CN111784638A - Pulmonary nodule false positive screening method and system based on convolutional neural network - Google Patents
Pulmonary nodule false positive screening method and system based on convolutional neural network Download PDFInfo
- Publication number
- CN111784638A CN111784638A CN202010502186.6A CN202010502186A CN111784638A CN 111784638 A CN111784638 A CN 111784638A CN 202010502186 A CN202010502186 A CN 202010502186A CN 111784638 A CN111784638 A CN 111784638A
- Authority
- CN
- China
- Prior art keywords
- image data
- candidate
- lung nodule
- neural network
- convolutional neural
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000013527 convolutional neural network Methods 0.000 title claims abstract description 42
- 238000012216 screening Methods 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000002685 pulmonary effect Effects 0.000 title claims abstract description 27
- 206010056342 Pulmonary mass Diseases 0.000 claims abstract description 87
- 238000012545 processing Methods 0.000 claims abstract description 42
- 238000012549 training Methods 0.000 claims abstract description 38
- 210000004072 lung Anatomy 0.000 claims abstract description 22
- 238000011176 pooling Methods 0.000 claims description 22
- 230000015572 biosynthetic process Effects 0.000 claims description 7
- 238000003786 synthesis reaction Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 230000004927 fusion Effects 0.000 claims description 4
- 238000001514 detection method Methods 0.000 abstract description 9
- 238000002591 computed tomography Methods 0.000 description 18
- 230000006870 function Effects 0.000 description 8
- 206010058467 Lung neoplasm malignant Diseases 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 5
- 201000005202 lung cancer Diseases 0.000 description 5
- 208000020816 lung neoplasm Diseases 0.000 description 5
- 230000007547 defect Effects 0.000 description 4
- 210000001519 tissue Anatomy 0.000 description 4
- 230000004913 activation Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000004083 survival effect Effects 0.000 description 2
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention discloses a method and a system for screening false positive of pulmonary nodules based on a convolutional neural network, wherein the method comprises the following steps: acquiring the coordinate position and the maximum radius value of a candidate lung nodule from lung CT image data; extracting original 3D image data of the candidate lung nodule from the lung CT image data according to the coordinate position and the maximum radius value, and performing interpolation processing on the original 3D image data; acquiring sample data of three planes corresponding to candidate lung nodule 3D image data obtained through interpolation, and performing scaling processing on the sample data of the three planes to form a training set; and training a convolutional neural network based on the training set, and performing false positive screening on the candidate lung nodule through a convolutional neural network model obtained through training. The embodiment of the invention can solve the problem of high false positive rate in the process of identifying the pulmonary nodules of the conventional end-to-end network and improve the accuracy of computer-aided pulmonary nodule automatic detection.
Description
Technical Field
The invention relates to the technical field of medical treatment, in particular to a pulmonary nodule false positive screening method and a system based on a convolutional neural network.
Background
Lung cancer is the most common malignant tumor in the world, and recently, newly discovered cases and death cases of lung cancer in China are far superior to those in other countries. The 5-year survival rate of lung cancer patients in China is only 16.1 percent, which is far lower than that of developed countries in the west, the discovery of too late is one of the main reasons, the early detection of early treatment is the only effective way for improving the survival rate, and the low-dose CT (computed tomography) is the only available early lung cancer screening means at present. Early lung cancer is mainly characterized by asymptomatic pulmonary nodules, which are difficult for even experienced physicians to make accurate judgments due to their complex morphology.
Currently, in the field of research related to pulmonary nodule detection, a method for automatically detecting a pulmonary nodule by computer aided based on a pulmonary CT image is proposed, which generally includes the following two key steps: firstly, a lung nodule candidate region is obtained through preliminary detection of a lung nodule, and a correct recognition result containing the lung nodule and a false positive object not containing the lung nodule exist in the lung nodule candidate region; secondly, a proper lung nodule classifier is trained to screen the detection result of the lung nodule candidate region so as to eliminate false positive lung nodules. The detection method developed according to the above two steps has the defect of solving: because the deep learning network cannot distinguish the characteristics of the false positive pulmonary nodules, a large number of false positive samples are generated, and the possibility of depending on manual identification is very low; the classification performance of the conventional two-dimensional classifier is similar to the recognition capability of a deep learning network, and the false positive samples generated by the deep learning network cannot be further screened; the existing three-dimensional classifier has the problems of difficult model training, overfitting of the model, large calculated amount and the like.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides a pulmonary nodule false positive screening method and a pulmonary nodule false positive screening system based on a convolutional neural network, which can solve the problem of high false positive rate in the process of identifying pulmonary nodules by the conventional end-to-end network and improve the accuracy of computer-aided pulmonary nodule automatic detection.
In order to solve the above problems, the present invention provides a lung nodule false positive screening method based on a convolutional neural network, the method including:
acquiring the coordinate position and the maximum radius value of a candidate lung nodule from lung CT image data;
extracting original 3D image data of the candidate lung nodule from the lung CT image data according to the coordinate position and the maximum radius value, and performing interpolation processing on the original 3D image data to obtain final candidate lung nodule 3D image data;
acquiring sample data of three planes corresponding to the candidate lung nodule 3D image data, and performing scaling processing on the sample data of the three planes to form a training set;
and training a convolutional neural network based on the training set, and performing false positive screening on the candidate lung nodule through a convolutional neural network model obtained through training.
Optionally, the extracting raw 3D image data of the lung nodule candidate from the lung CT image data according to the coordinate position and the maximum radius value includes:
setting a size of a processing region of the lung nodule candidate based on the maximum radius value, the processing region centered on the coordinate position;
calculating the number of slices contained in the original 3D image data of the candidate lung nodule to be 2N +1 based on the set slice interval;
and taking the processing area as an intermediate layer, acquiring continuous N slices right above the processing area and continuous N slices right below the processing area, and combining the slices into the original 3D image data according to the original slice sequence.
Optionally, the interpolating the original 3D image data includes:
and performing interpolation operation on 2N slice intervals existing on the Z axis of the original 3D image data by using the XY pixel interval of the original 3D image data as a reference through a Lagrange interpolation method.
Optionally, the obtaining sample data of three planes corresponding to the lung nodule candidate 3D image data includes:
acquiring maximum projection data of each plane in three planes corresponding to the candidate lung nodule 3D image data;
and respectively carrying out normalization processing on the maximum projection data of each plane in the three planes to obtain sample data of each plane.
Optionally, the normalizing the maximum projection data of each of the three planes respectively includes:
screening and replacing the maximum projection data of each plane in the three planes based on the human tissue density range;
and normalizing the maximum projection data of each plane after screening and replacing.
Optionally, the scaling the sample data of the three planes includes: and setting the sample data of each plane in the three planes to be equal in size.
Optionally, the network structure of the convolutional neural network includes four convolutional pooling layers, a feature synthesis layer, a full connection layer, and a Softmax function layer.
Optionally, the training a convolutional neural network based on the training set, and performing false positive screening on the candidate lung nodule through a convolutional neural network model obtained through the training includes:
sequentially inputting sample data of each plane in three planes corresponding to the candidate lung nodule 3D image data to the four convolution pooling layers to obtain a first feature map corresponding to each plane;
performing feature fusion on the first feature map corresponding to each plane through the feature synthesis layer to obtain a second feature map of the candidate lung nodule 3D image data;
and performing result classification on the second feature map based on the full-connection layer, and outputting the false positive probability of the candidate lung nodule according to the classified result based on the Softmax function layer.
The embodiment of the invention also provides a pulmonary nodule false positive screening system based on the convolutional neural network, which comprises:
the parameter acquisition module is used for acquiring the coordinate position and the maximum radius value of a candidate lung nodule from lung CT image data;
an image extraction module, configured to extract original 3D image data of the lung nodule candidate from the lung CT image data according to the coordinate position and the maximum radius value, and perform interpolation processing on the original 3D image data to obtain final lung nodule candidate 3D image data;
the data processing module is used for acquiring sample data of three planes corresponding to the candidate lung nodule 3D image data, and carrying out scaling processing on the sample data of the three planes to form a training set;
and the sample training module is used for training the convolutional neural network based on the training set and carrying out false positive screening on the candidate lung nodule through the convolutional neural network model obtained through training.
In the embodiment of the invention, in consideration of the defect that the deep learning network in the prior art cannot distinguish the characteristics of false positive pulmonary nodules, a novel convolutional neural network is provided for carrying out rapid and efficient characteristic extraction and classification screening on candidate pulmonary nodules, so that the accuracy of computer-aided pulmonary nodule automatic detection can be improved; in addition, the embodiment of the invention extracts the 3D image of the candidate lung nodule from the original lung CT image, and increases the sample data of the 3D image by using a Lagrange interpolation method, so that the integrity and the sufficiency of the input sample in the convolutional neural network can be ensured.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a lung nodule false positive screening method based on a convolutional neural network according to an embodiment of the present invention;
fig. 2 is a schematic structural composition diagram of a pulmonary nodule false positive screening system based on a convolutional neural network disclosed in the embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Specifically, fig. 1 shows a schematic flow chart of a lung nodule false positive screening method based on a convolutional neural network in an embodiment of the present invention, where the method includes the following steps:
s101, acquiring coordinate positions and maximum radius values of candidate lung nodules from lung CT image data;
in the embodiment of the present invention, since each of all lung nodule candidates included in the lung CT image data has unique labeling information, which includes a coordinate position and a maximum radius value, the required parameters can be obtained by direct search.
S102, extracting original 3D image data of the candidate lung nodule from the lung CT image data according to the coordinate position and the maximum radius value, and performing interpolation processing on the original 3D image data to obtain final candidate lung nodule 3D image data;
the specific implementation process comprises the following steps:
(1) setting a size of a processing region of the lung nodule candidate based on the maximum radius value, the processing region centered on the coordinate position;
specifically, in the embodiment of the present invention, the length and the width of the treatment region of the lung nodule candidate are both set to be 1.5R × 2 according to the known maximum radius value R.
(2) Based on the given slice interval, calculating the number of slices contained in the original 3D image data of the lung nodule candidate as 2N +1, wherein the calculation formula of the number of slices is:
x＝h÷d+1
where x is the number of slices, h is the height of the original 3D image data, and D is the slice spacing. It should be noted that, since the original 3D image of the lung nodule candidate is defined as a cube, the height of the original 3D image is 1.5R × 2, and the slice interval is obtained from the lung CT image data.
(3) Taking the processing region as an intermediate layer, acquiring N consecutive slices directly above the processing region and N consecutive slices directly below the processing region, and combining the slices into the original 3D image data according to an original slice sequence, wherein 2N slices have a size equal to that of the processing region of the lung nodule candidate, and the original slice sequence mentioned here is obtained from the lung CT image data;
(4) and performing interpolation operation on 2N slice intervals existing on the Z axis of the original 3D image data by using the XY pixel interval of the original 3D image data as a reference by using a Lagrange interpolation method so as to ensure that the pixel intervals of the original 3D image data on three coordinate axes, namely an X axis, a Y axis and a Z axis, are equal.
S103, acquiring sample data of three planes corresponding to the candidate lung nodule 3D image data, and performing scaling processing on the sample data of the three planes to form a training set;
the specific implementation process comprises the following steps:
(1) acquiring maximum projection data of each plane in three planes corresponding to the candidate lung nodule 3D image data;
specifically, maximum projection processing is performed on the lung nodule candidate 3D image data on an XY plane, an XZ plane, and a YZ plane, respectively, to obtain maximum projection data of the XY plane corresponding to the lung nodule candidate 3D image data as an example, where the maximum projection data on the XY plane is: y (i, j) ═ MAX (HU (i, j,0), HU (i, j,1), …, HU (i, j, N), …, HU (i, j, N-1)), where i is an X-axis coordinate value of projection data, j is a Y-axis coordinate value of projection data, N is the total number of slices of the lung nodule candidate 3D image data, and HU (i, j, N) is an HU value at (i, j) on the XY plane corresponding to the Z-axis coordinate in the lung nodule candidate 3D image data being N. Note that HU is a hounsfield unit, which represents the relative density of tissue structures on CT images of the lung.
(2) Respectively carrying out normalization processing on the maximum projection data of each plane in the three planes to obtain sample data of each plane;
specifically, based on the human tissue density range, screening and replacing the maximum projection data of each plane in the three planes; and then carrying out normalization processing on the maximum value projection data of each plane after screening and replacement, so that the pixel value corresponding to the maximum value projection data of each plane is in the range of [0,1 ].
Wherein, the screening replacement process is further explained as follows: since the relative density of air is-1000 HU and the relative density of human bone is 400HU, the human tissue density range is limited to [ -1000,400] in the embodiment of the present invention, and the screening and replacing of the maximum projection data of the XY plane corresponding to the 3D image data of the candidate lung nodule includes the following three cases: when the maximum value projection data of the XY plane is in the range of [ -1000,400], performing no replacement processing; when the maximum projection data of the XY surface is less than-1000 HU, replacing the maximum projection data with-1000 HU; when the maximum value projection data of the XY plane is larger than 400HU, it is replaced with 400 HU.
It should be noted that, for the sample data of each plane obtained after the normalization processing, one or more of the expansion manners such as affine, scaling, rotation, mirroring, flipping, translation, filtering, etc. need to be used to perform data expansion on the sample data, so as to ensure that the amount of data input to the convolutional neural network is sufficient.
(3) Scaling the sample data of each of the three planes, that is, setting the sample data of each of the three planes to have the same size, the embodiment of the present invention unifies the sample data of each plane into 128 × 128 pixels, where the sample data of each plane also includes sample data obtained by expansion.
And S104, training the convolutional neural network based on the training set, and performing false positive screening on the candidate lung nodule through the convolutional neural network model obtained through training.
In an embodiment of the present invention, the network structure of the convolutional neural network includes four convolutional pooling layers, a feature synthesis layer, a full connection layer, and a Softmax function layer, where the four convolutional pooling layers are respectively:
a first convolution pooling layer: contains 64 convolution kernels of size 3 × 3;
second convolution pooling layer: contains 128 convolution kernels of size 3 × 3;
third convolution pooling layer: contains 256 convolution kernels of size 3 × 3;
fourth convolution pooling layer: contains 512 convolution kernels of size 3 x 3.
It should be noted that, after the convolution calculation is performed on the sample data by the first convolution pooling layer, 64 different feature maps are obtained, 128 different feature maps are obtained after a part of the 64 different feature maps are subjected to secondary convolution by the second convolution pooling layer, 256 different feature maps are obtained after a part of the 128 different feature maps are subjected to tertiary convolution by the third convolution pooling layer, and 512 different feature maps are obtained after a part of the 256 different feature maps are subjected to four convolution by the fourth convolution pooling layer; and each convolution pooling layer takes LeakyRELU (linear unit with leakage correction) as an activation function, the corresponding convolution result needs to be input into the activation function for mapping processing, then 2 x 2 maximum pooling processing is executed, the original convolution result is subjected to feature extraction, and the computational complexity of data is reduced by deleting unimportant feature maps.
Specifically, the training process of the convolutional neural network based on the training set includes:
(1) sequentially inputting sample data of each plane in three planes corresponding to the candidate lung nodule 3D image data to the four convolution pooling layers to obtain a first feature map corresponding to each plane;
further, gradually inputting sample data of an XY plane corresponding to the lung nodule candidate 3D image data to the first convolution pooling layer, the second convolution pooling layer, the third convolution pooling layer and the fourth convolution pooling layer for processing, and outputting an 8 × 8 × 512 XY plane feature map; similarly, sample data of an XZ plane and a YZ plane corresponding to the lung nodule candidate 3D image data are processed by the four convolution pooling layers, and then an 8 × 8 × 512 XZ plane feature map and an 8 × 8 × 512 YZ plane feature map are correspondingly output.
(2) Performing feature fusion on the first feature map corresponding to each plane through the feature synthesis layer to obtain a second feature map of the candidate lung nodule 3D image data;
further, the XY plane feature map, the XZ plane feature map, and the YZ plane feature map are input to the feature synthesis layer to perform feature fusion, and an internal 1 × 1 convolution kernel is used to perform convolution operation, so as to obtain an 8 × 8 × 1024 second feature map corresponding to the lung nodule candidate 3D image data.
(3) And performing result classification on the second feature map based on the full-connection layer, and outputting the false positive probability of the candidate lung nodule according to the classified result based on the Softmax function layer.
Furthermore, 1024 feature results contained in the second feature map are highly purified feature results, so that basic features of the candidate pulmonary nodules can be accurately expressed, at the moment, the 1024 feature results are classified by using the full connection layer with the kernel size of 1024 × 2, similar feature results are gathered into one class, finally, the classification results are mapped to a [0,1] interval by using the Softmax function layer, and a node value with the maximum probability is selected from the false positive classes to serve as the false positive probability of the candidate pulmonary nodules.
It should be noted that, a binary cross entropy loss function is adopted in the convolutional neural network model to determine the closeness of the classification result output by the full-link layer to the expected output result, so as to feed back the prediction accuracy of the false positive probability of the lung nodule candidate.
Specifically, fig. 2 shows a convolutional neural network-based lung nodule false positive screening system in an embodiment of the present invention, where the system includes:
a parameter obtaining module 201, configured to obtain a coordinate position and a maximum radius value of a candidate lung nodule from lung CT image data;
an image extraction module 202, configured to extract original 3D image data of the lung nodule candidate from the lung CT image data according to the coordinate position and the maximum radius value, and perform interpolation processing on the original 3D image data to obtain final lung nodule candidate 3D image data;
the data processing module 203 is configured to acquire sample data of three planes corresponding to the candidate lung nodule 3D image data, and scale the sample data of the three planes to form a training set;
and the sample training module 204 is configured to train a convolutional neural network based on the training set, and perform false positive screening on the candidate lung nodule through a convolutional neural network model obtained through training.
For specific implementation of each module in the system, please refer to the method flowchart and specific implementation content shown in fig. 1, which are not described herein again.
In the embodiment of the invention, in consideration of the defect that the deep learning network in the prior art cannot distinguish the characteristics of false positive pulmonary nodules, a novel convolutional neural network is provided for carrying out rapid and efficient characteristic extraction and classification screening on candidate pulmonary nodules, so that the accuracy of computer-aided pulmonary nodule automatic detection can be improved; in addition, the embodiment of the invention extracts the 3D image of the candidate lung nodule from the original lung CT image, and increases the sample data of the 3D image by using a Lagrange interpolation method, so that the integrity and the sufficiency of the input sample in the convolutional neural network can be ensured.
Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by associated hardware instructed by a program, which may be stored in a computer-readable storage medium, and the storage medium may include: a Read Only Memory (ROM), a Random Access Memory (RAM), a magnetic or optical disk, or the like.
The method and the system for screening false positive of pulmonary nodule based on convolutional neural network provided by the embodiment of the present invention are introduced in detail, and a specific example is adopted herein to illustrate the principle and the implementation manner of the present invention, and the description of the above embodiment is only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (9)
1. A lung nodule false positive screening method based on a convolutional neural network is characterized by comprising the following steps:
acquiring the coordinate position and the maximum radius value of a candidate lung nodule from lung CT image data;
extracting original 3D image data of the candidate lung nodule from the lung CT image data according to the coordinate position and the maximum radius value, and performing interpolation processing on the original 3D image data to obtain final candidate lung nodule 3D image data;
acquiring sample data of three planes corresponding to the candidate lung nodule 3D image data, and performing scaling processing on the sample data of the three planes to form a training set;
and training a convolutional neural network based on the training set, and performing false positive screening on the candidate lung nodule through a convolutional neural network model obtained through training.
2. The method of claim 1, wherein the extracting raw 3D image data of the lung nodule candidate from the lung CT image data according to the coordinate position and the maximum radius value comprises:
setting a size of a processing region of the lung nodule candidate based on the maximum radius value, the processing region centered on the coordinate position;
calculating the number of slices contained in the original 3D image data of the candidate lung nodule to be 2N +1 based on the set slice interval;
and taking the processing area as an intermediate layer, acquiring continuous N slices right above the processing area and continuous N slices right below the processing area, and combining the slices into the original 3D image data according to the original slice sequence.
3. The lung nodule false positive screening method of claim 2, wherein the interpolating the raw 3D image data comprises:
and performing interpolation operation on 2N slice intervals existing on the Z axis of the original 3D image data by using the XY pixel interval of the original 3D image data as a reference through a Lagrange interpolation method.
4. The method according to claim 1, wherein the obtaining sample data of three planes corresponding to the 3D image data of the lung nodule candidate comprises:
acquiring maximum projection data of each plane in three planes corresponding to the candidate lung nodule 3D image data;
and respectively carrying out normalization processing on the maximum projection data of each plane in the three planes to obtain sample data of each plane.
5. The method of claim 4, wherein the normalizing the maximum projection data of each of the three planes comprises:
screening and replacing the maximum projection data of each plane in the three planes based on the human tissue density range;
and normalizing the maximum projection data of each plane after screening and replacing.
6. The method according to claim 4, wherein the scaling the sample data of the three planes comprises: and setting the sample data of each plane in the three planes to be equal in size.
7. The method of claim 1, wherein the network structure of the convolutional neural network comprises four convolutional pooling layers, one feature synthesis layer, one fully-connected layer, and one Softmax function layer.
8. The method of claim 7, wherein the training a convolutional neural network based on the training set, and the performing the false positive screening on the candidate lung nodule through the trained convolutional neural network model comprises:
sequentially inputting sample data of each plane in three planes corresponding to the candidate lung nodule 3D image data to the four convolution pooling layers to obtain a first feature map corresponding to each plane;
performing feature fusion on the first feature map corresponding to each plane through the feature synthesis layer to obtain a second feature map of the candidate lung nodule 3D image data;
and performing result classification on the second feature map based on the full-connection layer, and outputting the false positive probability of the candidate lung nodule according to the classified result based on the Softmax function layer.
9. A system for pulmonary nodule false positive screening based on convolutional neural network, the system comprising:
the parameter acquisition module is used for acquiring the coordinate position and the maximum radius value of a candidate lung nodule from lung CT image data;
an image extraction module, configured to extract original 3D image data of the lung nodule candidate from the lung CT image data according to the coordinate position and the maximum radius value, and perform interpolation processing on the original 3D image data to obtain final lung nodule candidate 3D image data;
the data processing module is used for acquiring sample data of three planes corresponding to the candidate lung nodule 3D image data, and carrying out scaling processing on the sample data of the three planes to form a training set;
and the sample training module is used for training the convolutional neural network based on the training set and carrying out false positive screening on the candidate lung nodule through the convolutional neural network model obtained through training.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010502186.6A CN111784638B (en) | 2020-06-04 | 2020-06-04 | Convolutional neural network-based lung nodule false positive screening method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010502186.6A CN111784638B (en) | 2020-06-04 | 2020-06-04 | Convolutional neural network-based lung nodule false positive screening method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111784638A true CN111784638A (en) | 2020-10-16 |
| CN111784638B CN111784638B (en) | 2024-06-25 |
Family
ID=72754624
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010502186.6A Active CN111784638B (en) | 2020-06-04 | 2020-06-04 | Convolutional neural network-based lung nodule false positive screening method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111784638B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
| CN113903030A (en) * | 2021-10-12 | 2022-01-07 | 杭州迪英加科技有限公司 | Liquid-based cell pathology image generation method based on weak supervised learning |
| CN116934757A (en) * | 2023-09-18 | 2023-10-24 | 电子科技大学 | Method, equipment and storage medium for lung nodule false positive pruning |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150141831A (en) * | 2014-06-10 | 2015-12-21 | 원광대학교산학협력단 | Apparatus for analysing medical image and method for classifying pulmonary vessel and pulmonary nodule |
| CN107154043A (en) * | 2017-06-05 | 2017-09-12 | 杭州健培科技有限公司 | A kind of Lung neoplasm false positive sample suppressing method based on 3DCNN |
| CN107464234A (en) * | 2017-07-24 | 2017-12-12 | 上海交通大学 | Lung neoplasm picture depth study identifying system and its method based on RGB channel stacking method |
| CN108230323A (en) * | 2018-01-30 | 2018-06-29 | 浙江大学 | A kind of Lung neoplasm false positive screening technique based on convolutional neural networks |
| 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 |
| CN110675398A (en) * | 2019-10-22 | 2020-01-10 | 深圳瀚维智能医疗科技有限公司 | Mammary gland ultrasonic screening method and device and computer equipment |
| CN110807788A (en) * | 2019-10-21 | 2020-02-18 | 腾讯科技（深圳）有限公司 | Medical image processing method, device, electronic equipment and computer storage medium |
-
2020
- 2020-06-04 CN CN202010502186.6A patent/CN111784638B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150141831A (en) * | 2014-06-10 | 2015-12-21 | 원광대학교산학협력단 | Apparatus for analysing medical image and method for classifying pulmonary vessel and pulmonary nodule |
| CN107154043A (en) * | 2017-06-05 | 2017-09-12 | 杭州健培科技有限公司 | A kind of Lung neoplasm false positive sample suppressing method based on 3DCNN |
| CN107464234A (en) * | 2017-07-24 | 2017-12-12 | 上海交通大学 | Lung neoplasm picture depth study identifying system and its method based on RGB channel stacking method |
| CN108230323A (en) * | 2018-01-30 | 2018-06-29 | 浙江大学 | A kind of Lung neoplasm false positive screening technique based on convolutional neural networks |
| 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 |
| CN110807788A (en) * | 2019-10-21 | 2020-02-18 | 腾讯科技（深圳）有限公司 | Medical image processing method, device, electronic equipment and computer storage medium |
| CN110675398A (en) * | 2019-10-22 | 2020-01-10 | 深圳瀚维智能医疗科技有限公司 | Mammary gland ultrasonic screening method and device and computer equipment |
Non-Patent Citations (2)
| Title |
|---|
| 张福玲;张少敏;: "应用于CT图像肺结节检测的深度学习方法综述", 计算机工程与应用, no. 13, 1 June 2020 (2020-06-01) * |
| 苗光;李朝锋;: "二维和三维卷积神经网络相结合的CT图像肺结节检测方法", 激光与光电子学进展, no. 05 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
| CN113903030A (en) * | 2021-10-12 | 2022-01-07 | 杭州迪英加科技有限公司 | Liquid-based cell pathology image generation method based on weak supervised learning |
| CN113903030B (en) * | 2021-10-12 | 2024-07-02 | 杭州迪英加科技有限公司 | Liquid-based cell pathology image generation method based on weak supervision learning |
| CN116934757A (en) * | 2023-09-18 | 2023-10-24 | 电子科技大学 | Method, equipment and storage medium for lung nodule false positive pruning |
| CN116934757B (en) * | 2023-09-18 | 2023-11-21 | 电子科技大学 | Method, equipment and storage medium for lung nodule false positive pruning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111784638B (en) | 2024-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109584252B (en) | Lung lobe segment segmentation method and device of CT image based on deep learning | |
| CN109102502B (en) | Pulmonary nodule detection method based on three-dimensional convolutional neural network | |
| CN107886514B (en) | Mammary gland molybdenum target image lump semantic segmentation method based on depth residual error network | |
| CN111784638A (en) | Pulmonary nodule false positive screening method and system based on convolutional neural network | |
| CN110889852B (en) | Liver segmentation method based on residual error-attention deep neural network | |
| CN111882560B (en) | Lung parenchyma CT image segmentation method based on weighted full convolution neural network | |
| CN112329871B (en) | Pulmonary nodule detection method based on self-correction convolution and channel attention mechanism | |
| CN112258461B (en) | Pulmonary nodule detection method based on convolutional neural network | |
| CN112819747A (en) | Method for automatically diagnosing benign and malignant nodules based on lung tomography image | |
| CN113744183A (en) | Pulmonary nodule detection method and system | |
| CN115409815A (en) | Pulmonary nodule detection method based on three-dimensional deformable transformer | |
| CN114581698A (en) | Target classification method based on space cross attention mechanism feature fusion | |
| CN111899850A (en) | Medical image information processing method, display method and readable storage medium | |
| CN114677511A (en) | Lung nodule segmentation method combining residual ECA channel attention UNet with TRW-S | |
| CN116883341A (en) | Liver tumor CT image automatic segmentation method based on deep learning | |
| Mridha et al. | A comprehensive survey on the progress, process, and challenges of lung cancer detection and classification | |
| CN117710760A (en) | Method for detecting chest X-ray focus by using residual noted neural network | |
| CN112233106A (en) | Thyroid cancer ultrasonic image analysis method based on residual capsule network | |
| CN114119447A (en) | Method, system, medium and device for constructing lung nodule multi-attribute classification model | |
| CN116402756A (en) | X-ray film lung disease screening system integrating multi-level characteristics | |
| Khouadja et al. | Lung Cancer Detection with Machine Learning and Deep Learning: A Narrative Review | |
| CN114693698B (en) | Computer-aided lung airway segmentation method based on neural network | |
| CN112541909B (en) | Lung nodule detection method and system based on three-dimensional neural network of slice perception | |
| CN111415350B (en) | Colposcope image identification method for detecting cervical lesions | |
| CN115880200A (en) | CT image lung nodule detection method and system based on 3D-Vnet and 3D-resNet |
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 |