CN113658657A - Lung nodule detection method and device combining depth network and optical flow mechanism - Google Patents
Lung nodule detection method and device combining depth network and optical flow mechanism Download PDFInfo
- Publication number
- CN113658657A CN113658657A CN202110975691.7A CN202110975691A CN113658657A CN 113658657 A CN113658657 A CN 113658657A CN 202110975691 A CN202110975691 A CN 202110975691A CN 113658657 A CN113658657 A CN 113658657A
- Authority
- CN
- China
- Prior art keywords
- lung
- image
- network
- classification
- nodule
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 206010056342 Pulmonary mass Diseases 0.000 title claims abstract description 99
- 238000001514 detection method Methods 0.000 title claims abstract description 92
- 230000003287 optical effect Effects 0.000 title claims abstract description 43
- 230000007246 mechanism Effects 0.000 title claims abstract description 33
- 210000004072 lung Anatomy 0.000 claims abstract description 72
- 238000007781 pre-processing Methods 0.000 claims abstract description 21
- 230000002685 pulmonary effect Effects 0.000 claims abstract description 18
- 238000000605 extraction Methods 0.000 claims abstract description 17
- 238000000034 method Methods 0.000 claims abstract description 13
- 238000004458 analytical method Methods 0.000 claims abstract description 9
- 238000012549 training Methods 0.000 claims description 23
- 238000013145 classification model Methods 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 13
- 238000010586 diagram Methods 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 4
- 238000003325 tomography Methods 0.000 claims description 3
- 238000009499 grossing Methods 0.000 claims description 2
- 238000013139 quantization Methods 0.000 claims description 2
- 238000012360 testing method Methods 0.000 description 10
- 239000002609 medium Substances 0.000 description 9
- 210000004027 cell Anatomy 0.000 description 5
- 206010058467 Lung neoplasm malignant Diseases 0.000 description 3
- 201000005202 lung cancer Diseases 0.000 description 3
- 208000020816 lung neoplasm Diseases 0.000 description 3
- 238000002790 cross-validation Methods 0.000 description 2
- 210000003743 erythrocyte Anatomy 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 208000024172 Cardiovascular disease Diseases 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 208000026106 cerebrovascular disease Diseases 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000002526 effect on cardiovascular system Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000003211 malignant effect Effects 0.000 description 1
- 239000012120 mounting media Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H15/00—ICT specially adapted for medical reports, e.g. generation or transmission thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention relates to a pulmonary nodule detection method and a device combining a depth network and an optical flow mechanism, wherein the method comprises the following steps: acquiring a 3D lung CT image, and preprocessing the 3D lung CT image; performing feature extraction and analysis on the 3D lung CT image; constructing a depth optical flow network to detect and identify lung nodules in the 3D lung CT image; obtaining a pulmonary nodule detection result and generating a detection report; the invention enhances the detection precision of the 3D lung CT image and improves the detection efficiency.
Description
Technical Field
The invention relates to the technical field of medical auxiliary diagnosis, in particular to a method and a device for detecting pulmonary nodules by combining a depth network and an optical flow mechanism.
Background
Early lung nodules are usually small, do not have a fixed shape and are therefore difficult to distinguish visually, and lung cancer results when benign lung nodules progress to malignant lung nodules. In reality, lung nodules are typically detected by CT images of the lungs. The lung nodules are used as early-stage manifestations of lung cancer, and detection of the lung nodules has great significance for prediction of the lung cancer. However, the lung nodule size is small, and the traditional C image lung nodule detection method is not only complicated in steps and slow in processing speed; in the clinical pulmonary nodule detection method, a doctor identifies whether a patient has pulmonary nodules by observing CT images of lungs, and generally, the number of complete CT sequences of one patient is large, the workload is large, the complete CT sequences are easy to miss, and the detection accuracy of the pulmonary nodules is low.
Disclosure of Invention
The invention aims to provide a lung nodule detection method and device combining a depth network and an optical flow mechanism, and aims to solve the problem that in the prior art, the detection accuracy of lung nodules in a lung CT image is low.
The technical purpose of the invention is realized by the following technical scheme:
in a first aspect, an embodiment of the present application provides a lung nodule detection method combining a depth network and an optical flow mechanism, the method including the following steps:
acquiring a 3D lung CT image, and preprocessing the 3D lung CT image;
performing feature extraction and analysis on the 3D lung CT image;
constructing a depth optical flow network to detect and identify lung nodules in the 3D lung CT image;
and obtaining a lung nodule detection result and generating a detection report.
In some embodiments, the constructing a depth optical flow network for detecting and identifying lung nodules in the 3D lung CT image includes:
automatically processing the 3D lung CT image into a CETS readable 3D lung image;
and detecting a lung nodule region of the 3D lung CT image by utilizing a TNet lung nodule 3D detection classification network, classifying the lung nodule, and inputting the extracted lung nodule into the TNet or TNet-VQ lung nodule 3D detection classification network.
In some embodiments, the feature extraction and analysis of the 3D pulmonary CT image comprises:
inputting the 3D lung CT image into a feature extraction network, and obtaining a suggested region of the nodule by judging whether the image is the two-classification output of the nodule and predicting a bounding box; inputting the obtained suggested area into a RoIploling layer to carry out size standardization processing on a characteristic diagram; outputting the data to a classification probability prediction layer and a boundary frame prediction layer through two full-connection layers;
and detecting the CT image to be detected by using the trained three suspected lung nodule detection models to obtain the classification probability and the bounding box of the suspected nodules, classifying by using the trained three weak classification models, and performing majority voting on the classification results of the three weak classification models to select a final classification result.
In some embodiments, before detecting a CT image to be detected by using the trained three suspected lung nodule detection models, the method further includes:
preprocessing positive and negative samples in the training data to balance the proportion of the positive and negative samples, and then classifying the preprocessed negative samples by using a pre-screening model to screen out the negative samples with wrong classification; and training three weak classification models by using the training data after the pretreatment and the screening.
In some of these embodiments, the pre-processing the 3D pulmonary CT image includes:
and carrying out quantization and smoothing processing on the 3D lung CT image.
In a second aspect, an embodiment of the present application provides a lung nodule detection apparatus combining a depth network and an optical flow mechanism, including:
the image acquisition module is used for acquiring a 3D lung CT image and preprocessing the 3D lung CT image;
the feature extraction module is used for extracting and analyzing features of the 3D lung CT image;
the network construction module is used for constructing a depth optical flow network to detect and identify lung nodules in the 3D lung CT image;
and the report generation module is used for obtaining a pulmonary nodule detection result and generating a detection report.
In some embodiments, the network building module comprises:
the preprocessing unit is used for automatically processing the 3D lung CT image into a CETS (computerized tomography System) readable 3D lung image;
and the classification unit is used for detecting the lung nodule region of the 3D lung CT image by utilizing a TNet lung nodule 3D detection classification network, classifying the lung nodules, and inputting the extracted lung nodules into the TNet or TNet-VQ lung nodule 3D detection classification network.
In a third aspect, embodiments of the present application provide a computer device, including a memory and one or more processors;
the memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement a lung nodule detection method combining a depth network and an optical flow mechanism as described in the first aspect.
In a fourth aspect, embodiments of the present application provide a storage medium containing computer-executable instructions for performing a combined depth network and optical flow mechanism lung nodule detection method as described in the first aspect when executed by a computer processor.
The invention has the beneficial effects that: the method comprises the steps of obtaining a 3D lung CT image, and preprocessing the 3D lung CT image; performing feature extraction and analysis on the 3D lung CT image; constructing a depth optical flow network to detect and identify lung nodules in the 3D lung CT image; obtaining a pulmonary nodule detection result and generating a detection report; the detection precision of the 3D lung CT image is enhanced, and the detection efficiency is improved.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a schematic diagram of the steps of a lung nodule detection method combining a depth network and an optical flow mechanism;
FIG. 2 is a schematic diagram of a lung nodule detection apparatus with a combination of a depth network and an optical flow mechanism;
FIG. 3 is a schematic diagram of a computer device for a lung nodule detection method combining a depth network and an optical flow mechanism.
Detailed Description
The present invention will be further described with reference to the accompanying drawings and the detailed description, and it should be noted that any combination of the embodiments or technical features described below can be used to form a new embodiment without conflict.
Referring to fig. 1, a lung nodule detection method combining a depth network and an optical flow mechanism according to the present invention is shown, the method includes the following steps:
100. and acquiring a 3D lung CT image, and preprocessing the 3D lung CT image.
Optionally, the 3D pulmonary CT image is quantized and smoothed.
200. And carrying out feature extraction and analysis on the 3D lung CT image.
Specifically, the 3D lung CT image is input into a feature extraction network, and a suggested region of a nodule is obtained by judging whether the image is a two-classification output of the nodule and predicting a bounding box; inputting the obtained suggested area into a RoIploling layer to carry out size standardization processing on a characteristic diagram; outputting the data to a classification probability prediction layer and a boundary frame prediction layer through two full-connection layers;
in the embodiment of the invention, for each 3D lung CT image in the training set, a slice image at the center of a nodule and two slice images adjacent to the center of the nodule are extracted according to the position of the labeled nodule, a suspected pulmonary nodule detection model is trained for each slice image, and three models are input into each slice in the test process to obtain the suspected nodule. The model diagram shown in fig. 1 is for one slice image and the other two slice images perform the same procedure to obtain three suspected lung nodule detection models. The three suspected lung nodule detection models are used for lung nodule detection, three slice images of a CT image to be detected are selected in the same way as in training during detection, then the three slice images are respectively input into the three suspected lung nodule detection models for lung nodule detection, finally a classification probability prediction result and a frame prediction result are obtained, and then whether the detected lung nodule is a lung nodule is determined according to the results.
In the embodiment of the invention, the existing data set comprises 2N subsets, 2N-1 subsets are selected from the existing data set as training sets to train a feature extraction network every time, the remaining 1 subset is used as a test set, a cross validation mode is adopted, and as 2N times of tests are carried out, and 1 subset is selected for testing each time to obtain a result, the 2N times of test results are finally merged, namely the union set is obtained, so that the test results of all the subsets are obtained.
For example, the existing data set may be data in LUNA16, which includes 10 subsets, we perform cross validation, take 9 subsets each time to train the network, 1 subset to perform a test, and finally merge the results of ten tests.
In the embodiment of the present invention, the roiploling (region of interest) region pooling layer first intercepts the region of the proposed region on the feature map corresponding to the deconvolution layer, and then performs Maxpooling (maximum pooling) on the intercepted region, so as to generate a feature map output with a fixed size (e.g., 7 × 7) for proposed regions with different sizes.
Specifically, preprocessing positive and negative samples in training data to balance the proportion of the positive and negative samples, and then classifying the preprocessed negative samples by using a pre-screening model to screen out the negative samples with wrong classification; and training three weak classification models by using the training data after the pretreatment and the screening.
In the embodiment of the invention, through preprocessing and screening, the preprocessed positive sample in each group of training data and the screened negative sample with wrong classification can be obtained, and the two types of sample data are used for subsequent weak classification model training. Firstly, selecting a first group of preprocessed and screened training data (namely preprocessed positive samples and screened negative samples with wrong classification) from three groups of training data, and training an AlexNet classification model; for the AlexNet classification model after training, carrying out model test by using the same training data, and selecting the data with wrong classification as the training data of a second classifier; secondly, retraining a second AlexNet classification model by using a second group of preprocessed and screened training data and the data with the last classification error, performing model test by using the same data, and selecting the data with the classification error as the training data of a third classifier; secondly, retraining a third AlexNet classification model by taking the data with the classification errors of the previous two times and the training data after the third group of preprocessing and screening; and finally, the three AlexNet classification models which are trained are three weak classification models.
Specifically, three trained suspected lung nodule detection models are used for detecting a CT image to be detected to obtain the classification probability and the bounding box of the suspected nodules, then three trained weak classification models are used for classification, and the classification results of the three weak classification models are subjected to majority voting to select a final classification result.
In the detection process, detecting a CT image to be detected by using the trained three suspected lung nodule detection models, detecting each slice of the CT image by using the three suspected lung nodule detection models, and outputting classification probability of each suspected nodule and data of a bounding box of the suspected nodule by each suspected lung nodule detection model; all the three data are output to three trained weak classification models, each weak classification model can output a classification result aiming at each part of data, namely one weak classification model outputs the classification results of the three data, the three weak classification models output 9 classification results in a modulus mode, and then the classification results of the three weak classification models are subjected to majority voting to select a final classification result.
300. And constructing a depth optical flow network to detect and identify lung nodules in the 3D lung CT image.
Specifically, a lung nodule region of the 3D lung CT image is detected by utilizing a TNet lung nodule 3D detection classification network, the lung nodules are classified, and the extracted lung nodules are input into the TNet or TNet-VQ lung nodule 3D detection classification network.
400. And obtaining a lung nodule detection result and generating a detection report.
In the invention, the 3D lung CT image is acquired, and the 3D lung CT image is preprocessed; performing feature extraction and analysis on the 3D lung CT image; constructing a depth optical flow network to detect and identify lung nodules in the 3D lung CT image; obtaining a pulmonary nodule detection result and generating a detection report; the detection precision of the 3D lung CT image is enhanced, and the detection efficiency is improved.
Referring to fig. 2, a lung nodule detecting apparatus with a combination of a depth network and an optical flow mechanism according to the present invention is shown, which includes: an image acquisition module 101, a feature extraction module 102, a network construction module 103, and a report generation module 104.
The image acquisition module is used for acquiring a 3D lung CT image and preprocessing the 3D lung CT image; the feature extraction module is used for extracting and analyzing features of the 3D lung CT image; the network construction module is used for constructing a depth optical flow network to detect and identify lung nodules in the 3D lung CT image; the report generation module is used for obtaining a pulmonary nodule detection result and generating a detection report.
In some embodiments, the network building module comprises: a preprocessing unit and a classification unit; the preprocessing unit is used for automatically processing the 3D lung CT image into a CETS (computerized tomography System) readable 3D lung image; the classification unit is used for detecting a lung nodule region of the 3D lung CT image by utilizing a TNet lung nodule 3D detection classification network, classifying the lung nodules, and inputting the extracted lung nodules into the TNet or TNet-VQ lung nodule 3D detection classification network.
The image scanning module is used for scanning the erythrocyte smear by using an electronic scanning mirror to obtain a scanned image; the image processing module is used for carrying out image preprocessing on the scanned image to obtain a cell image to be detected; the image classification module is used for inputting the cell image to be detected into the trained cell detection model to obtain a classification result corresponding to the cell image to be detected; the calculation and judgment module is used for calculating the erythrocyte aggregation index in the cell image to be detected according to the classification result so as to judge the probability of the cardiovascular and cerebrovascular diseases.
The embodiment of the present application further provides a computer device, which may integrate the lung nodule detection apparatus with the combination of the depth network and the optical flow mechanism provided in the embodiment of the present application. Fig. 3 is a schematic structural diagram of a computer device according to an embodiment of the present application. Referring to fig. 3, the computer apparatus includes: an input device 43, an output device 44, a memory 42, and one or more processors 41; the memory 42 for storing one or more programs; when executed by the one or more processors 41, cause the one or more processors 41 to implement a lung nodule detection method that combines a depth network and an optical flow mechanism as provided by the embodiments described above. Wherein the input device 43, the output device 44, the memory 42 and the processor 41 may be connected by a bus or other means, for example, in fig. 3.
The processor 41 executes various functional applications of the device and data processing by running software programs, instructions and modules stored in the memory 42, namely, implementing the lung nodule detection method combining the deep network and optical flow mechanism described above.
The computer device provided above can be used to execute the lung nodule detection method combining the depth network and the optical flow mechanism provided in the above embodiments, and has corresponding functions and advantages.
Embodiments of the present application further provide a storage medium containing computer-executable instructions which, when executed by a computer processor, perform a combined depth-network and optical-flow mechanism pulmonary nodule detection method, the combined depth-network and optical-flow mechanism pulmonary nodule detection method comprising: acquiring a 3D lung CT image, and preprocessing the 3D lung CT image; performing feature extraction and analysis on the 3D lung CT image; constructing a depth optical flow network to detect and identify lung nodules in the 3D lung CT image; and obtaining a lung nodule detection result and generating a detection report.
Storage medium-any of various types of memory devices or storage devices. The term "storage medium" is intended to include: mounting media such as CD-ROM, floppy disk, or tape devices; computer device memory or random access memory such as DRAM, DDRRAM, SRAM, EDORAM, Lanbus (Rambus) RAM, etc.; non-volatile memory such as flash memory, magnetic media (e.g., hard disk or optical storage); registers or other similar types of memory elements, etc. The storage medium may also include other types of memory or combinations thereof. In addition, the storage medium may be located in a first computer apparatus in which the program is executed, or may be located in a different second computer apparatus connected to the first computer apparatus through a network (such as the internet). The second computer device may provide program instructions to the first computer for execution. The term "storage medium" may include two or more storage media that may reside in different locations, such as in different computer devices that are connected by a network. The storage medium may store program instructions (e.g., embodied as a computer program) that are executable by one or more processors.
Of course, the storage medium provided by the embodiments of the present application contains computer-executable instructions, and the computer-executable instructions are not limited to the lung nodule detection method combining the depth network and the optical flow mechanism as described above, and may also perform related operations in the lung nodule detection method combining the depth network and the optical flow mechanism as provided by any embodiments of the present application.
The lung nodule detection apparatus, the storage medium and the computer device combining the depth network and the optical flow mechanism provided in the foregoing embodiments may perform the lung nodule detection method combining the depth network and the optical flow mechanism provided in any embodiments of the present application, and the technical details not described in detail in the foregoing embodiments may be referred to the lung nodule detection method combining the depth network and the optical flow mechanism provided in any embodiments of the present application.
The above embodiments are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited thereby, and any insubstantial changes and substitutions made by those skilled in the art based on the present invention are within the protection scope of the present invention.
Claims (9)
1. A lung nodule detection method combining a depth network and an optical flow mechanism is characterized in that: the method comprises the following steps:
acquiring a 3D lung CT image, and preprocessing the 3D lung CT image;
performing feature extraction and analysis on the 3D lung CT image;
constructing a depth optical flow network to detect and identify lung nodules in the 3D lung CT image;
and obtaining a lung nodule detection result and generating a detection report.
2. The lung nodule detection method based on the combination of the depth network and the optical flow mechanism as claimed in claim 1, wherein: the constructing of the depth optical flow network for detecting and identifying lung nodules in the 3D lung CT image comprises the following steps:
automatically processing the 3D lung CT image into a CETS readable 3D lung image;
and detecting a lung nodule region of the 3D lung CT image by utilizing a TNet lung nodule 3D detection classification network, classifying the lung nodule, and inputting the extracted lung nodule into the TNet or TNet-VQ lung nodule 3D detection classification network.
3. The lung nodule detection method based on the combination of the depth network and the optical flow mechanism as claimed in claim 1, wherein: the performing feature extraction and analysis on the 3D lung CT image comprises:
inputting the 3D lung CT image into a feature extraction network, and obtaining a suggested region of the nodule by judging whether the image is the two-classification output of the nodule and predicting a bounding box; inputting the obtained suggested area into a RoIploling layer to carry out size standardization processing on a characteristic diagram; outputting the data to a classification probability prediction layer and a boundary frame prediction layer through two full-connection layers;
and detecting the CT image to be detected by using the trained three suspected lung nodule detection models to obtain the classification probability and the bounding box of the suspected nodules, classifying by using the trained three weak classification models, and performing majority voting on the classification results of the three weak classification models to select a final classification result.
4. The lung nodule detection method based on the combination of the depth network and the optical flow mechanism as claimed in claim 3, wherein: before the three trained suspected pulmonary nodule detection models are used for detecting the CT image to be detected, the method further comprises the following steps:
preprocessing positive and negative samples in the training data to balance the proportion of the positive and negative samples, and then classifying the preprocessed negative samples by using a pre-screening model to screen out the negative samples with wrong classification; and training three weak classification models by using the training data after the pretreatment and the screening.
5. The lung nodule detection method based on the combination of the depth network and the optical flow mechanism as claimed in claim 1, wherein: the pre-processing the 3D lung CT image comprises:
and carrying out quantization and smoothing processing on the 3D lung CT image.
6. A lung nodule detection apparatus combining a depth network and an optical flow mechanism, comprising: the method comprises the following steps:
the image acquisition module is used for acquiring a 3D lung CT image and preprocessing the 3D lung CT image;
the feature extraction module is used for extracting and analyzing features of the 3D lung CT image;
the network construction module is used for constructing a depth optical flow network to detect and identify lung nodules in the 3D lung CT image;
and the report generation module is used for obtaining a pulmonary nodule detection result and generating a detection report.
7. The device of claim 6, wherein the lung nodule detection apparatus comprises: the network building module comprises:
the preprocessing unit is used for automatically processing the 3D lung CT image into a CETS (computerized tomography System) readable 3D lung image;
and the classification unit is used for detecting the lung nodule region of the 3D lung CT image by utilizing a TNet lung nodule 3D detection classification network, classifying the lung nodules, and inputting the extracted lung nodules into the TNet or TNet-VQ lung nodule 3D detection classification network.
8. A computer device, comprising: a memory and one or more processors;
the memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement a method for lung nodule detection with a combination of depth-network and optical-flow mechanisms as recited in any one of claims 1-5.
9. A storage medium containing computer-executable instructions for performing a combined depth network and optical flow mechanism lung nodule detection method as claimed in any one of claims 1 to 5 when executed by a computer processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110975691.7A CN113658657A (en) | 2021-08-24 | 2021-08-24 | Lung nodule detection method and device combining depth network and optical flow mechanism |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110975691.7A CN113658657A (en) | 2021-08-24 | 2021-08-24 | Lung nodule detection method and device combining depth network and optical flow mechanism |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113658657A true CN113658657A (en) | 2021-11-16 |
Family
ID=78481792
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110975691.7A Pending CN113658657A (en) | 2021-08-24 | 2021-08-24 | Lung nodule detection method and device combining depth network and optical flow mechanism |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113658657A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107103187A (en) * | 2017-04-10 | 2017-08-29 | 四川省肿瘤医院 | The method and system of Lung neoplasm detection classification and management based on deep learning |
| CN108664971A (en) * | 2018-05-22 | 2018-10-16 | 中国科学技术大学 | Pulmonary nodule detection method based on 2D convolutional neural networks |
| CN109003672A (en) * | 2018-07-16 | 2018-12-14 | 北京睿客邦科技有限公司 | A kind of early stage of lung cancer detection classification integration apparatus and system based on deep learning |
| 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 |
| CN113506288A (en) * | 2021-07-28 | 2021-10-15 | 中山仰视科技有限公司 | Lung nodule detection method and device based on transform attention mechanism |
-
2021
- 2021-08-24 CN CN202110975691.7A patent/CN113658657A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107103187A (en) * | 2017-04-10 | 2017-08-29 | 四川省肿瘤医院 | The method and system of Lung neoplasm detection classification and management based on deep learning |
| CN108664971A (en) * | 2018-05-22 | 2018-10-16 | 中国科学技术大学 | Pulmonary nodule detection method based on 2D convolutional neural networks |
| CN109003672A (en) * | 2018-07-16 | 2018-12-14 | 北京睿客邦科技有限公司 | A kind of early stage of lung cancer detection classification integration apparatus and system based on deep learning |
| 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 |
| CN113506288A (en) * | 2021-07-28 | 2021-10-15 | 中山仰视科技有限公司 | Lung nodule detection method and device based on transform attention mechanism |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Mahapatra et al. | Interpretability-driven sample selection using self supervised learning for disease classification and segmentation | |
| CN108464840B (en) | Automatic detection method and system for breast lumps | |
| EP4006831A1 (en) | Image processing method and apparatus, server, medical image processing device and storage medium | |
| CN108564026B (en) | Network construction method and system for thyroid tumor cytology smear image classification | |
| CN110807764A (en) | Lung cancer screening method based on neural network | |
| CN107280697A (en) | Lung neoplasm grading determination method and system based on deep learning and data fusion | |
| He et al. | Automated model design and benchmarking of deep learning models for covid-19 detection with chest ct scans | |
| CN109741312A (en) | A kind of Lung neoplasm discrimination method, device, equipment and medium | |
| CN111553892A (en) | Lung nodule segmentation calculation method, device and system based on deep learning | |
| CN112513927A (en) | Wind turbine blade defect inspection based on convolutional neural networks | |
| JP2023534448A (en) | Evaluation of feature heterogeneity in digital pathological images using machine learning techniques | |
| CN112990214A (en) | Medical image feature recognition prediction model | |
| US12008750B2 (en) | Generating annotation data of tissue images | |
| Milosevic | Different approaches to Imaging Mass Cytometry data analysis | |
| CN113764101B (en) | Novel auxiliary chemotherapy multi-mode ultrasonic diagnosis system for breast cancer based on CNN | |
| CN111028940A (en) | Multi-scale lung nodule detection method, device, equipment and medium | |
| CN114359671A (en) | Multi-target learning-based ultrasonic image thyroid nodule classification method and system | |
| CN111724356B (en) | Image processing method and system for CT image pneumonia recognition | |
| CN112801940A (en) | Model evaluation method, device, equipment and medium | |
| CN113658657A (en) | Lung nodule detection method and device combining depth network and optical flow mechanism | |
| US20220301689A1 (en) | Anomaly detection in medical imaging data | |
| Ge et al. | COVID-19 imaging-based AI research-a literature review | |
| WO2023102995A1 (en) | Intracranial hemorrhage area detection method and apparatus, computer device and storage medium | |
| CN110310897A (en) | The guidance inspection of semiconductor wafer based on space density analysis | |
| Ke et al. | An ultrasonic image recognition method for papillary thyroid carcinoma based on depth convolution neural network |
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 |