CN109949927A - A kind of intelligent diagnosing method and its system based on deep neural network - Google Patents
A kind of intelligent diagnosing method and its system based on deep neural network Download PDFInfo
- Publication number
- CN109949927A CN109949927A CN201910119744.8A CN201910119744A CN109949927A CN 109949927 A CN109949927 A CN 109949927A CN 201910119744 A CN201910119744 A CN 201910119744A CN 109949927 A CN109949927 A CN 109949927A
- Authority
- CN
- China
- Prior art keywords
- illness
- neural network
- deep neural
- submodule
- server
- 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
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 24
- 238000000034 method Methods 0.000 title claims abstract description 16
- 238000012549 training Methods 0.000 claims abstract description 47
- 238000003745 diagnosis Methods 0.000 claims abstract description 38
- 238000003780 insertion Methods 0.000 claims abstract description 20
- 230000037431 insertion Effects 0.000 claims abstract description 20
- 238000003062 neural network model Methods 0.000 claims abstract description 17
- 238000004364 calculation method Methods 0.000 claims abstract description 7
- 238000012790 confirmation Methods 0.000 claims description 5
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 5
- 201000010099 disease Diseases 0.000 claims description 4
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Abstract
The invention discloses a kind of intelligent diagnosing method based on deep neural network, deep neural network model is built, training data, and by the model insertion after training into server program；The illness of user terminal input is received, and is input in server program；Server illness database information is received, the term vector similarity of keyword in the illness keyword and server illness database of user terminal input is calculated；Tentative diagnosis is carried out according to similarity, and feeds back diagnostic result to user terminal.The invention also discloses a kind of intelligent diagnosis systems based on deep neural network, including data training module, illness receiving module, similarity calculation module and diagnostic feedback module.Use deep neural network algorithm of the invention innovative trains term vector model, and the key message that can be inputted with automatic identification user give doctor and customer to provide a preliminary diagnosis reference, raising diagnosis efficiency in conjunction with server illness database.
Description
Technical field
The present invention relates to data analysis technique fields, and in particular to a kind of intelligent diagnosing method based on deep neural network
And its system.
Background technique
During traditional diagnosis illness, generally requires patient and arrange in numerical order, be lined up into diagnosis room and show
Field diagnosis, it is time-consuming and laborious, greatly reduce the efficiency of diagnosis；Meanwhile the medical experience of patient is also significantly impacted, for
To the patient that itself disease condition is had gained some understanding, if not only can also according to hospital's row number, diagnosis, taking the processes such as medicine
Cause hospital in personnel's saturation state, also leverages the efficiency of diagnosis, for each patient, from the beginning doctor requires
Start to diagnose, causes doctor to be in manually medical state at any time, and then also result in the fatigue of doctor, the diagnosis for influencing doctor is quasi-
True rate.
Summary of the invention
Based on this, in view of the above-mentioned problems, it is necessary to propose a kind of key message that can be inputted with automatic identification user, in conjunction with
Database gives doctor and customer to provide a preliminary diagnosis reference, improve diagnosis efficiency based on deep neural network
Intelligent diagnosing method and its system.
The present invention provides a kind of intelligent diagnosing method based on deep neural network, and its technical solution is as follows:
A kind of intelligent diagnosing method based on deep neural network, comprising the following steps:
A, deep neural network model is built, training data, and by the model insertion after training into server program；
B, the illness of user terminal input is received, and is input in server program；
C, server illness database information is received, the illness keyword and server illness of user terminal input is calculated
The term vector similarity of keyword in database；
D, tentative diagnosis is carried out according to similarity, and feeds back diagnostic result to user terminal.
In the technical scheme, carry out training data using deep neural network model, can be inputted with automatic identification user
Key message gives user one preliminary diagnostic result in conjunction with server illness database, improves diagnosis efficiency.
Preferably, the step a the following steps are included:
A101, training data is collected, and carries out data preparation and classification；
A102, deep neural network model is built, trains training data that is collected and arranging, obtains term vector model；
A103, by term vector model insertion into server program.
It in the technical scheme, is subsequent calculating term vector phase using deep neural network model training term vector model
Model basis is provided like degree.
Preferably, the step b the following steps are included:
B101, the illness for receiving user terminal input, and the illness is scanned；
B102, participle operation is carried out to the illness information after scanning, obtain illness keyword；
B103, by the illness information input after participle into server program.
The key message of automatic identification user input, and carry out participle and extract keyword, in conjunction with server illness database,
So that the calculating of similarity is more accurate.
Preferably, in the step d carry out tentative diagnosis the following steps are included:
If similarity is more than or equal to 50%, preliminary to confirm illness, and obtain in server illness database with user terminal
The similar illness information of the illness of input；Conversely, not obtaining.
After judging similarity, confirm whether the illness of user's input is in server illness database according to similarity degree
Matched illness information, it is high if it is similarity, then the illness information in server illness database is obtained, and anti-
Feed patient and doctor, and preliminary diagnosis reference is provided for doctor, doctor is facilitated to carry out diagnosis confirmation.
The present invention also provides a kind of intelligent diagnosis systems based on deep neural network, and its technical solution is as follows:
A kind of intelligent diagnosis system based on deep neural network, including it is data training module, illness receiving module, similar
Spend computing module and diagnostic feedback module, in which:
Data training module for building deep neural network model, training data, and the model insertion after training is arrived
In server program；
Illness receiving module for receiving the illness description information of user terminal input, and is input in server program；
The illness of user terminal input is calculated for receiving server illness database information in similarity calculation module
The term vector similarity of keyword in keyword and server illness database；
Diagnostic feedback module for carrying out tentative diagnosis according to similarity, and feeds back diagnostic result to user terminal.
Preferably, the data training module includes data preparation submodule, model training submodule and model insertion
Module, in which:
Data preparation submodule for collecting training data, and carries out data preparation and classification；
Model training submodule is trained training data that is collected and arranging, is obtained for building deep neural network model
Take term vector model；
Model insertion submodule is used for term vector model insertion into server program.
Preferably, the illness receiving module includes illness scanning submodule, information participle submodule and data input
Module, in which:
Illness scans submodule, for receiving the illness of user terminal input, and is scanned to the illness；
Information segments submodule, for carrying out participle operation to the illness information after scanning, obtains illness keyword；
Data input submodule, for the illness information input after segmenting into server program.
Preferably, the diagnostic feedback module includes illness confirmation submodule, for judging that similarity is more than or equal to 50%
When, tentatively confirm illness, and obtain illness information similar with the illness of user terminal input in server illness database.
The beneficial effects of the present invention are:
Of the invention innovative use deep neural network algorithm trains term vector model, can be defeated with automatic identification user
The key message entered provides a preliminary diagnosis to doctor and customer and refers to, improve doctor in conjunction with server illness database
Diagnosis efficiency.
Detailed description of the invention
Fig. 1 is the flow chart of the intelligent diagnosing method described in the embodiment of the present invention based on deep neural network；
Fig. 2 is the functional block diagram of the intelligent diagnosis system described in the embodiment of the present invention based on deep neural network.
Description of symbols:
10- data training module；101- data preparation submodule；102- model training submodule；103- model insertion
Module；20- illness receiving module；201- illness scans submodule；202- information segments submodule；203- data input submodule
Block；30- similarity calculation module；40- diagnostic feedback module；401- illness confirms submodule.
Specific embodiment
The embodiment of the present invention is described in detail with reference to the accompanying drawing.
Embodiment 1
As shown in Figure 1, a kind of intelligent diagnosing method based on deep neural network, comprising the following steps:
A, deep neural network model is built, training data, and by the model insertion after training into server program；
B, the illness of user terminal input is received, and is input in server program；
C, server illness database information is received, the illness keyword and server illness of user terminal input is calculated
The term vector similarity of keyword in database；
D, tentative diagnosis is carried out according to similarity, and feeds back diagnostic result to user terminal.
In the present embodiment, carry out training data using deep neural network model, the pass that can be inputted with automatic identification user
Key information gives user one preliminary diagnostic result in conjunction with server illness database, improves diagnosis efficiency.
Embodiment 2
The present embodiment on the basis of embodiment 1, the step a the following steps are included:
A101, training data is collected, and carries out data preparation and classification；
A102, deep neural network model is built, trains training data that is collected and arranging, obtains term vector model；
A103, by term vector model insertion into server program.
It in the present embodiment, is that subsequent calculating term vector is similar using deep neural network model training term vector model
Degree provides model basis.
Embodiment 3
The present embodiment on the basis of embodiment 1, the step b the following steps are included:
B101, the illness for receiving user terminal input, and the illness is scanned；
B102, participle operation is carried out to the illness information after scanning, obtain illness keyword；
B103, by the illness information input after participle into server program.
The key message of automatic identification user input, and carry out participle and extract keyword, in conjunction with server illness database,
So that the calculating of similarity is more accurate.
Embodiment 4
The present embodiment on the basis of embodiment 1, in the step d carry out tentative diagnosis the following steps are included:
If similarity is more than or equal to 50%, preliminary to confirm illness, and obtain in server illness database with user terminal
The similar illness information of the illness of input；Conversely, not obtaining.
After judging similarity, confirm whether the illness of user's input is in server illness database according to similarity degree
Matched illness information, it is high if it is similarity, then the illness information in server illness database is obtained, and anti-
Feed patient and doctor, and preliminary diagnosis reference is provided for doctor, doctor is facilitated to carry out diagnosis confirmation.
Embodiment 5
The present embodiment is the system of embodiment 1, as shown in Fig. 2, a kind of intelligent diagnosis system based on deep neural network,
Including data training module 10, illness receiving module 20, similarity calculation module 30 and diagnostic feedback module 40, in which:
Data training module 10, for building deep neural network model, training data, and by the model insertion after training
Into server program；
Illness receiving module 20 for receiving the illness of user terminal input, and is input in server program；
The disease of user terminal input is calculated for receiving server illness database information in similarity calculation module 30
The term vector similarity of keyword in disease keyword and server illness database；
Diagnostic feedback module 40 for carrying out tentative diagnosis according to similarity, and feeds back diagnostic result to user terminal.
Embodiment 6
The present embodiment is the system of embodiment 2, and the data training module 10 includes data preparation submodule 101, model
Training submodule 102 and model insertion submodule 103, in which:
Data preparation submodule 101 for collecting training data, and carries out data preparation and classification；
Model training submodule 102 trains training number that is collected and arranging for building deep neural network model
According to acquisition term vector model；
Model insertion submodule 103 is used for term vector model insertion into server program.
Embodiment 7
The present embodiment is the system of embodiment 3, and the illness receiving module 20 includes that illness scans submodule 201, information
Segment submodule 202 and data input submodule 203, in which:
Illness scans submodule 201, for receiving the illness of user terminal input, and is scanned to the illness；
Information segments submodule 202, for carrying out participle operation to the illness information after scanning, obtains illness keyword；
Data input submodule 203, for the illness information input after segmenting into server program.
Embodiment 8
The present embodiment is the system of embodiment 4, and the diagnostic feedback module 40 includes that illness confirms submodule 401, is used for
Judge similarity be more than or equal to 50% when, tentatively confirmation illness, and obtain in server illness database with user terminal input
The similar illness information of illness.
A specific embodiment of the invention above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
Limitations on the scope of the patent of the present invention therefore cannot be interpreted as.It should be pointed out that for those of ordinary skill in the art
For, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to guarantor of the invention
Protect range.
Claims (8)
1. a kind of intelligent diagnosing method based on deep neural network, which comprises the following steps:
A, deep neural network model is built, training data, and by the model insertion after training into server program；
B, the illness of user terminal input is received, and is input in server program；
C, server illness database information is received, the illness keyword and server illness data of user terminal input is calculated
The term vector similarity of keyword in library；
D, tentative diagnosis is carried out according to similarity, and feeds back diagnostic result to user terminal.
2. the intelligent diagnosing method according to claim 1 based on deep neural network, which is characterized in that the step a
The following steps are included:
A101, training data is collected, and carries out data preparation and classification；
A102, deep neural network model is built, trains training data that is collected and arranging, obtains term vector model；
A103, by term vector model insertion into server program.
3. the intelligent diagnosing method according to claim 1 or 2 based on deep neural network, which is characterized in that the step
Rapid b the following steps are included:
B101, the illness for receiving user terminal input, and the illness is scanned；
B102, participle operation is carried out to the illness information after scanning, obtain illness keyword；
B103, by the illness information input after participle into server program.
4. the intelligent diagnosing method according to claim 1 or 2 based on deep neural network, which is characterized in that the step
Carry out tentative diagnosis in rapid d the following steps are included:
If similarity is more than or equal to 50%, illness is tentatively confirmed, and obtain in server illness database and input with user terminal
The similar illness information of illness；Conversely, not obtaining.
5. a kind of intelligent diagnosis system based on deep neural network, which is characterized in that received including data training module, illness
Module, similarity calculation module and diagnostic feedback module, in which:
Data training module, for building deep neural network model, training data, and by the model insertion after training to service
In device program；
Illness receiving module for receiving the illness of user terminal input, and is input in server program；
Similarity calculation module, for receiving server illness database information, the illness that user terminal input is calculated is crucial
The term vector similarity of keyword in word and server illness database；
Diagnostic feedback module for carrying out tentative diagnosis according to similarity, and feeds back diagnostic result to user terminal.
6. the intelligent diagnosis system according to claim 5 based on deep neural network, which is characterized in that the data instruction
Practicing module includes data preparation submodule, model training submodule and model insertion submodule, in which:
Data preparation submodule for collecting training data, and carries out data preparation and classification；
Model training submodule trains training data that is collected and arranging, obtains word for building deep neural network model
Vector model；
Model insertion submodule is used for term vector model insertion into server program.
7. the intelligent diagnosis system according to claim 5 or 6 based on deep neural network, which is characterized in that the disease
Disease receiving module includes illness scanning submodule, information participle submodule and data input submodule, in which:
Illness scans submodule, for receiving the illness of user terminal input, and is scanned to the illness；
Information segments submodule, for carrying out participle operation to the illness information after scanning, obtains illness keyword；
Data input submodule, for the illness information input after segmenting into server program.
8. the intelligent diagnosis system according to claim 5 or 6 based on deep neural network, which is characterized in that described to examine
Disconnected feedback module includes illness confirmation submodule, when for judging that similarity is more than or equal to 50%, tentatively confirms illness, and obtain
Illness information similar with the illness of user terminal input in server illness database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910119744.8A CN109949927A (en) | 2019-02-18 | 2019-02-18 | A kind of intelligent diagnosing method and its system based on deep neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910119744.8A CN109949927A (en) | 2019-02-18 | 2019-02-18 | A kind of intelligent diagnosing method and its system based on deep neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109949927A true CN109949927A (en) | 2019-06-28 |
Family
ID=67006748
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910119744.8A Pending CN109949927A (en) | 2019-02-18 | 2019-02-18 | A kind of intelligent diagnosing method and its system based on deep neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109949927A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111159977A (en) * | 2020-04-07 | 2020-05-15 | 深圳华大基因科技服务有限公司 | Method and device for determining standard human phenotype expression |
| CN113012804A (en) * | 2019-12-20 | 2021-06-22 | 中移(成都)信息通信科技有限公司 | Symptom determination method, apparatus, device and medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104102816A (en) * | 2014-06-20 | 2014-10-15 | 周晋 | Symptom match and machine learning-based automatic diagnosis system and method |
| CN105653840A (en) * | 2015-12-21 | 2016-06-08 | 青岛中科慧康科技有限公司 | Similar case recommendation system based on word and phrase distributed representation, and corresponding method |
| CN107368547A (en) * | 2017-06-28 | 2017-11-21 | 西安交通大学 | A kind of intelligent medical automatic question-answering method based on deep learning |
| CN107785075A (en) * | 2017-11-01 | 2018-03-09 | 杭州依图医疗技术有限公司 | Fever in children disease deep learning assistant diagnosis system based on text case history |
| CN107863147A (en) * | 2017-10-24 | 2018-03-30 | 清华大学 | The method of medical diagnosis based on depth convolutional neural networks |
| CN108198620A (en) * | 2018-01-12 | 2018-06-22 | 洛阳飞来石软件开发有限公司 | A kind of skin disease intelligent auxiliary diagnosis system based on deep learning |
| CN109192300A (en) * | 2018-08-17 | 2019-01-11 | 百度在线网络技术（北京）有限公司 | Intelligent way of inquisition, system, computer equipment and storage medium |
| US20190034810A1 (en) * | 2017-07-31 | 2019-01-31 | Boe Technology Group Co., Ltd. | Data processing method, medical term processing system and medical diagnostic system |
| US20190035506A1 (en) * | 2017-07-31 | 2019-01-31 | Hefei University Of Technology | Intelligent auxiliary diagnosis method, system and machine-readable medium thereof |
-
2019
- 2019-02-18 CN CN201910119744.8A patent/CN109949927A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104102816A (en) * | 2014-06-20 | 2014-10-15 | 周晋 | Symptom match and machine learning-based automatic diagnosis system and method |
| CN105653840A (en) * | 2015-12-21 | 2016-06-08 | 青岛中科慧康科技有限公司 | Similar case recommendation system based on word and phrase distributed representation, and corresponding method |
| CN107368547A (en) * | 2017-06-28 | 2017-11-21 | 西安交通大学 | A kind of intelligent medical automatic question-answering method based on deep learning |
| US20190034810A1 (en) * | 2017-07-31 | 2019-01-31 | Boe Technology Group Co., Ltd. | Data processing method, medical term processing system and medical diagnostic system |
| US20190035506A1 (en) * | 2017-07-31 | 2019-01-31 | Hefei University Of Technology | Intelligent auxiliary diagnosis method, system and machine-readable medium thereof |
| CN107863147A (en) * | 2017-10-24 | 2018-03-30 | 清华大学 | The method of medical diagnosis based on depth convolutional neural networks |
| CN107785075A (en) * | 2017-11-01 | 2018-03-09 | 杭州依图医疗技术有限公司 | Fever in children disease deep learning assistant diagnosis system based on text case history |
| CN108198620A (en) * | 2018-01-12 | 2018-06-22 | 洛阳飞来石软件开发有限公司 | A kind of skin disease intelligent auxiliary diagnosis system based on deep learning |
| CN109192300A (en) * | 2018-08-17 | 2019-01-11 | 百度在线网络技术（北京）有限公司 | Intelligent way of inquisition, system, computer equipment and storage medium |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113012804A (en) * | 2019-12-20 | 2021-06-22 | 中移(成都)信息通信科技有限公司 | Symptom determination method, apparatus, device and medium |
| CN113012804B (en) * | 2019-12-20 | 2024-03-19 | 中移(成都)信息通信科技有限公司 | Symptom determining method, device, equipment and medium |
| CN111159977A (en) * | 2020-04-07 | 2020-05-15 | 深圳华大基因科技服务有限公司 | Method and device for determining standard human phenotype expression |
| CN111159977B (en) * | 2020-04-07 | 2020-09-08 | 深圳华大基因科技服务有限公司 | Method and device for determining standard human phenotype expression |
| WO2021203694A1 (en) * | 2020-04-07 | 2021-10-14 | 深圳华大基因科技服务有限公司 | Human phenotypic standard term determination method, and related device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110232383A (en) | A kind of lesion image recognition methods and lesion image identifying system based on deep learning model | |
| CN103440421B (en) | medical data processing method and system | |
| CN106777873A (en) | Community's smart cloud health services system | |
| CN103324831A (en) | Remote interrogation system | |
| CN110473615A (en) | A kind of information processing method and device | |
| CN111738302A (en) | System for classifying and diagnosing Alzheimer disease based on multi-modal data | |
| CN109949927A (en) | A kind of intelligent diagnosing method and its system based on deep neural network | |
| CN109584229A (en) | A kind of real-time assistant diagnosis system of Endoscopic retrograde cholangio-pancreatiography art and method | |
| CN110338759B (en) | Facial pain expression data acquisition method | |
| CN117153431B (en) | Medical service system and method based on Internet | |
| CN107516004A (en) | The identifying processing method and device of medical image picture | |
| CN105611872A (en) | An apparatus and method for evaluating multichannel ECG signals | |
| CN110432860B (en) | Method and system for identifying lattice-change cracked holes in wide-area fundus map based on deep learning | |
| US20190027246A1 (en) | Method and apparatus for detecting and analyzing a workflow performed with an imaging apparatus | |
| CN114520062B (en) | Medical cloud communication system based on AI and letter creation | |
| CN110889847A (en) | Nuclear radiation damage assessment system and method based on infrared imaging | |
| CN110634559A (en) | Health examination intelligent guide system using smart phone | |
| CN114882983A (en) | Smart ward interaction method, system and storage medium | |
| CN115457422A (en) | Sampling process verification method and device and augmented reality function glasses | |
| Nugroho et al. | An automated detection and segmentation of thyroid nodules using Res-UNet | |
| CN113822917A (en) | Accurate registration method for liver cancer imaging omics images | |
| CN114300071A (en) | Real data-based informatization system and method | |
| CN113946217A (en) | Intelligent auxiliary evaluation system for enteroscope operation skills | |
| CN113436711A (en) | Western medicine inquiry system based on big data platform | |
| Boyadzhiev et al. | Deep learning image classification for pneumonia detection |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190628 |