CN109658411A - A kind of correlation analysis based on CT images feature Yu Patients with Non-small-cell Lung prognosis situation - Google Patents
A kind of correlation analysis based on CT images feature Yu Patients with Non-small-cell Lung prognosis situation Download PDFInfo
- Publication number
- CN109658411A CN109658411A CN201910052967.7A CN201910052967A CN109658411A CN 109658411 A CN109658411 A CN 109658411A CN 201910052967 A CN201910052967 A CN 201910052967A CN 109658411 A CN109658411 A CN 109658411A
- Authority
- CN
- China
- Prior art keywords
- feature
- analysis
- cell lung
- characteristic
- tumour
- 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
- 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
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
- G06T2207/20081—Training; Learning
-
- 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 provides a kind of correlation analysis based on CT images feature Yu Patients with Non-small-cell Lung prognosis Survival, belongs to computer-aided medical science field.The present invention is split first to lung tumors and feature extraction, optimization；Then, by the method for machine learning by tumor imaging feature in conjunction with survival of patients situation, construct the prognosis evaluation model of non-small cell lung cancer and predicted come the prognosis Survival to patient, while contrived experiment assesses the performance of prognostic analysis model.The present invention provides a kind of new scheme to seek the relationship of Features and patient's prognosis Survival.
Description
Technical field
The present invention relates to computer-aided medical science fields, are related to a kind of based on CT images feature and non-small cell lung cancer trouble
The correlation analysis of person's prognosis situation.
Background technique
The World Health Organization (WHO) international cancer research institution (IARC) issues newest report recently and claims, and lung cancer is the whole world
The fastest-rising malignant tumour of morbidity and mortality in range, it is contemplated that will cause 1,800,000 people dead for 2018, and account for estimated cancer
The 18.4% of total toll.Classified according to histological type, lung cancer is divided into non-small cell lung cancer and Small Cell Lung Cancer, wherein
Non-small cell lung cancer (non-small cell lung cancer, NSCLC) accounts for the 80%~85% of lung cancer illness total number of persons, packet
Include squamous cell carcinoma (squamous carcinoma), gland cancer, large cell carcinoma.For non-small cell lung cancer compared to Small Cell Lung Cancer, growth division is slower, expands
Scattered transfer is later, and lethal is also relatively weak, but due to there are a large amount of individuals between the lesion of different Patients with Non-small-cell Lung
Between difference so that there is very big difference in the development speed that different patient suffers from the cancer prognosis state of an illness.Epidemiology statistics show largely
Patients with Non-small-cell Lung fail to receive suitable treatment in time due to not obtaining the prediction of accurate progression of the disease so that should
The death rate of class patients with lung cancer is up to 75%.Therefore, there is an urgent need to effective survival of patients time prediction model come to treatment and
The selection of check scheme is assisted, and to improve the therapeutic effect of non-small cell lung cancer, and then is improved the cure rate of patient and is deposited
Motility rate.
Radiation group is an emerging field in medicine, and the birth of the technology and radiation genomics are in disease research
Great potential in the diagnosing and treating of disease of superperformance and medical image it is inseparable.Radiation group passes through feature
It extracts, by the information MAP of tumor region to high-dimensional feature space, the prognosis of disease is then constructed by the method for machine learning
Model predicts the future development of disease, to treatment to disease and check the selection of scheme and instruct.CT shadow
It is easy to the features such as comparing as data possess the easy and result of acquisition, as one of the important mode in radiation group database,
It is widely used in radiation group research.
In recent years, it is directed to computer-aided diagnosis (computer aided diagnosis, CAD) technology and essence both at home and abroad
The research of quasi- medical treatment (Precision Medicine) is more and more burning hoter.Cad technique and accurate medical treatment are both needed to learn to do by image
Section carries out quantitative analysis to tumour by extracting a large amount of Features, to achieve the purpose that adjuvant clinical diagnoses.And it utilizes
The Features of extraction carry out correlation analysis to patient's prognosis Survival, and building prognosis evaluation model prediction patient's is pre-
Survival afterwards, so that doctor preferably be instructed to select the treatment of patient and check method.
From the point of view of current domestic and international present Research, the research of non-small cell carcinoma prognostic analysis generally goes out from clinical angle
Hair, firstly, (clinical stages, smoking history, whether there is or not brain metastes, tumor marker, doctors according to the intuitive Clinical symptoms of case sample
Learn sign etc.) quantizating index as case sample；Then, by traditional statistical method to Clinical symptoms and Prognostic significance
Single factor test survival analysis is carried out, Clinical symptoms relevant to patient's prognosis is obtained；Finally, by single factor analysis with patient's prognosis
Relevant Clinical symptoms substitutes into COX regression model and carries out multiplicity, obtains the Prognostic Factors of non-small cell lung cancer, helps to cure
It takes root and carries out more accurate prognosis evaluation to Patients with Non-small-cell Lung according to Prognostic Factors, design preferably treatment and check
Scheme, to extend the survival of patients time.And such methods, there is also limitation, the clinical information type that can be utilized is less,
And the feature of medicine sign class only shows the portion forms characteristic of tumor region, and from iconography angle, it is available
The more abundant Features of more and type can effectively solve to swell to reflect the more implicit informations of tumour
Tumor heterogeneity is difficult to the problem of being quantitatively evaluated.For current research Shortcomings, the present invention devises new lung cancer for prognosis research
Method carries out analysis to the prognosis survival state of non-small cell lung cancer and probes into, obtains the non-small cell lung based on CT images feature
Cancer prognostic analysis model predicts the prognosis life span of patient；Meanwhile contrived experiment verifies experimental method, into
And improve presently, there are deficiencies, obtain better prognostic analysis effect.
Summary of the invention
The purpose of the present invention is propose a kind of non-small thin based on semi-automatic segmentation extraction for the deficiency in existing research
The correlation analysis of born of the same parents' lung cancer CT images feature and Patients with Non-small-cell Lung prognosis situation.
To achieve the above object, the technical solution adopted by the present invention the following steps are included:
Processing, characteristic processing, image feature and the association analysis of Survival and testing for experimental result of CT images
Card analysis.The detailed process of every part is described as follows:
The processing of one .CT image: firstly, passing through interactive medical image control system RadiAntViewer software from CT
Tumour is found in image and frame selects its approximate region, obtains tumour sequence image.Then, divide with semi-automatic partition method
Tumour out takes different types of tumour different splitting schemes: (1) stand alone tumour: take gray threshold segmentation algorithm and
The intersection of algorithm of region growing segmentation result；(2) adhesion lung wall-shaped tumour: firstly, based on Chain-Code-Method and " rolling ball method " to lung area
It is repaired at edge.Then, further divide referring to stand alone lesion segmentation approach, finally, above-mentioned segmentation result is checked, to mistake
Segmentation or less divided situation carry out manual repair；(3) adhesion vascular type tumour: firstly, being obtained with stand alone lesion segmentation approach
To image.Then, using the method for " blowing ball " and fuzzy C-means clustering, and bianry image is obtained based on threshold value, final point
Cut the intersection that result takes the two.Finally, check above-mentioned segmentation result, to over-segmentation or still there is the case where not removing blood vessel to carry out people
Work amendment.On the basis of tumor region segmentation, it is extracted and exists including gray feature, morphological feature, textural characteristics, medicine sign
Interior 258 kinds of quantitative characteristics describe tumour；
Two, characteristics processing: firstly, since the positive and negative sample imbalance of data used in machine learning will affect training
Model performance, in order to solve this problem, the present invention use SMOTE (Synthetic Minority Oversampling
Technique) algorithm balances positive negative sample；Then, in order to probe between CT images feature and patient's prognosis Survival
Correlation, it is necessary to which the quantitative target of multi-class extraction tumour multiple dimensioned as much as possible causes to avoid the omission for causing information
Analyze result inaccuracy.And the too high characteristic matching that often will lead to of intrinsic dimensionality extracted is excessively complicated, consumes system resource, and
And most homogenous characteristics describe the general character between variety classes tumour, have correlation between some feature
Property, it will cause the calculating of bulk redundancy using this category feature.To obtain disaggregated model of good performance, need to characterization redundancy letter
The characteristic of breath is filtered, while will have paraspecific feature to retain, therefore, to obtain of good performance point
Class model needs the characteristic to characterization redundancy to be filtered, while will have paraspecific feature to retain.
So using PCA principal component analysis (Principal ComponentAnalysis) method to the characteristic extracted into
Row dimensionality reduction, optimization, reduce error caused by redundancy, improve the performance of disaggregated model；Finally, returning to characteristic
The value interval of feature, is zoomed to the range of [0,1] by one change processing；
The association analysis of three, image features and prognosis situation: correlation analysis algorithm analysis image feature and prognosis are utilized
Correlation between situation is that index screening goes out statistically significant associated image feature and prognosis situation with P < 0.05；
The verifying of four, experimental results is analyzed: Various Classifiers on Regional being taken to establish Features and patient's prognosis Survival
Correlation models, in order to construct the disaggregated model of a best performance as much as possible, herein by a large amount of research and test,
Finally pick the representative classifier that following 7 kinds of generalization abilities are strong, can be used for Small Sample Database collection training, respectively decision tree
(Decision tree, DT), discriminatory analysis classifier (Discriminant Analysis Classifiers, DAC), logic
Recurrence classifier (Logistic Regression, LR), support vector machines (Support Vector Machine, SVM), K are close
Adjacent classifier (k-Nearest Neighbor, KNN), Ensemble classifier (Ensemble classifiers, EC), random forest
Classifier (Random Forest, RF) is classified and is tested to data.The phase is deposited with the triennial with clinical meaning as boundary
Two classification are carried out, to predict the prognosis situation of patient, and with accuracy rate (ACC), sensibility (SE), specificity (SP), ROC
Area under the curve (AUC) is used as index, and the performance of prognostic analysis model is verified using ten folding cross validation methods.
Detailed description of the invention
Fig. 1 is the process based on CT images feature Yu the correlation analysis of Patients with Non-small-cell Lung prognosis situation
Figure.
Table 1 is Features.
Table 2 is the table of comparisons of the experimental result of experimental result of the present invention and existing document
Specific embodiment
It elaborates with reference to the accompanying drawings of the specification to a specific embodiment of the invention.
The specific implementation process is as follows:
The processing of one .CT image: firstly, passing through interactive medical image control system RadiAntViewer software from CT
Tumour is found in image and frame selects its approximate region, obtains tumour sequence image.Then, divide with semi-automatic partition method
Tumour out takes different types of tumour different splitting schemes: (1) stand alone tumour: take gray threshold segmentation algorithm and
The intersection of algorithm of region growing segmentation result；(2) adhesion lung wall-shaped tumour: firstly, based on Chain-Code-Method and " rolling ball method " to lung area
It is repaired at edge.Then, further divide referring to stand alone lesion segmentation approach, finally, above-mentioned segmentation result is checked, to mistake
Segmentation or less divided situation carry out manual repair；(3) adhesion vascular type tumour: firstly, being obtained with stand alone lesion segmentation approach
To image.Then, using the method for " blowing ball " and fuzzy C-means clustering, and bianry image is obtained based on threshold value, final point
Cut the intersection that result takes the two.Finally, check above-mentioned segmentation result, to over-segmentation or still there is the case where not removing blood vessel to carry out people
Work amendment.On the basis of tumor region segmentation, it is extracted and exists including gray feature, morphological feature, textural characteristics, medicine sign
Interior 258 kinds of quantitative characteristics describe tumour；
Two, characteristics processing: firstly, since the positive and negative sample imbalance of data used in machine learning will affect training
Model performance, in order to solve this problem, using SMOTE (Synthetic Minority Oversampling
Technique) algorithm balances positive negative sample.The basic thought of SMOTE algorithm is analyzed minority class sample and according to few
Several classes of artificial synthesized new samples of sample are added in data set, and experiment flow is as follows:
1. using Euclidean distance as criterion calculation, it arrives minority class sample set S for each sample x in minority classminMiddle institute
There is the distance of sample, obtains its k neighbour.
2. multiplying power N is sampled according to one oversampling ratio of sample imbalance ratio setting to determine, for each minority class
Sample x randomly chooses several samples from its k neighbour, it is assumed that the neighbour selected is xn。
3. the neighbour x selected at random for eachn, new sample is constructed according to following formula with original sample respectively
xnew。
xnew=x+rand (0,1) × (x-xn) (1)
Then, in order to which the correlation probed between CT images feature and patient's prognosis Survival contacts, it is necessary to the greatest extent may be used
The quantitative target of the multiple dimensioned multi-class extraction tumour in energy ground causes to analyze result inaccuracy to avoid the omission for causing information.
And the intrinsic dimensionality extracted is too high often will lead to excessively complicated when characteristic matching, consumes system resource, and it is most together
Category feature describes the general character between variety classes tumour, has correlation between some feature, uses such spy
Sign will cause the calculating of bulk redundancy.Therefore, to obtain disaggregated model of good performance, the feature to characterization redundancy is needed
Data are filtered, while will have paraspecific feature to retain.So using PCA principal component analysis
The method of (Principal Component Analysis) carries out dimensionality reduction, optimization to the characteristic extracted, and reduces redundancy
Error caused by information improves the performance of disaggregated model.Finally, characteristic is normalized, by taking for feature
Value section zooms to the range of [0,1]；
PCA principal component analysis is a kind of more common dimensionality reduction technology, and main thought is to look for a hyperplane
To have sample is suitably expressed so that after existing sample is projected on hyperplane, each sample point to this hyperplane
Distance it is all close enough.Assuming that the new coordinate after projection is { w1,w2,w3···wd, w thereiniIt is orthonormal basis, d
For the dimension in former space, original partial coordinates are abandoned, dimension is reduced to d, original sample point xiBy { w1,w2,w3···
wdComposition space in projection be zi={ zi1,zi2,zi3···zid', have at this timeFor xiSeat after dimensionality reduction
Jth in mark system ties up coordinate.If the x after reconstructiFor xi', then haveThen xiWith xiThe distance between ' are as follows:
The target of principal component analysis is exactly to make tr (WTXXTW minimum value) is obtained as far as possible.Again by lagrange's method of multipliers,
Have:
XXTW=λ W (3)
Wherein λ is characterized value, preceding d maximum corresponding set { w1,w2,w3···wd'It is principal component analysis
Solution.
The specific operating process of PCA are as follows:
1) average value is removed, i.e. each feature subtracts respective average value.
2) covariance matrix is calculated.
3) eigen vector of covariance matrix is calculated.
4) characteristic value is sorted from large to small.
5) retain maximum a feature vector.
6) data are transformed into a new space for feature vector building.
Finally, characteristic is normalized, the value interval of feature is zoomed to the range of [0,1]；
The association analysis of three, image features and prognosis situation: image feature is analyzed using Pearson correlation analysis algorithm
It is that index screening goes out statistically significant associated image feature and prognosis feelings with P < 0.05 with the correlation between prognosis situation
Condition；
The verifying of four, experimental results is analyzed: Various Classifiers on Regional being taken to establish Features and patient's prognosis Survival
Correlation models, in order to construct a performance more preferably disaggregated model as much as possible, herein by a large amount of research and test,
Finally pick the representative classifier that following 7 kinds of generalization abilities are strong, can be used for Small Sample Database collection training, respectively decision tree
(Decision tree, DT), discriminatory analysis classifier (Discriminant Analysis Classifiers, DAC), logic
Recurrence classifier (Logistic Regression, LR), support vector machines (SupportVector Machine, SVM), K are close
Adjacent classifier (k-NearestNeighbor, KNN), Ensemble classifier (Ensemble classifiers, EC), random forest
Classifier (Random Forest, RF) is classified and is tested to data.Patients with lung cancer post-operative survival rates are shown by clinical research
Time is more than 3 years, and recurrence probability will substantially reduce；Life span is more than 5 years, then can regard as fully recovering, this kind of life span pair
Clinician designs patient's treatment and the auxiliary direction of the selection of check scheme is significant.Therefore the present invention is with clinical meaning
Triennial deposit the phase for boundary carry out two classification, to predict the prognosis situation of patient, and with accuracy rate (ACC), sensibility
(SE), area (AUC) is used as index under specific (SP), ROC curve, verifies prognostic analysis using ten folding cross validation methods
The performance of model.
Contrived experiment of the present invention carries out prognostic analysis research to Patients with Non-small-cell Lung, based on CT images feature to non-small
Cell lung cancer prognostic analysis model is constructed.According to typical radiation group research framework, non-small cell lung cancer data are carried out
Semi-automatic segmentation, feature extraction, characteristic optimization and classifier modeling, and the performance of prognostic analysis model evaluated.
By inventive algorithm and document [1] before " being studied based on lung cancer CT images radiation group prognostic model " and document
[2]《Predicting Outcomes ofNonsmall Cell Lung Cancer Using CT Image Features》
The prognostic analysis model algorithm result mentioned compares, with area under the accuracy rate (ACC) of prediction model and ROC curve
(AUC) it is used as algorithm evaluation index, comparing result is as shown in table 2.Comparison is as it can be seen that the feature extraction that algorithm used in the present invention uses
Type, the performance of feature optimization algorithm and prognosis classification model is compared with increasing.
Table 1
Table 2
Claims (5)
1. a kind of correlation analysis based on CT images feature Yu non-small cell lung cancer prognosis Survival, feature exist
In, comprising the following steps: processing, the characteristic processing, the association analysis and reality of Features and Survival of CT images
Test the verifying analysis of result.Detailed process is as follows:
The processing of 1.CT image: firstly, through interactive medical image control system RadiAntViewer software from CT images
It finds tumour and frame selects its approximate region, obtain tumour sequence image.Then, it is partitioned into semi-automatic partition method swollen
Tumor takes different types of tumour different splitting schemes: (1) stand alone tumour: taking gray threshold segmentation algorithm and region
The intersection of growth algorithm segmentation result；(2) adhesion lung wall-shaped tumour: firstly, based on Chain-Code-Method and " rolling ball method " to lung area edge
It is repaired.Then, further divide referring to stand alone lesion segmentation approach, finally, above-mentioned segmentation result is checked, to over-segmentation
Or less divided situation carries out manual repair；(3) adhesion vascular type tumour: firstly, obtaining figure with stand alone lesion segmentation approach
Picture.Then, using the method for " blowing ball " and fuzzy C-means clustering, and bianry image, final segmentation knot are obtained based on threshold value
Fruit takes the intersection of the two.Finally, check above-mentioned segmentation result, to over-segmentation or still there is the case where not removing blood vessel manually to be repaired
Just.On the basis of tumor region segmentation, it is extracted including gray feature, morphological feature, textural characteristics, medicine sign
258 quantitative characteristics describe tumour；
2. characteristic is handled: firstly, the positive and negative sample size of balance characteristics data set.Then, dimensionality reduction is carried out to characteristic, it is excellent
Change；Finally, the value interval of feature, is zoomed to the range of [0,1] by normalized；
3. the association analysis of image feature and prognosis situation: using Pearson correlation analysis algorithm analysis Features with
Correlation between prognosis situation, with P < 0.05 be index screening go out statistically with prognosis survival analysis model it is significantly associated
Features；
4. the verifying of experimental result is analyzed: multiple types classifier being taken to establish the correlation of Features with patient's Survival
Model selects performance more preferably disaggregated model, deposits the phase using the triennial with clinical meaning and classifies as boundary, to predict
The prognosis situation of patient out, and using area (AUC) under accuracy rate (ACC), sensibility (SE), specific (SP), ROC curve as
Index verifies the performance of prognostic analysis model using ten folding cross validation methods.
2. the correlation analysis according to claim 1 based between CT images feature and non-small cell lung oncogene expression
Method, which is characterized in that the gray feature totally 14, including mean value, standard deviation, maximum value minimum, maximum value, most
Small value, intermediate value, comentropy, kurtosis, gradient, contrast, energy, contrast, density.
3. the correlation point according to claim 1 based on CT images feature with Patients with Non-small-cell Lung prognosis situation
Analysis method, which is characterized in that the morphological feature totally 17, including volume, surface area, sphericity, three-dimensional longest diameter, table
Area-volume ratio, interlayer region are compact like circularity, interlayer region rectangular degree, interlayer region elongation, interlayer region
Degree, interlayer region profile size ratio, interlayer region perimeter, interlayer region concavity and convexity, 2,3,4 rank of interlayer region are not
Bending moment, interlayer region irregularity boundary degree, interlayer region boundary Fourier describe son.
4. the correlation analysis according to claim 1 based between CT images feature and non-small cell lung oncogene expression
Method, which is characterized in that the textural characteristics totally 52, in which:
1) gray level co-occurrence matrixes, there is 13 kinds of characteristic statistics, including with mean value, variance and variance, difference variance, inverse difference moment, right
Than degree, non-similarity, entropy and entropy, poor entropy, correlation, angular second moment, relevant information measurement, it is obtained based on image grayscale
The characteristic statistic extracted on the basis of the gray level co-occurrence matrixes in 4 directions of image (0 °, 45 °, 90 °, 135 °)；
2) gray scale run-length matrix has 17 kinds of characteristic statistics, including emphasizes the short distance of swimming, emphasizes the long distance of swimming, gray scale inhomogeneities, trip
Cheng Changdu inhomogeneities, run length distribution, distance of swimming ratio, emphasizes the low gray level distance of swimming, emphasizes high grade grey level grey level distribution
The distance of swimming emphasizes the low gray level of the short distance of swimming, emphasizes short distance of swimming high grade grey level, emphasize the low gray level of the long distance of swimming, emphasize long distance of swimming high ash
Spend grade, mean value, mean square deviation, energy, entropy；
3) neighborhood gray scale differential matrix has 5 characteristic statistics, including degree of rarefication, busy degree, complexity, contrast, texture power
Degree；
4) Gabor wavelet feature shares 2 kinds of characteristic statistics, including gabor texture mean value, gabor texture variance, it is base
In the top layer calling module for calling the Gabor filter of 4 directions (0 °, 45 °, 90 °, 135 °) to be filtered image and 9
The textural characteristics of subgraph layer；
5) local binary feature (LBP feature) shares 2 characteristic statistics, comprising: LBP characteristic mean, LBP characteristic standard
Difference；
6) Scale invariant features transform (SIFT feature) shares 72 characteristic statistics.
5. the correlation analysis according to claim 1 based between CT images feature and non-small cell lung oncogene expression
Method, which is characterized in that the medicine sign totally 7, including sign of lobulation, spicule sign, the uniformity of knurl density, vacuole
Sign, asterism sign, patch shape shadow and bubble property wind-puff, incisura identification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910052967.7A CN109658411A (en) | 2019-01-21 | 2019-01-21 | A kind of correlation analysis based on CT images feature Yu Patients with Non-small-cell Lung prognosis situation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910052967.7A CN109658411A (en) | 2019-01-21 | 2019-01-21 | A kind of correlation analysis based on CT images feature Yu Patients with Non-small-cell Lung prognosis situation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109658411A true CN109658411A (en) | 2019-04-19 |
Family
ID=66119017
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910052967.7A Pending CN109658411A (en) | 2019-01-21 | 2019-01-21 | A kind of correlation analysis based on CT images feature Yu Patients with Non-small-cell Lung prognosis situation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109658411A (en) |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110136078A (en) * | 2019-04-29 | 2019-08-16 | 天津大学 | The semi-automatic reparation complementing method of single plant corn image leaf destruction |
| CN110246584A (en) * | 2019-06-13 | 2019-09-17 | 吉林大学第一医院 | A kind of prediction and evaluation method and system of lung cancer sample |
| CN110309862A (en) * | 2019-06-11 | 2019-10-08 | 广东省人民医院(广东省医学科学院) | DME prognosis information forecasting system and its application method based on ensemble machine learning |
| CN110428396A (en) * | 2019-06-25 | 2019-11-08 | 郑州大学第一附属医院 | A kind of feature vectors dimensional down method and system based on CT images |
| CN110807785A (en) * | 2019-11-01 | 2020-02-18 | 上海理工大学 | Research method for postoperative recurrence correlation factors of non-small cell lung cancer patient |
| CN110916666A (en) * | 2019-12-11 | 2020-03-27 | 大连医科大学附属第一医院 | Imaging omics feature processing method for predicting recurrence of hepatocellular carcinoma after surgical resection based on multi-modal MRI (magnetic resonance imaging) images |
| CN110930402A (en) * | 2019-12-11 | 2020-03-27 | 大连医科大学附属第一医院 | Imaging group characteristic processing method of DKI sequence MK graph based on endometrial cancer |
| CN110930401A (en) * | 2019-12-11 | 2020-03-27 | 大连医科大学附属第一医院 | WHO/ISUP grading method for preoperative prediction of renal clear cell carcinoma by energy spectrum CT enhanced iodography imaging omics |
| CN111081317A (en) * | 2019-12-10 | 2020-04-28 | 山东大学 | Gene spectrum-based breast cancer lymph node metastasis prediction method and prediction system |
| CN111276191A (en) * | 2020-01-15 | 2020-06-12 | 范时浩 | Method, system, medium and device for statistical identification of molecular weight of sugar in pancreatic cancer blood |
| CN111340770A (en) * | 2020-02-22 | 2020-06-26 | 太原理工大学 | Method for constructing cancer prognosis model by combining global weighted LBP (local binary pattern) and texture analysis |
| CN111739645A (en) * | 2020-05-14 | 2020-10-02 | 上海依智医疗技术有限公司 | Training method of immune-related pneumonia prediction model |
| CN111738980A (en) * | 2020-05-14 | 2020-10-02 | 上海依智医疗技术有限公司 | Medical image display method, computer equipment and storage medium |
| CN111751263A (en) * | 2020-06-08 | 2020-10-09 | 西安交通大学 | Mark-free cell two-dimensional scattering image inversion method based on gray level co-occurrence matrix |
| CN112001896A (en) * | 2020-08-03 | 2020-11-27 | 什维新智医疗科技(上海)有限公司 | Thyroid gland border irregularity detection device |
| CN112184658A (en) * | 2020-09-24 | 2021-01-05 | 上海健康医学院 | Method, medium, and electronic device for predicting non-small cell lung cancer prognostic survival |
| CN112712875A (en) * | 2019-10-25 | 2021-04-27 | 中国医药大学附设医院 | Transfer learning aided prediction system, method and computer program product |
| CN112807008A (en) * | 2021-01-27 | 2021-05-18 | 山东大学齐鲁医院 | Method and system for identifying actual mycoplasma pneumoniae and streptococcus pneumoniae of children based on imaging omics |
| CN113256614A (en) * | 2021-06-22 | 2021-08-13 | 国家超级计算天津中心 | Medical image processing system |
| CN113361944A (en) * | 2021-06-21 | 2021-09-07 | 鑫安利中（北京）科技有限公司 | Safe production index abnormity rapid sensing method based on Internet of things and artificial intelligence |
| CN113516669A (en) * | 2021-06-23 | 2021-10-19 | 湖北英库科技有限公司 | Trachea extraction method, device and equipment based on CT image and storage medium |
| CN114438017A (en) * | 2022-02-18 | 2022-05-06 | 中山大学附属第一医院 | Correlation analysis method of endothelial progenitor cell function and cardiac function of sepsis patient and application of correlation analysis method in diagnosis of sepsis cardiomyopathy |
-
2019
- 2019-01-21 CN CN201910052967.7A patent/CN109658411A/en active Pending
Cited By (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110136078A (en) * | 2019-04-29 | 2019-08-16 | 天津大学 | The semi-automatic reparation complementing method of single plant corn image leaf destruction |
| CN110309862A (en) * | 2019-06-11 | 2019-10-08 | 广东省人民医院(广东省医学科学院) | DME prognosis information forecasting system and its application method based on ensemble machine learning |
| CN110246584A (en) * | 2019-06-13 | 2019-09-17 | 吉林大学第一医院 | A kind of prediction and evaluation method and system of lung cancer sample |
| CN110428396A (en) * | 2019-06-25 | 2019-11-08 | 郑州大学第一附属医院 | A kind of feature vectors dimensional down method and system based on CT images |
| CN112712875B (en) * | 2019-10-25 | 2022-12-02 | 台湾中国医药大学附设医院 | Auxiliary prediction system and method for transfer learning |
| CN112712875A (en) * | 2019-10-25 | 2021-04-27 | 中国医药大学附设医院 | Transfer learning aided prediction system, method and computer program product |
| CN110807785A (en) * | 2019-11-01 | 2020-02-18 | 上海理工大学 | Research method for postoperative recurrence correlation factors of non-small cell lung cancer patient |
| CN111081317A (en) * | 2019-12-10 | 2020-04-28 | 山东大学 | Gene spectrum-based breast cancer lymph node metastasis prediction method and prediction system |
| CN111081317B (en) * | 2019-12-10 | 2023-06-02 | 山东大学 | Gene spectrum-based breast cancer lymph node metastasis prediction method and prediction system |
| CN110930401A (en) * | 2019-12-11 | 2020-03-27 | 大连医科大学附属第一医院 | WHO/ISUP grading method for preoperative prediction of renal clear cell carcinoma by energy spectrum CT enhanced iodography imaging omics |
| CN110916666B (en) * | 2019-12-11 | 2023-03-21 | 大连医科大学附属第一医院 | Imaging omics feature processing method for predicting recurrence of hepatocellular carcinoma after surgical resection |
| CN110930402A (en) * | 2019-12-11 | 2020-03-27 | 大连医科大学附属第一医院 | Imaging group characteristic processing method of DKI sequence MK graph based on endometrial cancer |
| CN110930402B (en) * | 2019-12-11 | 2023-11-07 | 大连医科大学附属第一医院 | Image histology feature processing method based on DKI sequence MK (restriction enzyme) map of endometrial cancer |
| CN110916666A (en) * | 2019-12-11 | 2020-03-27 | 大连医科大学附属第一医院 | Imaging omics feature processing method for predicting recurrence of hepatocellular carcinoma after surgical resection based on multi-modal MRI (magnetic resonance imaging) images |
| CN110930401B (en) * | 2019-12-11 | 2023-11-10 | 大连医科大学附属第一医院 | Energy spectrum CT enhanced iodigram imaging preacademic prediction kidney transparent cell carcinoma WHO/ISUP grading method |
| CN111276191A (en) * | 2020-01-15 | 2020-06-12 | 范时浩 | Method, system, medium and device for statistical identification of molecular weight of sugar in pancreatic cancer blood |
| CN111276191B (en) * | 2020-01-15 | 2020-12-18 | 范时浩 | Method, system, medium and device for statistical identification of molecular weight of sugar in pancreatic cancer blood |
| CN111340770A (en) * | 2020-02-22 | 2020-06-26 | 太原理工大学 | Method for constructing cancer prognosis model by combining global weighted LBP (local binary pattern) and texture analysis |
| CN111738980A (en) * | 2020-05-14 | 2020-10-02 | 上海依智医疗技术有限公司 | Medical image display method, computer equipment and storage medium |
| CN111739645B (en) * | 2020-05-14 | 2024-01-30 | 北京深睿博联科技有限责任公司 | Training method of immune-related pneumonia prediction model |
| CN111738980B (en) * | 2020-05-14 | 2023-08-04 | 北京深睿博联科技有限责任公司 | Medical image display method, computer equipment and storage medium |
| CN111739645A (en) * | 2020-05-14 | 2020-10-02 | 上海依智医疗技术有限公司 | Training method of immune-related pneumonia prediction model |
| CN111751263A (en) * | 2020-06-08 | 2020-10-09 | 西安交通大学 | Mark-free cell two-dimensional scattering image inversion method based on gray level co-occurrence matrix |
| CN112001896A (en) * | 2020-08-03 | 2020-11-27 | 什维新智医疗科技(上海)有限公司 | Thyroid gland border irregularity detection device |
| CN112184658A (en) * | 2020-09-24 | 2021-01-05 | 上海健康医学院 | Method, medium, and electronic device for predicting non-small cell lung cancer prognostic survival |
| WO2022063200A1 (en) * | 2020-09-24 | 2022-03-31 | 上海健康医学院 | Non-small cell lung cancer prognosis survival prediction method, medium and electronic device |
| CN112184658B (en) * | 2020-09-24 | 2023-11-24 | 上海健康医学院 | Method, medium and electronic equipment for prognosis survival prediction of non-small cell lung cancer |
| CN112807008A (en) * | 2021-01-27 | 2021-05-18 | 山东大学齐鲁医院 | Method and system for identifying actual mycoplasma pneumoniae and streptococcus pneumoniae of children based on imaging omics |
| CN113361944B (en) * | 2021-06-21 | 2022-04-22 | 鑫安利中（北京）科技有限公司 | Safe production index abnormity rapid sensing method based on Internet of things and artificial intelligence |
| CN113361944A (en) * | 2021-06-21 | 2021-09-07 | 鑫安利中（北京）科技有限公司 | Safe production index abnormity rapid sensing method based on Internet of things and artificial intelligence |
| CN113256614A (en) * | 2021-06-22 | 2021-08-13 | 国家超级计算天津中心 | Medical image processing system |
| CN113516669B (en) * | 2021-06-23 | 2023-04-25 | 湖北英库科技有限公司 | CT image-based trachea extraction method, device, equipment and storage medium |
| CN113516669A (en) * | 2021-06-23 | 2021-10-19 | 湖北英库科技有限公司 | Trachea extraction method, device and equipment based on CT image and storage medium |
| CN114438017A (en) * | 2022-02-18 | 2022-05-06 | 中山大学附属第一医院 | Correlation analysis method of endothelial progenitor cell function and cardiac function of sepsis patient and application of correlation analysis method in diagnosis of sepsis cardiomyopathy |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109658411A (en) | A kind of correlation analysis based on CT images feature Yu Patients with Non-small-cell Lung prognosis situation | |
| CN109034045A (en) | A kind of leucocyte automatic identifying method based on convolutional neural networks | |
| Beevi et al. | A multi-classifier system for automatic mitosis detection in breast histopathology images using deep belief networks | |
| Hariraj et al. | Fuzzy multi-layer SVM classification of breast cancer mammogram images | |
| Hussain et al. | A comparison of SVM kernel functions for breast cancer detection | |
| Kharrat et al. | A hybrid approach for automatic classification of brain MRI using genetic algorithm and support vector machine | |
| Song et al. | Mammographic classification based on XGBoost and DCNN with multi features | |
| Abraham et al. | Automated grading of prostate cancer using convolutional neural network and ordinal class classifier | |
| CN104751178A (en) | Pulmonary nodule detection device and method based on shape template matching and combining classifier | |
| KR102058348B1 (en) | Apparatus and method for classification of angiomyolipoma wihtout visible fat and clear cell renal cell carcinoma in ct images using deep learning and sahpe features | |
| CN109887600A (en) | A kind of analysis method of pair of non-small cell lung cancer prognosis Survival | |
| CN109978880A (en) | Lung tumors CT image is carried out sentencing method for distinguishing using high dimensional feature selection | |
| CN111767952A (en) | Interpretable classification method for benign and malignant pulmonary nodules | |
| Li et al. | Mass classification in mammograms based on two-concentric masks and discriminating texton | |
| CN101551854A (en) | A processing system of unbalanced medical image and processing method thereof | |
| Liu et al. | Deep multiscale multi-instance networks with regional scoring for mammogram classification | |
| Liu et al. | Automated classification of cervical Lymph-Node-Level from ultrasound using depthwise separable convolutional swin transformer | |
| Wei et al. | Multi-feature fusion for ultrasound breast image classification of benign and malignant | |
| Lan et al. | A hybrid classifier for mammography cad | |
| Albashish et al. | Multi-scoring feature selection method based on SVM-RFE for prostate cancer diagnosis | |
| CN116228759A (en) | Computer-aided diagnosis system and apparatus for renal cell carcinoma type | |
| Bhaskar et al. | Pulmonary lung nodule detection and classification through image enhancement and deep learning | |
| Xu et al. | A high-precision classification method of mammary cancer based on improved DenseNet driven by an attention mechanism | |
| Valliappan et al. | A theoretical methodology and prototype implementation for detection segmentation classification of digital mammogram tumor by machine learning and problem solving approach | |
| de Brito Silva et al. | Classification of breast masses in mammograms using geometric and topological feature maps and shape distribution |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190419 |