CN111653356A - New coronary pneumonia screening method and new coronary pneumonia screening system based on deep learning - Google Patents
New coronary pneumonia screening method and new coronary pneumonia screening system based on deep learning Download PDFInfo
- Publication number
- CN111653356A CN111653356A CN202010312993.1A CN202010312993A CN111653356A CN 111653356 A CN111653356 A CN 111653356A CN 202010312993 A CN202010312993 A CN 202010312993A CN 111653356 A CN111653356 A CN 111653356A
- Authority
- CN
- China
- Prior art keywords
- probability
- classification
- coronary pneumonia
- new coronary
- lesion area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- 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
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- 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
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/80—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for detecting, monitoring or modelling epidemics or pandemics, e.g. flu
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a new coronary pneumonia screening method and a new coronary pneumonia screening system based on deep learning, wherein the new coronary pneumonia screening method comprises the following steps: the lung lesion region of CT is detected by a deep learning detection model, and then is sent to a three-classification network, wherein the three classifications comprise COVID-19, influenza A and no infection symptom. Then, the diagnosis result of CT and the probability of illness are output through calculation processing. The method is based on deep learning and automatic learning of the characteristics of CT images to distinguish COVID-19, influenza A and healthy people, the accuracy of the method is high, the total accuracy of the current test reaches 86.7%, the diagnosis speed is high, and only 30-60S is needed for one set of CT according to different slice numbers. The user can upload the CT image file through the new coronary pneumonia screening system, and calculate and output the diagnosis result and the illness probability of CT, so that the operation is convenient, the speed is high, and the detection rate of COVID-19 is greatly improved.
Description
Technical Field
The invention belongs to the technical field of medicine, and particularly relates to a new coronary pneumonia screening method and a new coronary pneumonia screening system based on deep learning.
Background
In the early screening of the novel coronary pneumonia (COVID-19), detection is mainly carried out by means of a kit at present, but the detection rate is low, and in order to improve the detection rate of the COVID-19 in the early screening, some other auxiliary methods need to be adopted for judgment, such as a CT image judgment method and the like.
Some methods for detecting lung lesion tissues exist in the prior art, for example, patent document No. CN110599448A discloses a migration learning lung lesion tissue detection system based on MaskScoring R-CNN network, which inputs a lung CT image to be detected, outputs the network to obtain an identified image, frames and masks the lesion tissues identified by mask, and marks the category of the lesion. However, during training, the mask of a diseased region needs to be calculated, a large amount of resources are consumed for marking, and the operation is complicated, so that the method is not suitable for detecting the novel coronary pneumonia (COVID-19) for improving the early screening detection rate of the COVID-19.
Disclosure of Invention
The invention provides a new coronary pneumonia screening method based on deep learning, which is used for identifying influenza A, new coronary and normal CT, is convenient to operate, outputs the whole infection probability and assists a doctor in diagnosing.
The technical scheme of the invention is as follows: a new coronary pneumonia screening method based on deep learning comprises the following steps:
s1, extracting a mask of the effective area of the lung;
s2, segmenting the effective region of the lung by using the extracted mask, reducing the influence of irrelevant regions, and normalizing the Hu value of the original CT;
s3, segmenting a lung lesion region by using a 3D convolutional neural network, and outputting position information of the lesion region, wherein the position information comprises coordinates (x, y, z) of a central point and the diameter D of the lesion region;
s4, detecting a two-dimensional lesion area image of the middle layer at the lesion area position, and simultaneously acquiring adjacent images above and below the two-dimensional lesion area image, namely acquiring three images in total;
s5, sending the classified images into a two-dimensional classification network, training, and judging that the images are COVID-19 and influenza A or normal;
s6, predicting the image cut out in the step S4 by using the trained model, and outputting the classification and the classification probability;
s7, calculating the classification and probability of a single lesion area for the three images divided from each lesion area;
and S8, calculating the classification and the probability of the single lesion area obtained in the step S7, and outputting the overall disease type and the probability of the single patient.
The invention judges the CT image of a patient by utilizing a deep learning technology, automatically learns the image characteristics of the patient and distinguishes influenza A from healthy people. The invention firstly utilizes a deep learning detection model to detect the lung lesion area of CT, and then sends the lung lesion area into a three-classification network, wherein the three classifications comprise COVID-19, influenza A and no infection symptom. Then, the diagnosis result of CT and the probability of illness are output through calculation processing.
Preferably, in step S2, the Hu value of the original CT is normalized to 0-255.
Preferably, in step S7, the classification and probability of a single lesion region are calculated by using a bayesian noise-or model.
Preferably, in step S8, the classification and probability of the single lesion region obtained in step S7 are calculated by using a bayesian noise-or model, and the overall disease type and probability of the single patient are output.
The invention also provides a new coronary pneumonia screening system based on the web interface, which comprises the following components:
the image acquisition module is used for acquiring CT original compressed files;
the decompression module is used for decompressing the CT original compressed file acquired by the image acquisition module;
the format judgment module is used for carrying out format judgment on the decompressed file decompressed by the decompression module and screening to obtain a CT image in a lung window format in CT;
the data processing module is used for calculating and processing the CT image in the lung window format in the CT obtained by screening of the format judging module and calculating the integral disease type and probability of a single patient;
and the display module is used for displaying the processing result of the data processing module.
The invention also develops an application system of the web application interface, which is used for uploading the original compressed CT file; decompressing the uploaded CT compressed file, judging whether the format of the file is correct or not, and screening out a set of CT images in a lung window format in CT; sending the arranged CT image into a data processing module for processing; and finally, obtaining a result calculated by the data processing module and displaying the result. The user can upload the CT image file through the web interface, the background algorithm side calculates, the diagnosis result and the sick probability of the CT are output, and the detection rate is improved.
Preferably, the data processing module calculates the classification and the probability of a single lesion area by using a Bayesian noise-or model, then calculates the classification and the probability of the single lesion area by using the Bayesian noise-or model, and outputs the whole disease type and the probability of a single patient.
Compared with the prior art, the invention has the beneficial effects that:
(1) the lung lesion area of the CT is detected by using a deep learning detection model, and then is sent to a three-classification network, wherein the three classifications comprise COVID-19, influenza A and no infection symptom; then, the diagnosis result of CT and the probability of illness are output through calculation processing. The method is based on deep learning and automatic learning of the characteristics of CT images to distinguish COVID-19, influenza A and healthy people, the accuracy of the method is high, the total accuracy of the current test reaches 86.7%, the diagnosis speed is high, and only 30-60S is needed for one set of CT according to different slice numbers.
(2) The invention also develops an application system of the web application interface, which can be used by doctors and the like, and users can upload CT image files through the web interface of the invention, and the background algorithm side can calculate and output the diagnosis result and the illness probability of CT, thereby not only having convenient operation, but also having high speed and greatly improving the detection rate of COVID-19.
(3) The lung lesion tissue detection method is convenient to operate, can directly output the trained lung lesion tissue detection model to the lesion tissue, and is more efficient and concise compared with the existing lung lesion tissue detection method.
Detailed Description
The technical solution in the embodiments of the present invention is clearly and completely described below. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the invention without inventive step, are within the scope of the invention.
Example 1
A new coronary pneumonia screening method based on deep learning specifically comprises the following steps:
s1 extracting mask of effective area of lung
S2, segmenting the effective region of the lung by using the extracted mask, reducing the influence of irrelevant regions, and normalizing the Hu value of the original CT to 0-255;
s3, segmenting a lung lesion region by using a 3D convolutional neural network, and outputting position information of the lesion region, wherein the position information comprises coordinates (x, y, z) of a central point and the diameter D of the lesion region;
s4, detecting the two-dimensional lesion area image of the middle layer of the lesion area position, and simultaneously acquiring the adjacent images above and below the two-dimensional lesion area image, wherein the total three images are
S5, sending the classified images into a two-dimensional classification network for training, and judging whether the images are COVID-19, influenza A or normal;
s6, predicting the image cut out in the step 4 by using the trained model, and outputting the classification and the classification probability;
s7, calculating the three pictures divided from each lesion region by using a Bayesian noise-or model, and calculating the classification and probability of each lesion region;
and S8, calculating the classification and the probability of the single lesion area obtained in the step S7 by using a Bayesian noise-or model, and outputting the overall disease type and the probability of the single patient.
In the embodiment, a deep learning detection model is used for detecting a lung lesion region of CT, and then the lung lesion region is sent into a three-classification network, wherein the three classifications comprise COVID-19, influenza A and no infection symptom; then, the diagnosis result of CT and the probability of illness are output through calculation processing. The total accuracy rate of the method reaches 86.7 percent, the diagnosis speed is high, and a set of CT only needs 30 to 60 seconds according to different number of slices.
Example 2
This embodiment is a new crown pneumonia screening system based on web interface, which is designed on the basis of embodiment 1, and includes:
the image acquisition module is used for acquiring CT original compressed files;
the decompression module is used for decompressing the CT original compressed file acquired by the image acquisition module;
the format judgment module is used for carrying out format judgment on the decompressed file decompressed by the decompression module and screening to obtain a CT image in a lung window format in CT;
the data processing module is used for calculating and processing the CT image in the lung window format in the CT obtained by screening of the format judging module and calculating the integral disease type and probability of a single patient;
and the display module is used for displaying the processing result of the data processing module.
The embodiment develops a web application interface, which is used for uploading a CT original compressed file, decompressing the uploaded CT compressed file and judging whether the format of the uploaded CT compressed file is correct or not; then screening out a set of CT images in a lung window format in CT; sending the sorted CT image into an algorithm module for processing; and finally, obtaining a result calculated by the algorithm module and displaying the result.
The invention is not limited to the above alternative embodiments, and any other various forms of products can be obtained by anyone in the light of the present invention, but any changes in shape or structure thereof, which fall within the scope of the present invention as defined in the claims, fall within the scope of the present invention.
Claims (6)
1. A new coronary pneumonia screening method based on deep learning is characterized by comprising the following steps:
s1, extracting a mask of the effective area of the lung;
s2, segmenting the effective region of the lung by using the extracted mask, reducing the influence of irrelevant regions, and normalizing the Hu value of the original CT;
s3, segmenting a lung lesion region by using a 3D convolutional neural network, and outputting position information of the lesion region, wherein the position information comprises coordinates (x, y, z) of a central point and the diameter D of the lesion region;
s4, detecting a two-dimensional lesion area image of the middle layer at the lesion area position, and simultaneously acquiring adjacent images above and below the two-dimensional lesion area image, namely acquiring three images in total;
s5, sending the classified images into a two-dimensional classification network, training, and judging that the images are COVID-19 and influenza A or normal;
s6, predicting the image cut out in the step S4 by using the trained model, and outputting the classification and the classification probability;
s7, calculating the classification and probability of a single lesion area for the three images divided from each lesion area;
and S8, calculating the classification and the probability of the single lesion area obtained in the step S7, and outputting the overall disease type and the probability of the single patient.
2. The deep learning-based new coronary pneumonia screening method of claim 1, wherein the Hu value of the original CT is normalized to 0 ~ 255 in the step S2.
3. The method for screening new coronary pneumonia according to claim 1 or 2, wherein the classification of single lesion area and its probability are calculated by using Bayesian noise-or model in step S7.
4. The method for screening new coronary pneumonia based on deep learning as claimed in claim 3, wherein the classification and probability of single lesion area obtained in step S7 in step S8 are calculated by using Bayesian noise-or model, and the overall disease type and probability of single patient are output.
5. A web interface-based new crown pneumonia screening system, comprising:
the image acquisition module is used for acquiring CT original compressed files;
the decompression module is used for decompressing the CT original compressed file acquired by the image acquisition module;
the format judgment module is used for carrying out format judgment on the decompressed file decompressed by the decompression module and screening to obtain a CT image in a lung window format in CT;
the data processing module is used for calculating and processing the CT image in the lung window format in the CT obtained by screening of the format judging module and calculating the integral disease type and probability of a single patient;
and the display module is used for displaying the processing result of the data processing module.
6. The web interface-based new coronary pneumonia screening system of claim 5, wherein the data processing module calculates the classification and probability of a single lesion area by using a Bayesian noise-or model, and then calculates the classification and probability of a single lesion area by using the Bayesian noise-or model, and outputs the overall disease type and probability of a single patient.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010312993.1A CN111653356A (en) | 2020-04-20 | 2020-04-20 | New coronary pneumonia screening method and new coronary pneumonia screening system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010312993.1A CN111653356A (en) | 2020-04-20 | 2020-04-20 | New coronary pneumonia screening method and new coronary pneumonia screening system based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111653356A true CN111653356A (en) | 2020-09-11 |
Family
ID=72346501
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010312993.1A Pending CN111653356A (en) | 2020-04-20 | 2020-04-20 | New coronary pneumonia screening method and new coronary pneumonia screening system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111653356A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111932540A (en) * | 2020-10-14 | 2020-11-13 | 北京信诺卫康科技有限公司 | CT image contrast characteristic learning method for clinical typing of new coronary pneumonia |
| CN112102945A (en) * | 2020-11-09 | 2020-12-18 | 电子科技大学 | Device for predicting severe condition of COVID-19 patient |
| CN112216401A (en) * | 2020-10-12 | 2021-01-12 | 中国石油大学(华东) | DF-based new coronary pneumonia federal learning detection method |
| CN112786189A (en) * | 2021-01-05 | 2021-05-11 | 重庆邮电大学 | Intelligent diagnosis system for new coronary pneumonia based on deep learning |
| CN112951412A (en) * | 2021-03-11 | 2021-06-11 | 深圳大学 | Auxiliary diagnosis method and application thereof |
| CN112967778A (en) * | 2021-03-17 | 2021-06-15 | 复旦大学附属华山医院 | Accurate medication method and system for inflammatory bowel disease based on machine learning |
| CN113138250A (en) * | 2021-04-23 | 2021-07-20 | 西湖大学 | Non-diagnostic method for typing covid-19 grade by using characteristic urine protein and application |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955703A (en) * | 2014-04-25 | 2014-07-30 | 杭州电子科技大学 | Medical image disease classification method based on naive Bayes |
| CN106909778A (en) * | 2017-02-09 | 2017-06-30 | 北京市计算中心 | A kind of Multimodal medical image recognition methods and device based on deep learning |
| CN107507184A (en) * | 2017-09-26 | 2017-12-22 | 上海辉明软件有限公司 | Method for building up, device and the electronic equipment of focus model |
| CN108985345A (en) * | 2018-06-25 | 2018-12-11 | 重庆知遨科技有限公司 | A kind of detection device based on the classification of lung's Medical image fusion |
| CN109003672A (en) * | 2018-07-16 | 2018-12-14 | 北京睿客邦科技有限公司 | A kind of early stage of lung cancer detection classification integration apparatus and system based on deep learning |
| CN109036568A (en) * | 2018-09-03 | 2018-12-18 | 浪潮软件集团有限公司 | Method for establishing prediction model based on naive Bayes algorithm |
| CN109145838A (en) * | 2018-08-29 | 2019-01-04 | 常州市第二人民医院 | Clear cell carcinoma of kidney diagnostic method based on random Gaussian field neural network aiding |
| WO2019157214A2 (en) * | 2018-02-07 | 2019-08-15 | Ai Technologies Inc. | Deep learning-based diagnosis and referral of diseases and disorders |
| CN110634132A (en) * | 2019-08-30 | 2019-12-31 | 浙江大学 | Method for automatically generating tuberculosis quantitative diagnosis report based on deep learning 3D CT image |
| CN110807788A (en) * | 2019-10-21 | 2020-02-18 | 腾讯科技（深圳）有限公司 | Medical image processing method, device, electronic equipment and computer storage medium |
-
2020
- 2020-04-20 CN CN202010312993.1A patent/CN111653356A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955703A (en) * | 2014-04-25 | 2014-07-30 | 杭州电子科技大学 | Medical image disease classification method based on naive Bayes |
| CN106909778A (en) * | 2017-02-09 | 2017-06-30 | 北京市计算中心 | A kind of Multimodal medical image recognition methods and device based on deep learning |
| CN107507184A (en) * | 2017-09-26 | 2017-12-22 | 上海辉明软件有限公司 | Method for building up, device and the electronic equipment of focus model |
| WO2019157214A2 (en) * | 2018-02-07 | 2019-08-15 | Ai Technologies Inc. | Deep learning-based diagnosis and referral of diseases and disorders |
| CN108985345A (en) * | 2018-06-25 | 2018-12-11 | 重庆知遨科技有限公司 | A kind of detection device based on the classification of lung's Medical image fusion |
| CN109003672A (en) * | 2018-07-16 | 2018-12-14 | 北京睿客邦科技有限公司 | A kind of early stage of lung cancer detection classification integration apparatus and system based on deep learning |
| CN109145838A (en) * | 2018-08-29 | 2019-01-04 | 常州市第二人民医院 | Clear cell carcinoma of kidney diagnostic method based on random Gaussian field neural network aiding |
| CN109036568A (en) * | 2018-09-03 | 2018-12-18 | 浪潮软件集团有限公司 | Method for establishing prediction model based on naive Bayes algorithm |
| CN110634132A (en) * | 2019-08-30 | 2019-12-31 | 浙江大学 | Method for automatically generating tuberculosis quantitative diagnosis report based on deep learning 3D CT image |
| CN110807788A (en) * | 2019-10-21 | 2020-02-18 | 腾讯科技（深圳）有限公司 | Medical image processing method, device, electronic equipment and computer storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 张巧丽;赵地;迟学斌;: "基于深度学习的医学影像诊断综述", 计算机科学, no. 2 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112216401A (en) * | 2020-10-12 | 2021-01-12 | 中国石油大学(华东) | DF-based new coronary pneumonia federal learning detection method |
| CN111932540A (en) * | 2020-10-14 | 2020-11-13 | 北京信诺卫康科技有限公司 | CT image contrast characteristic learning method for clinical typing of new coronary pneumonia |
| CN112102945A (en) * | 2020-11-09 | 2020-12-18 | 电子科技大学 | Device for predicting severe condition of COVID-19 patient |
| CN112102945B (en) * | 2020-11-09 | 2021-02-05 | 电子科技大学 | Device for predicting severe condition of COVID-19 patient |
| CN112786189A (en) * | 2021-01-05 | 2021-05-11 | 重庆邮电大学 | Intelligent diagnosis system for new coronary pneumonia based on deep learning |
| CN112786189B (en) * | 2021-01-05 | 2022-07-01 | 重庆邮电大学 | Intelligent diagnosis system for new coronary pneumonia based on deep learning |
| CN112951412A (en) * | 2021-03-11 | 2021-06-11 | 深圳大学 | Auxiliary diagnosis method and application thereof |
| CN112967778A (en) * | 2021-03-17 | 2021-06-15 | 复旦大学附属华山医院 | Accurate medication method and system for inflammatory bowel disease based on machine learning |
| CN112967778B (en) * | 2021-03-17 | 2024-02-02 | 复旦大学附属华山医院 | Accurate medicine application method and system for inflammatory bowel disease based on machine learning |
| CN113138250A (en) * | 2021-04-23 | 2021-07-20 | 西湖大学 | Non-diagnostic method for typing covid-19 grade by using characteristic urine protein and application |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111653356A (en) | New coronary pneumonia screening method and new coronary pneumonia screening system based on deep learning | |
| CN110223281B (en) | Pulmonary nodule image classification method when data set contains uncertain data | |
| Shakeel et al. | Lung cancer detection from CT image using improved profuse clustering and deep learning instantaneously trained neural networks | |
| CN108898595B (en) | Construction method and application of positioning model of focus region in chest image | |
| CN107730489A (en) | Wireless capsule endoscope small intestine disease variant computer assisted detection system and detection method | |
| JP6885517B1 (en) | Diagnostic support device and model generation device | |
| CN108062749B (en) | Identification method and device for levator ani fissure hole and electronic equipment | |
| CN112365973B (en) | Pulmonary nodule auxiliary diagnosis system based on countermeasure network and fast R-CNN | |
| CN112365464A (en) | GAN-based medical image lesion area weak supervision positioning method | |
| CN114358194A (en) | Gesture tracking based detection method for abnormal limb behaviors of autism spectrum disorder | |
| CN111598868B (en) | Lung ultrasonic image identification method and system | |
| Hatanaka et al. | Improvement of automated detection method of hemorrhages in fundus images | |
| CN111833321A (en) | Window-adjusting optimization-enhanced intracranial hemorrhage detection model and construction method thereof | |
| CN110782441A (en) | DR image pulmonary tuberculosis intelligent segmentation and detection method based on deep learning | |
| CN110634132A (en) | Method for automatically generating tuberculosis quantitative diagnosis report based on deep learning 3D CT image | |
| CN114972272A (en) | Grad-CAM-based segmentation method for new coronary pneumonia lesions | |
| CN101404062A (en) | Automatic screening method for digital galactophore image based on decision tree | |
| CN111724356A (en) | Image processing method and system for CT image pneumonia identification | |
| Bhaskar et al. | Pulmonary lung nodule detection and classification through image enhancement and deep learning | |
| Gaber et al. | Comprehensive assessment of facial paralysis based on facial animation units | |
| CN115409812A (en) | CT image automatic classification method based on fusion time attention mechanism | |
| CN114023436A (en) | Artificial intelligence-based pulmonary tuberculosis classification diagnosis model training method | |
| CN113130050A (en) | Medical information display method and system | |
| CN112614091A (en) | Ultrasonic multi-section data detection method for congenital heart disease | |
| Misra et al. | Automatic lung segmentation: a comparison of anatomical and machine learning approaches |
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 |