Lung's Medical image fusion categorization module is used
In carrying out integrated classification differentiation to CT lung scans image, and it is transferred to lesion test section by result is distinguished and judges；Lesion
Test section is judged to obtain the doubtful of the patient for distinguishing result to the Fourier transformation value combination row integrated classification
Focal area.Lung as shown in Figure 1 Medical image fusion categorization module comprises the following modules:
Data preprocessing module: obtaining the CT lung scans image of the patient, carries out including that the removal lung CT is swept
Pre-treatment step including baseline and high-frequency noise of the tracing as in, and it is classified as CT lung scans gray scale anatomic image and puppet
Color function CT lung scans image, is respectively labeled asWith
Multi-resolution decomposition module: during multi-resolution decomposition, by change propagate filter in Gaussian convolution core it is big
It is small, to the two width source images to step 1WithIt is smoothed, using smoothing processing to signalWithCarry out multiple dimensioned point
Solution obtains smoothed image and detail pictures；Fusion is weighted using comentropy for smoothed image S to obtain merging smooth figure
Picture, and detail pictures are merged to obtain fusion detail pictures F using multiple featuresS；Propagate the prototype of filter are as follows: Sp=
1/Zp∑t∈N(p)wp,t×It, wherein SpFor smoothed out image, ZpFor normalization factor, N (p) is using p as center pixel
Neighborhood, i indicate pixel sequence number, ItFor in the image pixel value of coordinate position t, wp,tTo propagate filter weights；Wherein, the first item of filter weights is propagated
It indicates to utilize the pixel the distance between adjacent with pixel p and pixel tWith Gaussian convolution core σaGaussian function；It passes
Broadcast the Section 2 of filter weightsIt indicates to utilize all pixels distance and Gaussian convolution adjacent with pixel p
Core σrGaussian function；
Reconstruct Fusion Module: using the simple operation in Pixel-level to fused smoothed image FDWith detail pictures FSInto
Row reconstructs to obtain blending image F based on ADMM；Export final blending image F；
Categorization module: and classified using improved Least risk Bayes Multiple Classifier Fusion image F, wherein
AdaBoost method is as follows:
Network data is inputted with a matrix type, initializes weight i=1,2 ... ..., n, is executed and is recycled m=1,2 ... ...,
M, by ωiValue substitute into AdaBoost frame in, be trained by Least risk Bayes classifier；Assuming that P:X ∈ y i,
Classifier traverses entire data set, and P is marked to classify the sample of correct sample and classification error, according to overall sample
This quantity calculates the classification error rate α of P come the sample number that judges incorrectlym, by classification error rate αmIt updates, obtains training sample
Weight beContinue the circulation of beginning next round, until M circulation knot
Beam；By repeatedly recycling, the Least risk Bayes sorting algorithm based on AdaBoost can sum up M classifier Pm out, pass through
Algorithm obtainsFinal P (x) is namely based in the in-depth filtration algorithm of content by M times
Obtained final classification device after study, classification obtains the classification of lung images.
The AdaBoost algorithm of optimization is the Least risk Bayes in-depth filtration algorithm based on AdaBoost algorithm.With
Training frame of the AdaBoost algorithm as classifier, is replaced in AdaBoost algorithm with Least risk Bayes sorting algorithm
Weak Classifier finally achieves the combination of two algorithms as the classifier of AdaBoost.Least risk Bayes sorting algorithm is just
It is to solve the problems, such as error rate based on Bayes and naive Bayesian, is the optimization in minimal error rate meaning.Pattra leaves
This sorting algorithm is the prior probability model by certain object, calculates its posterior probability using Bayesian formula.To obtain
The theme (selecting the class with maximum a posteriori probability as theme belonging to object source) of object source.Pass through training set of source data
It closes, each data information is obtained in inhomogeneous probability size by Bayesian Classification Arithmetic, constructs Bayesian Classification Model, Piao
Plain Bayes be exactly in Bayesian Classification Model error rate it is the smallest, and its needed for estimation parameter it is seldom, realize algorithm very
Simply.AdaBoost algorithm is a kind of iterative algorithm, and core concept is the classifier different for the training of the same training set
(Weak Classifier) then gets up these weak classifier sets, finally constitutes a strongest final classification device (strong classifier).
Preferably, the multi-resolution decomposition module is merged to obtain smoothed image for smoothed image S using comentropy
FDWeight computing is carried out using the comentropy that Shannon proposes, wherein information entropy functionThe sequence of weight is mainly carried out using the probability distribution of pixel value, the same value
Sum of all pixels more multilist shows that the pixel is more important and assigns the pixel biggish weight, on the contrary then assign the pixel lesser power
Weight.
Preferably, the pretreatment portion includes image smoothing and de-noising, image sharpening, image background segmentation and image two-value
Change step.The preferred embodiment of the present invention has been described in detail above.It should be appreciated that those skilled in the art are not necessarily to
Creative work, which according to the present invention can be conceived, makes many modifications and variations.Therefore, all technology people in the art
Member passes through the available skill of logical analysis, reasoning, or a limited experiment on the basis of existing technology under this invention's idea
Art scheme, all should be within the scope of protection determined by the claims.
Claims (3)
1. a kind of detection device based on the classification of lung's Medical image fusion, obtains for handling the CT lung scans image of patient
To the suspected abnormality region of the patient characterized by comprising acquisition unit obtains the CT lung scans image of the patient；
Pretreatment portion is pre-processed to obtain binaryzation gray level image to the CT lung scans image after acquisition；Morphological erosion
With bulge, the image after morphological erosion operation is corroded is carried out to the binaryzation gray level image, to the binaryzation
Gray level image carries out the image after dilation operation is expanded；Operational part carries out opening operation to the image after the corrosion and obtains
Opening operation figure carries out closed operation to the image after the expansion and obtains closed operation figure；Fourier transformation portion, to the opening operation figure
Fourier transformation, which is carried out, with the closed operation figure obtains Fourier transformation value；Lung's Medical image fusion categorization module, for pair
CT lung scans image carries out integrated classification differentiation, and differentiation result is transferred to lesion test section and is judged；Lesion detection
Portion is judged to obtain the suspected abnormality of the patient for distinguishing result to the Fourier transformation value combination row integrated classification
Region.Lung's Medical image fusion categorization module comprises the following modules:
Data preprocessing module: obtaining the CT lung scans image of the patient, carries out including the removal CT lung scans figure
The pre-treatment step including baseline and high-frequency noise as in, and it is classified as CT lung scans gray scale anatomic image and pseudo-colours
Function CT lung scans image, is respectively labeled asWith
Multi-resolution decomposition module: during multi-resolution decomposition, the size of Gaussian convolution core in filter is propagated by changing, right
To two width source images of step 1WithIt is smoothed, using smoothing processing to signalWithMulti-resolution decomposition is carried out, is obtained
Obtain smoothed image and detail pictures；Fusion is weighted using comentropy for smoothed image S and obtains fusion smoothed image, and it is right
It is merged to obtain fusion detail pictures F using multiple features in detail picturesS；Propagate the prototype of filter are as follows: Sp=1/Zp
∑t∈N(p)wp,t×It, wherein SpFor smoothed out image, ZpFor normalization factor, N (p) is using p as the neighbour of center pixel
Domain, i indicate pixel sequence number, ItFor in the image pixel value of coordinate position t, wp,tTo propagate filter weights；Wherein, the first item of filter weights is propagated
It indicates to utilize the pixel the distance between adjacent with pixel p and pixel tWith Gaussian convolution core σaGaussian function；It passes
Broadcast the Section 2 of filter weightsIt indicates to roll up using all pixels distance and Gauss adjacent with pixel p
Product core σrGaussian function；
Reconstruct Fusion Module: using the simple operation in Pixel-level to fused smoothed image FDWith detail pictures FSCarry out base
It reconstructs to obtain blending image F in ADMM；Export final blending image F；
Categorization module: and being classified using improved Least risk Bayes Multiple Classifier Fusion image F, wherein the side AdaBoost
Method is as follows:
Network data is inputted with a matrix type, initializes weight i=1,2 ... ..., n, executes circulation m=1,2 ... ..., M, it will
ωiValue substitute into AdaBoost frame in, be trained by Least risk Bayes classifier；Assuming that P:X ∈ yi, will classify
Device traverses entire data set, and P is marked to classify the sample of correct sample and classification error, according to the number of population sample
Amount calculates the classification error rate α of P come the sample number that judges incorrectlym, by classification error rate αmIt updates, obtains the weight of training sample
ForContinue the circulation of beginning next round, until M circulation terminates；Pass through
Repeatedly circulation, the Least risk Bayes sorting algorithm based on AdaBoost can sum up M classifier Pm out, obtain by algorithmFinal P (x) is namely based in the in-depth filtration algorithm of content the institute after M study
Obtained final classification device, classification obtain the classification of lung images.
2. the detection device according to claim 1 based on the classification of lung's Medical image fusion, which is characterized in that described more
Scale Decomposition module is merged to obtain smoothed image F for smoothed image S using comentropyDThe letter proposed using Shannon
It ceases entropy and carries out weight computing, wherein information entropy functionThe main probability for utilizing pixel value
It is distributed to carry out the sequence of weight, the sum of all pixels of the same value more multilist shows the pixel, and more important to assign the pixel biggish
Weight, it is on the contrary then assign the pixel lesser weight.
3. the detection device according to claim 1 based on the classification of lung's Medical image fusion, which is characterized in that described pre-
Processing unit includes image smoothing and de-noising, image sharpening, image background segmentation and image binaryzation step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810662571.XA CN108985345B (en) | 2018-06-25 | 2018-06-25 | Detection apparatus based on lung medical image fusion classification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810662571.XA CN108985345B (en) | 2018-06-25 | 2018-06-25 | Detection apparatus based on lung medical image fusion classification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108985345A true CN108985345A (en) | 2018-12-11 |
| CN108985345B CN108985345B (en) | 2020-09-18 |
Family
ID=64538720
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810662571.XA Active CN108985345B (en) | 2018-06-25 | 2018-06-25 | Detection apparatus based on lung medical image fusion classification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108985345B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110197722A (en) * | 2019-05-31 | 2019-09-03 | 贵州精准健康数据有限公司 | AI-CPU system platform |
| CN110246109A (en) * | 2019-05-15 | 2019-09-17 | 清华大学 | Merge analysis system, method, apparatus and the medium of CT images and customized information |
| CN110415200A (en) * | 2019-07-26 | 2019-11-05 | 西南科技大学 | A kind of bone cement implant CT image layer interpolation method |
| CN111325700A (en) * | 2020-02-26 | 2020-06-23 | 无锡久仁健康云科技有限公司 | Multi-dimensional fusion algorithm and system based on color images |
| CN111653356A (en) * | 2020-04-20 | 2020-09-11 | 浙江大学 | New coronary pneumonia screening method and new coronary pneumonia screening system based on deep learning |
| CN111724360A (en) * | 2020-06-12 | 2020-09-29 | 深圳技术大学 | Lung lobe segmentation method and device and storage medium |
| CN112349425A (en) * | 2020-02-10 | 2021-02-09 | 胡秋明 | Novel artificial intelligent rapid screening system for coronavirus infection pneumonia |
| CN113763297A (en) * | 2021-06-30 | 2021-12-07 | 安徽省立医院(中国科学技术大学附属第一医院) | Acromioclavicular joint CT image processing method |
| CN115861303A (en) * | 2023-02-16 | 2023-03-28 | 四川大学 | EGFR gene mutation detection method and system based on lung CT image |
| CN116823767A (en) * | 2023-06-27 | 2023-09-29 | 无锡市人民医院 | Method for judging lung transplantation activity grade based on image analysis |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101339653A (en) * | 2008-01-30 | 2009-01-07 | 西安电子科技大学 | Infrared and colorful visual light image fusion method based on color transfer and entropy information |
| CN102068271A (en) * | 2011-02-22 | 2011-05-25 | 南方医科大学 | Method for retrospectively classifying chest or abdomen computed tomography (CT) images based on respiratory phase |
| CN106600587A (en) * | 2016-12-09 | 2017-04-26 | 上海理工大学 | Lung CT image auxiliary detection processing device |
-
2018
- 2018-06-25 CN CN201810662571.XA patent/CN108985345B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101339653A (en) * | 2008-01-30 | 2009-01-07 | 西安电子科技大学 | Infrared and colorful visual light image fusion method based on color transfer and entropy information |
| CN102068271A (en) * | 2011-02-22 | 2011-05-25 | 南方医科大学 | Method for retrospectively classifying chest or abdomen computed tomography (CT) images based on respiratory phase |
| CN106600587A (en) * | 2016-12-09 | 2017-04-26 | 上海理工大学 | Lung CT image auxiliary detection processing device |
Non-Patent Citations (4)
| Title |
|---|
| JEN-HAO RICK CHANG ET AL.: "Propagated Image Filtering", 《IEEE》 * |
| YISHU PENG ET AL.: "Detail Enhancement for Infrared Images Based on Propagated Image Filter", 《HINDAWI》 * |
| 李茹 等: "基于AdaBoost的最小风险贝叶斯的垃圾邮件过滤算法", 《济南大学学报（自然科学版）》 * |
| 杜娇: "像素级多尺度医学图像融合方法研究", 《中国博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 |
|---|---|---|---|---|
| CN110246109A (en) * | 2019-05-15 | 2019-09-17 | 清华大学 | Merge analysis system, method, apparatus and the medium of CT images and customized information |
| CN110197722A (en) * | 2019-05-31 | 2019-09-03 | 贵州精准健康数据有限公司 | AI-CPU system platform |
| CN110415200A (en) * | 2019-07-26 | 2019-11-05 | 西南科技大学 | A kind of bone cement implant CT image layer interpolation method |
| CN110415200B (en) * | 2019-07-26 | 2022-03-08 | 西南科技大学 | Method for interpolating among CT (computed tomography) image layers of bone cement implant |
| CN112349425A (en) * | 2020-02-10 | 2021-02-09 | 胡秋明 | Novel artificial intelligent rapid screening system for coronavirus infection pneumonia |
| CN111325700A (en) * | 2020-02-26 | 2020-06-23 | 无锡久仁健康云科技有限公司 | Multi-dimensional fusion algorithm and system based on color images |
| CN111653356A (en) * | 2020-04-20 | 2020-09-11 | 浙江大学 | New coronary pneumonia screening method and new coronary pneumonia screening system based on deep learning |
| CN111724360A (en) * | 2020-06-12 | 2020-09-29 | 深圳技术大学 | Lung lobe segmentation method and device and storage medium |
| CN111724360B (en) * | 2020-06-12 | 2023-06-02 | 深圳技术大学 | Lung lobe segmentation method, device and storage medium |
| CN113763297A (en) * | 2021-06-30 | 2021-12-07 | 安徽省立医院(中国科学技术大学附属第一医院) | Acromioclavicular joint CT image processing method |
| CN115861303A (en) * | 2023-02-16 | 2023-03-28 | 四川大学 | EGFR gene mutation detection method and system based on lung CT image |
| CN115861303B (en) * | 2023-02-16 | 2023-04-28 | 四川大学 | EGFR gene mutation detection method and system based on lung CT image |
| CN116823767A (en) * | 2023-06-27 | 2023-09-29 | 无锡市人民医院 | Method for judging lung transplantation activity grade based on image analysis |
| CN116823767B (en) * | 2023-06-27 | 2024-03-01 | 无锡市人民医院 | Method for judging lung transplantation activity grade based on image analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108985345B (en) | 2020-09-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108985345A (en) | A kind of detection device based on the classification of lung's Medical image fusion | |
| Huang et al. | Segmentation of breast ultrasound image with semantic classification of superpixels | |
| Schlemper et al. | Attention gated networks: Learning to leverage salient regions in medical images | |
| van Ginneken | Fifty years of computer analysis in chest imaging: rule-based, machine learning, deep learning | |
| Lessmann et al. | Automatic calcium scoring in low-dose chest CT using deep neural networks with dilated convolutions | |
| Jung et al. | An automatic nuclei segmentation method based on deep convolutional neural networks for histopathology images | |
| Wang et al. | A hybrid CNN feature model for pulmonary nodule malignancy risk differentiation | |
| CN110310287B (en) | Automatic organ-at-risk delineation method, equipment and storage medium based on neural network | |
| Mostapha et al. | Role of deep learning in infant brain MRI analysis | |
| Rao et al. | Brain tumor detection and segmentation using conditional random field | |
| CN108898593B (en) | Detection apparatus based on abdomen CT medical image fusion classification | |
| Taha et al. | Automatic polyp detection in endoscopy videos: A survey | |
| Maicas et al. | Pre and post-hoc diagnosis and interpretation of malignancy from breast DCE-MRI | |
| Li et al. | Human treelike tubular structure segmentation: A comprehensive review and future perspectives | |
| Dudovitch et al. | Deep learning automatic fetal structures segmentation in MRI scans with few annotated datasets | |
| Li et al. | BUSnet: A deep learning model of breast tumor lesion detection for ultrasound images | |
| Hasan et al. | A combined approach using image processing and deep learning to detect pneumonia from chest X-ray image | |
| Pei | Emphysema classification using convolutional neural networks | |
| Carmo et al. | A systematic review of automated segmentation methods and public datasets for the lung and its lobes and findings on computed tomography images | |
| Suhail et al. | Automatic detection of abnormalities in mammograms | |
| Srinivasa et al. | Identifying lung nodules on MRR connected feature streams for tumor segmentation | |
| CN108898173A (en) | A kind of the electrocardiogram Medical image fusion and classification method of multiple dimensioned multiple features | |
| Elmenabawy et al. | Deep segmentation of the liver and the hepatic tumors from abdomen tomography images | |
| Vijila Rani et al. | Detection of cervix tumor using an intelligent system accompanied with PNN classification approach | |
| Priya et al. | A Hybrid Deep Learning based Classification of Brain Lesion Classification in CT Image using Convolutional Neural Networks |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200824 Address after: No. 100, Shunhe Road, Longcheng Town, Xiaoxian County, Suzhou City, Anhui Province Applicant after: Anhui Beitai Photoelectric Technology Co.,Ltd. Address before: No. 12, No. 12, Lake Yun street, Chongqing, Chongqing Applicant before: CHONGQING ZHIAO TECHNOLOGY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |