CN107563997B - Skin disease diagnosis system, construction method, classification method and diagnosis device - Google Patents
Skin disease diagnosis system, construction method, classification method and diagnosis device Download PDFInfo
- Publication number
- CN107563997B CN107563997B CN201710735310.1A CN201710735310A CN107563997B CN 107563997 B CN107563997 B CN 107563997B CN 201710735310 A CN201710735310 A CN 201710735310A CN 107563997 B CN107563997 B CN 107563997B
- Authority
- CN
- China
- Prior art keywords
- classifier
- skin
- image
- skin disease
- training image
- 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
Abstract
The invention provides a skin disease diagnosis system, a construction method, a diagnosis method and a diagnosis device. The construction method comprises the following steps: acquiring a training image, the training image comprising: a first training image corresponding to a first skin condition, a second training image corresponding to a second skin condition, and a third training image corresponding to a third skin condition; constructing a first classifier for identifying a first skin disease by using a first neural network, and taking the three training images as input data of the first neural network to train the first classifier; constructing a second classifier for identifying a second skin disease by using a second neural network, and taking the three training images as input data of the second neural network to train the second classifier; constructing a skin condition diagnostic system based on the first classifier and the second classifier. The invention constructs a skin disease diagnosis system based on an image recognition technology, and can automatically diagnose skin diseases.
Description
Technical Field
The invention relates to the technical field of image recognition application, in particular to a skin disease diagnosis system, a construction method, a diagnosis method and a diagnosis device.
Background
With the advent of the big data age, deep learning techniques are increasingly being applied to the application of image recognition. Deep learning is a powerful technique that stems from artificial neural networks. The artificial neural network is inspired by the neural network of the natural life, and the capability of accurately recognizing images similar to human beings is achieved by constructing multiple layers of neurons and repeatedly training a large amount of data.
The current image recognition products are more used for recognizing living images such as human faces, license plates, moving targets and the like. For the medical field, especially skin-like diseases, image recognition classes have not been cited yet for non-manual diagnosis.
Disclosure of Invention
The invention aims to provide a technical scheme for realizing self-diagnosis of skin diseases based on an image recognition technology.
To achieve the above object, in one aspect, an embodiment of the present invention provides a method for constructing a skin disease diagnosis system, including:
acquiring a training image, the training image comprising: a first training image corresponding to a first skin condition, a second training image corresponding to a second skin condition, and a third training image corresponding to a third skin condition;
constructing a first classifier for identifying a first skin disease by using a first neural network, and taking the first training image, the second training image and the third training image as input data of the first neural network to train the first classifier;
constructing a second classifier for identifying a second skin disease by using a second neural network, and taking the first training image, the second training image and the third training image as input data of the first neural network to train the second classifier;
constructing a skin disease diagnosis system for diagnosing the first skin disease, the second skin disease, and the third skin disease based on at least the first classifier and the second classifier.
Wherein, the construction method further comprises the following steps:
constructing a third classifier for distinguishing a first skin disease from a second skin disease using a third neural network, and using the first training image and the second training image as input data of the third neural network to train the third classifier;
wherein the step of constructing the skin disease diagnosis system specifically comprises:
constructing a skin disease diagnosis system for diagnosing the first skin disease, the second skin disease, and the third skin disease based on the first classifier, the second classifier, and the third classifier.
Wherein the first skin disease, the second skin disease and the third skin disease are each one of the following skin diseases:
melanoma, keratosis, and nevi.
After the training image is obtained and before the first classifier, the second classifier and the third classifier are constructed, the construction method further includes:
carrying out preprocessing operation on the training image;
the preprocessing operation comprises at least one of the following modes:
clipping the invalid recognition area of the training image;
the display sizes of the different training images are normalized.
After the training image is obtained and before the first classifier, the second classifier and the third classifier are constructed, the construction method further includes:
performing data enhancement operation on the training image;
the data enhancement operation comprises at least one of the following modes:
optimizing saturation and/or brightness and/or contrast of the training image;
flipping and/or rotating the training image.
On the other hand, the embodiment of the invention also provides a skin disease diagnosis system which is constructed by the construction method provided by the invention.
In addition, an embodiment of the present invention further provides a method for diagnosing a skin disease, which is applied to the skin disease diagnosis system provided by the present invention, and includes:
acquiring a target skin image to be diagnosed;
using the target skin image as input data for a first classifier and a second classifier in the skin disease diagnosis system;
if the first classifier identifies that the target skin image corresponds to a first skin disease and the second classifier does not identify that the target skin image corresponds to a second skin disease, determining that the target skin image is affected by the first skin disease;
if the first classifier does not recognize that the target skin image corresponds to a first skin disease and the second classifier recognizes that the target skin image corresponds to a second skin disease, determining that the target skin image has the second skin disease;
and if the first classifier does not recognize that the target skin image corresponds to the first skin disease and the second classifier does not recognize that the target skin image corresponds to the second skin disease, determining that the target skin image has the third skin disease.
Wherein the skin disease diagnosis system further comprises a third classifier, and the diagnosis method further comprises:
if the first classifier identifies that the target skin image corresponds to a first skin disease and the second classifier identifies that the target skin image corresponds to a second skin disease, taking the target skin image as input data of the third classifier;
and if the third classifier identifies that the target skin image corresponds to the first skin disease, determining that the target skin image has the first skin disease.
And if the third classifier identifies that the target skin image corresponds to the second skin disease, determining that the target skin image has the second skin disease.
Further, an embodiment of the present invention also provides a diagnostic apparatus for skin diseases, including:
the image acquisition module is used for acquiring a target skin image to be diagnosed;
a data processing module storing a computer program for executing the steps of the diagnostic method as described above.
Wherein the diagnostic device further comprises:
and the diagnosis result output module is used for outputting the diagnosis result of the data processing module.
The scheme of the invention has the following beneficial effects:
the scheme of the invention is based on the image recognition technology, constructs a skin disease diagnosis system, can automatically diagnose the skin disease or assist doctors to diagnose the skin disease, and can reduce the probability of misdiagnosis, thereby having high practical value in the medical field.
Drawings
FIG. 1 is a schematic diagram illustrating steps of a method for constructing a skin disease diagnosis system according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating the steps of a diagnostic method according to an embodiment of the present invention;
FIG. 3 is a schematic structural diagram of a skin disease diagnosis system provided in an embodiment of the present invention in practical application;
fig. 4 is a schematic structural diagram of a diagnostic apparatus according to an embodiment of the present invention.
Detailed Description
In order to make the technical problems, technical solutions and advantages of the present invention more apparent, the following detailed description is given with reference to the accompanying drawings and specific embodiments.
The invention provides a technical scheme for diagnosing skin diseases based on an image recognition technology.
In one aspect, an embodiment of the present invention provides a method for constructing a skin disease diagnosis system, as shown in fig. 1, including:
In practical applications, the training images are the case images of the three skin diseases, and can be obtained from clinical or simulation.
And 120, constructing a first classifier for identifying the first skin disease by using the first neural network, and training the first classifier by taking the first training image, the second training image and the third training image as input data of the second neural network.
It should be noted here that, the method for constructing a classifier using a neural network is a prior art, and the first classifier can be constructed by using the latest CNN neural network models such as current inclusion-ResNet, google net v3, and the detailed implementation process is not repeated for example because the step is only required to perform some adaptive parameter modifications on the basis of the existing neural network model.
similarly, the second classifier can be constructed by using the existing neural network model.
And step 140, constructing a diagnosis model for diagnosing the first skin disease, the second skin disease and the third skin disease at least based on the first classifier and the second classifier to form a skin disease diagnosis system.
The construction method of the embodiment is used for constructing a diagnostic system capable of diagnosing skin diseases, can automatically diagnose the skin diseases or assist doctors in diagnosing the skin diseases, and can reduce the probability of misdiagnosis, so that the construction method has high practical value in the medical field.
Further, when the target skin image to be diagnosed is input into the skin disease diagnosis system of the present embodiment, if the first classifier identifies that the target skin image corresponds to the first skin disease and the second classifier does not identify that the target skin image corresponds to the second skin disease, the diagnosis may be confirmed as having the first skin disease; if the first classifier does not recognize that the target skin image corresponds to the first skin disease and the second classifier recognizes that the target skin image corresponds to the second skin disease, the patient can be determined to have the second skin disease; and if the first classifier does not identify that the target skin image corresponds to the first skin disease and the second classifier does not identify that the target skin image corresponds to the second skin disease, determining that the target skin image has the third skin disease.
Based on the diagnosis principle of the skin disease diagnosis system, it can be known that the first classifier and the second classifier of the present embodiment can perform double verification by combined classification, thereby being capable of diagnosing three skin diseases. That is, the skin disease diagnosis system of the present embodiment uses fewer classifiers, and realizes a diagnosis function for more categories of skin diseases.
Specifically, on the basis of the above, as a preferable scheme, the construction method of the embodiment further includes:
step 131, constructing a third classifier for distinguishing the first skin disease and the second skin disease by using a third neural network, and taking the first training image and the second training image as input data of the third neural network to train the third classifier;
in step 140, a skin disease diagnosis system for diagnosing the first skin disease, the second skin disease and the third skin disease is constructed based on the first classifier, the second classifier and the third classifier.
In the diagnosis application, if the first classifier identifies that the target skin image corresponds to a first skin disease and the second classifier identifies that the target skin image corresponds to a second skin disease, the target skin image is used as input data of a third classifier;
if the third classifier identifies that the target skin image corresponds to the first skin disease, the first skin disease is determined to be present.
And if the third classifier identifies that the target skin image corresponds to the second skin disease, the second skin disease is determined to be suffered.
Obviously, the third classifier of the present embodiment is dedicated to distinguishing the first skin disease from the second skin disease. Therefore, when the first classifier and the second classifier confirm and diagnose the target skin image as the respective corresponding skin diseases at the same time, the target skin image can be further input to the third classifier for more accurate classification. The third classifier can distinguish whether the target skin image corresponds to the first skin disease or the second skin disease, thereby ensuring the accuracy of diagnosis.
In addition, in a specific practical application, the embodiment trains the first classifier, the second classifier and the third classifier by using a large amount of training images.
In order to ensure that the first classifier and the second classifier can be trained normally, as a preferable scheme, after the training image is acquired and before the first classifier, the second classifier and the third classifier are constructed, the embodiment may further perform a preprocessing operation on the training image, where the preprocessing operation includes at least one of the following manners:
clipping the invalid recognition area of the training image;
the display sizes of the different training images are normalized.
Obviously, based on the preprocessing operation, various training images obtained by different acquisition paths can have uniform image specifications, so that the first classifier and the second classifier can maintain the same standard to train the image features of the corresponding skin diseases.
In addition, as a preferable scheme, after the training image is acquired and before the first classifier, the second classifier and the third classifier are constructed, data enhancement processing may be performed on the training image to enhance generalization capability and recognized capability of the training image.
Specifically, the data enhancement processing mode may include at least one of the following:
enhancing data of colors of the training image through a preset image processing algorithm, and optimizing the aspects including saturation, brightness, contrast and the like of the colors;
and turning or rotating the training image through a preset image processing algorithm to optimize the display visual angle of the training image.
It should be noted that the data enhancement processing can be implemented by current image processing software, and is not described herein again because it is prior art.
In addition, in practical applications, the skin disease diagnosis system of the present embodiment can be used to detect melanoma, keratosis, and nevus. That is, the first skin disease, the second skin disease, and the third skin disease are one of melanoma, keratosis, and nevus, respectively.
It can be known that the characteristics of the three diseases, melanoma, keratosis and nevus, on the skin are similar, and are difficult for non-professional medical personnel to distinguish, and the skin disease diagnosis system obtained based on the construction method of the embodiment can accurately diagnose the three skin diseases, and help patients to take correct treatment measures in time.
On the other hand, the embodiment of the invention also provides a skin disease diagnosis system which is constructed by the construction method provided by the invention.
It can be seen that the skin disease diagnosis system of the present embodiment can diagnose skin diseases by itself or assist doctors in diagnosing skin diseases, and can reduce the probability of misdiagnosis, so that the system has a high practical value in the medical field.
In addition, an embodiment of the present invention further provides a method for diagnosing skin diseases, which is applied to the system for diagnosing skin diseases provided by the present invention, as shown in fig. 2, and includes:
In step 230, if the first classifier identifies that the target skin image corresponds to the first skin disease and the second classifier does not identify that the target skin image corresponds to the second skin disease, the patient is determined to have the first skin disease.
And 240, if the first classifier does not recognize that the target skin image corresponds to the first skin disease and the second classifier recognizes that the target skin image corresponds to the second skin disease, determining that the target skin image has the second skin disease.
And step 250, if the first classifier does not identify that the target skin image corresponds to the first skin disease and the second classifier does not identify that the target skin image corresponds to the second skin disease, determining that the patient has the third skin disease.
Based on the diagnosis principle of the skin disease diagnosis system, it can be known that the first classifier and the second classifier of the present embodiment can perform double verification by combined classification, thereby being capable of diagnosing three skin diseases.
Specifically, in order to enhance the accuracy of targeting, the skin disease diagnosis system applied in the present embodiment further includes a third classifier, and the corresponding diagnosis method further includes:
step 260, if the first classifier identifies that the target skin image corresponds to the first skin disease and the second classifier identifies that the target skin image corresponds to the second skin disease, the target skin image is used as input data of a third classifier;
in step 270, if the third classifier identifies that the target skin image corresponds to the first skin disease, the skin image is determined to have the first skin disease.
And step 280, if the third classifier identifies that the target skin image corresponds to the second skin disease, determining that the target skin image has the second skin disease.
Obviously, the third classifier of the present embodiment is dedicated to distinguishing the first skin disease from the second skin disease. Therefore, when the first classifier and the second classifier confirm and diagnose the target skin image as the respective corresponding skin diseases at the same time, the target skin image can be further input to the third classifier for more accurate classification. The third classifier can distinguish whether the target skin image corresponds to the first skin disease or the second skin disease, thereby ensuring the accuracy of diagnosis.
The diagnostic mechanism of the present embodiment will be described in detail with reference to a practical application.
Illustratively, the skin disease diagnosis system applied in the present embodiment is used for diagnosing three skin diseases, i.e., melanoma, keratosis, and nevus.
Namely, the skin disease diagnosis system 300 includes:
a melanoma classifier 301 for training lesion images of the three skin diseases to detect melanoma and other lesions;
a keratosis classifier 302 for training data on lesion images of the three skin diseases to distinguish keratosis from other lesions for detection purposes;
the melanoma and keratosis classifier 303 trains data to be lesion images of melanoma and keratosis, and the detection purpose is to distinguish melanoma from keratosis.
Firstly, a melanoma classifier 301 and a keratosis classifier 302 are combined, a target skin image to be diagnosed is simultaneously input into the melanoma classifier 301 and the keratosis classifier 302 for identification, and different processing is carried out on classification results:
and if the melanoma classifier identifies the melanoma and the keratosis classifier identifies other lesions, outputting the classification result as the melanoma.
If the melanoma classifier identifies other lesions and the keratosis classifier identifies other lesions, the classification result is output as nevi.
And if the melanoma classifier identifies other lesions and the keratosis classifier identifies keratosis, outputting a classification result as the keratosis.
If the melanoma classifier identifies melanoma and the keratosis classifier identifies keratosis, the images are input into the melanoma and keratosis classifier 303 for further identification and classification;
the melanoma and keratosis classifier 303 identifies melanoma, and outputs the classification result as melanoma; the melanoma and keratosis classifier 303 identifies keratosis, and outputs the classification result as keratosis.
Wherein the classification result is a diagnosis result.
In addition, an embodiment of the present invention also provides a skin disease diagnosis apparatus 400, as shown in fig. 4, including:
an image acquisition module 401, configured to acquire a target skin image to be diagnosed;
a data processing module 402 storing a computer program for executing the steps of the diagnostic method according to the present invention;
and a diagnosis result output module 403, configured to output a diagnosis result of the data processing module.
In practical applications, the diagnostic apparatus 400 of the present embodiment may be a dedicated medical device, or may be a user terminal device.
Taking a user terminal device as an example, the image acquisition module 401 may be a camera of the user terminal device, the data processing module 402 may be a processor of the user terminal device, the processor is implemented based on preset value calculation and a program, or the processing area is implemented based on a cloud processor or a server that establishes a connection with the user terminal device, and the diagnosis result output module 403 is a display of the user terminal device.
Obviously, based on the above practical applications, the user can diagnose the skin diseases through the personal terminal device (such as mobile phone, PAD), thereby lowering the diagnosis threshold. Therefore, the diagnostic device of the present embodiment has high practicability.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.
Unless defined otherwise, technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention belongs. The use of "first," "second," and similar terms in this disclosure is not intended to indicate any order, quantity, or importance, but rather is used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that the element or item listed before the word covers the element or item listed after the word and its equivalents, but does not exclude other elements or items.
Claims (9)
1. A method of constructing a skin disease diagnostic system, comprising:
acquiring a training image, the training image comprising: a first training image corresponding to a first skin condition, a second training image corresponding to a second skin condition, and a third training image corresponding to a third skin condition;
constructing a first classifier for identifying a first skin disease by using a first neural network, and taking the first training image, the second training image and the third training image as input data of the first neural network to train the first classifier;
constructing a second classifier for identifying a second skin disease by using a second neural network, and taking the first training image, the second training image and the third training image as input data of the second neural network to train the second classifier;
constructing a skin disease diagnosis system for diagnosing the first skin disease, the second skin disease and the third skin disease based on at least the first classifier and the second classifier;
further comprising:
constructing a third classifier for distinguishing a first skin disease from a second skin disease using a third neural network, and using the first training image and the second training image as input data of the third neural network to train the third classifier;
wherein the step of constructing the skin disease diagnosis system specifically comprises:
constructing a skin disease diagnosis system for diagnosing the first skin disease, the second skin disease, and the third skin disease based on the first classifier, the second classifier, and the third classifier.
2. The construction method according to claim 1,
the first skin disease, the second skin disease and the third skin disease are respectively one of the following skin diseases:
melanoma, keratosis, and nevi.
3. The construction method according to claim 1, after acquiring the training image and before constructing the first classifier, the second classifier, and the third classifier, further comprising:
carrying out preprocessing operation on the training image;
the preprocessing operation comprises at least one of the following modes:
clipping the invalid recognition area of the training image;
the display sizes of the different training images are normalized.
4. The construction method according to claim 1, after acquiring the training image and before constructing the first classifier, the second classifier, and the third classifier, further comprising:
performing data enhancement operation on the training image;
the data enhancement operation comprises at least one of the following modes:
optimizing saturation and/or brightness and/or contrast of the training image;
flipping and/or rotating the training image.
5. A skin disease diagnosis system constructed by the construction method according to any one of claims 1 to 4.
6. A skin image classification method applied to the skin disease diagnosis system according to claim 5, comprising:
acquiring a target skin image to be diagnosed;
using the target skin image as input data for a first classifier and a second classifier in the skin disease diagnosis system;
if the first classifier identifies that the target skin image corresponds to a first classification result and the second classifier does not identify that the target skin image corresponds to a second classification result, determining that the target skin image belongs to the first classification result;
if the first classifier does not recognize that the target skin image corresponds to a first classification result and the second classifier recognizes that the target skin image corresponds to a second classification result, determining that the target skin image belongs to the second classification result;
and if the first classifier does not recognize that the target skin image corresponds to the first classification result and the second classifier does not recognize that the target skin image corresponds to the second classification result, determining that the target skin image belongs to the third classification result.
7. The skin image classification method according to claim 6, characterized in that the skin disease diagnosis system further includes a third classifier, the classification method further comprising:
if the first classifier identifies that the target skin image corresponds to a first classification result and the second classifier identifies that the target skin image corresponds to a second classification result, taking the target skin image as input data of the third classifier;
if the third classifier identifies that the target skin image corresponds to a first classification result, determining that the target skin image belongs to the first classification result;
and if the third classifier identifies that the target skin image corresponds to a second classification result, determining that the target skin image belongs to the second classification result.
8. A diagnostic device for skin conditions, comprising:
the image acquisition module is used for acquiring a target skin image to be diagnosed;
data processing means holding a computer program for performing the steps of the classification method according to claim 6 or 7.
9. The diagnostic apparatus for skin diseases according to claim 8, further comprising:
and the diagnosis result output module is used for outputting the diagnosis result of the data processing module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710735310.1A CN107563997B (en) | 2017-08-24 | 2017-08-24 | Skin disease diagnosis system, construction method, classification method and diagnosis device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710735310.1A CN107563997B (en) | 2017-08-24 | 2017-08-24 | Skin disease diagnosis system, construction method, classification method and diagnosis device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107563997A CN107563997A (en) | 2018-01-09 |
| CN107563997B true CN107563997B (en) | 2020-06-02 |
Family
ID=60976826
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710735310.1A Active CN107563997B (en) | 2017-08-24 | 2017-08-24 | Skin disease diagnosis system, construction method, classification method and diagnosis device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107563997B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108154509B (en) * | 2018-01-12 | 2022-11-11 | 平安科技（深圳）有限公司 | Cancer identification method, device and storage medium |
| CN108198620B (en) * | 2018-01-12 | 2022-03-22 | 洛阳飞来石软件开发有限公司 | Skin disease intelligent auxiliary diagnosis system based on deep learning |
| CN109949272B (en) * | 2019-02-18 | 2023-06-20 | 四川拾智联兴科技有限公司 | Data acquisition method and system for identifying skin disease types and acquiring human skin pictures |
| JP2022536808A (en) * | 2019-06-18 | 2022-08-18 | デジタル ダイアグノスティックス インコーポレイテッド | Using a Set of Machine Learning Diagnostic Models to Determine a Diagnosis Based on a Patient's Skin Tone |
| CN110648751A (en) * | 2019-10-30 | 2020-01-03 | 中南大学湘雅三医院 | System and method for delineating possible diseases by utilizing skin CT |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101517602A (en) * | 2006-09-22 | 2009-08-26 | 皇家飞利浦电子股份有限公司 | Methods for feature selection using classifier ensemble based genetic algorithms |
| CN101517614A (en) * | 2006-09-22 | 2009-08-26 | 皇家飞利浦电子股份有限公司 | Advanced computer-aided diagnosis of lung nodules |
| CN102221363A (en) * | 2011-04-12 | 2011-10-19 | 东南大学 | Fault-tolerant combined method of strapdown inertial integrated navigation system for underwater vehicles |
| CN102509122A (en) * | 2011-11-25 | 2012-06-20 | 广东威创视讯科技股份有限公司 | Intelligent pen color identifying method applied to interactive touch screen |
| CN103246888A (en) * | 2012-02-03 | 2013-08-14 | 通用电气公司 | System and method for diagnosing lung disease by computer |
| CN103914064A (en) * | 2014-04-01 | 2014-07-09 | 浙江大学 | Industrial process fault diagnosis method based on multiple classifiers and D-S evidence fusion |
| WO2015054666A1 (en) * | 2013-10-10 | 2015-04-16 | Board Of Regents, The University Of Texas System | Systems and methods for quantitative analysis of histopathology images using multi-classifier ensemble schemes |
| CN106971160A (en) * | 2017-03-23 | 2017-07-21 | 西京学院 | Winter jujube disease recognition method based on depth convolutional neural networks and disease geo-radar image |
| CN107368670A (en) * | 2017-06-07 | 2017-11-21 | 万香波 | Stomach cancer pathology diagnostic support system and method based on big data deep learning |
-
2017
- 2017-08-24 CN CN201710735310.1A patent/CN107563997B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101517602A (en) * | 2006-09-22 | 2009-08-26 | 皇家飞利浦电子股份有限公司 | Methods for feature selection using classifier ensemble based genetic algorithms |
| CN101517614A (en) * | 2006-09-22 | 2009-08-26 | 皇家飞利浦电子股份有限公司 | Advanced computer-aided diagnosis of lung nodules |
| CN102221363A (en) * | 2011-04-12 | 2011-10-19 | 东南大学 | Fault-tolerant combined method of strapdown inertial integrated navigation system for underwater vehicles |
| CN102509122A (en) * | 2011-11-25 | 2012-06-20 | 广东威创视讯科技股份有限公司 | Intelligent pen color identifying method applied to interactive touch screen |
| CN103246888A (en) * | 2012-02-03 | 2013-08-14 | 通用电气公司 | System and method for diagnosing lung disease by computer |
| WO2015054666A1 (en) * | 2013-10-10 | 2015-04-16 | Board Of Regents, The University Of Texas System | Systems and methods for quantitative analysis of histopathology images using multi-classifier ensemble schemes |
| CN103914064A (en) * | 2014-04-01 | 2014-07-09 | 浙江大学 | Industrial process fault diagnosis method based on multiple classifiers and D-S evidence fusion |
| CN106971160A (en) * | 2017-03-23 | 2017-07-21 | 西京学院 | Winter jujube disease recognition method based on depth convolutional neural networks and disease geo-radar image |
| CN107368670A (en) * | 2017-06-07 | 2017-11-21 | 万香波 | Stomach cancer pathology diagnostic support system and method based on big data deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107563997A (en) | 2018-01-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107563997B (en) | Skin disease diagnosis system, construction method, classification method and diagnosis device | |
| TWI709852B (en) | System and method for anomaly detection via a multi-prediction-model architecture | |
| WO2021143353A1 (en) | Gesture information processing method and apparatus, electronic device, and storage medium | |
| JP6947759B2 (en) | Systems and methods for automatically detecting, locating, and semantic segmenting anatomical objects | |
| WO2019052062A1 (en) | Medical image report generation system and method based on artificial intelligence | |
| CN104274191B (en) | A kind of Psychological Evaluation method and system thereof | |
| US10916348B2 (en) | Machine learning to identify locations of brain injury | |
| KR102162683B1 (en) | Reading aid using atypical skin disease image data | |
| Yang et al. | More to less (M2L): Enhanced health recognition in the wild with reduced modality of wearable sensors | |
| CN116942149B (en) | Lumbar vertebra monitoring method, device, equipment and storage medium based on millimeter wave radar | |
| Pan et al. | BP neural network classification for bleeding detection in wireless capsule endoscopy | |
| CN111150369A (en) | Medical assistance apparatus, medical assistance detection apparatus and method | |
| Zhang et al. | Quantification of advanced dementia patients’ engagement in therapeutic sessions: An automatic video based approach using computer vision and machine learning | |
| CN115035086A (en) | Intelligent tuberculosis skin test screening and analyzing method and device based on deep learning | |
| CN115317304A (en) | Intelligent massage method and system based on physiological characteristic detection | |
| Kalbande et al. | An artificial intelligence approach for the recognition of early stages of eczema | |
| Karnowski et al. | Retina image analysis and ocular telehealth: the Oak Ridge National Laboratory-Hamilton Eye Institute case study | |
| Chin et al. | CWD2GAN: Generative adversarial network of chronic wound depth detection for predicting chronic wound depth | |
| CN112487980A (en) | Micro-expression-based treatment method, device, system and computer-readable storage medium | |
| CN111340778A (en) | Glaucoma image processing method and equipment | |
| Patel et al. | IoT based Smart Health Monitoring System with Patient Identification using Face Recognition | |
| Grant et al. | TCM Tongue Segmentation and Analysis with a Mobile Device | |
| CN117253096B (en) | Finger-adaptive health index monitoring method, device, equipment and storage medium | |
| CN111222374A (en) | Lie detection data processing method and device, computer equipment and storage medium | |
| CN113116299B (en) | Pain degree evaluation method, pain degree evaluation device, apparatus, and storage medium |
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 |