CN107910061A - A kind of medical data processing method and system - Google Patents
A kind of medical data processing method and system Download PDFInfo
- Publication number
- CN107910061A CN107910061A CN201711245775.5A CN201711245775A CN107910061A CN 107910061 A CN107910061 A CN 107910061A CN 201711245775 A CN201711245775 A CN 201711245775A CN 107910061 A CN107910061 A CN 107910061A
- Authority
- CN
- China
- Prior art keywords
- client
- recognition result
- server
- training
- medical
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/248—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2455—Query execution
- G06F16/24552—Database cache management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses a kind of medical data processing method and system, comprise the following steps：Training data is received, loads deep learning model, the training data is inputted into the deep learning model, and training pattern is obtained using transfer learning method, and is preserved in the server, and the training pattern list L preserved in server is sent to client；Client loads the training pattern, gathers medical image, identifies the medical image using the training pattern, obtains recognition result R；The recognition result R is sent to server, the server retrieves the corresponding diseases of the recognition result R according to recognition result R in disease database, obtains final auxiliary diagnosis result Y.The present invention uses the deep learning method in artificial intelligence, and the processing accuracy of medical data is improved by way of artificial intelligence.
Description
Technical field
The present invention relates to medical data process field, particularly a kind of medical data processing method and system.
Background technology
At present, the problem of medical resource is insufficient has become a global problem.In some populous development
Middle country, such as China, India, medical resource is at full stretch, particularly medical practitioner resource.The high doctor people of medical technology content
Number is simultaneously few, is largely focused on big and medium-sized cities.This makes the doctor in some samll cities and towns and rural area can not be to some more complicated diseases
Disease carries out professional diagnosis, is easy to cause mistaken diagnosis, many patients is not obtained medical treatment in time.And cultivate the cycle of medical practitioner very
It is long, it is difficult to solve medical shortage problem in a short time.Recently as the development of artificial intelligence technology, Artificial Intelligence is in medical treatment
The application in field is more and more extensive.Artificial intelligence has very big prospect in medical domain, particularly with professional knowledge and height
In terms of the medical diagnosis robot of diagnostic accuracy.The appearance of artificial intelligence technology provides one for medical practitioner shortage of resources
Good solution.Artificial intelligence can be applied to many medical fields.For example, in terms of radiology, Chinese medicine image
Data are increased with 30% speed every year, and the growth rate of imaging practitioners only has 4%.Therefore, artificial intelligence technology will cure imaging
Life can break away from heavy work.Them can also be helped to reduce misdiagnosis rate, improve accuracy rate, wherein deep learning is artificial intelligence
Can the most popular technology in field.
The concept of deep learning comes from the research of artificial neural network.Multilayer perceptron containing more hidden layers is exactly a kind of depth
Learning structure.Deep learning forms more abstract high-rise expression attribute classification or feature by combining low-level feature, to find
The distributed nature of data represents.The concept of deep learning was proposed by Hinton et al. in 2006.Based on depth confidence network
(DBN) propose non-supervisory greed successively training algorithm, bring hope to solve the relevant optimization problem of deep structure, then propose
Multilayer autocoder deep structure.In addition the convolutional neural networks that Lecun et al. is proposed are first real sandwich constructions
Algorithm is practised, it reduces number of parameters to improve training performance using spatial correlation.Doctor is special using image viewing disease
Property influence, to determine disease and therapeutic scheme.This provides a large amount of data that can be used for training and learn, and makes artificial intelligence
Identification disease and completion diagnosis are possibly realized.At present, many researchers do in terms of use is studied in depth with classification of diseases
Extensive work, as Stamford university research team has been carried out training cutaneum carcinoma image benign to distinguish using deep learning
And chromoma.Team of Zhongshan University proposes to use deep learning Diagnosis of Congenital cataract.The work of these people proves, deep
It is very effective for Medical Images Classification and identification to enter study, it might even be possible to reaches the level of medical practitioner.
The content of the invention
The technical problems to be solved by the invention are, in view of the shortcomings of the prior art, providing a kind of medical data processing method
And system, improve the accuracy rate that medical data is handled, save medical resources.
In order to solve the above technical problems, the technical solution adopted in the present invention is：A kind of medical data processing method, including
Following steps：
1) training data is received, loads deep learning model, the training data is inputted into the deep learning model, and
Training pattern is obtained using transfer learning method, and is preserved in the server, and the training pattern list L that will be preserved in server
It is sent to client；
2) client loads the training pattern, gathers medical image, is schemed using the training pattern identification medical treatment
Picture, obtains recognition result R；
3) the recognition result R is sent to server, the server is examined according to recognition result R in disease database
The corresponding diseases of Suo Suoshu recognition results R, obtain final auxiliary diagnosis result Y.
After step 3), also it is handled as follows：
4) the auxiliary diagnosis result Y is sent to client by the server, and client shows the auxiliary diagnosis knot
Fruit.
The client identifies the medical image using tensorflow modules, obtains recognition result R.
The deep learning model is inceptionV3 models.
Correspondingly, present invention also offers a kind of medical treatment data processing system, including：
Server, for receiving training data, loads deep learning model, and the training data is inputted the depth
Model is practised, and preserves the training pattern, the training pattern list L of preservation is sent to client；Identified and tied according to client
Fruit R retrieves the corresponding diseases of the recognition result R in disease database, obtains final auxiliary diagnosis result Y；
Client, for loading the training pattern, gathers medical image, and the medical treatment is identified using the training pattern
Image, obtains recognition result R；The recognition result R is sent to server.
The client is additionally operable to show the auxiliary diagnosis result.
The client includes cache module, for caching the training pattern downloaded.
Compared with prior art, the advantageous effect of present invention is that：The present invention uses the depth in artificial intelligence
Learning method, improves the processing accuracy of medical data by way of artificial intelligence, it was proved that, after loading training pattern
Recognition methods to can reach general hospital medical practitioner horizontal, greatly alleviate medical resource it is nervous with distribution is unbalanced asks
Topic, can improve diagnosis efficiency, reduce misdiagnosis rate.
Brief description of the drawings
Fig. 1 is fundamental diagram of the present invention.
Fig. 2 is mobile client identification process figure of the present invention.
Fig. 3 trains flow chart for server-side of the present invention.
Embodiment
As Fig. 1, present invention specific implementation step are as follows：
1) model is trained to by training module and is stored in server and is deposited by received server-side to training data first
Store up in equipment, and the model list L preserved in server is sent to client.
2) client starts to identify, according to model list L, selects training pattern, such as：Model X, is first searched from local cache
Rope model X, if not having, goes server end to search for simultaneously download model X in storing.
3) after download model X, client starts stress model, and collection image starts to identify.
4) after end of identification, result R is sent to server end by parameter recognition result R, client, and server end is according to knot
Diseases of the fruit R in disease database corresponding to retrieval result R, obtains last diagnostic result Y, then Y is sent to client.
5) client receives diagnostic result Y, and result is shown.
Client identification process is as shown in Figure 2：
1) enter client, select the disease to be identified, it is locally slow into selection identification model interface, client-side search
Whether have model A, if 2) no model A is entered step, if entering step 3) if depositing
2) model A is not found, client initiates request, and search server end model cache list, is looked for
To model A, download it in client, download completes, and reminds user cache success, enters step 3)
3) selection is buffered in local identification model, client tensorflow modules, load identification model A, and carries out
Initialization, calls the camera of mobile equipment to gather an image, image is sent in tensorflow modules, is loading afterwards
Identification model in calculate, obtain as a result, entering step 4)
4) after decision-making module obtains result, server end is sent the result to, corresponding disease is matched in disease database,
The information and therapeutic modality of corresponding disease are returned, is finally shown on client screen, completes identification process.
Fig. 3 is server end workflow, and server end training process is as follows：
1) user uploads the training data of prescribed form, after data are passed to server end, verify form, passes through rear progress
Data cleansing is handled, and training data is loaded into training module afterwards, is entered step 2)
2) training pattern loading deep learning model, such as inceptionV3, is trained using transfer learning mode, obtained
3) identification model after to initial training, enters step
3) model of initial training is loaded into data processing module, is standardized compiling, generation is applicable in and movement
The identification module of equipment tensorflow, is finally stored in data memory module and waits client downloads.
4) identification model that user directly uploads after standardization by can also be stored directly in server.
Process and depth learning technology feature are diagnosed with reference to traditional medical, the present invention uses tensorflow frames and depth
Learn inceptionV3 models, using transfer learning technology, the medical image produced in medical procedure is arranged as training sample
This, and training in inception models is placed on, the medical diagnosismode of corresponding disease is obtained, finally loads diagnostic model
To identification client, doctor diagnoses the medical image of the disease using client, can also be adjusted after identifying successfully
With the disease database of specialty, the most scientific therapeutic scheme of the disease is provided, so as to help doctor to carry out diagnosis reference.The technology
Implementation includes two parts, diagnostic clients end, with training server end.
Client, using Mobile Development technology and tensorflow frames, is mainly used in android mobile platforms, bag
Include data acquisition module, decision-making module, model cache module, tensorflow modules.Data acquisition module is to move number of devices
It is mainly video camera based on sampling instrument, when receiving identification instruction, calls shooting api collections data to be identified, general number
According to being that picture is artwork after collection, its size is commonly more than 1M according to clarity, and picture size is excessive to be had for efficiency of transmission
Influenced, therefore the module also compresses the cutting that intelligence is carried out to the picture after collection, to improve the operational efficiency that identification is.
The decision-making module of client includes two aspect functions, first, to be used for according to different application scenarios, i.e., according to being identified
Various disease selects corresponding identification model.Second, for being analyzed after being completed in identification recognition result, and then
The disease database in high in the clouds is called, obtains corresponding disease treatment scheme.
Cache module：Due to mobile equipment in the wireless context, the limitation of the bandwidth and stability of network, and need from clothes
Substantial amounts of model data is loaded on business device, and downloading process takes long, therefore is not suitable for downloading identification in real time in identification process
Model, many mobile terminal smart machines can be embedded in the local memory device of one piece of certain capacity at present in practice.Based on
Upper reason, adds cache module in the client, can cache some according to hardware device and the needs of user and downloaded
Identification model.System identification speed can be improved to local memory device management by the cache module of client.
Tensorflow modules, TensorFlow are the second generation artificial intelligence that Google is researched and developed based on DistBelief
Learning system frame, is generally used for mainframe computing devices, and with the development of mobile equipment, mobile equipment cpu cores are not
Disconnected progress, it has increasingly outstanding computing capability, and tensorflow is exactly transplanted to mobile equipment, used by this programme
The computing capability of mobile equipment calculates to be identified, and thus difference and traditional high in the clouds recognition mode, visitor is placed on by identification
Family end, to improve service efficiency, the serious forgiveness effectively improved.Tensorflow modules are the cores that client is identified,
It, which can be directly invoked, is cached to local identification model from server end, and its label included recombines calculating figure, right
The picture to be measured of collecting device collection is calculated, and finally draws recognition result.
Server end includes, data memory module, training module, data processing module, and data storage network mainly stores
(1) algorithm and model (3) trained identification model (4) disease data of training data (2) deep learning and machine learning is treated
Storehouse, for the interpretation of result after identification.Training module uses tensorflow frames, by calling deep learning model, such as
Inception3, by way of transfer learning, performs training mission to incoming training data, obtains its corresponding identification
Model, and preserve data memory module.System to user open teaching interface, as long as user by training data according to regulation lattice
Similar training image, its entitled class name of file, you can the training for being uploaded to us is put down are stored in formula, i.e. a file
Platform, is trained, without spending resource to go oneself to build training environment again.If user possesses the training equipment of oneself at the same time,
Trained identification model can also be uploaded, facilitate other users to be downloaded use.A kind of crowdsourcing in this way
Mode makes to widen the cognition border of system so that the system is constantly evolved, more and more intelligently.
Claims (7)
1. a kind of medical data processing method, it is characterised in that comprise the following steps：
1) training data is received, loads deep learning model, the training data is inputted into the deep learning model, and utilize
Transfer learning method obtains training pattern, and preserves in the server, and the training pattern list L preserved in server is sent
To client；
2) client loads the training pattern, gathers medical image, identifies the medical image using the training pattern, obtains
To recognition result R；
3) the recognition result R is sent to server, the server retrieves institute according to recognition result R in disease database
The corresponding diseases of recognition result R are stated, obtain final auxiliary diagnosis result Y.
2. medical data processing method according to claim 1, it is characterised in that after step 3), also located as follows
Reason：
4) the auxiliary diagnosis result Y is sent to client by the server, and client shows the auxiliary diagnosis result.
3. medical data processing method according to claim 1, it is characterised in that the client utilizes tensorflow
Module identifies the medical image, obtains recognition result R.
4. medical data processing method according to claim 1, it is characterised in that the deep learning model is
InceptionV3 models.
A kind of 5. medical treatment data processing system, it is characterised in that including：
Server, for receiving training data, loads deep learning model, and the training data is inputted the deep learning mould
Type, and the training pattern is preserved, the training pattern list L of preservation is sent to client；Existed according to client recognition result R
The corresponding diseases of the recognition result R are retrieved in disease database, obtain final auxiliary diagnosis result Y；
Client, for loading the training pattern, gathers medical image, is schemed using the training pattern identification medical treatment
Picture, obtains recognition result R；The recognition result R is sent to server.
6. medical treatment data processing system according to claim 5, it is characterised in that the client is additionally operable to described in display
Auxiliary diagnosis result.
7. medical treatment data processing system according to claim 5, it is characterised in that the client includes cache module,
For caching the training pattern downloaded.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711245775.5A CN107910061A (en) | 2017-12-01 | 2017-12-01 | A kind of medical data processing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711245775.5A CN107910061A (en) | 2017-12-01 | 2017-12-01 | A kind of medical data processing method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107910061A true CN107910061A (en) | 2018-04-13 |
Family
ID=61848415
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711245775.5A Pending CN107910061A (en) | 2017-12-01 | 2017-12-01 | A kind of medical data processing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107910061A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108538390A (en) * | 2018-04-28 | 2018-09-14 | 中南大学 | A kind of increment type processing method towards medical data |
| CN109003660A (en) * | 2018-06-27 | 2018-12-14 | 南京邮电大学 | Intelligent medical service prediction management method and system, readable storage medium storing program for executing and terminal |
| CN109086789A (en) * | 2018-06-08 | 2018-12-25 | 四川斐讯信息技术有限公司 | A kind of image-recognizing method and system |
| CN109242836A (en) * | 2018-08-27 | 2019-01-18 | 百度在线网络技术（北京）有限公司 | Disease forecasting method, apparatus and storage medium |
| CN109893240A (en) * | 2019-03-18 | 2019-06-18 | 武汉大学 | A kind of hyperplasia of prostate bipolar electric resection operation method for early warning based on artificial intelligence |
| CN109948412A (en) * | 2018-12-27 | 2019-06-28 | 中南大学 | Prohibited items identification method based on depth convolutional neural networks |
| CN110009007A (en) * | 2019-03-18 | 2019-07-12 | 武汉大学 | A kind of artificial intelligence surgical assistant system towards polymorphic type disease |
| CN110458756A (en) * | 2019-06-25 | 2019-11-15 | 中南大学 | Fuzzy video super-resolution method and system based on deep learning |
| CN110491509A (en) * | 2019-07-01 | 2019-11-22 | 珠海格力电器股份有限公司 | Medical Robot, medical service method and storage medium |
| CN111199794A (en) * | 2018-11-19 | 2020-05-26 | 复旦大学附属眼耳鼻喉科医院 | Operation intelligent decision-making system suitable for high myopia cataract and establishing method thereof |
| CN111276254A (en) * | 2020-01-13 | 2020-06-12 | 印迹信息科技（北京）有限公司 | Medical open platform system and diagnosis and treatment data processing method |
| WO2020164270A1 (en) * | 2019-02-15 | 2020-08-20 | 平安科技（深圳）有限公司 | Deep-learning-based pedestrian detection method, system and apparatus, and storage medium |
| CN112309564A (en) * | 2019-07-26 | 2021-02-02 | 深圳百诺明医说科技有限公司 | Artificial intelligence diagnostic system and intelligent robot |
| CN112349425A (en) * | 2020-02-10 | 2021-02-09 | 胡秋明 | Novel artificial intelligent rapid screening system for coronavirus infection pneumonia |
| CN113485306A (en) * | 2021-07-28 | 2021-10-08 | 成都斯普智和信息技术有限公司 | Lightweight equipment fault diagnosis mobile terminal and diagnosis method thereof |
| CN113707289A (en) * | 2021-07-16 | 2021-11-26 | 联影智能医疗科技(北京)有限公司 | Medical artificial intelligence platform and construction method thereof |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955608A (en) * | 2014-04-24 | 2014-07-30 | 上海星华生物医药科技有限公司 | Intelligent medical information remote processing system and processing method |
| CN106934235A (en) * | 2017-03-09 | 2017-07-07 | 中国科学院软件研究所 | Patient's similarity measurement migratory system between a kind of disease areas based on transfer learning |
| CN106991439A (en) * | 2017-03-28 | 2017-07-28 | 南京天数信息科技有限公司 | Image-recognizing method based on deep learning and transfer learning |
| CN107133942A (en) * | 2017-04-24 | 2017-09-05 | 南京天数信息科技有限公司 | A kind of medical image processing method based on deep learning |
| US20170308656A1 (en) * | 2016-03-10 | 2017-10-26 | Siemens Healthcare Gmbh | Content-based medical image rendering based on machine learning |
| CN107403201A (en) * | 2017-08-11 | 2017-11-28 | 强深智能医疗科技（昆山）有限公司 | Tumour radiotherapy target area and jeopardize that organ is intelligent, automation delineation method |
-
2017
- 2017-12-01 CN CN201711245775.5A patent/CN107910061A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955608A (en) * | 2014-04-24 | 2014-07-30 | 上海星华生物医药科技有限公司 | Intelligent medical information remote processing system and processing method |
| US20170308656A1 (en) * | 2016-03-10 | 2017-10-26 | Siemens Healthcare Gmbh | Content-based medical image rendering based on machine learning |
| CN106934235A (en) * | 2017-03-09 | 2017-07-07 | 中国科学院软件研究所 | Patient's similarity measurement migratory system between a kind of disease areas based on transfer learning |
| CN106991439A (en) * | 2017-03-28 | 2017-07-28 | 南京天数信息科技有限公司 | Image-recognizing method based on deep learning and transfer learning |
| CN107133942A (en) * | 2017-04-24 | 2017-09-05 | 南京天数信息科技有限公司 | A kind of medical image processing method based on deep learning |
| CN107403201A (en) * | 2017-08-11 | 2017-11-28 | 强深智能医疗科技（昆山）有限公司 | Tumour radiotherapy target area and jeopardize that organ is intelligent, automation delineation method |
Non-Patent Citations (1)
| Title |
|---|
| KEHUA GUO 等: "MADP: An Open and Scalable Medical Auxiliary Diagnosis Platform", 《COMPUTING IN SCIENCE & ENGINEERING 》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108538390A (en) * | 2018-04-28 | 2018-09-14 | 中南大学 | A kind of increment type processing method towards medical data |
| CN109086789A (en) * | 2018-06-08 | 2018-12-25 | 四川斐讯信息技术有限公司 | A kind of image-recognizing method and system |
| CN109003660A (en) * | 2018-06-27 | 2018-12-14 | 南京邮电大学 | Intelligent medical service prediction management method and system, readable storage medium storing program for executing and terminal |
| CN109242836A (en) * | 2018-08-27 | 2019-01-18 | 百度在线网络技术（北京）有限公司 | Disease forecasting method, apparatus and storage medium |
| CN111199794A (en) * | 2018-11-19 | 2020-05-26 | 复旦大学附属眼耳鼻喉科医院 | Operation intelligent decision-making system suitable for high myopia cataract and establishing method thereof |
| CN111199794B (en) * | 2018-11-19 | 2024-03-01 | 复旦大学附属眼耳鼻喉科医院 | Intelligent surgical decision making system suitable for high myopia cataract and establishing method thereof |
| CN109948412B (en) * | 2018-12-27 | 2022-09-16 | 中南大学 | Forbidden article identification method based on deep convolutional neural network |
| CN109948412A (en) * | 2018-12-27 | 2019-06-28 | 中南大学 | Prohibited items identification method based on depth convolutional neural networks |
| WO2020164270A1 (en) * | 2019-02-15 | 2020-08-20 | 平安科技（深圳）有限公司 | Deep-learning-based pedestrian detection method, system and apparatus, and storage medium |
| CN109893240A (en) * | 2019-03-18 | 2019-06-18 | 武汉大学 | A kind of hyperplasia of prostate bipolar electric resection operation method for early warning based on artificial intelligence |
| CN110009007A (en) * | 2019-03-18 | 2019-07-12 | 武汉大学 | A kind of artificial intelligence surgical assistant system towards polymorphic type disease |
| CN110458756A (en) * | 2019-06-25 | 2019-11-15 | 中南大学 | Fuzzy video super-resolution method and system based on deep learning |
| CN110491509A (en) * | 2019-07-01 | 2019-11-22 | 珠海格力电器股份有限公司 | Medical Robot, medical service method and storage medium |
| CN112309564A (en) * | 2019-07-26 | 2021-02-02 | 深圳百诺明医说科技有限公司 | Artificial intelligence diagnostic system and intelligent robot |
| CN111276254A (en) * | 2020-01-13 | 2020-06-12 | 印迹信息科技（北京）有限公司 | Medical open platform system and diagnosis and treatment data processing method |
| CN112349425A (en) * | 2020-02-10 | 2021-02-09 | 胡秋明 | Novel artificial intelligent rapid screening system for coronavirus infection pneumonia |
| CN113707289A (en) * | 2021-07-16 | 2021-11-26 | 联影智能医疗科技(北京)有限公司 | Medical artificial intelligence platform and construction method thereof |
| CN113707289B (en) * | 2021-07-16 | 2023-11-10 | 联影智能医疗科技(北京)有限公司 | Medical artificial intelligence platform and construction method thereof |
| CN113485306A (en) * | 2021-07-28 | 2021-10-08 | 成都斯普智和信息技术有限公司 | Lightweight equipment fault diagnosis mobile terminal and diagnosis method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107910061A (en) | A kind of medical data processing method and system | |
| CN111353542B (en) | Training method and device for image classification model, computer equipment and storage medium | |
| Yue et al. | Cardiac segmentation from LGE MRI using deep neural network incorporating shape and spatial priors | |
| CN110349156A (en) | The recognition methods of lesion characteristics and device, storage medium in the picture of eyeground | |
| US20220262162A1 (en) | Face detection method, apparatus, and device, and training method, apparatus, and device for image detection neural network | |
| CN112861624A (en) | Human body posture detection method, system, storage medium, equipment and terminal | |
| CN107767935A (en) | Medical image specification processing system and method based on artificial intelligence | |
| CN110491480A (en) | A kind of medical image processing method, device, electromedical equipment and storage medium | |
| CN103765421B (en) | Contents controlling method, content control device and program | |
| CN110532996A (en) | The method of visual classification, the method for information processing and server | |
| CN110808096B (en) | Automatic heart lesion detection system based on convolutional neural network | |
| CN110472737A (en) | Training method, device and the magic magiscan of neural network model | |
| CN107529222A (en) | A kind of WiFi indoor locating systems based on deep learning | |
| CN110321956B (en) | Grass pest control method and device based on artificial intelligence | |
| CN110503077A (en) | A kind of real-time body's action-analysing method of view-based access control model | |
| CN107358023A (en) | Generate medical report, method, system and the equipment of interaction medical report | |
| CN109597907A (en) | Dress ornament management method and device, electronic equipment, storage medium | |
| CN115238173B (en) | Behavior analysis and medical service pushing method, equipment and medium based on big data | |
| CN110037680A (en) | The recognition methods of electrocardiogram and device, storage medium, computer equipment | |
| CN113435236A (en) | Home old man posture detection method, system, storage medium, equipment and application | |
| CN110009007A (en) | A kind of artificial intelligence surgical assistant system towards polymorphic type disease | |
| CN112613548A (en) | User customized target detection method, system and storage medium based on weak supervised learning | |
| CN116543429A (en) | Tongue image recognition system and method based on depth separable convolution | |
| CN108764233A (en) | A kind of scene character recognition method based on continuous convolution activation | |
| WO2018000248A1 (en) | Multi-user management method and system for smart body building apparatus |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180413 |
|
| RJ01 | Rejection of invention patent application after publication |