CN107516004A - The identifying processing method and device of medical image picture - Google Patents
The identifying processing method and device of medical image picture Download PDFInfo
- Publication number
- CN107516004A CN107516004A CN201710547048.8A CN201710547048A CN107516004A CN 107516004 A CN107516004 A CN 107516004A CN 201710547048 A CN201710547048 A CN 201710547048A CN 107516004 A CN107516004 A CN 107516004A
- Authority
- CN
- China
- Prior art keywords
- picture
- gray level
- medical image
- medical
- matrix
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/63—Scene text, e.g. street names
Abstract
The invention discloses a kind of identifying processing method and device of medical image picture, the above method includes：Medical image picture is detected, filters out the character area of the medical image picture；Distribution statisticses are carried out to the pixel of the character area, multiple key message points are selected according to data gradient change；External matrix corresponding to the character area is built using the multiple key message point and cut；External matrix after cutting is converted into gray level image；Picture recognition is carried out to the gray level image and automatically extracts information into text box.By technical scheme provided by the invention, the accuracy of information extraction can be greatly improved, and improves interrogation efficiency.
Description
Technical field
The present invention relates to the communications field, in particular to a kind of identifying processing method and device of medical image picture.
Background technology
With scientific and technological fast development, internet it is highly popular, greatly shorten interaction distance and the time of person to person,
So as to drive the emergence of internet hospital.Internet hospital brings many facilities to us, benefits masses, particularly compares
Remote backward areas, allow popular wide and other regional famous experts, the professor of being gone up north with interrogation that can stay indoors；It is not only
Masses have saved round travel charge, it is often more important that shorten medical, the medical speed of patient, be the timely medical treatment of many patients
Offer convenience.
Internet hospital, it is meant that interrogation networking, include the electronization of inspection result, laboratory test report etc..It is main at present real
Existing mode is the secondary interrogation based on basic hospital, i.e. patient does preliminary interrogation and inspection in basic hospital, due to basic hospital
It is equipped with and doctor such as lacks experience at the reason to difficult and complicated illness, causes to the therapeutic effect unobvious of patient, it is necessary to seek help experience more
The big doctor expert in more big cities, and then initiate basic unit doctor and the process of the accomplished expert internet consultation of doctors.
Fig. 1 is the consultation process figure of correlation technique, as shown in figure 1, the content in figure in dotted line frame is using medical analysis mould
Formula, i.e., being uploaded the patient on pictorial informations such as all inspection results of this consultation of doctors and laboratory test reports in base level patient end
Into the personal case of network hospital platform, big doctor is after every inspection result is checked, to being cured after illness analysis with basic unit
It is raw to carry out the network consultation of doctors, therapeutic scheme is drawn, and treated, the course for the treatment of can carry out new inspection after terminating, and obtain new chemical examination
As a result picture, carry out new analysis for both sides doctor and further treat or draw treatment results.It is crucial in whole flow process
Any is the document in kind that laboratory test report, case history provided under this pattern etc. is all based on entity basic hospital, swift electron
Operation be exactly upload of taking a picture, wherein, electronic health record for internet hospital platform staff according to the case history figure of upload
Some other information that piece and basic unit provide, carries out manual entry, completes electronization；Secondly, laboratory test report etc. checks that picture can only
Doctor carries out manual reviews by both party, and substantial amounts of result of laboratory test takes also long for pathological analysis.From this two
For individual aspect, interrogation efficiency is greatly diminished, also can not reasonably utilize expert time, effective resource, while
Add the input of internet hospital platform backstage cost of labor.
The content of the invention
It is a primary object of the present invention to disclose a kind of identifying processing method and device of medical image picture, with least
Solves the document in kind that laboratory test report, case history etc. in correlation technique are all based on entity basic hospital, by internet hospital platform
Some other information that staff provides according to the case history picture of upload and basic unit, carry out manual entry and realize electronization, and
It is time-consuming also long and laboratory test report etc. checks picture doctor's progress manual reviews by both party, cause asking for interrogation efficiency reduction
Topic.
A kind of according to an aspect of the invention, there is provided identifying processing method of medical image picture.
Included according to the identifying processing method of the medical image picture of the present invention：Medical image picture is detected, sieved
Select the character area of the medical image picture；Distribution statisticses are carried out to the pixel of the character area, according to data gradient
Multiple key message points are selected in change；External matrix corresponding to the character area is built simultaneously using the multiple key message point
Cut；External matrix after cutting is converted into gray level image；Picture recognition is carried out to the gray level image and is carried automatically
Breath is won the confidence into text box.
Preferably, medical image picture is detected, filtering out the character area of the medical image picture includes：Adopt
Medical image picture is detected with canny edge detection algorithms, filters out the character area of the medical image picture.
Preferably, include to the external matrix after cutting is converted into gray level image：When the external matrix is RGB
During RGB three-dimensional matrices, RGB three-dimensional matrices superposition is converted into a matrix；Two-value is carried out to the matrix after the conversion
Change is handled, wherein, the numerical value in the matrix is contrasted with the threshold value pre-set respectively, when the numerical value is more than or waits
When the threshold value, the numerical value is arranged to 1, when the numerical value is less than the threshold value, the numerical value is arranged to 0.
Preferably, after to the external matrix after cutting is converted into gray level image, in addition to：The gray level image is gone out
Existing noise carries out denoising.
Preferably, picture recognition is carried out to the gray level image and automatically extracts information to text box and includes：To described
Gray level image is identified, and the word that will identify that is matched with the word prestored in medical science word database；Will matching
Obtained medical guidelines extract with analysis data, and automatic input is into the text box, and the doctor that will be extracted
Index and analysis data is learned to store into the corresponding database of user.
Preferably, the medical guidelines extracted and analysis data are stored into the corresponding database of user it
Afterwards, in addition to：Part or all of analysis data of the user in each stage is extracted in database corresponding to the user；Root
Move towards tendency chart according to the partly or entirely analysis data drawing data and analyzed.
According to another aspect of the present invention, there is provided a kind of recognition process unit of medical image picture.
Included according to the recognition process unit of the medical image picture of the present invention：Screening module is detected, for medical science shadow
As picture is detected, the character area of the medical image picture is filtered out；Module is chosen, for the character area
Pixel carries out distribution statisticses, and multiple key message points are selected according to data gradient change；Module is built, for using the multiple
Key message point builds external matrix corresponding to the character area and cut；Modular converter, outside will be after cutting
Matrix conversion is connect into gray level image；Extraction module is identified, for carrying out picture recognition to the gray level image and automatically extracting letter
Breath is into text box.
Preferably, the modular converter includes：Converting unit, for being RGB RGB three-dimensional squares when the external matrix
During battle array, RGB three-dimensional matrices superposition is converted into a matrix；Processing unit, for being carried out to the matrix after the conversion
Binary conversion treatment, wherein, the numerical value in the matrix is contrasted with the threshold value pre-set respectively, when the numerical value is more than
Or during equal to the threshold value, the numerical value is arranged to 1, when the numerical value is less than the threshold value, the numerical value is arranged to 0.
Preferably, the identification extraction module includes：Matching unit, for the gray level image to be identified, it will know
The word not gone out is matched with the word prestored in medical science word database；Typing unit is extracted, for matching to be obtained
Medical guidelines extracted with analysis data, automatic input refers into the text box, and by the medical science extracted
Mark is stored into the corresponding database of user with analysis data.
Preferably, said apparatus also includes：Data extraction module, should for the extraction in database corresponding to the user
Part or all of analysis data of the user in each stage；Analysis module, for being painted according to the partly or entirely analysis data
Data processed are moved towards tendency chart and analyzed.
Compared with prior art, the embodiment of the present invention at least has advantages below：On the basis of OCR identifications, for figure
The different problems of piece, design new picture processing algorithm and pretreatment processing is done to picture, figure is carried out by OCR after pre-processing
Piece identifies, can greatly improve the accuracy of information extraction, and improve interrogation efficiency.
Brief description of the drawings
Fig. 1 is the consultation process figure according to correlation technique；
Fig. 2 is the flow chart of the identifying processing method of medical image picture according to embodiments of the present invention；
Fig. 3 is the schematic diagram of medical image Pictures Electronics according to embodiments of the present invention；
Fig. 4 is the schematic diagram of quick chemical examination achievement data storage according to embodiments of the present invention；
Fig. 5 is the schematic diagram of pathological index trending analysis according to embodiments of the present invention；
Fig. 6 is the flow chart of the identifying processing method of medical image picture according to the preferred embodiment of the invention；
Fig. 7 is the structured flowchart of the recognition process unit of medical image picture according to embodiments of the present invention；And
Fig. 8 is the structured flowchart of the recognition process unit of medical image picture according to the preferred embodiment of the invention.
Embodiment
The specific implementation of the present invention is made a detailed description with reference to Figure of description.
Fig. 2 is the flow chart of the identifying processing method of medical image picture according to embodiments of the present invention.As shown in Fig. 2
The identifying processing method of the medical image picture includes：
Step S201：Medical image picture is detected, filters out the character area of above-mentioned medical image picture；
Step S203：Distribution statisticses are carried out to the pixel of above-mentioned character area, multiple passes are selected according to data gradient change
Key information point；
Step S205：External matrix corresponding to above-mentioned character area is built using above-mentioned multiple key message points and is cut out
Cut；
Step S207：External matrix after cutting is converted into gray level image；
Step S209：Picture recognition is carried out to above-mentioned gray level image and automatically extracts information into text box.
In correlation technique, using picture recognition technology (for example, optical character identification (Optical Character
Recognition, referred to as OCR)) extract the electronic information of medical image picture, but recognition effect is unsatisfactory.It is based on
The picture recognition and information extraction of medical record and laboratory test report, picture are artificial shooting gained, therefore by the very big (angle of subjective impact
Degree, light, definition etc.).Due to these influences, often identification is less than text information in figure, so as to cause extraction to fail.Using
Method shown in Fig. 1, on the basis of OCR identifications, for the different problems of picture, new picture processing algorithm is designed to picture
Pretreatment processing is done, picture recognition is carried out by OCR after pre-processing, the accuracy of information extraction can be greatly improved, and carry
High interrogation efficiency.
Preferably, medical image picture is detected, filtering out the character area of above-mentioned medical image picture can enter
One step includes：Medical image picture is detected using canny edge detection algorithms, filters out above-mentioned medical image picture
Character area.
During being preferable to carry out, picture recognition is carried out, emphasis is not whole photo, but text filed, shown in Fig. 1
User region interested first can be cut out by pretreating scheme in advance, can so substantially reduce the error rate of identification.
Pretreatment process is specific as follows:Canny operator detections are carried out to picture, filter out word segment；Carry out pixel distribution statistics, root
Change according to data gradient, select text filed multiple (for example, 4) key message points；One is built to the key message point of selection
Individual text external matrix interested, is cut.
Preferably, may further include to the external matrix after cutting is converted into gray level image：When above-mentioned external square
When battle array is RGB RGB three-dimensional matrices, the superposition of above-mentioned RGB three-dimensional matrices is converted into a matrix；To the square after above-mentioned conversion
Battle array carries out binary conversion treatment, wherein, the numerical value in above-mentioned matrix is contrasted with the threshold value pre-set respectively, when above-mentioned number
When value is more than or equal to above-mentioned threshold value, the numerical value is arranged to 1, when above-mentioned numerical value is less than above-mentioned threshold value, the numerical value set
For 0.
During being preferable to carry out, the actual superposition for three two-dimensional matrixs of the digital information based on RGB pictures, wherein R
Red is represented, G represents green, and B represents blueness, and its value is all 0-255 sections.And each pixel is tri- kinds of RGB in picture
Color numeral is formed by stacking.For example, white point, it is that RGB is respectively 255,255,255, black 0,0,0 is worth folded
Add.Actual pretreatment is exactly that the digitalized data of picture is handled.Picture format is but different except RGB, also YUV
Form can uniformly change into gray-scale map and be handled, i.e., RGB three-dimensional matrice superposition is converted into a matrix (is worth between 0-
255), without color, then unify to carry out binaryzation again.So-called binaryzation, it is that each pixel is non-black i.e. white, number
It is 0,1 according to value.Certainly, 0-255 numerical value binaryzation, a threshold value is necessarily referred to, line of demarcation is that is to say, herein using basis
The adaptive threshold of picture concrete condition, so energy flexible adaptation are also beneficial to automation and realized in majority of case.
Preferably, after to the external matrix after cutting is converted into gray level image, can also include：To above-mentioned gray-scale map
As the noise occurred carries out denoising.
During being preferable to carry out, necessarily occurs noise after binary conversion treatment, now increasing a denoising can
To obtain more clearly pre-processing picture.
Preferably, picture recognition is carried out to above-mentioned gray level image and automatically extracts information can further wrap into text box
Include：Above-mentioned gray level image is identified, the word that will identify that and the word progress prestored in medical science word database
Match somebody with somebody；The medical guidelines that matching obtains are extracted with analysis data, automatic input will extract into above-mentioned text box
Above-mentioned medical guidelines and analysis data store into the corresponding database of user.
, can be to substantial amounts of electronics on the basis of solving the problems, such as picture recognition accuracy during being preferable to carry out
Picture is identified and information extraction, and the information that will be extracted, and directly applies to the typing of backstage case history and result of laboratory test etc.
System, the operating efficiency of internet hospital platform is drastically increased, and then can also reduce human cost.For example, uploading pictures
Afterwards, click recognition button, the relevant information of needs can be extracted in Fig. 3 in the text box on right side automatically, directly preserved,
Which greatly enhances efficiency.Similar cases similarly operate, and manually triggering identification, automatic identification be not more efficient.
On the basis of picture recognition and electronization, the data of Pictures Electronics are obtained, this is extracted for Back ground Information
It has been met that, but for the human body indicators data of laboratory test report, simply scattered individual data, can not intelligent storage for subsequently making
With.For example, there is " cholesterol " analysis data in laboratory test report, but what is identified is only three words, " courage ", " Gu ", " alcohol ",
For algorithm, program can not know that this is the global index information for needing to extract, if presetting this word and matching
It can obtain, but efficiency is too low, it is flux matched greatly systematic function to be caused not catch up with.So segmentation methods are introduced to carry out laboratory test report
The extraction of middle achievement data, for example, above-mentioned gray level image is identified, the word " courage " that will identify that, " Gu ", " alcohol " and doctor
The word " cholesterol " prestored in word database is learned to be matched；Obtained medical guidelines and analysis data extraction will be matched
Out, automatic input is stored to user couple into above-mentioned text box, and by the above-mentioned medical guidelines extracted and analysis data
In the database answered.Index name, also analysis data are not only extracted, in addition to rapid extraction goes out related data electronization, can also
Store data on platform in the database of the user, this can provide help well for follow-up interrogation analysis.Such as Fig. 4
It is shown, using quick identification extracting method provided by the invention, after being extracted based on participle, extracted the laboratory indexes that need and
Concrete numerical value, this be based on identification after significant enhancement, actually can apply to every Internet service.
Preferably, the above-mentioned medical guidelines extracted and analysis data are stored into the corresponding database of user it
Afterwards, can also include：Part or all of chemical examination number of the user in each stage is extracted in database corresponding to above-mentioned user
According to；Move towards tendency chart according to above-mentioned partly or entirely analysis data drawing data and analyzed.
During being preferable to carry out, obtain the data such as the inspection laboratory test report of user refers specifically to scale value, and this is doctor to disease
The important evidence of people's diagnosis.No matter entity hospital or internet hospital, service pattern is essentially identical, by course of therapy, according to
Therapeutic effect and chemical examination forms data instruct follow-up therapeutic scheme., can basis after the laboratory indexes data of user are extracted
The data of the different phase of each user do a data analysis and drawing data moves towards tendency chart, so as to doctor's quicklook
The variation tendency of user's body pathology index of correlation over the course for the treatment of is seen on ground, provides facility for interrogation, improves interrogation efficiency.
As shown in figure 5, after rapid extraction to the multiple diagnosis and treatment chemical examination electronic data of user, one can be done to the body index parameter of patient
The analysis of individual intuitively trending, scheme, medicine adjustment that each diagnosis and treatment use can also be refined as, can allow doctor very intuitively from
Diagnosis and treatment scheme, medication etc. lead the obstructed effect brought with patient, intelligent analysis interrogation to see corresponding to.
Above-mentioned preferred embodiment is further described below in conjunction with Fig. 6.
Fig. 6 is the flow chart of the identifying processing method of medical image picture according to the preferred embodiment of the invention.Such as Fig. 6 institutes
Show, the identifying processing method of the medical image picture includes：
Step S601：Receive the relevant medical image picture that patient uploads.
Step S603：Above-mentioned medical image picture is pre-processed.
For example, being detected to medical image picture, the character area of above-mentioned medical image picture is filtered out；To above-mentioned text
The pixel in block domain carries out distribution statisticses, and multiple key message points are selected according to data gradient change；Using above-mentioned multiple keys
Information point builds external matrix corresponding to above-mentioned character area and cut；External matrix after cutting is converted into gray-scale map
Picture.
Step S605：Picture recognition is carried out to above-mentioned gray level image using OCR picture recognitions technology.
Step S607：Information is extracted into text box using segmentation methods.
Step S609：Assistance platform is filed.
For example, by the index name of extraction, also analysis data, in addition to rapid extraction goes out related data electronization, can also
Store data on platform in the database of the user, this can provide help well for follow-up interrogation analysis.
Step S611：Doctor administers according to the course for the treatment of to patient.
Step S613：Receive the new medical image picture of correlation that patient uploads.
Step S615：Above-mentioned medical image picture is pre-processed, using OCR picture recognitions technology to above-mentioned gray-scale map
As carrying out picture recognition, information is extracted into text box using segmentation methods.
Step S617：Trending analysis is carried out to pathology.
Step S619：Therapeutic effect is judged whether to meet to require, if not, step S621 is performed, if it is, flow knot
Beam.
Step S621：Doctor is notified to adjust therapeutic scheme.
On the basis of OCR identifications, for the different problems of picture, design new picture processing algorithm and picture is carried out in advance
Processing operation, carries out picture recognition by OCR after pre-processing, can greatly improve the accuracy of information extraction.It was verified that
Can solve the problems, such as to identify that mistake and information extraction ability are low more than 93% picture region under application scenarios, be whole Intelligent Recognition
System provides solid foundation.
Fig. 7 is the structured flowchart of the recognition process unit of medical image picture according to embodiments of the present invention.Such as Fig. 7 institutes
Show, the recognition process unit of the medical image picture includes：Screening module 70 is detected, for being examined to medical image picture
Survey, filter out the character area of above-mentioned medical image picture；Module 72 is chosen, for dividing the pixel of above-mentioned character area
Cloth is counted, and multiple key message points are selected according to data gradient change；Module 74 is built, for using above-mentioned multiple key messages
Point builds external matrix corresponding to above-mentioned character area and cut；Modular converter 76, for by the external matrix after cutting
It is converted into gray level image；Extraction module 78 is identified, for carrying out picture recognition to above-mentioned gray level image and automatically extracting information extremely
In text box.
Using the device shown in Fig. 7, on the basis of OCR identifications, for the different problems of picture, design at new picture
Adjustment method does pretreatment processing to picture, carries out picture recognition by OCR after pre-processing, can greatly improve information extraction
Accuracy.
Preferably, as shown in figure 8, above-mentioned modular converter 76 may further include：Converting unit 760, for when above-mentioned
When external matrix is RGB RGB three-dimensional matrices, the superposition of above-mentioned RGB three-dimensional matrices is converted into a matrix；Processing unit
762, for carrying out binary conversion treatment to the matrix after above-mentioned conversion, wherein, by the numerical value in above-mentioned matrix respectively with setting in advance
The threshold value put is contrasted, and when above-mentioned numerical value is more than or equal to above-mentioned threshold value, the numerical value is arranged into 1, when above-mentioned numerical value is small
When above-mentioned threshold value, the numerical value is arranged to 0.
Preferably, as shown in figure 8, above-mentioned identification extraction module 78 may further include：Matching unit 780, for pair
Above-mentioned gray level image is identified, and the word that will identify that is matched with the word prestored in medical science word database；Carry
Typing unit 782 is taken, is extracted for obtained medical guidelines will to be matched with analysis data, automatic input to above-mentioned text box
In, and the above-mentioned medical guidelines extracted and analysis data are stored into the corresponding database of user.
Preferably, as shown in figure 8, said apparatus can also include：Data extraction module 80, for corresponding in above-mentioned user
Database in extract part or all of analysis data of the user in each stage；Analysis module 82, for according to above-mentioned portion
Divide or whole analysis data drawing datas move towards tendency chart and analyzed.
In summary, by above-described embodiment provided by the invention, case history picture and laboratory test report picture to upload are carried out
Pretreatment, the quick useful information identified in picture, Pictures Electronics, and the result of laboratory test that can be directed to laboratory test report carry out single
Solely extraction, and trend assistant analysis is carried out for course for the treatment of data several times, analyzed for doctor and bring faster more intuitively reference,
The medical efficiency of lifting.
Disclosed above is only several specific embodiments of the present invention, and still, the present invention is not limited to this, any ability
What the technical staff in domain can think change should all fall into protection scope of the present invention.
Claims (10)
- A kind of 1. identifying processing method of medical image picture, it is characterised in that including：Medical image picture is detected, filters out the character area of the medical image picture；Distribution statisticses are carried out to the pixel of the character area, multiple key message points are selected according to data gradient change；External matrix corresponding to the character area is built using the multiple key message point and cut；External matrix after cutting is converted into gray level image；Picture recognition is carried out to the gray level image and automatically extracts information into text box.
- 2. according to the method for claim 1, it is characterised in that medical image picture is detected, filters out the doctor Learning the character area of image picture includes：Medical image picture is detected using canny edge detection algorithms, filters out institute State the character area of medical image picture.
- 3. according to the method for claim 1, it is characterised in that to the external matrix after cutting is converted into gray level image bag Include：When the external matrix is RGB RGB three-dimensional matrices, RGB three-dimensional matrices superposition is converted into a matrix；Binary conversion treatment is carried out to the matrix after the conversion, wherein, by numerical value in the matrix respectively with pre-setting Threshold value is contrasted, and when the numerical value is more than or equal to the threshold value, the numerical value is arranged into 1, when the numerical value is less than institute When stating threshold value, the numerical value is arranged to 0.
- 4. according to the method for claim 1, it is characterised in that to by the external matrix after cutting be converted into gray level image it Afterwards, in addition to：Denoising is carried out to the noise that the gray level image occurs.
- 5. according to the method for claim 1, it is characterised in that picture recognition is carried out to the gray level image and is automatically extracted Information to text box includes：The gray level image is identified, the word that will identify that and the word progress prestored in medical science word database Match somebody with somebody；The medical guidelines that matching obtains are extracted with analysis data, automatic input will extract into the text box The medical guidelines and the analysis data come is stored into the corresponding database of user.
- 6. according to the method for claim 5, it is characterised in that deposit the medical guidelines and the analysis data that extract After storage is into database corresponding to user, in addition to：Part or all of analysis data of the user in each stage is extracted in database corresponding to the user；Move towards tendency chart according to the partly or entirely analysis data drawing data and analyzed.
- A kind of 7. recognition process unit of medical image picture, it is characterised in that including：Screening module is detected, for being detected to medical image picture, filters out the character area of the medical image picture；Module is chosen, for carrying out distribution statisticses to the pixel of the character area, multiple passes are selected according to data gradient change Key information point；Module is built, for building external matrix corresponding to the character area using the multiple key message point and being cut out Cut；Modular converter, for the external matrix after cutting to be converted into gray level image；Extraction module is identified, for carrying out picture recognition to the gray level image and automatically extracting information into text box.
- 8. device according to claim 7, it is characterised in that the modular converter includes：Converting unit, for when the external matrix is RGB RGB three-dimensional matrices, RGB three-dimensional matrices superposition to be turned Change a matrix into；Processing unit, for carrying out binary conversion treatment to the matrix after the conversion, wherein, the numerical value in the matrix is distinguished Contrasted with the threshold value pre-set, when the numerical value is more than or equal to the threshold value, the numerical value is arranged to 1, works as institute When stating numerical value and being less than the threshold value, the numerical value is arranged to 0.
- 9. device according to claim 7, it is characterised in that the identification extraction module includes：Matching unit, for the gray level image to be identified, the word that will identify that with depositing in advance in medical science word database The word of storage is matched；Typing unit is extracted, is extracted for obtained medical guidelines will to be matched with analysis data, automatic input to the text In this frame, and the medical guidelines extracted and analysis data are stored into the corresponding database of user.
- 10. device according to claim 9, it is characterised in that also include：Data extraction module, it is part or all of in each stage for extracting the user in database corresponding to the user Analysis data；Analysis module, for moving towards tendency chart according to the partly or entirely analysis data drawing data and being analyzed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710547048.8A CN107516004A (en) | 2017-07-06 | 2017-07-06 | The identifying processing method and device of medical image picture |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710547048.8A CN107516004A (en) | 2017-07-06 | 2017-07-06 | The identifying processing method and device of medical image picture |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107516004A true CN107516004A (en) | 2017-12-26 |
Family
ID=60721708
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710547048.8A Withdrawn CN107516004A (en) | 2017-07-06 | 2017-07-06 | The identifying processing method and device of medical image picture |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107516004A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109360450A (en) * | 2018-12-05 | 2019-02-19 | 福州大学 | A kind of vessel traffic service method and system of wisdom navigation mark |
| CN110348346A (en) * | 2019-06-28 | 2019-10-18 | 苏宁云计算有限公司 | A kind of bill classification recognition methods and system |
| CN110957016A (en) * | 2019-11-21 | 2020-04-03 | 山东鲁能软件技术有限公司 | Physical examination data intelligent recognition system and method based on health cloud management platform |
| CN111913930A (en) * | 2019-05-10 | 2020-11-10 | 上海中晶科技有限公司 | Species data analysis method, system and computer program product |
| CN112149523A (en) * | 2020-09-04 | 2020-12-29 | 开普云信息科技股份有限公司 | Method and device for OCR recognition and picture extraction based on deep learning and co-searching algorithm, electronic equipment and storage medium |
| CN113257409A (en) * | 2021-06-04 | 2021-08-13 | 杭州云呼医疗科技有限公司 | Clinical decision support system based on patient disease symptoms and medical examination reports |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110075901A1 (en) * | 2009-09-30 | 2011-03-31 | Fujifilm Corporation | Medical image processing apparatus and method and computer-readable recording medium |
| CN104268814A (en) * | 2014-10-11 | 2015-01-07 | 北京网秦天下科技有限公司 | Method and device for generating electronic medical record |
| CN104282025A (en) * | 2014-10-17 | 2015-01-14 | 中山大学 | Biomedical image feature extraction method |
| CN104408449A (en) * | 2014-10-27 | 2015-03-11 | 西安电子科技大学宁波信息技术研究院 | Intelligent mobile terminal scene character processing method |
| CN104835098A (en) * | 2015-05-15 | 2015-08-12 | 上海翼依信息技术有限公司 | Medical record electronic data identification method and system thereof |
| CN106355177A (en) * | 2016-08-17 | 2017-01-25 | 天津德闻数据科技有限公司 | Method for automatically extracting item information from medical examination reports |
| CN106383988A (en) * | 2016-08-30 | 2017-02-08 | 杭州筹图科技有限公司 | Generation method and device of medical examination report |
| CN106446881A (en) * | 2016-07-29 | 2017-02-22 | 北京交通大学 | Method for extracting lab test result from medical lab sheet image |
| CN106845480A (en) * | 2017-01-13 | 2017-06-13 | 河海大学 | A kind of method that car plate is recognized from picture |
| CN106874823A (en) * | 2015-12-10 | 2017-06-20 | 中国移动通信集团公司 | A kind of information identifying method and device |
-
2017
- 2017-07-06 CN CN201710547048.8A patent/CN107516004A/en not_active Withdrawn
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110075901A1 (en) * | 2009-09-30 | 2011-03-31 | Fujifilm Corporation | Medical image processing apparatus and method and computer-readable recording medium |
| CN104268814A (en) * | 2014-10-11 | 2015-01-07 | 北京网秦天下科技有限公司 | Method and device for generating electronic medical record |
| CN104282025A (en) * | 2014-10-17 | 2015-01-14 | 中山大学 | Biomedical image feature extraction method |
| CN104408449A (en) * | 2014-10-27 | 2015-03-11 | 西安电子科技大学宁波信息技术研究院 | Intelligent mobile terminal scene character processing method |
| CN104835098A (en) * | 2015-05-15 | 2015-08-12 | 上海翼依信息技术有限公司 | Medical record electronic data identification method and system thereof |
| CN106874823A (en) * | 2015-12-10 | 2017-06-20 | 中国移动通信集团公司 | A kind of information identifying method and device |
| CN106446881A (en) * | 2016-07-29 | 2017-02-22 | 北京交通大学 | Method for extracting lab test result from medical lab sheet image |
| CN106355177A (en) * | 2016-08-17 | 2017-01-25 | 天津德闻数据科技有限公司 | Method for automatically extracting item information from medical examination reports |
| CN106383988A (en) * | 2016-08-30 | 2017-02-08 | 杭州筹图科技有限公司 | Generation method and device of medical examination report |
| CN106845480A (en) * | 2017-01-13 | 2017-06-13 | 河海大学 | A kind of method that car plate is recognized from picture |
Non-Patent Citations (2)
| Title |
|---|
| 季丽琴: "彩色图像内文字的自动提取与去除的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 梅亚敏: "融合先验知识的场景文本识别应用研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109360450A (en) * | 2018-12-05 | 2019-02-19 | 福州大学 | A kind of vessel traffic service method and system of wisdom navigation mark |
| CN111913930A (en) * | 2019-05-10 | 2020-11-10 | 上海中晶科技有限公司 | Species data analysis method, system and computer program product |
| CN110348346A (en) * | 2019-06-28 | 2019-10-18 | 苏宁云计算有限公司 | A kind of bill classification recognition methods and system |
| CN110957016A (en) * | 2019-11-21 | 2020-04-03 | 山东鲁能软件技术有限公司 | Physical examination data intelligent recognition system and method based on health cloud management platform |
| CN110957016B (en) * | 2019-11-21 | 2023-08-08 | 山东鲁能软件技术有限公司 | Physical examination data intelligent identification system and method based on health cloud management platform |
| CN112149523A (en) * | 2020-09-04 | 2020-12-29 | 开普云信息科技股份有限公司 | Method and device for OCR recognition and picture extraction based on deep learning and co-searching algorithm, electronic equipment and storage medium |
| CN113257409A (en) * | 2021-06-04 | 2021-08-13 | 杭州云呼医疗科技有限公司 | Clinical decision support system based on patient disease symptoms and medical examination reports |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107516004A (en) | The identifying processing method and device of medical image picture | |
| CN110909780B (en) | Image recognition model training and image recognition method, device and system | |
| CN109871735B (en) | Image analysis method and device and manufacturing method for learning deep learning algorithm | |
| CN110647875B (en) | Method for segmenting and identifying model structure of blood cells and blood cell identification method | |
| CN108830149B (en) | Target bacterium detection method and terminal equipment | |
| US10140713B1 (en) | Morphology identification in tissue samples based on comparison to named feature vectors | |
| CN113076987B (en) | Osteophyte identification method, device, electronic equipment and storage medium | |
| CN110766656B (en) | Method, device, equipment and storage medium for screening fundus macular region abnormality | |
| CN113223041B (en) | Method, system and storage medium for automatically extracting target area in image | |
| EP3564857A1 (en) | Pattern recognition method of autoantibody immunofluorescence image | |
| CN109948429A (en) | Image analysis method, device, electronic equipment and computer-readable medium | |
| CN113450305B (en) | Medical image processing method, system, equipment and readable storage medium | |
| CN111882559B (en) | ECG signal acquisition method and device, storage medium and electronic device | |
| Jayalath et al. | Identification of medicinal plants by visual characteristics of leaves and flowers | |
| CN115661580A (en) | Convolutional neural network-based traditional Chinese medicine decoction piece image identification method and system | |
| CN110766665A (en) | Tongue picture data analysis method based on strong supervision algorithm and deep learning network | |
| CN116109818A (en) | Traditional Chinese medicine pulse condition distinguishing system, method and device based on facial video | |
| CN113237881B (en) | Detection method and device for specific cells and pathological section detection system | |
| CN108985302A (en) | A kind of skin lens image processing method, device and equipment | |
| KR100915922B1 (en) | Methods and System for Extracting Facial Features and Verifying Sasang Constitution through Image Recognition | |
| CN114120035A (en) | Medical image recognition training method | |
| Bindhu et al. | Analytical study on digital image processing applications | |
| CN110942493A (en) | Method for automatically generating scanning area in CT system | |
| P Mathai | Sasyasneha–An Approach for Plant Leaf Disease Detection | |
| CN117237351B (en) | Ultrasonic image analysis method and related device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20171226 |