CN107016665A - A kind of CT pulmonary nodule detection methods based on depth convolutional neural networks - Google Patents

A kind of CT pulmonary nodule detection methods based on depth convolutional neural networks Download PDF

Info

Publication number
CN107016665A
CN107016665A CN201710083606.XA CN201710083606A CN107016665A CN 107016665 A CN107016665 A CN 107016665A CN 201710083606 A CN201710083606 A CN 201710083606A CN 107016665 A CN107016665 A CN 107016665A
Authority
CN
China
Prior art keywords
sample
lung neoplasm
tubercle
dimensional
images
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
Application number
CN201710083606.XA
Other languages
Chinese (zh)
Other versions
CN107016665B (en
Inventor
金弘晟
李宗曜
童若锋
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.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
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 Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to CN201710083606.XA priority Critical patent/CN107016665B/en
Publication of CN107016665A publication Critical patent/CN107016665A/en
Application granted granted Critical
Publication of CN107016665B publication Critical patent/CN107016665B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • 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
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • 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

Abstract

The invention discloses a kind of CT pulmonary nodule detection methods based on depth convolutional neural networks, comprise the following steps：1) to CT image preprocessings, make pixel separation unified, picture contrast is unified；2) two-dimensional convolution neutral net U net are trained, Lung neoplasm segmentation figure picture are predicted, based on Lung neoplasm segmentation figure as recommended candidate tubercle；3) three dimensional depth residual error neutral net Resnet3D is trained, the true and false positive probability of Lung neoplasm is predicted, screens out false-positive nodule.The CT pulmonary nodule detection methods that the present invention is provided, have given full play to the advantage of deep learning, can be with more efficient, the more accurately automatic detection pulmonary nodule in CT images, and have stronger adaptability to medical big data.

Description

A kind of CT pulmonary nodule detection methods based on depth convolutional neural networks
Technical field
Depth convolutional Neural is based on the present invention relates to a kind of pulmonary nodule detection method for CT images, more particularly to one kind The CT pulmonary nodule detection methods of network.
Background technology
Lung cancer is the main cause of global cancer related mortality, and it is a kind of that inspection is carried out to High risk group using CT scan The effective means for finding the early stage of lung cancer, and such crowd's substantial amounts, the workload of image department doctor increased dramatically, therefore Computer-aided diagnosis is just played a very important role.
Currently, computer aided detection Lung neoplasm field is utilized in CT images, based on traditional statistical machine learning side Numerous studies work has been carried out in method, and achieves certain achievement.Detecting step is generally divided into two steps, and the first step is to recommend to wait Tubercle is selected, detects to there may be the region of tubercle in lung CT image, second step is to screen out false-positive nodule, in the first step Detect that obtained region is identified, judge whether suspection target therein is tubercle, false-positive nodule is reduced as far as possible, in addition also It may determine that whether tubercle occurs canceration etc..However, the method based on conventional statistics machine learning, extracts fixed in advance in the picture The texture morphological feature of justice, such as area, effective diameter, gradient, these features are not enough to represent tubercle exactly, caused The false positive Lung neoplasm quantity detected is still a lot.
In recent years, deep learning has attracted substantial amounts of research interest, and achieving conventional method in various fields can not reach Achievement.Similarly, deep learning is also proved to be maximally efficient means in medical image analysis field, current main flow Computer-aided detection system is the method for having used deep learning.Olaf Ronneberger et al. were proposed in 2015 “U-Net:Convolutional Networks for Biomedical Image Segmentation ", this method propose A kind of medical image cutting method based on convolutional neural networks, this method is applied in plurality of medical image segmentation task, Such as blood vessel segmentation, cell segmentation, good effect is achieved.In terms of pulmonary nodule detection method, Setio et al. was carried in 2016 " Pulmonary Nodule Detection in CT Images are gone out:False Positive Reduction Using Traditional method is combined by Multi-ViewConvolutional Networks ", this method with deep learning method, is being pushed away Recommend candidate nodule still to use conventional methods, a kind of two-dimensional convolution nerve net of various visual angles is devised when screening out false-positive nodule Network, achieves good Detection results.Qi et al. proposed " Multi-level Contextual 3D CNNs in 2016 For False Positive Reduction in Pulmonary Nodule Detection ", this method propose one kind Three dimensional convolution neutral net screens out false-positive nodule, reduces false positive Lung neoplasm recall rate.With two-dimensional convolution neutral net phase Than three-dimensional convolutional Neural neutral net can catch more spatial informations, can extract more rich characteristics of image, to a certain degree On reduce the recall rate of false positive Lung neoplasm.However, this method planned network structure is shallower, only three-layer coil lamination, it is necessary to The network model of multiple yardsticks is trained to be merged.And this method still using conventional method come recommended candidate tubercle, fail fully Utilize the advantage of deep learning.The present invention is the further accuracy and robust for improving computer aided detection CT Lung neoplasm methods Property, it is proposed that a kind of method based on depth convolutional neural networks detects Lung neoplasm in CT images.
The content of the invention
Present invention technical problem of interest is：How to be detected automatically, efficiently and accurately in CT images using computer Lung neoplasm.
In order to solve the above technical problems, detecting lung knot in CT using depth convolutional neural networks the invention provides one kind The method of section, it gives full play in recommended candidate tubercle and the method for employing deep learning when screening out false-positive nodule The advantage of deep learning method, and can be while ensureing that false-positive nodule recall rate is lower, it is ensured that there is higher look into tubercle Full rate.It specifically includes following steps：
(1) to CT image preprocessings, make whole CT image pixels intervals unified, picture contrast is unified；
(2) two-dimensional convolution neutral net U-net is trained, Lung neoplasm segmentation figure picture is predicted, is pushed away based on Lung neoplasm segmentation figure picture Recommend candidate nodule；
(3) three dimensional depth residual error neutral net Resnet3D is trained, the true and false positive probability of Lung neoplasm is predicted, false sun is screened out Property tubercle.
Wherein, step (1) is to CT image preprocessings
The CT images that different instrument varying environments are collected pixel separation (between i.e. adjacent two pixel it is actual away from From), picture contrast aspect is very different.This step is pre-processed by linear interpolation, the normalized means of numerical value CT images, to obtain the CT body data that each side is consistent.The spatial information of CT images and strong in the step of this causes subsequent detection Degree information is consistent, and ensure that follow-up machine learning step can extract useful feature, obtains more preferable effect；
Its specific method is：
1) unified pixel interval
The pixel spacing information of whole CT images is counted first, and the diameter information of tubercle formulates unified pixel separation, so Operation is zoomed in and out to original CT image by way of linear interpolation afterwards so that the pixel separation system of all CT images One；
2) unified CT picture contrasts
The average for calculating the i-th width CT images is Meani, standard deviation is Stdi；Pass through the pixel to each pixel in CT images Value IxyzIt is normalized by such as following formula (1), so that the contrast of unified CT images；
Ixyz=(Ixyz-Meani)/Stdi (1)。
It is specially in step (2)：The Standard Segmentation image of recommended candidate tubercle step training sample is first generated as label, Two-dimensional convolution neutral net U-net is trained, two-dimensional convolution neutral net U-net network model is obtained.Further according to two Dimension convolutional neural networks U-net network model is predicted acquisition Lung neoplasm segmentation figure picture to test sample, to Lung neoplasm point Cut image and carry out binary conversion treatment, it is background or tubercle to distinguish pixel, and binarization segmentation image is grasped using morphological erosion Make reduction noise, and calculate the barycentric coodinates in the three-dimensional communication region of tubercle pixel, the center for the candidate's Lung neoplasm as recommended, Merge the Lung neoplasm center that three dimensions Euclidean distance is less than 3cm, it is to avoid the repetition detection of same Lung neoplasm.
Step (3) is specifically adopted with the following method：
1) data prepare and pretreatment
The training sample for screening out false-positive nodule step is first obtained after recommended candidate tubercle using the method for step (2), Above-mentioned training sample and all test samples are copied to the 3-D view block of CT images according to the nodule center of recommendation, to instruction Practice the Lung neoplasm position that sample is marked according to specialist in these training sample original images, to each 3-D view block It is labeled, distinguishes true positives tubercle and false positive Lung neoplasm；True positives tubercle sample is extended；
2) network structure is built
Build in three dimensional depth residual error neural network structure Resnet3D, the structure using Three dimensional convolution layer, three-dimensional residual error Block, three-dimensional pond layer；
3) network model is trained
The classification of true and false positive nodule is trained based on constructed three dimensional depth residual error neural network structure Resnet3D Device, obtains three dimensional depth residual error neutral net Resnet3D network model；
4) pulmonary nodule is predicted
The network model obtained using training predicts the true and false positive nodule probability of test sample, according to set in advance general Rate threshold value distinguishes true and false positive nodule, records the position coordinates of true positives tubercle, is examined so as to obtain Lung neoplasm in each CT images Survey result.
In the concrete scheme of above-mentioned steps (3), further, step 1) in true positives tubercle sample is extended can be with Operated using translation, scaling and the horizontal rotation made at random to the 3-D view block of original sample several times, obtain several new 3-D view block, i.e., new true positives nodule image block expands training data.
Further, step 2) described in three-dimensional pond layer in can use P＆C (Pooling+Cropping) pond Layer, described P＆C ponds layer refers to that the output simultaneously to last layer carries out pond (Pooling) and cuts (Cropping) behaviour Make, and the Chi Huayu results for cutting operation are connected in passage dimension, be used as next layer of input；Described pondization operation Refer to that the down-sampled maximum pond for original 1/8 size will be inputted, the described operation that cuts is intercepted in the characteristic pattern of input Go out 1/8 part of middle.
Further, step 3) in network model training can be based on stochastic gradient descent algorithm, and can be first using random Gradient descent algorithm carries out pre-training, then using the optimisation strategy of the difficult sample training of online dynamic select, i.e., each iteration has N Test the loss of N number of sample before individual sample, iteration with current network first, sorted, take out K maximum sample of loss, This K sample losses function back-propagation is updated to the weights of network.
Further, step 4) in sample predictions can to test sample make several times translate, scale and level revolve The extended operation turned, is predicted to extending obtained sample standard deviation every time, and the arithmetic predicted the outcome of all extension samples is put down Average predicts the outcome as final sample.The invention has the advantages that：
(1) a set of method that Lung neoplasm in CT is detected using depth convolutional neural networks is provided, computer can be made more Efficiently and accurately aid in detecting Lung neoplasm in CT images；
(2) a kind of P＆C (Pooling+Cropping) new pond Rotating fields are provided, wherein Pooling operations are obtained The global information in characteristic pattern is taken, and Cropping operations obtain the central tubercle information in characteristic pattern.The different scale of generation Characteristic pattern can help to catch the feature of multiple dimensioned Lung neoplasm, obtain more accurately Detection results.The structure causes not simultaneously With all convolutional calculations before have shared between the tubercle of yardstick, the side proposed compared to Qi mentioned in background technology et al. Method, separately handles multi-Scale Data, multiple models is trained, with higher efficiency.
(3) a kind of optimisation strategy of the difficult sample training of online dynamic select is provided, most of be easier to dynamically is eliminated The sample of differentiation, greatly accelerates the efficiency of network training.Simultaneously so that the sample of more difficult classification has obtained sufficient training, Improve the predictive ability of model.
(4) said in nature from method, due to the method that the present invention employs deep learning completely, involved deep learning Network structure can learn characteristics of image automatically from substantial amounts of CT images, there is stronger adaptability to medical big data, can be with Obtain more accurately Detection results.
Brief description of the drawings
Fig. 1 is the flow chart of method provided by the present invention；
Fig. 2 is the schematic network structure of depth two-dimensional convolution neutral net U-Net in recommended candidate tubercle of the present invention；
Fig. 3 is the network structure signal that the present invention screens out depth Three dimensional convolution neutral net Resnet3D in false-positive nodule Figure；
Fig. 4 is Lung neoplasm testing result (FROC curves) schematic diagram of embodiment of the present invention；
Fig. 5 is Lung neoplasm testing result disclosed in Setio et al. methods (FROC curves) schematic diagram；
Fig. 6 is Lung neoplasm testing result disclosed in Qi et al. methods (FROC curves) schematic diagram.
Embodiment
With reference to the accompanying drawings and examples, the embodiment to the present invention is described in further detail.Implement below Example is used to illustrate the present invention, but is not limited to the scope of the present invention：
Data used come from LIDC/IDRI databases in specific experiment of the present invention, include the CT figures of 888 patients Picture, altogether comprising 1086 tubercles.The data are that 4 experienced breast radiation section doctors carry out two in the CT images to the patient It is blind mark for the first time obtained by secondary image labeling, is to refer to other doctor's correction results for the second time.But, this 888 CT images Collected by different instruments, the pixel separation of CT images is different, and excursion is larger, the pixel separation of Z axis dimension Scope is 0.45-2.50mm, and X, the pixel separation scope of Y-axis dimension are 0.46-0.97mm.Meanwhile, the HU models of these CT images Enclose and contrast is also different.
It is the processing of the CT pulmonary nodule detection methods provided by the present invention based on depth convolutional neural networks as shown in Figure 1 Flow chart.Comprise the following steps：
1.CT image preprocessings
This step is by pre-processing the CT images that different instrument varying environments are collected, to obtain the CT that each side is consistent Volume data so that the machine learning method in the step of subsequent detection can obtain more preferable effect, specifically include following sub-step：
1.1st, unified pixel interval
The pixel spacing information of whole CT images is counted first, and the diameter information of tubercle is formulated unified pixel separation, made Its with original image is unlikely differs too far, while ensureing there is suitable length in pixels scope in any dimension superior thyroid tubercle.Then lead to Cross the mode of linear interpolation to zoom in and out original CT image operation, to cause the pixel separation of all CT images to unite One.So the spatial information of tubercle can be avoided inconsistent, among the forward-propagating process of network, it is to avoid the feature letter of tubercle Breath produces fuzzy, the accuracy of influence subsequent step.Pixel separation employed in specific experiment of the present invention：X, Y dimension is 0.75mm, Z-dimension is 1.25mm.
1.2nd, unified CT picture contrasts
The average for calculating the i-th width CT images is Meani, standard deviation is Stdi.Pass through the pixel to each pixel in CT images Value IxyzIt is normalized by such as following formula (1), so that the contrast of unified CT images.The signal intensity of tubercle can so be avoided It is inconsistent, have influence on the extraction of the characteristic information of tubercle.
Ixyz=(Ixyz-Meani)/Stdi (1)
2. recommended candidate tubercle
This step is based on U-net two-dimensional convolutions neutral net as shown in Figure 2, trains U-net two-dimensional convolution nerve nets Network, predicts Lung neoplasm segmentation figure picture, based on Lung neoplasm segmentation figure as recommended candidate tubercle；Specifically include following sub-step：
2.1st, the Standard Segmentation image of training sample is generated
For recommended candidate tubercle step training sample, it is divided into cross section two dimensional image by Z axis, z layers each The pixel value I of pixelXy, zCalculated by formula (2), wherein PxyzThe pixel coordinate,For marked in the CT images n-th The pixel coordinate of individual tubercle, rnFor the tubercle radius.
2.2nd, U-net networks are trained
The two-dimensional convolution neutral net network structure U-net proposed using 01af Ronneberger et al., the network knot Structure in specific experiment of the present invention as shown in Fig. 2 add after every layer of convolutional layer Dropout layers to mitigate over-fitting.And it is straight Connect using the optimisation strategy of stochastic gradient descent, the training method of cross validation and obtain its network model to train, and predict CT Every layer of Lung neoplasm segmentation figure picture in volume data.
2.3rd, candidate nodule center is extracted
First, acquisition Lung neoplasm is predicted to test sample according to two-dimensional convolution neutral net U-net network model Segmentation figure picture, binary conversion treatment is carried out to Lung neoplasm segmentation figure picture, and 0 represents that the pixel is background, and 1 represents that the pixel belongs to knot Section.Then, to binarization segmentation image using morphological erosion operation reduce noise, and calculate each pixel value be 1 three Tie up the barycentric coodinates of connected region, the center for the candidate's Lung neoplasm as recommended.Finally, three dimensions distance is merged too near (logical Three dimensions Euclidean distance can often be set less than 3cm) Lung neoplasm center, it is to avoid the repetition detection of same Lung neoplasm.
3. screen out false-positive nodule
False-positive nodule in the candidate nodule of recommendation is more, and this step uses three dimensional depth residual error neutral net Resnet3D mode obtains the feature of Lung neoplasm, and then screens out false-positive nodule.Specifically include following sub-step：
3.1st, data prepare and pretreatment
The nodule center recommended in step 2 is used to the training sample for screening out false-positive nodule step, according to the knot of recommendation Section center copies above-mentioned training sample and all test samples the 3-D view block (image patch) of CT images, this hair The tile size used in bright specific experiment is 44x44x28 pixel.For training sample, according to specialist original The Lung neoplasm position marked in image, represents true positives tubercle, 0 represents false sun to be labeled to each 3-D view block, 1 Property Lung neoplasm.In order that obtaining positive and negative sample equilibrium in training data, and in order to be able to obtain more preferable classifying quality, extend true positives Tubercle sample data, the i.e. random 3-D view block to original sample are made translation small several times, scaling and horizontal rotation and operated, Several new 3-D view blocks are obtained, i.e., new true positives nodule image block expands training data.
3.2nd, network structure is built
Build three dimensional depth residual error neural network structure Resnet3D as shown in Figure 3 to screen false-positive nodule, the knot The convolutional layer of three-dimensional is employed in structure, the convolution compared to ordinary two dimensional can obtain more abundant spatial information, and network can be carried Take out more representative feature.The structure has used for reference the thought of existing two-dimentional residual error network, devises three-dimensional residual error Block, depth residual error network often has more preferable data expression capability compared to common convolutional neural networks.Designed in the structure Unique P＆C (Pooling+Cropping) pond layer is to merge multiple dimensioned information, i.e., while the output to last layer is entered Tieed up with cutting (Cropping) operation, the result that pondization is operated with the result for cutting operation in passage in row pond (Pooling) On connect, be used as next layer of input.Pondization operation is will to input the down-sampled maximum pond for original 1/8 size, is cut out It is 1/8 part that middle is intercepted out in the characteristic pattern of input to cut operation.Because Three dimensional convolution neutral net is to computing resource It is required that higher, Considering experimental data are with computing resource, setting the three-dimensional residual block between two layers of pond layer in this example Number N=3, employs 27 layers of three dimensional depth residual error neutral net.
3.3rd, network model is trained
True and false positive nodule is trained based on constructed three dimensional depth residual error neural network structure Resnet3D in 3.2 Grader, obtains three dimensional depth residual error neutral net Resnet3D network model.Network training is calculated based on stochastic gradient descent Method, it is using the optimisation strategy for taking the difficult sample training of online dynamic select, i.e., each on the basis of stochastic gradient descent algorithm Iteration has the loss for testing N number of sample before N number of sample, iteration with current network first, is sorted, and takes out the maximum K of loss Individual sample, this K sample losses function back-propagation is updated the weights of network.
This optimisation strategy can increase the training difficulty of network to a certain extent, to ensure that network convergence is normal, Ke Yijin One step uses a kind of pre-training skill, i.e., first directly use stochastic gradient descent algorithm training network, now the convergence speed of network Degree is faster, after certain iterations, continues to train the network using above-mentioned optimisation strategy, now network can be more easy to instruction Practice, convergence can be guaranteed.
3.4th, pulmonary nodule is predicted
The network model obtained using training in 3.3 predicts the true and false positive nodule probability of test sample, according to setting in advance Fixed probability threshold value distinguishes true and false positive nodule, the position coordinates of true positives tubercle is recorded, so as to obtain Lung neoplasm in each CT Testing result.In order to obtain predicting the outcome for more robust, an optimization method is further also used in sample predictions, i.e., Make translation small several times, scaling and horizontal rotation extended operation to test sample, enter to extending obtained sample standard deviation every time Row prediction, the final sample predicts the outcome as the arithmetic mean of instantaneous values predicted the outcome of all extension samples.This method can be with Effectively reduce influence of the network over-fitting to result, with similar, both sides of merging the result of model of multiple identical networks Method can lift the accuracy finally predicted the outcome, but the extension test data time spent with repeatedly test is much smaller than The time for training multiple models to spend, thus the optimization method repeatedly tested can practicality with higher.
FROC (the free recipient's operating characteristic curve) curve synoptic diagrams of embodiment of the present invention are as shown in figure 4, can be with Find out, relative to Setio et al. using method and Qi of the two-dimensional convolution neutral net of various visual angles et al. using shallow three-dimensional volume Product neutral net method come improve the result for screening out false-positive nodule step and being obtained (such as Fig. 5, Fig. 6, the two be use with What identical data set of the present invention was carried out, referring specifically to the corresponding document of background section), the method using the present invention is obvious More accurately Detection results can be obtained.

Claims (9)

1. a kind of CT pulmonary nodule detection methods based on depth convolutional neural networks, it is characterised in that comprise the following steps：
(1) to CT image preprocessings, make whole CT image pixels intervals unified, picture contrast is unified；
(2) two-dimensional convolution neutral net U-net is trained, Lung neoplasm segmentation figure picture is predicted, is waited based on Lung neoplasm segmentation image recommendation Select tubercle；
(3) three dimensional depth residual error neutral net Resnet3D is trained, the true and false positive probability of Lung neoplasm is predicted, screens out false positive knot Section.
2. the CT pulmonary nodule detection methods as claimed in claim 1 based on depth convolutional neural networks, it is characterised in that step (1) following steps are specifically included in：
1) unified pixel interval
The pixel spacing information of whole CT images is counted first, and the diameter information of tubercle formulates unified pixel separation, Ran Houtong Cross the mode of linear interpolation to zoom in and out original CT image operation so that the pixel separation of all CT images is unified；
2) unified CT picture contrasts
The average for calculating the i-th width CT images is Meani, standard deviation is Stdi；Pass through the pixel value to each pixel in CT images IxyzIt is normalized by such as following formula (1), so that the contrast of unified CT images；
Ixyz=(Ixyz-Meani)/Stdi (1)
3. the CT pulmonary nodule detection methods according to claim 1 based on depth convolutional neural networks, it is characterised in that step Suddenly the Standard Segmentation image of recommended candidate tubercle step training sample is first generated in (2) as label, to two-dimensional convolution nerve net Network U-net is trained, and obtains two-dimensional convolution neutral net U-net network model.
4. the CT pulmonary nodule detection methods according to claim 3 based on depth convolutional neural networks, it is characterised in that step Suddenly Lung neoplasm segmentation figure is based in (2) as recommended candidate tubercle, its specific method is as follows：First according to two-dimensional convolution neutral net U- Net network model is predicted acquisition Lung neoplasm segmentation figure picture to test sample, and binaryzation is carried out to Lung neoplasm segmentation figure picture Processing, it is background or tubercle to distinguish pixel, to binarization segmentation image using morphological erosion operation reduction noise, and is calculated The barycentric coodinates in the three-dimensional communication region of tubercle pixel, the center for the candidate's Lung neoplasm as recommended merges three dimensions Euclidean Distance is less than 3cm Lung neoplasm center, it is to avoid the repetition detection of same Lung neoplasm.
5. the CT pulmonary nodule detection methods according to claim 1 based on depth convolutional neural networks, it is characterised in that step Suddenly (3) specific method is as follows：
1) data prepare and pretreatment
The training sample for screening out false-positive nodule step is first obtained after recommended candidate tubercle using the method for step (2), according to The nodule center of recommendation copies above-mentioned training sample and all test samples the 3-D view block of CT images, to training sample This Lung neoplasm position marked according to specialist in these training sample original images, is carried out to each 3-D view block Mark, distinguishes true positives tubercle and false positive Lung neoplasm；True positives tubercle sample is extended；
2) network structure is built
Build in three dimensional depth residual error neural network structure Resnet3D, the structure using Three dimensional convolution layer, three-dimensional residual block, three Wei Chiization layer；
3) network model is trained
The grader of true and false positive nodule is trained based on constructed three dimensional depth residual error neural network structure Resnet3D, obtained To three dimensional depth residual error neutral net Resnet3D network model；
4) pulmonary nodule is predicted
The network model obtained using training predicts the true and false positive nodule probability of test sample, according to probability threshold set in advance Value distinguishes true and false positive nodule, records the position coordinates of true positives tubercle, so as to obtain Lung neoplasm detection knot in each CT images Really.
6. the CT pulmonary nodule detection methods according to claim 5 based on depth convolutional neural networks, it is characterised in that step It is rapid 1) in true positives tubercle sample is extended translation several times, scaling is made using the random 3-D view block to original sample With horizontal rotation operation, several new 3-D view blocks are obtained, i.e., new true positives nodule image block expands training data.
7. the CT pulmonary nodule detection methods according to claim 5 based on depth convolutional neural networks, it is characterised in that step It is rapid 2) described in three-dimensional pond layer in use P＆C ponds layer, described P＆C ponds layer refers to while entering to the output of last layer Row Chi Huayu cuts operation, and the Chi Huayu results for cutting operation are connected in passage dimension, is used as next layer of input； Described pondization, which is operated, refers to that the down-sampled maximum pond for original 1/8 size will be inputted, and the described operation that cuts is in input Characteristic pattern in intercept out 1/8 part of middle.
8. the CT pulmonary nodule detection methods according to claim 5 based on depth convolutional neural networks, it is characterised in that step It is rapid 3) in network model training be based on stochastic gradient descent algorithm, pre-training is first carried out using stochastic gradient descent algorithm, then adopted With the optimisation strategy of the difficult sample training of online dynamic select, i.e., each iteration uses current network first before having N number of sample, iteration The loss of N number of sample is tested, is sorted, K maximum sample of loss is taken out, by this K sample losses function back-propagation Update the weights of network.
9. the CT pulmonary nodule detection methods according to claim 5 based on depth convolutional neural networks, it is characterised in that step It is rapid 4) in sample predictions to test sample make several times translate, scale and horizontal rotation extended operation, to every time extension Obtained sample standard deviation is predicted, using it is all extension samples the arithmetic mean of instantaneous values predicted the outcome as final sample prediction knot Really.
CN201710083606.XA 2017-02-16 2017-02-16 CT pulmonary nodule detection method based on deep convolutional neural network Active CN107016665B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710083606.XA CN107016665B (en) 2017-02-16 2017-02-16 CT pulmonary nodule detection method based on deep convolutional neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710083606.XA CN107016665B (en) 2017-02-16 2017-02-16 CT pulmonary nodule detection method based on deep convolutional neural network

Publications (2)

Publication Number Publication Date
CN107016665A true CN107016665A (en) 2017-08-04
CN107016665B CN107016665B (en) 2021-05-04

Family

ID=59440399

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710083606.XA Active CN107016665B (en) 2017-02-16 2017-02-16 CT pulmonary nodule detection method based on deep convolutional neural network

Country Status (1)

Country Link
CN (1) CN107016665B (en)

Cited By (129)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107247971A (en) * 2017-06-28 2017-10-13 中国人民解放军总医院 The intelligent analysis method and system of a kind of ultrasonic thyroid nodule risk indicator
CN107274406A (en) * 2017-08-07 2017-10-20 北京深睿博联科技有限责任公司 A kind of method and device of detection sensitizing range
CN107424152A (en) * 2017-08-11 2017-12-01 联想(北京)有限公司 The detection method and electronic equipment of organ lesion and the method and electronic equipment for training neuroid
CN107480665A (en) * 2017-08-09 2017-12-15 北京小米移动软件有限公司 Character detecting method, device and computer-readable recording medium
CN107507197A (en) * 2017-08-18 2017-12-22 东北大学 A kind of pulmonary parenchyma extracting method based on clustering algorithm and convolutional neural networks
CN107590443A (en) * 2017-08-23 2018-01-16 上海交通大学 Limiter stage live video automatic testing method and system based on the study of depth residual error
CN107610129A (en) * 2017-08-14 2018-01-19 四川大学 A kind of multi-modal nasopharyngeal carcinima joint dividing method based on CNN
CN107665736A (en) * 2017-09-30 2018-02-06 百度在线网络技术（北京）有限公司 Method and apparatus for generating information
CN107680088A (en) * 2017-09-30 2018-02-09 百度在线网络技术（北京）有限公司 Method and apparatus for analyzing medical image
CN107808377A (en) * 2017-10-31 2018-03-16 北京青燕祥云科技有限公司 The localization method and device of focus in a kind of lobe of the lung
CN107886510A (en) * 2017-11-27 2018-04-06 杭州电子科技大学 A kind of prostate MRI dividing methods based on three-dimensional full convolutional neural networks
CN107895369A (en) * 2017-11-28 2018-04-10 腾讯科技（深圳）有限公司 Image classification method, device, storage medium and equipment
CN107977963A (en) * 2017-11-30 2018-05-01 北京青燕祥云科技有限公司 Decision method, device and the realization device of Lung neoplasm
CN108038860A (en) * 2017-11-30 2018-05-15 杭州电子科技大学 Spine segmentation method based on the full convolutional neural networks of 3D
CN108078581A (en) * 2017-12-12 2018-05-29 北京青燕祥云科技有限公司 The good pernicious method of discrimination system of lung cancer and realization device based on convolutional neural networks
CN108090903A (en) * 2017-12-29 2018-05-29 苏州体素信息科技有限公司 Lung neoplasm detection model training method and device, pulmonary nodule detection method and device
CN108154164A (en) * 2017-11-15 2018-06-12 上海微波技术研究所（中国电子科技集团公司第五十研究所） Signal of communication modulation classification system and method based on deep learning
CN108154468A (en) * 2018-01-12 2018-06-12 平安科技（深圳）有限公司 Lung neoplasm detection method, application server and computer readable storage medium
CN108171694A (en) * 2017-12-28 2018-06-15 深圳开立生物医疗科技股份有限公司 Nodule detection methods, system and equipment based on convolutional neural networks
CN108257128A (en) * 2018-01-30 2018-07-06 浙江大学 A kind of method for building up of the Lung neoplasm detection device based on 3D convolutional neural networks
CN108257139A (en) * 2018-02-26 2018-07-06 中国科学院大学 RGB-D three-dimension object detection methods based on deep learning
CN108288271A (en) * 2018-02-06 2018-07-17 上海交通大学 Image detecting system and method based on three-dimensional residual error network
CN108389190A (en) * 2018-02-08 2018-08-10 贵州联科卫信科技有限公司 A kind of Lung neoplasm automatic testing method based on deep learning method
CN108389201A (en) * 2018-03-16 2018-08-10 北京推想科技有限公司 The good pernicious sorting technique of Lung neoplasm based on 3D convolutional neural networks and deep learning
CN108447062A (en) * 2018-02-01 2018-08-24 浙江大学 A kind of dividing method of the unconventional cell of pathological section based on multiple dimensioned mixing parted pattern
CN108470185A (en) * 2018-02-12 2018-08-31 北京佳格天地科技有限公司 The atural object annotation equipment and method of satellite image
CN108510502A (en) * 2018-03-08 2018-09-07 华南理工大学 Melanoma picture tissue segmentation methods based on deep neural network and system
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
CN108596247A (en) * 2018-04-23 2018-09-28 南方医科大学 A method of fusion radiation group and depth convolution feature carry out image classification
CN108615237A (en) * 2018-05-08 2018-10-02 上海商汤智能科技有限公司 A kind of method for processing lung images and image processing equipment
CN108629764A (en) * 2018-04-17 2018-10-09 杭州依图医疗技术有限公司 A kind of good pernicious method and device of determining Lung neoplasm
CN108648172A (en) * 2018-03-30 2018-10-12 四川元匠科技有限公司 A kind of CT figure Lung neoplasm detecting systems based on 3D-Unet
CN108647568A (en) * 2018-03-30 2018-10-12 电子科技大学 Grassland degeneration extraction method based on full convolutional neural networks
CN108648178A (en) * 2018-04-17 2018-10-12 杭州依图医疗技术有限公司 A kind of method and device of image nodule detection
CN108648192A (en) * 2018-05-17 2018-10-12 杭州依图医疗技术有限公司 A kind of method and device of detection tubercle
CN108670285A (en) * 2018-06-05 2018-10-19 胡晓云 A kind of CT pulmonary tuberculosis detection artificial intelligence diagnosis and therapy system
CN108764263A (en) * 2018-02-12 2018-11-06 北京佳格天地科技有限公司 The atural object annotation equipment and method of remote sensing image
CN108765369A (en) * 2018-04-20 2018-11-06 平安科技（深圳）有限公司 Detection method, device, computer equipment and the storage medium of Lung neoplasm
CN108742679A (en) * 2018-06-29 2018-11-06 上海联影医疗科技有限公司 Nodule detection device and method
CN108805209A (en) * 2018-06-14 2018-11-13 清华大学深圳研究生院 A kind of Lung neoplasm screening method based on deep learning
CN108846835A (en) * 2018-05-31 2018-11-20 西安电子科技大学 The image change detection method of convolutional network is separated based on depth
CN108876779A (en) * 2018-06-22 2018-11-23 中山仰视科技有限公司 Lung cancer method for early prediction, electronic equipment based on deep learning
CN108876772A (en) * 2018-06-05 2018-11-23 南华大学 A kind of Lung Cancer Images diagnostic system and method based on big data
CN108898595A (en) * 2018-06-27 2018-11-27 慧影医疗科技（北京）有限公司 A kind of construction method of thoracopathy detection model and application
CN108986085A (en) * 2018-06-28 2018-12-11 深圳视见医疗科技有限公司 CT image pulmonary nodule detection method, device, equipment and readable storage medium storing program for executing
CN108986067A (en) * 2018-05-25 2018-12-11 上海交通大学 Pulmonary nodule detection method based on cross-module state
CN109035234A (en) * 2018-07-25 2018-12-18 腾讯科技（深圳）有限公司 A kind of nodule detection methods, device and storage medium
CN109064477A (en) * 2018-07-06 2018-12-21 陕西师范大学 With the method for improved U-Net detection cell nuclear periphery
CN109086782A (en) * 2018-08-21 2018-12-25 广东工业大学 Feature Descriptor generation method, device, equipment and computer readable storage medium
CN109087703A (en) * 2018-08-24 2018-12-25 南京大学 Abdominal cavity CT image peritonaeum metastatic marker method based on depth convolutional neural networks
CN109102506A (en) * 2018-08-20 2018-12-28 东北大学 A kind of automatic division method carrying out abdominal CT hepatic disease image based on three-stage cascade network
CN109190682A (en) * 2018-08-13 2019-01-11 北京安德医智科技有限公司 A kind of classification method and equipment of the brain exception based on 3D nuclear magnetic resonance image
CN109191476A (en) * 2018-09-10 2019-01-11 重庆邮电大学 The automatic segmentation of Biomedical Image based on U-net network structure
CN109190481A (en) * 2018-08-06 2019-01-11 中国交通通信信息中心 A kind of remote sensing image road material extracting method and system
CN109215014A (en) * 2018-07-02 2019-01-15 中国科学院深圳先进技术研究院 Training method, device, equipment and the storage medium of CT image prediction model
CN109242962A (en) * 2018-09-28 2019-01-18 北京青燕祥云科技有限公司 The 3D display method, apparatus and electronic equipment of Lung neoplasm
CN109300136A (en) * 2018-08-28 2019-02-01 众安信息技术服务有限公司 It is a kind of to jeopardize organs automatic segmentation method based on convolutional neural networks
CN109308695A (en) * 2018-09-13 2019-02-05 镇江纳兰随思信息科技有限公司 Based on the cancer cell identification method for improving U-net convolutional neural networks model
CN109389584A (en) * 2018-09-17 2019-02-26 成都信息工程大学 Multiple dimensioned rhinopharyngeal neoplasm dividing method based on CNN
CN109410185A (en) * 2018-10-10 2019-03-01 腾讯科技（深圳）有限公司 A kind of image partition method, device and storage medium
CN109426773A (en) * 2017-08-24 2019-03-05 浙江宇视科技有限公司 A kind of roads recognition method and device
CN109447960A (en) * 2018-10-18 2019-03-08 神州数码医疗科技股份有限公司 A kind of object identifying method and device
CN109492547A (en) * 2018-10-24 2019-03-19 腾讯科技（深圳）有限公司 A kind of tubercle recognition methods, device and storage medium
CN109523547A (en) * 2018-12-21 2019-03-26 四川大学华西医院 A kind of method and device of image tubercle detection
CN109544528A (en) * 2018-11-19 2019-03-29 上海馨孚科技有限公司 A kind of small pulmonary artery image-recognizing method and device
CN109544545A (en) * 2018-11-30 2019-03-29 济南浪潮高新科技投资发展有限公司 A kind of salt mine intelligent detecting method and system based on convolutional neural networks
CN109544510A (en) * 2018-10-24 2019-03-29 广州大学 A kind of three-dimensional Lung neoplasm recognition methods based on convolutional neural networks
CN109584252A (en) * 2017-11-03 2019-04-05 杭州依图医疗技术有限公司 Lobe of the lung section dividing method, the device of CT images based on deep learning
CN109636788A (en) * 2018-12-11 2019-04-16 中国石油大学(华东) A kind of CT image gall stone intelligent measurement model based on deep neural network
CN109685768A (en) * 2018-11-28 2019-04-26 心医国际数字医疗系统(大连)有限公司 Lung neoplasm automatic testing method and system based on lung CT sequence
CN109829880A (en) * 2018-12-07 2019-05-31 清影医疗科技（深圳）有限公司 A kind of CT image detecting method based on deep learning, device and control equipment
CN109886307A (en) * 2019-01-24 2019-06-14 西安交通大学 A kind of image detecting method and system based on convolutional neural networks
CN109919961A (en) * 2019-02-22 2019-06-21 北京深睿博联科技有限责任公司 A kind of processing method and processing device for aneurysm region in encephalic CTA image
CN110008992A (en) * 2019-02-28 2019-07-12 合肥工业大学 A kind of deep learning method for prostate cancer auxiliary diagnosis
CN110021430A (en) * 2019-04-09 2019-07-16 科大讯飞股份有限公司 A kind of the attribute information prediction technique and device of lesion
CN110059697A (en) * 2019-04-29 2019-07-26 上海理工大学 A kind of Lung neoplasm automatic division method based on deep learning
CN110084810A (en) * 2019-05-06 2019-08-02 成都医云科技有限公司 A kind of Lung neoplasm image detecting method, model training method, device and storage medium
CN110084813A (en) * 2019-05-13 2019-08-02 山东大学 A kind of good pernicious prediction technique of Lung neoplasm based on three dimensional depth learning network
CN110119736A (en) * 2018-02-07 2019-08-13 浙江宇视科技有限公司 License plate location recognition method, device and electronic equipment
CN110123347A (en) * 2019-03-22 2019-08-16 杭州深睿博联科技有限公司 Image processing method and device for breast molybdenum target
CN110148114A (en) * 2019-04-02 2019-08-20 成都真实维度科技有限公司 A kind of deep learning model training method based on 2D faulted scanning pattern data set
CN110148467A (en) * 2019-05-16 2019-08-20 东北大学 A kind of Lung neoplasm device of computer aided diagnosis and method based on improvement CNN
CN110148112A (en) * 2019-04-02 2019-08-20 成都真实维度科技有限公司 A method of it acquires and marks the progress data set foundation of tomoscan diagram data
WO2019174376A1 (en) * 2018-03-14 2019-09-19 大连理工大学 Lung texture recognition method for extracting appearance and geometrical feature based on deep neural network
CN110309860A (en) * 2019-06-06 2019-10-08 昆明理工大学 The method classified based on grade malignancy of the convolutional neural networks to Lung neoplasm
CN110473168A (en) * 2019-07-09 2019-11-19 天津大学 A kind of Lung neoplasm automatic checkout system squeezing-motivate residual error network based on 3D
WO2019218774A1 (en) * 2018-05-14 2019-11-21 腾讯科技（深圳）有限公司 Object recognition method and device, and storage medium
CN110660045A (en) * 2019-08-30 2020-01-07 杭州电子科技大学 Lymph node identification semi-supervision method based on convolutional neural network
CN110689547A (en) * 2019-09-25 2020-01-14 重庆大学 Pulmonary nodule segmentation method based on three-dimensional CT image
CN110689112A (en) * 2019-09-02 2020-01-14 杭州深睿博联科技有限公司 Data processing method and device
CN110705570A (en) * 2019-09-20 2020-01-17 桂林电子科技大学 Image feature identification method
CN110706200A (en) * 2019-09-02 2020-01-17 杭州深睿博联科技有限公司 Data prediction method and device
CN110717916A (en) * 2019-09-29 2020-01-21 华中科技大学 Pulmonary embolism detection system based on convolutional neural network
CN110766713A (en) * 2019-10-30 2020-02-07 上海微创医疗器械（集团）有限公司 Lung image segmentation method and device and lung lesion region identification equipment
CN110766693A (en) * 2018-09-06 2020-02-07 北京连心医疗科技有限公司 Method for jointly predicting radiotherapy structure position based on multi-model neural network
CN110890143A (en) * 2019-11-21 2020-03-17 重庆邮电大学 2D convolution method introducing spatial information
CN110910404A (en) * 2019-11-18 2020-03-24 西南交通大学 Anti-noise data breast ultrasonic nodule segmentation method
CN110969623A (en) * 2020-02-28 2020-04-07 北京深睿博联科技有限责任公司 Lung CT multi-symptom automatic detection method, system, terminal and storage medium
CN111080625A (en) * 2019-12-18 2020-04-28 北京推想科技有限公司 Training method and training device for lung image strip and rope detection model
CN111079821A (en) * 2019-12-12 2020-04-28 哈尔滨市科佳通用机电股份有限公司 Derailment automatic braking pull ring falling fault image identification method
WO2020108009A1 (en) * 2018-11-26 2020-06-04 Guangdong Oppo Mobile Telecommunications Corp., Ltd. Method, system, and computer-readable medium for improving quality of low-light images
CN111310624A (en) * 2020-02-05 2020-06-19 腾讯科技（深圳）有限公司 Occlusion recognition method and device, computer equipment and storage medium
CN111340756A (en) * 2020-02-13 2020-06-26 北京深睿博联科技有限责任公司 Medical image lesion detection and combination method, system, terminal and storage medium
CN111340825A (en) * 2020-03-05 2020-06-26 上海市肺科医院（上海市职业病防治院） Method and system for generating mediastinal lymph node segmentation model
CN111476766A (en) * 2020-03-31 2020-07-31 哈尔滨商业大学 Lung nodule CT image detection system based on deep learning
CN111583271A (en) * 2020-04-13 2020-08-25 华东师范大学 Method for automatically predicting gene expression categories based on cancer CT images
CN111656372A (en) * 2017-11-20 2020-09-11 皇家飞利浦有限公司 Training neural network model
CN111768382A (en) * 2020-06-30 2020-10-13 重庆大学 Interactive segmentation method based on lung nodule growth form
CN111784593A (en) * 2020-06-04 2020-10-16 广东省智能制造研究所 Lung nodule CT image data enhancement method and system for deep learning
CN111815591A (en) * 2020-06-29 2020-10-23 郑州大学 Pulmonary nodule detection method based on CT image
CN112184657A (en) * 2020-09-24 2021-01-05 上海健康医学院 Pulmonary nodule automatic detection method, device and computer system
CN112241948A (en) * 2020-09-23 2021-01-19 深圳视见医疗科技有限公司 Lung nodule detection and analysis method and system capable of adapting to thickness of layer
WO2021027135A1 (en) * 2019-08-15 2021-02-18 平安科技（深圳）有限公司 Cell detection model training method and apparatus, computer device and storage medium
CN112400187A (en) * 2018-07-17 2021-02-23 国际商业机器公司 Knockout autoencoder for detecting anomalies in biomedical images
CN112529870A (en) * 2020-12-10 2021-03-19 重庆大学 Multi-scale CNNs (CNNs) lung nodule false positive elimination method based on combination of source domain and frequency domain
CN112541909A (en) * 2020-12-22 2021-03-23 南开大学 Lung nodule detection method and system based on three-dimensional neural network of slice perception
CN112581436A (en) * 2020-12-11 2021-03-30 佛山市普世医学科技有限责任公司 Lung nodule identification and segmentation method and system based on deep learning
CN112669284A (en) * 2020-12-29 2021-04-16 天津大学 Method for realizing pulmonary nodule detection by generating confrontation network
CN112669319A (en) * 2021-03-22 2021-04-16 四川大学 Multi-view multi-scale lymph node false positive inhibition modeling method
CN112819747A (en) * 2019-10-31 2021-05-18 深圳市华嘉生物智能科技有限公司 Method for automatically diagnosing benign and malignant nodules based on lung tomography image
CN112991269A (en) * 2021-02-07 2021-06-18 复旦大学 Identification and classification method for lung CT image
CN113113132A (en) * 2021-03-25 2021-07-13 华南理工大学 Method for diagnosing congenital megacolon and homologous diseases based on deep convolutional neural network
CN113168912A (en) * 2018-09-04 2021-07-23 艾登斯 Ip 有限公司 Determining growth rate of objects in 3D data sets using deep learning
CN113256634A (en) * 2021-07-13 2021-08-13 杭州医策科技有限公司 Cervical carcinoma TCT slice vagina arranging method and system based on deep learning
CN113658162A (en) * 2021-08-24 2021-11-16 中山仰视科技有限公司 Refined lung nodule detection method, system, equipment and medium
US11188773B2 (en) 2018-11-26 2021-11-30 Shanghai United Imaging Intelligence Co., Ltd. Systems and methods for detecting region of interset in image
TWI760657B (en) * 2019-11-14 2022-04-11 和碩聯合科技股份有限公司 Training method for anomaly detection model and electronic device using the same
CN114677383A (en) * 2022-03-03 2022-06-28 西北工业大学 Pulmonary nodule detection and segmentation method based on multi-task learning
CN116012355A (en) * 2023-02-07 2023-04-25 重庆大学 Adaptive false positive lung nodule removing method based on deep learning

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1462884A (en) * 2003-06-24 2003-12-24 南京大学 Method of recognizing image of lung cancer cells with high accuracy and low rate of false negative
CN102542562A (en) * 2011-11-23 2012-07-04 首都医科大学 Method for extracting texture based on lung nodules three orthogonality position computed tomography (CT) image and method for forecasting lung cancers
CN102842132A (en) * 2012-07-12 2012-12-26 上海联影医疗科技有限公司 CT pulmonary nodule detection method
CN104700118A (en) * 2015-03-18 2015-06-10 中国科学院自动化研究所 Pulmonary nodule benignity and malignancy predicting method based on convolutional neural networks
CN105005714A (en) * 2015-06-18 2015-10-28 中国科学院自动化研究所 Non-small cell lung cancer prognosis method based on tumor phenotypic characteristics
CN105160361A (en) * 2015-09-30 2015-12-16 东软集团股份有限公司 Image identification method and apparatus
CN105469419A (en) * 2016-01-07 2016-04-06 于翠妮 Pulmonary nodule detection method based on CT image
CN106203327A (en) * 2016-07-08 2016-12-07 清华大学 Lung tumor identification system and method based on convolutional neural networks
CN106372390A (en) * 2016-08-25 2017-02-01 姹ゅ钩 Deep convolutional neural network-based lung cancer preventing self-service health cloud service system

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1462884A (en) * 2003-06-24 2003-12-24 南京大学 Method of recognizing image of lung cancer cells with high accuracy and low rate of false negative
CN102542562A (en) * 2011-11-23 2012-07-04 首都医科大学 Method for extracting texture based on lung nodules three orthogonality position computed tomography (CT) image and method for forecasting lung cancers
CN102842132A (en) * 2012-07-12 2012-12-26 上海联影医疗科技有限公司 CT pulmonary nodule detection method
CN104700118A (en) * 2015-03-18 2015-06-10 中国科学院自动化研究所 Pulmonary nodule benignity and malignancy predicting method based on convolutional neural networks
CN105005714A (en) * 2015-06-18 2015-10-28 中国科学院自动化研究所 Non-small cell lung cancer prognosis method based on tumor phenotypic characteristics
CN105160361A (en) * 2015-09-30 2015-12-16 东软集团股份有限公司 Image identification method and apparatus
CN105469419A (en) * 2016-01-07 2016-04-06 于翠妮 Pulmonary nodule detection method based on CT image
CN106203327A (en) * 2016-07-08 2016-12-07 清华大学 Lung tumor identification system and method based on convolutional neural networks
CN106372390A (en) * 2016-08-25 2017-02-01 姹ゅ钩 Deep convolutional neural network-based lung cancer preventing self-service health cloud service system

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
OLAF RONNEBERGER ET AL: "U-Net: Convolutional Networks for Biomedical Image Segmentation", 《MICCAI 2015》 *
冬虫夏草: "Multi-level Contextual 3D CNNs for False Positive Reduction in Pulmonary Nodule Detection", 《HTTP://WWW.DOCIN.COM/P-1842833617.HTML》 *
金奇樑: "基于CT图像的肺结节自动识别系统研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 *

Cited By (189)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107247971B (en) * 2017-06-28 2020-10-09 中国人民解放军总医院 Intelligent analysis method and system for ultrasonic thyroid nodule risk index
CN107247971A (en) * 2017-06-28 2017-10-13 中国人民解放军总医院 The intelligent analysis method and system of a kind of ultrasonic thyroid nodule risk indicator
CN107274406A (en) * 2017-08-07 2017-10-20 北京深睿博联科技有限责任公司 A kind of method and device of detection sensitizing range
CN107480665A (en) * 2017-08-09 2017-12-15 北京小米移动软件有限公司 Character detecting method, device and computer-readable recording medium
CN107480665B (en) * 2017-08-09 2020-08-11 北京小米移动软件有限公司 Character detection method and device and computer readable storage medium
CN107424152A (en) * 2017-08-11 2017-12-01 联想(北京)有限公司 The detection method and electronic equipment of organ lesion and the method and electronic equipment for training neuroid
CN107610129A (en) * 2017-08-14 2018-01-19 四川大学 A kind of multi-modal nasopharyngeal carcinima joint dividing method based on CNN
CN107507197A (en) * 2017-08-18 2017-12-22 东北大学 A kind of pulmonary parenchyma extracting method based on clustering algorithm and convolutional neural networks
CN107507197B (en) * 2017-08-18 2020-11-24 东北大学 Lung parenchyma extraction method based on clustering algorithm and convolutional neural network
CN107590443A (en) * 2017-08-23 2018-01-16 上海交通大学 Limiter stage live video automatic testing method and system based on the study of depth residual error
CN109426773A (en) * 2017-08-24 2019-03-05 浙江宇视科技有限公司 A kind of roads recognition method and device
CN107680088A (en) * 2017-09-30 2018-02-09 百度在线网络技术（北京）有限公司 Method and apparatus for analyzing medical image
CN107665736B (en) * 2017-09-30 2021-05-25 百度在线网络技术（北京）有限公司 Method and apparatus for generating information
US10810735B2 (en) 2017-09-30 2020-10-20 Baidu Online Network Technology (Beijing) Co., Ltd. Method and apparatus for analyzing medical image
CN107665736A (en) * 2017-09-30 2018-02-06 百度在线网络技术（北京）有限公司 Method and apparatus for generating information
CN107808377A (en) * 2017-10-31 2018-03-16 北京青燕祥云科技有限公司 The localization method and device of focus in a kind of lobe of the lung
CN107808377B (en) * 2017-10-31 2019-02-12 北京青燕祥云科技有限公司 The positioning device of lesion in a kind of lobe of the lung
CN111709953B (en) * 2017-11-03 2023-04-07 杭州依图医疗技术有限公司 Output method and device in lung lobe segment segmentation of CT (computed tomography) image
CN109584252B (en) * 2017-11-03 2020-08-14 杭州依图医疗技术有限公司 Lung lobe segment segmentation method and device of CT image based on deep learning
CN111709953A (en) * 2017-11-03 2020-09-25 杭州依图医疗技术有限公司 Output method and device in lung lobe segment segmentation of CT (computed tomography) image
CN109584252A (en) * 2017-11-03 2019-04-05 杭州依图医疗技术有限公司 Lobe of the lung section dividing method, the device of CT images based on deep learning
CN108154164A (en) * 2017-11-15 2018-06-12 上海微波技术研究所（中国电子科技集团公司第五十研究所） Signal of communication modulation classification system and method based on deep learning
CN111656372A (en) * 2017-11-20 2020-09-11 皇家飞利浦有限公司 Training neural network model
CN107886510A (en) * 2017-11-27 2018-04-06 杭州电子科技大学 A kind of prostate MRI dividing methods based on three-dimensional full convolutional neural networks
CN107895369A (en) * 2017-11-28 2018-04-10 腾讯科技（深圳）有限公司 Image classification method, device, storage medium and equipment
CN107895369B (en) * 2017-11-28 2022-11-22 腾讯科技（深圳）有限公司 Image classification method, device, storage medium and equipment
CN108038860A (en) * 2017-11-30 2018-05-15 杭州电子科技大学 Spine segmentation method based on the full convolutional neural networks of 3D
CN107977963A (en) * 2017-11-30 2018-05-01 北京青燕祥云科技有限公司 Decision method, device and the realization device of Lung neoplasm
CN108078581A (en) * 2017-12-12 2018-05-29 北京青燕祥云科技有限公司 The good pernicious method of discrimination system of lung cancer and realization device based on convolutional neural networks
CN108078581B (en) * 2017-12-12 2019-02-12 北京青燕祥云科技有限公司 The good pernicious judgement system of lung cancer and realization device based on convolutional neural networks
CN108171694A (en) * 2017-12-28 2018-06-15 深圳开立生物医疗科技股份有限公司 Nodule detection methods, system and equipment based on convolutional neural networks
CN108090903A (en) * 2017-12-29 2018-05-29 苏州体素信息科技有限公司 Lung neoplasm detection model training method and device, pulmonary nodule detection method and device
CN108154468B (en) * 2018-01-12 2022-03-01 平安科技（深圳）有限公司 Pulmonary nodule detection method, application server and computer readable storage medium
CN108154468A (en) * 2018-01-12 2018-06-12 平安科技（深圳）有限公司 Lung neoplasm detection method, application server and computer readable storage medium
WO2019136922A1 (en) * 2018-01-12 2019-07-18 平安科技（深圳）有限公司 Pulmonary nodule detection method, application server, and computer-readable storage medium
CN108257128B (en) * 2018-01-30 2020-09-18 浙江大学 Establishment method of pulmonary nodule detection device based on 3D convolutional neural network
CN108257128A (en) * 2018-01-30 2018-07-06 浙江大学 A kind of method for building up of the Lung neoplasm detection device based on 3D convolutional neural networks
CN108447062B (en) * 2018-02-01 2021-04-20 浙江大学 Pathological section unconventional cell segmentation method based on multi-scale mixed segmentation model
CN108447062A (en) * 2018-02-01 2018-08-24 浙江大学 A kind of dividing method of the unconventional cell of pathological section based on multiple dimensioned mixing parted pattern
CN108288271A (en) * 2018-02-06 2018-07-17 上海交通大学 Image detecting system and method based on three-dimensional residual error network
CN110119736A (en) * 2018-02-07 2019-08-13 浙江宇视科技有限公司 License plate location recognition method, device and electronic equipment
CN110119736B (en) * 2018-02-07 2021-12-28 浙江宇视科技有限公司 License plate position identification method and device and electronic equipment
CN108389190A (en) * 2018-02-08 2018-08-10 贵州联科卫信科技有限公司 A kind of Lung neoplasm automatic testing method based on deep learning method
CN108389190B (en) * 2018-02-08 2022-12-16 贵州联科卫信科技有限公司 Pulmonary nodule automatic detection method based on deep learning method
CN108764263A (en) * 2018-02-12 2018-11-06 北京佳格天地科技有限公司 The atural object annotation equipment and method of remote sensing image
CN108470185A (en) * 2018-02-12 2018-08-31 北京佳格天地科技有限公司 The atural object annotation equipment and method of satellite image
CN108257139B (en) * 2018-02-26 2020-09-08 中国科学院大学 RGB-D three-dimensional object detection method based on deep learning
CN108257139A (en) * 2018-02-26 2018-07-06 中国科学院大学 RGB-D three-dimension object detection methods based on deep learning
CN108510502A (en) * 2018-03-08 2018-09-07 华南理工大学 Melanoma picture tissue segmentation methods based on deep neural network and system
CN108510502B (en) * 2018-03-08 2020-09-22 华南理工大学 Melanoma image tissue segmentation method and system based on deep neural network
WO2019174376A1 (en) * 2018-03-14 2019-09-19 大连理工大学 Lung texture recognition method for extracting appearance and geometrical feature based on deep neural network
US11170502B2 (en) 2018-03-14 2021-11-09 Dalian University Of Technology Method based on deep neural network to extract appearance and geometry features for pulmonary textures classification
CN108389201A (en) * 2018-03-16 2018-08-10 北京推想科技有限公司 The good pernicious sorting technique of Lung neoplasm based on 3D convolutional neural networks and deep learning
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
CN108648172A (en) * 2018-03-30 2018-10-12 四川元匠科技有限公司 A kind of CT figure Lung neoplasm detecting systems based on 3D-Unet
CN108647568A (en) * 2018-03-30 2018-10-12 电子科技大学 Grassland degeneration extraction method based on full convolutional neural networks
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
WO2019201204A1 (en) * 2018-04-17 2019-10-24 杭州依图医疗技术有限公司 Method and device for detecting nodule in image
CN108648178A (en) * 2018-04-17 2018-10-12 杭州依图医疗技术有限公司 A kind of method and device of image nodule detection
CN108629764A (en) * 2018-04-17 2018-10-09 杭州依图医疗技术有限公司 A kind of good pernicious method and device of determining Lung neoplasm
CN108765369A (en) * 2018-04-20 2018-11-06 平安科技（深圳）有限公司 Detection method, device, computer equipment and the storage medium of Lung neoplasm
WO2019200740A1 (en) * 2018-04-20 2019-10-24 平安科技（深圳）有限公司 Pulmonary nodule detection method and apparatus, computer device, and storage medium
CN108596247A (en) * 2018-04-23 2018-09-28 南方医科大学 A method of fusion radiation group and depth convolution feature carry out image classification
CN108615237A (en) * 2018-05-08 2018-10-02 上海商汤智能科技有限公司 A kind of method for processing lung images and image processing equipment
CN108615237B (en) * 2018-05-08 2021-09-07 上海商汤智能科技有限公司 Lung image processing method and image processing equipment
WO2019218774A1 (en) * 2018-05-14 2019-11-21 腾讯科技（深圳）有限公司 Object recognition method and device, and storage medium
US11488302B2 (en) 2018-05-14 2022-11-01 Tencent Technology (Shenzhen) Company Limited Object recognition method and device, and storage medium
CN108648192A (en) * 2018-05-17 2018-10-12 杭州依图医疗技术有限公司 A kind of method and device of detection tubercle
CN108986067B (en) * 2018-05-25 2020-08-14 上海交通大学 Cross-modality-based pulmonary nodule detection method
CN108986067A (en) * 2018-05-25 2018-12-11 上海交通大学 Pulmonary nodule detection method based on cross-module state
CN108846835B (en) * 2018-05-31 2020-04-14 西安电子科技大学 Image change detection method based on depth separable convolutional network
CN108846835A (en) * 2018-05-31 2018-11-20 西安电子科技大学 The image change detection method of convolutional network is separated based on depth
CN108670285A (en) * 2018-06-05 2018-10-19 胡晓云 A kind of CT pulmonary tuberculosis detection artificial intelligence diagnosis and therapy system
CN108876772A (en) * 2018-06-05 2018-11-23 南华大学 A kind of Lung Cancer Images diagnostic system and method based on big data
CN108805209B (en) * 2018-06-14 2021-11-19 清华大学深圳研究生院 Lung nodule screening method based on deep learning
CN108805209A (en) * 2018-06-14 2018-11-13 清华大学深圳研究生院 A kind of Lung neoplasm screening method based on deep learning
CN108876779A (en) * 2018-06-22 2018-11-23 中山仰视科技有限公司 Lung cancer method for early prediction, electronic equipment based on deep learning
CN108898595B (en) * 2018-06-27 2021-02-19 慧影医疗科技（北京）有限公司 Construction method and application of positioning model of focus region in chest image
CN108898595A (en) * 2018-06-27 2018-11-27 慧影医疗科技（北京）有限公司 A kind of construction method of thoracopathy detection model and application
CN108986085A (en) * 2018-06-28 2018-12-11 深圳视见医疗科技有限公司 CT image pulmonary nodule detection method, device, equipment and readable storage medium storing program for executing
CN108986085B (en) * 2018-06-28 2021-06-01 深圳视见医疗科技有限公司 CT image pulmonary nodule detection method, device and equipment and readable storage medium
CN108742679A (en) * 2018-06-29 2018-11-06 上海联影医疗科技有限公司 Nodule detection device and method
CN109215014A (en) * 2018-07-02 2019-01-15 中国科学院深圳先进技术研究院 Training method, device, equipment and the storage medium of CT image prediction model
CN109215014B (en) * 2018-07-02 2022-03-04 中国科学院深圳先进技术研究院 Training method, device and equipment of CT image prediction model and storage medium
CN109064477A (en) * 2018-07-06 2018-12-21 陕西师范大学 With the method for improved U-Net detection cell nuclear periphery
CN112400187A (en) * 2018-07-17 2021-02-23 国际商业机器公司 Knockout autoencoder for detecting anomalies in biomedical images
CN109035234A (en) * 2018-07-25 2018-12-18 腾讯科技（深圳）有限公司 A kind of nodule detection methods, device and storage medium
CN109035234B (en) * 2018-07-25 2020-12-01 腾讯科技（深圳）有限公司 Nodule detection method, device and storage medium
CN109190481A (en) * 2018-08-06 2019-01-11 中国交通通信信息中心 A kind of remote sensing image road material extracting method and system
CN109190481B (en) * 2018-08-06 2021-11-23 中国交通通信信息中心 Method and system for extracting road material of remote sensing image
CN109190682A (en) * 2018-08-13 2019-01-11 北京安德医智科技有限公司 A kind of classification method and equipment of the brain exception based on 3D nuclear magnetic resonance image
CN109102506B (en) * 2018-08-20 2021-08-13 东北大学 Automatic segmentation method for abdominal CT liver lesion image based on three-level cascade network
CN109102506A (en) * 2018-08-20 2018-12-28 东北大学 A kind of automatic division method carrying out abdominal CT hepatic disease image based on three-stage cascade network
CN109086782A (en) * 2018-08-21 2018-12-25 广东工业大学 Feature Descriptor generation method, device, equipment and computer readable storage medium
CN109087703B (en) * 2018-08-24 2022-06-07 南京大学 Peritoneal transfer marking method of abdominal cavity CT image based on deep convolutional neural network
CN109087703A (en) * 2018-08-24 2018-12-25 南京大学 Abdominal cavity CT image peritonaeum metastatic marker method based on depth convolutional neural networks
CN109300136A (en) * 2018-08-28 2019-02-01 众安信息技术服务有限公司 It is a kind of to jeopardize organs automatic segmentation method based on convolutional neural networks
CN113168912B (en) * 2018-09-04 2023-12-01 艾登斯 Ip 有限公司 Determining growth rate of objects in 3D dataset using deep learning
CN113168912A (en) * 2018-09-04 2021-07-23 艾登斯 Ip 有限公司 Determining growth rate of objects in 3D data sets using deep learning
CN110766693B (en) * 2018-09-06 2022-06-21 北京连心医疗科技有限公司 Method for jointly predicting radiotherapy structure position based on multi-model neural network
CN110766693A (en) * 2018-09-06 2020-02-07 北京连心医疗科技有限公司 Method for jointly predicting radiotherapy structure position based on multi-model neural network
CN109191476A (en) * 2018-09-10 2019-01-11 重庆邮电大学 The automatic segmentation of Biomedical Image based on U-net network structure
CN109191476B (en) * 2018-09-10 2022-03-11 重庆邮电大学 Novel biomedical image automatic segmentation method based on U-net network structure
CN109308695A (en) * 2018-09-13 2019-02-05 镇江纳兰随思信息科技有限公司 Based on the cancer cell identification method for improving U-net convolutional neural networks model
CN109389584A (en) * 2018-09-17 2019-02-26 成都信息工程大学 Multiple dimensioned rhinopharyngeal neoplasm dividing method based on CNN
CN109242962A (en) * 2018-09-28 2019-01-18 北京青燕祥云科技有限公司 The 3D display method, apparatus and electronic equipment of Lung neoplasm
CN109410185A (en) * 2018-10-10 2019-03-01 腾讯科技（深圳）有限公司 A kind of image partition method, device and storage medium
CN109447960A (en) * 2018-10-18 2019-03-08 神州数码医疗科技股份有限公司 A kind of object identifying method and device
CN109544510A (en) * 2018-10-24 2019-03-29 广州大学 A kind of three-dimensional Lung neoplasm recognition methods based on convolutional neural networks
CN109492547A (en) * 2018-10-24 2019-03-19 腾讯科技（深圳）有限公司 A kind of tubercle recognition methods, device and storage medium
CN109544510B (en) * 2018-10-24 2021-10-26 广州大学 Three-dimensional lung nodule identification method based on convolutional neural network
CN109492547B (en) * 2018-10-24 2022-03-08 腾讯科技（深圳）有限公司 Nodule identification method and device and storage medium
CN109544528A (en) * 2018-11-19 2019-03-29 上海馨孚科技有限公司 A kind of small pulmonary artery image-recognizing method and device
US11694086B2 (en) 2018-11-26 2023-07-04 Shanghai United Imaging Intelligence Co., Ltd. Systems and methods for detecting region of interset in image
WO2020108009A1 (en) * 2018-11-26 2020-06-04 Guangdong Oppo Mobile Telecommunications Corp., Ltd. Method, system, and computer-readable medium for improving quality of low-light images
US11188773B2 (en) 2018-11-26 2021-11-30 Shanghai United Imaging Intelligence Co., Ltd. Systems and methods for detecting region of interset in image
US20210272246A1 (en) * 2018-11-26 2021-09-02 Guangdong Oppo Mobile Telecommunications Corp., Ltd. Method, system, and computer-readable medium for improving quality of low-light images
US11741578B2 (en) 2018-11-26 2023-08-29 Guangdong Oppo Mobile Telecommunications Corp., Ltd. Method, system, and computer-readable medium for improving quality of low-light images
CN109685768A (en) * 2018-11-28 2019-04-26 心医国际数字医疗系统(大连)有限公司 Lung neoplasm automatic testing method and system based on lung CT sequence
CN109544545A (en) * 2018-11-30 2019-03-29 济南浪潮高新科技投资发展有限公司 A kind of salt mine intelligent detecting method and system based on convolutional neural networks
CN109829880A (en) * 2018-12-07 2019-05-31 清影医疗科技（深圳）有限公司 A kind of CT image detecting method based on deep learning, device and control equipment
CN109636788A (en) * 2018-12-11 2019-04-16 中国石油大学(华东) A kind of CT image gall stone intelligent measurement model based on deep neural network
CN109523547A (en) * 2018-12-21 2019-03-26 四川大学华西医院 A kind of method and device of image tubercle detection
CN109886307A (en) * 2019-01-24 2019-06-14 西安交通大学 A kind of image detecting method and system based on convolutional neural networks
CN109919961A (en) * 2019-02-22 2019-06-21 北京深睿博联科技有限责任公司 A kind of processing method and processing device for aneurysm region in encephalic CTA image
CN110008992B (en) * 2019-02-28 2023-04-11 合肥工业大学 Deep learning method for prostate cancer auxiliary diagnosis
CN110008992A (en) * 2019-02-28 2019-07-12 合肥工业大学 A kind of deep learning method for prostate cancer auxiliary diagnosis
CN110123347A (en) * 2019-03-22 2019-08-16 杭州深睿博联科技有限公司 Image processing method and device for breast molybdenum target
CN110148114A (en) * 2019-04-02 2019-08-20 成都真实维度科技有限公司 A kind of deep learning model training method based on 2D faulted scanning pattern data set
CN110148112A (en) * 2019-04-02 2019-08-20 成都真实维度科技有限公司 A method of it acquires and marks the progress data set foundation of tomoscan diagram data
CN110021430A (en) * 2019-04-09 2019-07-16 科大讯飞股份有限公司 A kind of the attribute information prediction technique and device of lesion
CN110059697A (en) * 2019-04-29 2019-07-26 上海理工大学 A kind of Lung neoplasm automatic division method based on deep learning
CN110084810A (en) * 2019-05-06 2019-08-02 成都医云科技有限公司 A kind of Lung neoplasm image detecting method, model training method, device and storage medium
CN110084813A (en) * 2019-05-13 2019-08-02 山东大学 A kind of good pernicious prediction technique of Lung neoplasm based on three dimensional depth learning network
CN110148467A (en) * 2019-05-16 2019-08-20 东北大学 A kind of Lung neoplasm device of computer aided diagnosis and method based on improvement CNN
CN110309860A (en) * 2019-06-06 2019-10-08 昆明理工大学 The method classified based on grade malignancy of the convolutional neural networks to Lung neoplasm
CN110473168A (en) * 2019-07-09 2019-11-19 天津大学 A kind of Lung neoplasm automatic checkout system squeezing-motivate residual error network based on 3D
WO2021027135A1 (en) * 2019-08-15 2021-02-18 平安科技（深圳）有限公司 Cell detection model training method and apparatus, computer device and storage medium
CN110660045A (en) * 2019-08-30 2020-01-07 杭州电子科技大学 Lymph node identification semi-supervision method based on convolutional neural network
CN110660045B (en) * 2019-08-30 2021-12-10 杭州电子科技大学 Lymph node identification semi-supervision method based on convolutional neural network
CN110689112A (en) * 2019-09-02 2020-01-14 杭州深睿博联科技有限公司 Data processing method and device
CN110706200A (en) * 2019-09-02 2020-01-17 杭州深睿博联科技有限公司 Data prediction method and device
CN110705570A (en) * 2019-09-20 2020-01-17 桂林电子科技大学 Image feature identification method
CN110705570B (en) * 2019-09-20 2022-06-21 桂林电子科技大学 Image feature identification method
CN110689547A (en) * 2019-09-25 2020-01-14 重庆大学 Pulmonary nodule segmentation method based on three-dimensional CT image
CN110717916A (en) * 2019-09-29 2020-01-21 华中科技大学 Pulmonary embolism detection system based on convolutional neural network
CN110717916B (en) * 2019-09-29 2022-08-30 华中科技大学 Pulmonary embolism detection system based on convolutional neural network
CN110766713A (en) * 2019-10-30 2020-02-07 上海微创医疗器械（集团）有限公司 Lung image segmentation method and device and lung lesion region identification equipment
CN112819747A (en) * 2019-10-31 2021-05-18 深圳市华嘉生物智能科技有限公司 Method for automatically diagnosing benign and malignant nodules based on lung tomography image
TWI760657B (en) * 2019-11-14 2022-04-11 和碩聯合科技股份有限公司 Training method for anomaly detection model and electronic device using the same
US11756179B2 (en) 2019-11-14 2023-09-12 Pegatron Corporation Training method for anomaly detection model and electronic device using the same
CN110910404A (en) * 2019-11-18 2020-03-24 西南交通大学 Anti-noise data breast ultrasonic nodule segmentation method
CN110890143B (en) * 2019-11-21 2022-03-08 重庆邮电大学 2D convolution method introducing spatial information
CN110890143A (en) * 2019-11-21 2020-03-17 重庆邮电大学 2D convolution method introducing spatial information
CN111079821A (en) * 2019-12-12 2020-04-28 哈尔滨市科佳通用机电股份有限公司 Derailment automatic braking pull ring falling fault image identification method
CN111080625A (en) * 2019-12-18 2020-04-28 北京推想科技有限公司 Training method and training device for lung image strip and rope detection model
CN111310624A (en) * 2020-02-05 2020-06-19 腾讯科技（深圳）有限公司 Occlusion recognition method and device, computer equipment and storage medium
CN111310624B (en) * 2020-02-05 2023-11-21 腾讯科技（深圳）有限公司 Occlusion recognition method, occlusion recognition device, computer equipment and storage medium
CN111340756A (en) * 2020-02-13 2020-06-26 北京深睿博联科技有限责任公司 Medical image lesion detection and combination method, system, terminal and storage medium
CN111340756B (en) * 2020-02-13 2023-11-28 北京深睿博联科技有限责任公司 Medical image lesion detection merging method, system, terminal and storage medium
CN110969623A (en) * 2020-02-28 2020-04-07 北京深睿博联科技有限责任公司 Lung CT multi-symptom automatic detection method, system, terminal and storage medium
CN111340825B (en) * 2020-03-05 2023-05-09 上海市肺科医院（上海市职业病防治院） Method and system for generating mediastinum lymph node segmentation model
CN111340825A (en) * 2020-03-05 2020-06-26 上海市肺科医院（上海市职业病防治院） Method and system for generating mediastinal lymph node segmentation model
CN111476766A (en) * 2020-03-31 2020-07-31 哈尔滨商业大学 Lung nodule CT image detection system based on deep learning
CN111476766B (en) * 2020-03-31 2023-08-22 哈尔滨商业大学 Lung nodule CT image detection system based on deep learning
CN111583271A (en) * 2020-04-13 2020-08-25 华东师范大学 Method for automatically predicting gene expression categories based on cancer CT images
CN111784593A (en) * 2020-06-04 2020-10-16 广东省智能制造研究所 Lung nodule CT image data enhancement method and system for deep learning
CN111815591A (en) * 2020-06-29 2020-10-23 郑州大学 Pulmonary nodule detection method based on CT image
CN111815591B (en) * 2020-06-29 2024-02-27 郑州大学 Lung nodule detection method based on CT image
CN111768382A (en) * 2020-06-30 2020-10-13 重庆大学 Interactive segmentation method based on lung nodule growth form
CN111768382B (en) * 2020-06-30 2023-08-15 重庆大学 Interactive segmentation method based on lung nodule growth morphology
CN112241948A (en) * 2020-09-23 2021-01-19 深圳视见医疗科技有限公司 Lung nodule detection and analysis method and system capable of adapting to thickness of layer
CN112184657A (en) * 2020-09-24 2021-01-05 上海健康医学院 Pulmonary nodule automatic detection method, device and computer system
CN112529870A (en) * 2020-12-10 2021-03-19 重庆大学 Multi-scale CNNs (CNNs) lung nodule false positive elimination method based on combination of source domain and frequency domain
CN112529870B (en) * 2020-12-10 2024-04-12 重庆大学 Multi-scale CNNs lung nodule false positive eliminating method based on combination of source domain and frequency domain
CN112581436A (en) * 2020-12-11 2021-03-30 佛山市普世医学科技有限责任公司 Lung nodule identification and segmentation method and system based on deep learning
CN112541909A (en) * 2020-12-22 2021-03-23 南开大学 Lung nodule detection method and system based on three-dimensional neural network of slice perception
CN112669284A (en) * 2020-12-29 2021-04-16 天津大学 Method for realizing pulmonary nodule detection by generating confrontation network
CN112991269A (en) * 2021-02-07 2021-06-18 复旦大学 Identification and classification method for lung CT image
CN112669319B (en) * 2021-03-22 2021-11-16 四川大学 Multi-view multi-scale lymph node false positive inhibition modeling method
CN112669319A (en) * 2021-03-22 2021-04-16 四川大学 Multi-view multi-scale lymph node false positive inhibition modeling method
CN113113132A (en) * 2021-03-25 2021-07-13 华南理工大学 Method for diagnosing congenital megacolon and homologous diseases based on deep convolutional neural network
CN113256634B (en) * 2021-07-13 2021-11-23 杭州医策科技有限公司 Cervical carcinoma TCT slice vagina arranging method and system based on deep learning
CN113256634A (en) * 2021-07-13 2021-08-13 杭州医策科技有限公司 Cervical carcinoma TCT slice vagina arranging method and system based on deep learning
CN113658162A (en) * 2021-08-24 2021-11-16 中山仰视科技有限公司 Refined lung nodule detection method, system, equipment and medium
CN114677383B (en) * 2022-03-03 2024-03-15 西北工业大学 Pulmonary nodule detection and segmentation method based on multitask learning
CN114677383A (en) * 2022-03-03 2022-06-28 西北工业大学 Pulmonary nodule detection and segmentation method based on multi-task learning
CN116012355B (en) * 2023-02-07 2023-11-21 重庆大学 Adaptive false positive lung nodule removing method based on deep learning
CN116012355A (en) * 2023-02-07 2023-04-25 重庆大学 Adaptive false positive lung nodule removing method based on deep learning

Also Published As

Publication number Publication date
CN107016665B (en) 2021-05-04

Similar Documents

Publication Publication Date Title
CN107016665A (en) A kind of CT pulmonary nodule detection methods based on depth convolutional neural networks
CN109102502B (en) Pulmonary nodule detection method based on three-dimensional convolutional neural network
WO2020253629A1 (en) Detection model training method and apparatus, computer device, and storage medium
WO2022063200A1 (en) Non-small cell lung cancer prognosis survival prediction method, medium and electronic device
JP2022529557A (en) Medical image segmentation methods, medical image segmentation devices, electronic devices and computer programs
CN109685768B (en) Pulmonary nodule automatic detection method and system based on pulmonary CT sequence
CN110766051A (en) Lung nodule morphological classification method based on neural network
EP3452959A1 (en) Model construction in a neural network for object detection
Hashemi et al. Mass detection in lung CT images using region growing segmentation and decision making based on fuzzy inference system and artificial neural network
CN110084237A (en) Detection model construction method, detection method and the device of Lung neoplasm
CN109389129A (en) A kind of image processing method, electronic equipment and storage medium
US20230230241A1 (en) System and method for detecting lung abnormalities
CN115546605A (en) Training method and device based on image labeling and segmentation model
Kousalya et al. Improved the detection and classification of breast cancer using hyper parameter tuning
Khan Identification of lung cancer using convolutional neural networks based classification
Li et al. Image segmentation based on improved unet
Tyagi et al. LCSCNet: A multi-level approach for lung cancer stage classification using 3D dense convolutional neural networks with concurrent squeeze-and-excitation module
CN112488996A (en) Inhomogeneous three-dimensional esophageal cancer energy spectrum CT (computed tomography) weak supervision automatic labeling method and system
Zhang et al. Multicontext 3D residual CNN for false positive reduction of pulmonary nodule detection
Heeneman et al. Lung nodule detection by using Deep Learning
CN109313802A (en) Biological object detection
Zhuang et al. An interpretable multi-task system for clinically applicable COVID-19 diagnosis using CXR
Pan et al. A review of machine learning approaches, challenges and prospects for computational tumor pathology
Park et al. Classification of cervical cancer using deep learning and machine learning approach
Yang et al. Leveraging auxiliary information from emr for weakly supervised pulmonary nodule detection

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
GR01 Patent grant
GR01 Patent grant