CN114821176A - Children brain MR image viral encephalitis classification system - Google Patents
Children brain MR image viral encephalitis classification system Download PDFInfo
- Publication number
- CN114821176A CN114821176A CN202210463034.9A CN202210463034A CN114821176A CN 114821176 A CN114821176 A CN 114821176A CN 202210463034 A CN202210463034 A CN 202210463034A CN 114821176 A CN114821176 A CN 114821176A
- Authority
- CN
- China
- Prior art keywords
- image
- convolution
- viral encephalitis
- model
- sub
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
A children brain MR image viral encephalitis classification system comprises a computer memory, a computer processor and a computer program which is stored in the computer memory and can be executed on the computer processor, wherein a trained classification model is stored in the computer memory, the classification model adopts an improved SE ResNet network model and totally comprises four convolution parts, each convolution part consists of a plurality of sub-module groups, each sub-module group comprises an Incepration sub-module and an SE Res sub-module, and finally, the final classification result is obtained through a full connection layer; the computer processor, when executing the computer program, performs the steps of: and inputting the MR images of the brain of the child to be classified into the trained classification model to obtain a viral encephalitis classification result. By using the method and the device, the learning ability of the model to different dimensional characteristics can be improved, and the efficiency and the accuracy of the viral encephalitis diagnosis of children are greatly improved.
Description
Technical Field
The invention belongs to the field of medical artificial intelligence, and particularly relates to a children brain MR image viral encephalitis classification system.
Background
Childhood encephalitis is a relatively common disease in pediatrics. Generally, comprehensive judgment can be made by clinical symptoms, laboratory examinations, and imaging and electroencephalogram tests. If the diagnosis is confirmed, the patient needs to be treated specifically under the guidance of a professional doctor.
At present, doctor diagnosis is mainly carried out by methods such as clinical symptoms, laboratory examination (cerebrospinal fluid examination), imaging, electroencephalogram detection and the like, but the clinical symptoms are not accurate; imaging and electroencephalogram detection can only visually observe a diseased region when severe; the cerebrospinal fluid examination is accurate, but takes a long time, and the cerebrospinal fluid needs to be extracted, which causes trauma and pain to children.
With the development of artificial intelligence and deep learning, in the medical field, many researchers have attempted to automatically identify electroencephalographic data using intelligent algorithms.
For example, chinese patent publication No. CN112561863A discloses a granular classification recognition system based on deep learning for granulocytic images; the positioning module is used for extracting features of an input granulocyte picture by utilizing a Hourglass network model, respectively positioning all cells in the granulocyte picture, cutting the positioned cells out, leaving single complete cells, and carrying out size normalization processing on all the cut cells; the classification module classifies the granulocytes positioned by the positioning module by adopting the constructed deep learning classification model; the system can assist clinicians in accurately and efficiently completing granulocyte classification, identification and counting tasks, reduce errors caused by subjectivity, reduce the workload of doctors, and assist the doctors in making disease judgment; the system can effectively solve the cell classification under the unbalanced data and the fine-grained classification among granular cells, and improves the network classification and identification effects.
Chinese patent publication No. CN112132808A discloses a method and apparatus for detecting breast X-ray image lesions based on normal model learning. The method comprises segmenting a breast region from a mammographic image; extracting image blocks and carrying out brightness normalization processing; selecting a part of normal area image blocks as a training set, inputting the training set into a dual-depth convolution neural network model for training to obtain a normal model; selecting a plurality of normal area image blocks from the training set as templates, and inputting the templates into a normal model to obtain feature vectors of the template images; inputting the test set into a normal model to obtain a feature vector of a test image; and inputting the feature vectors of the template image and the test image into a nearest neighbor classifier to perform classification to obtain a test result.
However, the characteristics of the imaging data are not obvious for the viral encephalitis of children, and the accurate diagnosis of whether children have the viral encephalitis is difficult by using the conventional deep learning method.
Disclosure of Invention
The invention provides a children brain MR image viral encephalitis classification system, which can be used for diagnosing viral encephalitis only based on MR images in addition to lumbar puncture cerebrospinal fluid examination and clinical examination, and has higher accuracy.
A children brain MR image viral encephalitis classification system comprises a computer memory, a computer processor and a computer program which is stored in the computer memory and can be executed on the computer processor, wherein a trained classification model is stored in the computer memory, the classification model adopts an improved SE ResNet network model and totally comprises four convolution parts, each convolution part consists of a plurality of sub-module groups, each sub-module group comprises an Incepration sub-module and an SE Res sub-module, and finally, the final classification result is obtained through a full connection layer;
the Incep sub-module improves learning capacity of features with different sizes through convolution with different scales; the SE Res submodule comprises an SE part and an Res part, the SE part improves the learning capacity of the model to effective characteristics by compressing and expanding the number of channels, and the Res part connects an input characteristic matrix X and an output characteristic matrix of the SE part through jumping
Splicing is carried out, and the learning ability of the model to different dimensional characteristics is improved;
the computer processor, when executing the computer program, performs the steps of:
and inputting the MR images of the brain of the child to be classified into the trained classification model to obtain a viral encephalitis classification result.
Further, the Incep sub-module has the following structure: after obtaining the data input X of the previous layer, the data enters the multi-core convolution layer L incep This layer designs three convolution kernels of different sizes and one pooling kernel, i.e. C incep ＝[C 1 ,C 2 ,C 3 ,P 1 ](ii) a Wherein, C 1 ,C 2 ,C 3 The convolution kernel sizes of (1 x 1), (3 x 3), (5 x 5), respectively, P 1 The nucleus size of (a) is 3 x 3; four different features are obtained by these convolution kernels
F＝[F 1 ,F 2 ,F 3 ,F 4 ]Then all the characteristics are spliced to finally obtain the multi-core convolution layer L incep Output of (F) is Concat (F) 1 ,F 2 ,F 3 ,F 4 )。
In the SE Res submodule, the SE section has the following structure: acquiring input data X (c w h) of the previous layer, wherein c, w and h respectively represent the channel number, width and height of the feature matrix; firstly, performing pooling by using a global pooling layer with the size of 1 x 1 to obtain a pooled feature matrix F 1 Gobelpool (x) with size c w h; then using 1/16 c channel number to make full-connection convolution to obtain F 2 ＝FC(F 1 ) (ii) its size is c/16 w h; then convolution is carried out by the convolution layer with the number of c channels to obtain F 3 ＝FC(F 2 ) The size is c 1; then, normalization operation is carried out by using sigmoid activation function, the weight is normalized to be between 0 and 1, and F is obtained 4 ＝Sigmoid(F 3 ) The size is c 1; finally, the result is used to weight the input data to obtain
Its size is c w h.
The structure of the Res part is as follows: the feature matrix X (c w h) is obtained after passing through the SE part
The number of the two channels is c, and the two feature matrixes are spliced through jump connection, so that a new feature matrix is obtained
Its size is 2c w h.
The training process of the classification model is as follows:
(1) collecting MR image data of T1W sequences of patients with viral encephalitis and normal children, and preprocessing the image data;
(2) dividing the preprocessed image data into a training set, a verification set and a test set;
(3) and sending the training set into the constructed classification model for training, evaluating the performance of the classification model by using the verification set, adjusting the hyper-parameters of the model according to the evaluation effect, and finally obtaining the classification model with the performance reaching the standard through repeated training and verification.
In the step (1), the preprocessing comprises zooming the image, selecting the maximum slice number as a standard, and supplementing the data which does not reach the slice number by copying head and tail slices to keep the input data of each case consistent; meanwhile, the image is normalized in scale, and noise is filtered by adopting a Gaussian filter.
In the step (2), the preprocessed image data is divided into a training set, a verification set and a test set according to the ratio of 7:1: 2.
And (3) training the classification model by adopting a supervision training method.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention innovatively provides the judgment of whether the child is infected with the viral encephalitis or not by utilizing the child brain MR image, and does not need to check the cerebrospinal fluid through the waist, so that the operation is not needed, the pain of the child patient is reduced, and the diagnosis efficiency is greatly improved.
2. The classification model adopts an improved SE ResNet network model, an SE Res module is added on the basis of the inclusion network model, the inclusion network model improves the learning capacity of the model to the features with different sizes through convolution with different scales, the SE Res module firstly improves the learning capacity of the model to effective features through compressing and expanding the number of channels, and then improves the learning capacity of the model to the features with different dimensions through jump connection; the efficiency and the accuracy of diagnosis are greatly improved.
Drawings
FIG. 1 is a flowchart of an embodiment of a system for classifying children brain MR images viral encephalitis according to the present invention;
FIG. 2 is an overall block diagram of a classification model according to the present invention;
FIG. 3 is a block diagram of each convolution portion of the classification model;
FIG. 4 is a schematic diagram of a network structure of an inclusion sub-module in the classification model of the present invention;
FIG. 5 is a network structure diagram of the SE Res sub-module in the classification model of the present invention.
Detailed Description
The invention will be described in further detail below with reference to the drawings and examples, which are intended to facilitate the understanding of the invention without limiting it in any way.
A children brain MR image viral encephalitis classification system comprises a computer memory, a computer processor and a computer program which is stored in the computer memory and can be executed on the computer processor, wherein a trained classification model is stored in the computer memory.
As shown in fig. 1, the whole system is implemented as follows:
1. image pre-processing
Collecting MR image data of T1W stage of patients with viral encephalitis and normal children, zooming the images, wherein the maximum slice number is selected as a standard because the number of MR slices scanned by each case is inconsistent, the data which do not reach the slice number are supplemented by copying head and tail slices, so that the input data of each case are kept consistent, and the method also comprises the steps of carrying out scale normalization on the images and filtering noise by adopting a Gaussian filter.
2. Data packet
70% of the data set was used as the training set, 10% of the data set was used as the validation set, and 20% of the data set was used as the test set.
3. Model construction
And constructing a classification model, wherein the classification model adopts an improved SE ResNet network model, and the SE Res module is added to the network on the basis of the inclusion network model. The Incepration network model improves the learning capability of the model on the features with different sizes through convolution with different scales, the SE Res module firstly improves the learning capability of the model on effective features through compressing and expanding the number of channels, and then improves the learning capability of the model on the features with different dimensions through jumping connection.
As shown in fig. 2 and fig. 3, the model includes four volume blocks, volume block 1-convolution block 4, each volume block is composed of several sub-module groups, and each sub-module group includes an inclusion sub-module and an SE Res sub-module.
The structure of the inclusion sub-modules is shown in fig. 4, and features of different dimensions are obtained through the inclusion sub-modules. After obtaining the data input X of the previous layer, the data enters the multi-core convolution layer L incep This layer designs three convolution kernels of different sizes and one pooling kernel, i.e. C incep ＝[C 1 ,C 2 ,C 3 ,P 1 ]Wherein, C 1 ,C 2 ,C 3 The convolution kernel sizes of (1 x 1), (3 x 3), (5 x 5), respectively, P 1 The size of the nuclei of (a) is 3 x 3. Four different features can be obtained by these convolution kernels
F＝[F 1 ,F 2 ,F 3 ,F 4 ]Then all the characteristics are spliced to finally obtain the multi-core convolution layer L incep Output of (F) is Concat (F) 1 ,F 2 ,F 3 ,F 4 )。
Because the number of features finally acquired by the Incep submodule is large, in order to improve the calculation speed and the model precision, the SE Res submodule is adopted for carrying out weighted screening on the features in the method.
The structure of the SE Res submodule is shown in fig. 5, and the SE Res submodule includes two parts, SE and Res. In the SE section, the input data X (c × w × h) of the previous layer is obtained, where c, w, and h represent the number of channels, width, and height of the feature matrix, respectively. Firstly, performing pooling by using a global pooling layer with the size of 1 x 1 to obtain a pooled feature matrix F 1 Gobelpool (x) of size c w h, followed by full concatenated convolution with a channel number of 1/16 c, yielding F 2 ＝FC(F 1 ) The size is c/16 w h, and then convolution is performed by convolution layers with the number of c channels to obtain F 3 ＝FC(F 2 ) The size is c x 1, then normalization operation is carried out by using sigmoid activation function, the weight is normalized to be between 0 and 1,obtaining F 4 ＝Sigmoid(F 3 ) C x 1, and finally weighting the input data by using the result to obtain
Its size is c w h. And weighting each channel through the SE part, and further screening out the channels with correlation to the result. In the Res part, the feature matrix X (c w h) can be obtained after passing through an SE module
The number of the two channels is c, and the two feature matrixes are spliced to obtain a new feature matrix
Its size is 2c w h.
And finally, obtaining the probability of whether the input case is viral encephalitis or not through the full-connection convolution layer.
4. Model training and classification testing
When the segmentation model is trained, a training set is sent into a classification model; the verification set adjusts the hyper-parameters of the model, an optimizer is used for updating the parameters, the network is optimized, the learning rate is automatically adjusted, and a trained classification network is obtained; the test set is used to estimate the generalization ability of the model after the learning process is complete.
5. Evaluation phase
On the test set, the classification effect of the model is evaluated: for the evaluation of classification tasks, the Precision (Precision) and Recall (Recall) of each class needs to be calculated. The accuracy of each class is divided by the number of cases correctly classified into the class (TP + FP), and when cases not belonging to the class are classified into the class by the model, False positives are counted (FP). The recall rate of each class is correctly classified into cases of this class (TP) divided by the number of real cases of this class (TP + TN), and when cases belonging to this class are classified into other classes by the model, false Negative is counted (TN). And finally, evaluating the classification performance AUC of the intelligent encephalitis diagnosis classification model on the test set, wherein an AU curve is an area below an ROC (receiver Operating characterization) curve with a false positive rate (FP _ rate) and a false negative rate (TP _ rate) as axes.
The technical solutions and advantages of the present invention have been described in detail with reference to the above embodiments, it should be understood that the above embodiments are only specific examples of the present invention and should not be construed as limiting the present invention, and any modifications, additions and equivalents made within the scope of the principles of the present invention should be included in the scope of the present invention.
Claims (8)
1. A system for classifying a viral encephalitis in an MR image of a child's brain comprising a computer memory, a computer processor and a computer program stored in said computer memory and executable on said computer processor, characterized in that:
the computer memory is stored with a trained classification model, the classification model adopts an improved SE ResNet network model and comprises four convolution parts, each convolution part is composed of a plurality of sub-module groups, each sub-module group comprises an inclusion sub-module and an SE Res sub-module, and finally, the final classification result is obtained through a full connection layer;
the Incep sub-module improves learning capacity of features with different sizes through convolution with different scales; the SE Res submodule comprises an SE part and an Res part, the SE part improves the learning capacity of the model to effective characteristics by compressing and expanding the number of channels, and the Res part connects an input characteristic matrix X and an output characteristic matrix of the SE part through jumping
Splicing is carried out, and the learning ability of the model to different dimensional characteristics is improved;
the computer processor, when executing the computer program, performs the steps of:
and inputting the MR images of the brain of the child to be classified into the trained classification model to obtain a viral encephalitis classification result.
2. The system for classifying children brain MR image viral encephalitis according to claim 1, wherein the Incep sub-module has the following structure:
after obtaining the data input X of the previous layer, the data enters the multi-core convolution layer L incep This layer designs three convolution kernels of different sizes and one pooling kernel, i.e. C incep ＝[C 1 ,C 2 ,C 3 ,P 1 ](ii) a Wherein, C 1 ,C 2 ,C 3 The convolution kernel sizes of (1 x 1), (3 x 3), (5 x 5), respectively, P 1 The nucleus size of (a) is 3 x 3; four different features are obtained by these convolution kernels
Then all the characteristics are spliced to finally obtain the multi-core convolution layer L incep Output of (F) is equal to Concat (F) 1 ,F 2 ,F 3 ,F 4 )。
3. The system for classifying children brain MR image viral encephalitis according to claim 1, wherein in the SE Res submodule, the structure of the SE part is as follows:
acquiring input data X (c w h) of the previous layer, wherein c, w and h respectively represent the channel number, width and height of the feature matrix; firstly, performing pooling by using a global pooling layer with the size of 1 x 1 to obtain a pooled feature matrix F 1 Gobelpool (x) with size c w h; then using 1/16 c channel number to make full-connection convolution to obtain F 2 ＝FC(F 1 ) The size is c/16 w h; then convolution is carried out by the convolution layer with the number of c channels to obtain F 3 ＝FC(F 2 ) The size is c 1; then, normalization operation is carried out by using sigmoid activation function, the weight is normalized to be between 0 and 1, and F is obtained 4 ＝Sigmoid(F 3 ) The size is c 1; finally, the result is used to weight the input data to obtain
Its size is c w h.
4. The system for classifying children brain MR image viral encephalitis according to claim 1, wherein in the SE Res submodule, the structure of Res part is as follows:
the feature matrix X (c w h) is obtained after passing through the SE part
The number of the two channels is c, and the two feature matrixes are spliced through jump connection, so that a new feature matrix is obtained
Its size is 2c w h.
5. The system for classifying children brain MR image viral encephalitis according to claim 1, wherein the training process of the classification model is as follows:
(1) collecting MR image data of T1W sequences of patients with viral encephalitis and normal children, and preprocessing the image data;
(2) dividing the preprocessed image data into a training set, a verification set and a test set;
(3) and sending the training set into the constructed classification model for training, evaluating the performance of the classification model by using the verification set, adjusting the hyper-parameters of the model according to the evaluation effect, and finally obtaining the classification model with the performance reaching the standard through repeated training and verification.
6. The system for classifying children brain MR image viral encephalitis according to claim 5, wherein in step (1), the preprocessing includes scaling the image, selecting the maximum slice number as the standard, and supplementing the data which does not reach the slice number by copying the head and tail slices to keep the input data of each case consistent; meanwhile, the image is normalized in scale, and noise is filtered by adopting a Gaussian filter.
7. The system for classifying children brain MR image viral encephalitis according to claim 5, wherein in step (2), the preprocessed image data are divided into training set, verification set and test set according to 7:1: 2.
8. The system for classifying children's brain MR image viral encephalitis according to claim 5, wherein in step (3), the classification model is trained by adopting a supervised training method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210463034.9A CN114821176B (en) | 2022-04-28 | 2022-04-28 | Viral encephalitis classification system for MR (magnetic resonance) images of children brain |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210463034.9A CN114821176B (en) | 2022-04-28 | 2022-04-28 | Viral encephalitis classification system for MR (magnetic resonance) images of children brain |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114821176A true CN114821176A (en) | 2022-07-29 |
| CN114821176B CN114821176B (en) | 2022-11-01 |
Family
ID=82509353
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210463034.9A Active CN114821176B (en) | 2022-04-28 | 2022-04-28 | Viral encephalitis classification system for MR (magnetic resonance) images of children brain |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114821176B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116363438A (en) * | 2023-04-14 | 2023-06-30 | 浙江大学 | Child encephalitis classification system based on multiple modes |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0918458B1 (en) * | 1996-05-08 | 2006-12-27 | Meryl Squires | Antimicrobial treatment for herpes simplex virus and other infectious diseases |
| CN108171232A (en) * | 2017-11-15 | 2018-06-15 | 中山大学 | The sorting technique of bacillary and viral children Streptococcus based on deep learning algorithm |
| CN108427920A (en) * | 2018-02-26 | 2018-08-21 | 杭州电子科技大学 | A kind of land and sea border defense object detection method based on deep learning |
| CN109071610A (en) * | 2016-05-03 | 2018-12-21 | 无微华斯生物科技有限公司 | Method and system for diagnosing and treating virus infection |
| US20200000362A1 (en) * | 2018-06-29 | 2020-01-02 | Mayo Foundation For Medical Education And Research | Systems, methods, and media for automatically diagnosing intraductal papillary mucinous neosplasms using multi-modal magnetic resonance imaging data |
| CN110739070A (en) * | 2019-09-26 | 2020-01-31 | 南京工业大学 | brain disease diagnosis method based on 3D convolutional neural network |
| CN111104961A (en) * | 2019-10-31 | 2020-05-05 | 太原理工大学 | Method for classifying breast cancer based on improved MobileNet network |
| CN111223553A (en) * | 2020-01-03 | 2020-06-02 | 大连理工大学 | Two-stage deep migration learning traditional Chinese medicine tongue diagnosis model |
| CN111709446A (en) * | 2020-05-14 | 2020-09-25 | 天津大学 | X-ray chest radiography classification device based on improved dense connection network |
| CN111868080A (en) * | 2018-02-27 | 2020-10-30 | 磨石肿瘤生物技术公司 | Identification of neoantigens using pan-allelic models |
| CN112116005A (en) * | 2020-09-18 | 2020-12-22 | 推想医疗科技股份有限公司 | Training method and device of image classification model, storage medium and electronic equipment |
| CN112906748A (en) * | 2021-01-25 | 2021-06-04 | 浙江大学 | 12-lead ECG arrhythmia detection classification model construction method based on residual error network |
| WO2021216467A1 (en) * | 2020-04-20 | 2021-10-28 | The Regents Of The University Of California | A nano-enabled vaccination approach for coronavirus disease (covid-19) and other viral diseases |
-
2022
- 2022-04-28 CN CN202210463034.9A patent/CN114821176B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0918458B1 (en) * | 1996-05-08 | 2006-12-27 | Meryl Squires | Antimicrobial treatment for herpes simplex virus and other infectious diseases |
| CN109071610A (en) * | 2016-05-03 | 2018-12-21 | 无微华斯生物科技有限公司 | Method and system for diagnosing and treating virus infection |
| CN108171232A (en) * | 2017-11-15 | 2018-06-15 | 中山大学 | The sorting technique of bacillary and viral children Streptococcus based on deep learning algorithm |
| CN108427920A (en) * | 2018-02-26 | 2018-08-21 | 杭州电子科技大学 | A kind of land and sea border defense object detection method based on deep learning |
| CN111868080A (en) * | 2018-02-27 | 2020-10-30 | 磨石肿瘤生物技术公司 | Identification of neoantigens using pan-allelic models |
| US20200000362A1 (en) * | 2018-06-29 | 2020-01-02 | Mayo Foundation For Medical Education And Research | Systems, methods, and media for automatically diagnosing intraductal papillary mucinous neosplasms using multi-modal magnetic resonance imaging data |
| CN110739070A (en) * | 2019-09-26 | 2020-01-31 | 南京工业大学 | brain disease diagnosis method based on 3D convolutional neural network |
| CN111104961A (en) * | 2019-10-31 | 2020-05-05 | 太原理工大学 | Method for classifying breast cancer based on improved MobileNet network |
| CN111223553A (en) * | 2020-01-03 | 2020-06-02 | 大连理工大学 | Two-stage deep migration learning traditional Chinese medicine tongue diagnosis model |
| WO2021216467A1 (en) * | 2020-04-20 | 2021-10-28 | The Regents Of The University Of California | A nano-enabled vaccination approach for coronavirus disease (covid-19) and other viral diseases |
| CN111709446A (en) * | 2020-05-14 | 2020-09-25 | 天津大学 | X-ray chest radiography classification device based on improved dense connection network |
| CN112116005A (en) * | 2020-09-18 | 2020-12-22 | 推想医疗科技股份有限公司 | Training method and device of image classification model, storage medium and electronic equipment |
| CN112906748A (en) * | 2021-01-25 | 2021-06-04 | 浙江大学 | 12-lead ECG arrhythmia detection classification model construction method based on residual error network |
Non-Patent Citations (4)
| Title |
|---|
| WENLI WU等: "《Differentiation of Glioma Mimicking Encephalitis and Encephalitis Using Multiparametric MR-Based Deep Learning》", 《FRONTIERS IN ONCOLOGY》 * |
| 吴蒙蒙 等: "《CT和MRI对小儿病毒性脑炎的诊断价值（80例）》", 《医疗装备》 * |
| 李哲明 等: "《基于人工智能技术的儿童慢病管理平台的研制与应用》", 《中国医疗设备》 * |
| 王佳 等: "《基于改进Inception-ResNet-v2的城市交通路面状态识别算法》", 《科学技术与工程》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116363438A (en) * | 2023-04-14 | 2023-06-30 | 浙江大学 | Child encephalitis classification system based on multiple modes |
| CN116363438B (en) * | 2023-04-14 | 2023-09-15 | 浙江大学 | Child encephalitis classification system based on multiple modes |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114821176B (en) | 2022-11-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Omonigho et al. | Breast cancer: tumor detection in mammogram images using modified alexnet deep convolution neural network | |
| CN111009324B (en) | Auxiliary diagnosis system and method for mild cognitive impairment through multi-feature analysis of brain network | |
| CN106780475A (en) | A kind of image processing method and device based on histopathologic slide's image organizational region | |
| CN108511055A (en) | Ventricular premature beat identifying system and method based on Multiple Classifier Fusion and diagnostic rule | |
| CN112784856A (en) | Channel attention feature extraction method and identification method of chest X-ray image | |
| Rachmatullah et al. | Convolutional neural network for semantic segmentation of fetal echocardiography based on four-chamber view | |
| CN114926477A (en) | Brain tumor multi-modal MRI (magnetic resonance imaging) image segmentation method based on deep learning | |
| Vij et al. | A novel deep transfer learning based computerized diagnostic Systems for Multi-class imbalanced diabetic retinopathy severity classification | |
| CN110570419A (en) | Method and device for acquiring characteristic information and storage medium | |
| CN114821176B (en) | Viral encephalitis classification system for MR (magnetic resonance) images of children brain | |
| Shrivastava et al. | Control or autism-classification using convolutional neural networks on functional MRI | |
| Sanjana et al. | Automated diabetic retinopathy detection using transfer learning models | |
| Archana et al. | Deep convolutional neural networks for multiclass cervical cell classification | |
| Chak et al. | Neural network and svm based kidney stone based medical image classification | |
| WO2023198224A1 (en) | Method for constructing magnetic resonance image preliminary screening model for mental disorders | |
| Nizarudeen et al. | Multi-Layer ResNet-DenseNet architecture in consort with the XgBoost classifier for intracranial hemorrhage (ICH) subtype detection and classification | |
| Kanse et al. | HG-SVNN: harmonic genetic-based support vector neural network classifier for the glaucoma detection | |
| CN115205599B (en) | Multi-age-segment child chest image pneumonia classification system based on domain generalization model | |
| CN116433973A (en) | Child pneumonia classification system and method based on layered multi-scale feature fusion | |
| Dafwen Toresa et al. | Automated Detection and Counting of Hard Exudates for Diabetic Retinopathy by using Watershed and Double Top-Bottom Hat Filtering Algorithm | |
| CN114305387A (en) | Magnetic resonance imaging-based method, equipment and medium for classifying small cerebral vascular lesion images | |
| CN116363438B (en) | Child encephalitis classification system based on multiple modes | |
| Hoteit et al. | Breast Abnormalities' Classification Using Convolutional Neural Network | |
| CN113139627B (en) | Mediastinal lump identification method, system and device | |
| Srirangam | Artificial Intelligence Based Hyper-parameter Adjustment on Deep Neural Networks: An Application of Detection and Classification of COVID-19 Diseases |
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 |