CN109377477B - Image classification method and device and computer readable storage medium - Google Patents
Image classification method and device and computer readable storage medium Download PDFInfo
- Publication number
- CN109377477B CN109377477B CN201811122177.3A CN201811122177A CN109377477B CN 109377477 B CN109377477 B CN 109377477B CN 201811122177 A CN201811122177 A CN 201811122177A CN 109377477 B CN109377477 B CN 109377477B
- Authority
- CN
- China
- Prior art keywords
- image
- images
- subset
- layers
- classification
- 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.)
- Active
Links
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- 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
- G06T2207/10132—Ultrasound image
-
- 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
Abstract
The invention discloses a method and a device for classifying images and a computer readable storage medium, which relate to the field of computers, in particular to the field of medical image processing, wherein limited medical images and medical images obtained by transformation are used as training data sets, and for different organs, a deep learning network method is used for image classification training to obtain classifiers of corresponding layers of the different organs; the new input image is finally classified by the classifier. The invention also provides an image classification device and a computer readable storage medium. The invention completes the training of the classifier by using less data and finally realizes the automatic classification of the images, thereby greatly improving the image classification efficiency.
Description
Technical Field
The present invention relates to the field of computer image processing/artificial intelligence, and in particular, to a method and an apparatus for image classification and a computer-readable storage medium.
Background
Digital medical image technology is one of the most important diagnostic means in modern medicine, and with the continuous development and popularization of digital medical image diagnostic technology, the data volume of the generated medical images is larger and larger, and the use frequency of doctors is higher and higher. When medical image information of a patient is acquired, a part of the medical image information is generally scanned more to completely cover the position of a suspected lesion, and when diagnosis is performed, an image containing the suspected lesion needs to be called out for diagnosis and analysis.
With the continuous improvement of medical information, hospitals begin to use his (medical information system) system, pacs (picture Archiving and Communication systems) system, etc., and the storage and transmission of medical image data with large data volume needs to occupy more storage space and transmission bandwidth.
Disclosure of Invention
In view of the above, the present invention provides a method, an apparatus and a computer-readable storage medium for image classification, to solve the problems: the medical images are classified, so that the time for a doctor to screen the images is reduced, and redundant invalid images can be removed, so that the image data volume is reduced, and the transmission is facilitated.
According to some embodiments of the present disclosure, there is provided a method of image classification, comprising: acquiring a medical image; screening, for different organs, a subset image of the medical image that contains all images of the corresponding organ; dividing the subset image into a number of layers of images; carrying out classification training on the layered images to obtain classifiers of corresponding layers of different organs; and classifying the images to be classified through a classifier.
In some embodiments, the medical image acquired comprises an electron computed tomography, CT, image, or a magnetic resonance, MRI, or B-mode image.
In some embodiments, the acquiring of the medical image further comprises processing the medical image by the acquired image, the processing method comprising: and (3) performing rotation of any angle or movement of any size or noise increasing/decreasing operation on the medical image, and obtaining a group of new images once or by a combination of rotation of one angle or movement of one size or noise increasing.
In some embodiments, the medical image and the new image are screened for different organs, including a subset image of all images of the corresponding organ, having the subset image a total of N layers, the first layer of which corresponds to the top layer of the organ and the last layer of which corresponds to the bottom most layer of the organ.
In some embodiments, the partial images include one or more layers of images, the partial images are Q partial images, and the method for dividing the subset of images includes: dividing Q parts (Q is less than or equal to N) from N layers, wherein the 1 st part comprises a top layer, the Q part comprises a bottom layer, and the j (1< j < Q) th part in the middle is evenly distributed; the number of layers of the subset image included in the j-th part is: j =1+ ((Q-1)/(N-1)) + Ni, Q being the total number of Q segments, N being the total number of layers of the subset image, Ni being the ith layer of the subset image, rounded up, Q, N, l, j being a natural number greater than or equal to 1.
In some embodiments, the part images 1, 2, …, i, …, (Q-1), Q part are used as training data sets of corresponding parts, and are respectively input into an image classifier, and the classifier training of the corresponding parts is performed by using a deep learning network method (such as CNN, SVM), so as to respectively obtain Q part image classifiers of the layers 1, 2, …, i, …, (Q-1), Q.
In some embodiments, the classifying the image to be classified by the classifier specifically includes: and sequentially inputting the images to be classified into the Q partial image classifier respectively, judging whether the images belong to the part, and outputting the part of the images to be classified and the probability thereof, wherein the part of the images to be classified has the maximum image attribution probability.
According to other embodiments of the present disclosure, there is provided an apparatus for image classification, including: a medical image acquisition module for acquiring a first medical image; a medical image processing module for processing the medical image to obtain a second medical image; the subset image screening module is used for screening the subset images aiming at different organs; the subset image dividing module is used for dividing the subset image into a plurality of partial images, the partial images comprise one layer or a plurality of layers of images, the partial images are Q partial images, Q parts are divided from N layers, the 1 st part comprises a top layer, the Q part comprises a bottom layer, the j th part in the middle is evenly distributed, and j is more than 1 and less than Q; the number of layers of the subset image included in the j-th part is: j =1+ ((Q-1)/(N-1)) + Ni, Q being the total number of Q segments, N being the total number of layers of the subset image, Ni being the ith layer of the subset image, rounded, Q, N, i, j being a natural number greater than or equal to 1; an image training module, configured to perform classification training on the partial images to obtain corresponding partial classifiers, where the 1 st, 2 nd, … th, i, …, (Q-1) th, Q-sections of the partial images are respectively used as training data sets of corresponding layers, and are input into corresponding partial image classifiers, and the machine learning method and/or the deep learning method are used to perform classifier training on the corresponding partial classifiers to obtain the 1 st, 2 nd, … th, i, …, (Q-1) th, Q-section Q-partial image classifiers, respectively; (ii) a And the image classification module is used for classifying the images to be classified through the classifier.
After the medical image acquisition module acquires a medical image, the medical image is transmitted to the medical image processing module for image processing to generate a new image, or the medical image is transmitted to the subset image screening module, the subset image screening module screens out subset images of different organs after receiving the medical image transmitted by the medical image acquisition module and/or the medical image processing module and transmits the subset images to the subset image dividing module, the subset image dividing module divides the subset images into a plurality of partial images after receiving the subset images transmitted by the subset image screening module, and transmits the partial images to the image training module for training classifiers of corresponding parts of the different organs and then transmits the trained classifiers to the image classification module, and finally, inputting the image to be classified into the image classification module to finish classification.
According to other embodiments of the present disclosure, there is provided an apparatus for image classification, including: a memory; and a processor coupled to the memory, the memory connected to the processor by a bus, the processor connected to the medical image acquisition module, the medical image processing module, the subset image screening module, the subset image partitioning module, the image training module, and the image classification module by a bus, respectively, the processor configured to perform the method of image classification as in any one of the preceding embodiments based on instructions stored in the memory device.
According to further embodiments of the present disclosure, there is provided an image classification computer-readable storage medium, on which a computer program is stored, characterized in that the program, when executed by a processor, implements the steps of the method of image classification according to any one of the preceding embodiments.
The image classifier is trained by acquiring fewer images and generating new images, so that the requirement on the data volume is reduced, the trained image classifier can identify and classify effective images, and invalid images are removed, so that the image data volume is reduced.
Drawings
In order to more clearly illustrate the embodiments of the present disclosure 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 disclosure, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 illustrates a flow diagram of a method of image classification of some embodiments of the present disclosure.
Fig. 2 illustrates a schematic structural diagram of an apparatus for image classification according to some embodiments of the present disclosure.
Fig. 3 shows a schematic structural diagram of an image classification apparatus according to further embodiments of the present disclosure.
Fig. 4 is a schematic structural diagram of an image classification apparatus according to still other embodiments of the present disclosure.
Detailed Description
The technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all of the embodiments. The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the disclosure, its application, or uses. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
The method for classifying images provided by the present disclosure will be described in conjunction with fig. 1.
A flow diagram of a method of image classification of some embodiments is shown in fig. 1. The method comprises the following steps of S101-S105:
s101, acquiring a medical image;
in some embodiments, the medical image acquired is an electron computed tomography, CT, image.
In some embodiments, the medical image acquired is a magnetic resonance MRI image.
In some embodiments, the medical image acquired is a B-mode ultrasound image.
In some embodiments, the medical image further comprises an obtained image for processing the medical image by rotating the medical image by an arbitrary angle, i.e. obtaining a set of new images.
In some embodiments, the medical image further comprises an obtained image that is processed on the medical image by shifting the medical image by an arbitrary size, i.e. by obtaining a set of new images.
In some embodiments, the medical image further comprises an acquired image for processing the medical image by applying a noise-adding operation to the medical image, i.e. obtaining a set of new images.
In some embodiments, the medical image further comprises an acquired image obtained by processing the medical image by moving the medical image by an arbitrary size and adding noise, i.e. obtaining a set of new images.
S102, aiming at different organs, screening subset images of all images of the medical image containing the corresponding organ;
in some embodiments, the medical image and the new image are screened for a different organ, such as the liver, and a subset of the images that includes exactly the entire image of the liver organ, such as a CT image with a 2mm slice spacing, the subset having a total of 85 slices, with slice 1 corresponding to the top slice of the organ and slice 85 corresponding to the bottom slice of the organ.
And S103, dividing the subset image into a plurality of partial images.
In some embodiments, the number of layers of images are 20 partial images, each partial image includes a number of layers of images, and the method for dividing the subset of images includes: dividing 20 parts from 85 layers, wherein the 1 st part comprises a top layer, the 20 th part comprises a bottom layer, and the j (1< j <20) th part in the middle is evenly distributed; the number of layers of the subset image corresponding to the j-th part is as follows: j =1+ (Q-1)/(N-1) × (Ni), Q being the total number of Q parts, N being the total number of layers of the subset image, and Ni being the ith layer of the subset image, rounded up. For example, the 1 st, 2 nd, 3 rd layers of the subset image included in the 1 st part of the plurality of partial images are: when j =1, 1+ (20-1)/(85-1) × Ni, Ni are 1, 2, 3 respectively, after rounding up, 1+ (20-1)/(85-1) × Ni are all 1, namely 1, 2, 3 th layer is divided into 1 st part; also included in section 13 are layers 1, 2, and 3 of the subset image, namely: when j =12, 1+ (20-1)/(85-1) × Ni, Ni are 48, 49, 50, 51 respectively, and after rounding up, 1+ (20-1)/(85-1) × Ni is 12, that is, the 48 th, 49 th, 50 th, 51 th layer is divided into 12 th parts.
S104, carrying out classification training on the layered images to obtain classifiers of corresponding layers of different organs;
in some embodiments, the 1 st, 2 nd, … th, i, … th, 19 th, 20 th parts of the subset images are used as training data sets of corresponding parts, and are respectively input into image classifiers, and deep learning network methods (such as SVM) are used for training the classifiers of the corresponding parts, so as to respectively obtain the image classifiers of the 1 st, 2 nd, … th, i, … th, 19 th, 20 th parts.
And S105, classifying the images to be classified through the classifier.
In some embodiments, the classification method of the classifier includes: and sequentially inputting the images to be classified (if 127 layers exist) into the 20 partial image classifier respectively, judging whether the images belong to the part respectively, and outputting the part of the images to be classified and the probability thereof, wherein the part of the images to be classified has the maximum image attribution probability. If the probability that layer 3 belongs to part 1 is greater than the probability of belonging to other parts, then layer 3 belongs to part 1.
The present disclosure also provides an image classification apparatus, which is described in conjunction with fig. 2.
Fig. 2 is a block diagram of some embodiments of the image classification apparatus of the present disclosure, and as shown in fig. 2, the image classification apparatus 20 of the embodiment includes:
a medical image acquisition module 201 for acquiring a first medical image;
a subset image screening module 203, configured to screen the subset images for different organs;
a layered image dividing module 204, configured to divide the subset image into a plurality of partial images, where the plurality of partial images include one or more layers of images, and make the plurality of partial images be Q partial images, and divide Q parts from N layers, where the 1 st part includes a top layer, the Q th part includes a bottom layer, and the j th part in the middle is evenly distributed, where 1< j < Q; the number of layers of the subset image included in the j-th part is: j =1+ ((Q-1)/(N-1)) + Ni, Q being the total number of Q segments, N being the total number of layers of the subset image, Ni being the ith layer of the subset image, rounded, Q, N, i, j being a natural number greater than or equal to 1;
an image training module 205, configured to perform classification training on the partial images to obtain classifiers for corresponding parts, where the 1 st, 2 nd, … th, i, …, (Q-1) th, Q-th parts of the partial images are used as training data sets of corresponding layers, respectively, and are input into image classifiers of corresponding parts, and perform classifier training of corresponding parts by using a machine learning method and/or a deep learning method to obtain Q-th, 2 nd, … th, i, …, (Q-1), and Q-th part image classifiers, respectively;
and the image classification module 206 is used for classifying the images to be classified through the classifier.
In particular, the apparatus for image classification further comprises a medical image processing module 202 for processing the medical image to obtain a second medical image.
After the medical image obtaining module 201 obtains a medical image, the medical image is transmitted to the medical image processing module 202 for image processing to generate a new image, or the medical image is transmitted to the subset image screening module 203, the subset image screening module screens out subset images of different organs after receiving the medical image transmitted by the medical image obtaining module and/or the medical image processing module, and transmits the subset images to the subset image dividing module 204, the subset image dividing module 204 divides the subset images into a plurality of partial images after receiving the subset images transmitted by the subset image screening module 203, and then transmits the plurality of partial images to the image training module 205 for training classifiers of corresponding parts of the different organs, and then transmits the trained classifiers to the image classification module 206, finally, the image to be classified is input to the image classification module 206, and classification is completed.
The means for image classification in embodiments of the present disclosure may each be implemented by various computing devices or computer systems, described below in conjunction with fig. 3 and 4.
FIG. 3 is a block diagram of some embodiments of an apparatus for image classification according to the present disclosure. As shown in fig. 3, the apparatus 30 of this embodiment includes: a memory 301 and a processor 302 coupled to the memory 301, the memory 301 being connected to the processor 302 by a bus, the processor 302 being connected to the medical image acquisition module 201, the medical image processing module 202, the subset image filtering module 203, the subset image dividing module 204, the image training module 205 and the image classification module 206 by a bus, respectively, the processor 302 being configured to perform the method of image classification in any of the embodiments of the present disclosure based on instructions stored in the memory 301.
Fig. 4 is a block diagram of another embodiment of an apparatus for image classification according to the present disclosure. As shown in fig. 4, the apparatus 40 of this embodiment includes: the bus 401, memory 404, and processor 402 are similar to the memory 301 and processor 302, respectively. An input output interface 403, a storage interface 405, a network interface 406, etc. may also be included. These interfaces 403, 405, 406 and the memory 404 may be connected to the processor 402, for example, via a bus 401. The input/output interface 403 provides a connection interface for input/output devices such as a display, a mouse, a keyboard, and a touch screen. The storage interface 405 provides a connection interface for external storage devices such as an SD card and a usb disk. The network interface 406 provides a connection interface for various networked devices, such as may connect to a database server or a cloud storage server, among others.
The present disclosure also provides a computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, realizes the steps of the method of image classification of any of the preceding embodiments.
As will be appreciated by one skilled in the art, embodiments of the present disclosure may be provided as a method, system, or computer program product. Accordingly, the present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable non-transitory storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the disclosure. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only exemplary of the present disclosure and is not intended to limit the present disclosure, which is to be construed in any way as imposing limitations thereon, such as the appended claims, and all changes and equivalents that fall within the true spirit and scope of the present disclosure.
Claims (8)
1. A method of image classification, comprising:
acquiring a medical image;
screening, for different organs, a subset image of the medical image that contains all images of the corresponding organ;
dividing the subset image into a plurality of parts, specifically:
the method for dividing the subset images comprises the following steps of:
dividing Q parts from N layers, wherein the 1 st part comprises a top layer, the Q part comprises a bottom layer, and the j part in the middle is evenly distributed, wherein j is 1< Q;
the number of layers of the subset image included in the j-th part is: j =1+ ((Q-1)/(N-1)) + Ni, Q being the total number of Q segments, N being the total number of layers of the subset image, Ni being the ith layer of the subset image, rounded, Q, N, i, j being a natural number greater than or equal to 1;
and carrying out classification training on the plurality of partial images to obtain classifiers of corresponding layers of different organs, which specifically comprises the following steps:
the 1 st, 2 nd, … th, i, …, (Q-1) th, Q-part of the partial images are respectively used as training data sets of corresponding layers and input into image classifiers of corresponding parts, and classifier training of the corresponding parts is carried out by utilizing a machine learning method and/or a deep learning method, so that the Q-part image classifiers of the 1 st, 2 nd, … th, i, …, (Q-1) th, Q-part are respectively obtained;
and classifying the images to be classified through a classifier.
2. An image classification method as claimed in claim 1, characterized in that the medical image acquired comprises an electron-computed tomography CT image, or a magnetic resonance MRI image, or a B-mode ultrasound image.
3. An image classification method as claimed in claim 1, characterized in that the acquiring of the medical image further comprises processing the medical image to obtain an image, the processing method comprising:
and (3) performing rotation of any angle or movement of any size or noise increasing/decreasing operation on the medical image, and obtaining a group of new images once or by a combination of rotation of one angle or movement of one size or noise increasing.
4. An image classification method as claimed in claim 3, characterized in that for different organs, the medical image and the new image are screened to include a subset image of all images of the corresponding organ, and the subset images have a total of N layers, the first layer of which corresponds to the top layer of the organ and the last layer of which corresponds to the bottom layer of the organ.
5. The method for classifying images according to claim 1, wherein the classifying the images to be classified by the classifier specifically comprises: and inputting the images to be classified into Q image classifiers, respectively judging whether the images belong to the part, and outputting the part of the images to be classified and the probability of the part, wherein the image attribution probability is the maximum part.
6. An apparatus for image classification, comprising:
a medical image acquisition module for acquiring a first medical image;
a subset image screening module for screening the subset image for a first medical image of a different organ;
the subset image dividing module is used for dividing the subset image into a plurality of partial images, the partial images comprise one layer or a plurality of layers of images, the partial images are Q partial images, Q parts are divided from N layers, the 1 st part comprises a top layer, the Q part comprises a bottom layer, the j th part in the middle is evenly distributed, and j is more than 1 and less than Q; the number of layers of the subset image included in the j-th part is: j =1+ ((Q-1)/(N-1)) + Ni, Q being the total number of Q segments, N being the total number of layers of the subset image, Ni being the ith layer of the subset image, rounded, Q, N, i, j being a natural number greater than or equal to 1;
an image training module, configured to perform classification training on the partial images to obtain corresponding partial classifiers, where the 1 st, 2 nd, … th, i, …, (Q-1) th, Q-sections of the partial images are respectively used as training data sets of corresponding layers, and are input into corresponding partial image classifiers, and the machine learning method and/or the deep learning method are used to perform classifier training on the corresponding partial classifiers to obtain the 1 st, 2 nd, … th, i, …, (Q-1) th, Q-section Q-partial image classifiers, respectively;
and the image classification module is used for classifying the images to be classified through the classifier.
7. An apparatus for image classification as claimed in claim 6, comprising:
a memory; and
a processor coupled to the memory, the memory connected to the processor by a bus, the processor connected to the medical image acquisition module, the medical image processing module, the subset image screening module, the subset image partitioning module, the image training module and the image classification module by a bus, respectively, the processor configured to perform the method of image classification of any of claims 1-5 based on instructions stored in the memory device.
8. An image classification computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811122177.3A CN109377477B (en) | 2018-09-26 | 2018-09-26 | Image classification method and device and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811122177.3A CN109377477B (en) | 2018-09-26 | 2018-09-26 | Image classification method and device and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109377477A CN109377477A (en) | 2019-02-22 |
| CN109377477B true CN109377477B (en) | 2020-09-18 |
Family
ID=65401871
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811122177.3A Active CN109377477B (en) | 2018-09-26 | 2018-09-26 | Image classification method and device and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109377477B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993206A (en) * | 2019-03-01 | 2019-07-09 | 福建工程学院 | A kind of vegetables recognition methods and terminal |
| CN110399913A (en) * | 2019-07-12 | 2019-11-01 | 杭州依图医疗技术有限公司 | The classification method and device at position are shot in a kind of medical image |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6466687B1 (en) * | 1997-02-12 | 2002-10-15 | The University Of Iowa Research Foundation | Method and apparatus for analyzing CT images to determine the presence of pulmonary tissue pathology |
| CN102034117A (en) * | 2010-12-17 | 2011-04-27 | 东软集团股份有限公司 | Image classification method and apparatus |
| CN102135606A (en) * | 2010-12-13 | 2011-07-27 | 电子科技大学 | KNN (K-Nearest Neighbor) sorting algorithm based method for correcting and segmenting grayscale nonuniformity of MR (Magnetic Resonance) image |
| CN108471995A (en) * | 2015-09-30 | 2018-08-31 | 上海联影医疗科技有限公司 | The system and method for determining breast area in medical image |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017092615A1 (en) * | 2015-11-30 | 2017-06-08 | 上海联影医疗科技有限公司 | Computer aided diagnosis system and method |
-
2018
- 2018-09-26 CN CN201811122177.3A patent/CN109377477B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6466687B1 (en) * | 1997-02-12 | 2002-10-15 | The University Of Iowa Research Foundation | Method and apparatus for analyzing CT images to determine the presence of pulmonary tissue pathology |
| CN102135606A (en) * | 2010-12-13 | 2011-07-27 | 电子科技大学 | KNN (K-Nearest Neighbor) sorting algorithm based method for correcting and segmenting grayscale nonuniformity of MR (Magnetic Resonance) image |
| CN102034117A (en) * | 2010-12-17 | 2011-04-27 | 东软集团股份有限公司 | Image classification method and apparatus |
| CN108471995A (en) * | 2015-09-30 | 2018-08-31 | 上海联影医疗科技有限公司 | The system and method for determining breast area in medical image |
Non-Patent Citations (2)
| Title |
|---|
| Symbolic features for classification of medical X-ray body organ images;Amir Rajaei et al;《2012 12th International Conference on Hybrid Intelligent Systems (HIS)》;20121207;第378-383页 * |
| 结合全卷积网络和GrowCut的肾皮质分割算法;时永刚等;《中国图象图形学报》;20171031(第10期);第1418-1427页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109377477A (en) | 2019-02-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11615878B2 (en) | Systems and methods for integrating neural network image analyses into medical image viewing applications | |
| Chartrand et al. | Deep learning: a primer for radiologists | |
| US10417788B2 (en) | Anomaly detection in volumetric medical images using sequential convolutional and recurrent neural networks | |
| US9760990B2 (en) | Cloud-based infrastructure for feedback-driven training and image recognition | |
| US11646119B2 (en) | Systems and methods for automated analysis of medical images | |
| CN103514597A (en) | Image processing device | |
| JP6727176B2 (en) | Learning support device, method of operating learning support device, learning support program, learning support system, and terminal device | |
| US11151703B2 (en) | Artifact removal in medical imaging | |
| CN109377477B (en) | Image classification method and device and computer readable storage medium | |
| Maity et al. | Automatic lung parenchyma segmentation using a deep convolutional neural network from chest X-rays | |
| CN109447120A (en) | A kind of method, apparatus and computer readable storage medium of Image Automatic Segmentation | |
| US11132793B2 (en) | Case-adaptive medical image quality assessment | |
| Liz et al. | Deep learning for understanding multilabel imbalanced Chest X-ray datasets | |
| Panayides et al. | The promise of big data technologies and challenges for image and video analytics in healthcare | |
| CN109345545A (en) | A kind of method, apparatus and computer readable storage medium of segmented image generation | |
| AU2019204365B1 (en) | Method and System for Image Segmentation and Identification | |
| Horry et al. | Factors determining generalization in deep learning models for scoring COVID-CT images | |
| CN109410217B (en) | Image segmentation method and device and computer readable storage medium | |
| Marathe et al. | Detection of Lung Cancer using CT Scans with Deep Learning Approach | |
| CN114402395A (en) | VRDS 4D medical image-based spine disease identification method and related device | |
| US11334976B1 (en) | Medical imaging with functional architecture tracking | |
| CN113393445B (en) | Breast cancer image determination method and system | |
| KR102390347B1 (en) | Method for assisting in generation of pathology report from pathology image and apparatus using the same | |
| Vidhya et al. | A Systematic Approach for Constructing 3D MRI Brain Image over 2D Images | |
| Anand et al. | RoMIA: a framework for creating Robust Medical Imaging AI models for chest radiographs |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200807 Address after: 215028 unit 301, floor 3, No.288, Dongping street, Suzhou Industrial Park, Suzhou area, China (Jiangsu) pilot Free Trade Zone, Suzhou City, Jiangsu Province Applicant after: Suzhou Liulian Technology Co.,Ltd. Address before: 523000 Guangdong province Dongguan City Chashan town water Tong Kok Wai Industrial Zone Applicant before: GUANGDONG EDA TECHNOLOGY Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |