CN1845096A - Establishment of multi-dimensional enhanced cross-marked medical information database and search method thereof - Google Patents
Establishment of multi-dimensional enhanced cross-marked medical information database and search method thereof Download PDFInfo
- Publication number
- CN1845096A CN1845096A CNA2006100342644A CN200610034264A CN1845096A CN 1845096 A CN1845096 A CN 1845096A CN A2006100342644 A CNA2006100342644 A CN A2006100342644A CN 200610034264 A CN200610034264 A CN 200610034264A CN 1845096 A CN1845096 A CN 1845096A
- Authority
- CN
- China
- Prior art keywords
- image
- medical information
- information
- medical
- information database
- 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
Images
Landscapes
- Medical Treatment And Welfare Office Work (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention relates to a method for building and searching a multidimensional strengthened intercross mark medical information database. Wherein, it describes a method for building the organic relationship between known medical information of present of HIS, RIS, and PACS systems, to build a new medical information database with strong inquire method, with which user can easily search the medical information stored in said systems; the known information of HIS or RIS system is the criterion of treating and inquiring. The inventive method also can be used in the information manage system of image, work, video, etc.
Description
Technical field:
The present invention relates to medicine technology field, is a kind of establishment and searching method thereof of multi-dimensional enhanced cross-marked medical information database specifically.
Background technology:
1. hospital information management system (Hospital Information System, HIS), dept. of radiology's information management system (Radiology Information System, RIS) and medical image management and communication system (Picture Archive and Communication Systems PACS) has become the requisite ingredient of modern digital hospital.All there is every day increasing patient cases to pour into various dissimilar medical information management systems.Each system all has its approaches to IM at present, and the exchange of the information between system is subjected to a lot of restrictions.For example, to from HIS or RIS, go to consult and be stored in medical image information huge in the PACS system, the extremely limited inquiry means that the user must use this system to provide such as the image that only is certain inspection of the known patient of reading, thereby are subjected to a lot of restrictions.The invention describes and a kind ofly can go to set up the method and the process of the organic connections of the medical information between them from the known information of existing HIS, RIS and PACS system, thereby set up a medical information database, thereby make the user can search for the medical information that stores in these systems easily with a new generation of strong inquiry means.Here the Given information of HIS or RIS system with processed and as the inquiry criterion.
In addition, the various features that the present invention has also described with medical image/image is the management and the inquiry means in the medical image data storehouse of foundation, and it provides the instrument of new search relevant medical information for the user.
All these methods all are based upon on the basis of construction method of multi-dimensional enhanced cross-marked medical information database, extract the feature of all medical informations in this database (the various information that comprise image, image and literal), and as the criterion of inquiry, thereby a kind of brand-new medical information management and the method and system of search are provided for the user.Because versatility of the present invention, same thinking also can be applied in other information management system relevant with image, image and literal with method.
Summary of the invention:
Accompanying drawings summary of the invention of the present invention describes from following several respects, as shown in the figure:
Fig. 1 is the identification process of medical information system of new generation and sets up the process sketch.
Fig. 2 is the diagrammatic representation sketch with data structure of multidimensional sign.
Fig. 3 is the workflow sketch of data search.
1. strengthen the PACS of intersection identification type
By patient/case/series/image (Patient/Study/Series/Image, PSSI) form substantially by four-layer structure for the data of traditional PACS.If the user is to find their conceivable medical information very easily by numbering (ID) Query Information of above structure and known PSSI.But under most situation, the result is really not so.For most medical personnels and part patient, condition and criterion that they inquire about data often are not these ID, but following these information: such as, information such as patient name, check data, diagnostic code, anatomical position, even further how much of medical image/image to be diagnosed on hand, pathology, lesion information and image/image feature information to them.The feature of the inquiry that some of them are the most frequently used does not usually have definite quantitative and definition, but the notion of some inquiries (Fussy Search) of bluring relatively for example, is searched close with the position of this piece focus generation, or size is similar, close case that density is similar or the like.This information management system of will seeking medical advice has more strong inquiry means, that is to say and carries out more effective sign and tissue.Up to the present, existing part medical information management system has increased some identification means and technology, and it belongs to words identification all, and the neither one system can provide above-described fuzzy query function based on medical image/image information.In this section, we will describe function and the composition of this novel PACS in detail.
Strengthen identification technology 1.1. intersect automatically
Automatically advanced (PUSH) behind the DICOM receiver of PACS a new DICOM case, these image/images will be checked by Automatic Program, and give its unique system numbering (ID).After this, the program of particular design will be carried out following operation among the present invention:
1. read the up-to-date patient/fox message relevant from HIS and RIS with this case.
2. extract all essential information from the DICOM file header.
3. analyze above information, carry out the sign of case (inspection) level, include but not limited to: date, patient information, information about doctor, fox message or the like.
4. carry out the sign of serial level, include but not limited to: train type, image/image type, image space, imaging orientation and image/picture contrast or the like.
5. carry out image/image pre-service, include but not limited to: filtration of the minor structure formation of the elimination of imaging background noise, imaging displacement/rotation difference correction, image/image and adjustment, image/image or the like based on series layer time sign.
6. carry out extraction based on the relevant image/characteristics of image of image/image level
7. to the automatic registration (Auto-registration) of this case/series/image/image.
After above step was finished, all image/images of this case all will be by this novel PACS system registry, and can have access in its online storage area.The identification information at all levels that is produced by Automatic Program will be deposited in the form of corresponding database standby.
1.2. intersecting, semi-automatic image strengthens identification technology
In some cases, program can't realize robotization fully, needs the intervention of artificial or artificial intelligence, and for example, the project in some necessary DICOM file header (Header) is not received in or the like when scanning imagery.In addition, the sign of accurate three-dimensional (3D) anatomical position also can only be realized under the help of manual intervention at present.The operation of this section is carried out before or after can saving (1.3) at next.Modal manual intervention operation comprises following:
1. the project of all losses in the DICOM file header will be checked and identify to the Automatic Logos program, need manual intervention to replenish the project of these losses when program can't be handled.
2. increase some essential signs, such as, mark some special functional key word and explanation (Comments).
3. manually increase the sign of some anatomical positions, will work as two dimension (2D) image of pre-treatment and the two dimension (2D) of system's use or the virtual manikin of three-dimensional (3D) and link together, show thereby set up three-dimensional accurately (3D) image.
1.3. intersecting, semi-automatic diagnosis strengthens identification technology
Medical diagnosis strengthens sign should be by having the to a certain degree personage of professional medical knowledge, such as, radiologist, radiographer or other medical professions operate to guarantee its accuracy.Relevant medical diagnosis report and the following content will be added in the database:
1. medical science standard diagnostics code is to strengthen the speed and the accuracy of information inquiry.
2. some keywords in the medical diagnosis report are made super be connected (Hyper-link) with phrase and relative single or multiple medical images/image.
3. (Region Of Interesting ROI) indicates, and gives corresponding explanation and diagnostic code with relevant area-of-interest.
4. with some keywords in the medical diagnosis report and phrase and its corresponding single or multiple area-of-interest (Region Of Interesting, ROI) super connect (Hyper-link) of work.
2. the medical science case of multidimensional sign
2.1. have the diagrammatic representation of the data structure of multidimensional sign
In the data structure of the PACS with intersection enhancing sign, not only indicated the numbering (ID) of traditional PSSI, and more most other useful information and parameter are handled and identified, thereby increased the query capability of user information in the system.That the user can obtain will be more, means are gone to inquire about them and wanted information inquiring more easily.Fig. 2 is the diagrammatic representation with data structure of multidimensional sign, intersection enhancing sign.
2.2. based on the PACS of PSSI with based on the HIS/RIS of PS
In traditional PACS, its basic structure is by four layers of formation of PSSI (Patient/Study/Series/Image).Each layer all has the unique number (ID) of oneself.And each DICOM file all has an absolute unique numbering (SOP-ID).So-called unique number among four layers of the PACS just is only unique at this level and following level usually.The main designation system of fundamental sum among Here it is PACS and the HIS/RIS, no matter its all information (comprising literal and image/image information) is online or off-line, all with this denotation approach tissue, management and inquiry.Native system also has this function.
2.3. date information
Here have a lot of very useful date information to can be used as the criterion of sign and inquiry: such as (but being not limited to), patient enrollment time, inspection requirements time, supervision time, case filling time, case are checked time, admission date/time, the date of leaving hospital and some other date commonly used or the like.The information major part of this group mark all can be extracted from the DICOM file header, or obtains from the database of HIS/RIS.
2.4. patient's personal information
Most of patient's personal information sign commonly used includes but not limited to: name, age, sex, race, body weight, height or the like.The information major part of this group mark also all can be extracted from the DICOM file header, or obtains from the database of HIS/RIS.
2.5. doctor/information for hospital
The Information sign relevant with doctor/hospital includes but not limited to: hospital, section office, imaging device and device number thereof, the doctor that changes the place of examination, diagnostician, operation technician or the like.Equally, the information major part of this group mark also all can be extracted from the DICOM file header, or obtains from the database of HIS/RIS.
2.6. clinical medicine information
This group is one of most important flag information group that is used for literal and image/image information inquiry.The Information sign relevant with clinical medicine includes but not limited to: diagnostic code, medical diagnosis report, super COMMENT that connects (Hyper-link), keyword, crucial image/image and doctor or the like.Part in the information of this group mark can obtain from the database of HIS/RIS, but needs the work of part post-processed.
2.7. anatomical position information
This group is the important sign information that is used for 3-dimensional image/image information reconstruct and inquiry.The Information sign relevant with anatomical position includes but not limited to: the anatomical position registration of the serial image/image registrations of three-dimensional or 2.5 dimensions and identification, crucial image/image and identification, the registration of ROI anatomical position and identification, image/image anatomic region or the registration of organ attribute and identification or the like.Post-processed/reconstruction that the informational needs of this group mark is a large amount of.
2.8. the information that image/image is relevant
This group also is one of most important flag information group that is used for literal and image/image information inquiry.The Information sign relevant with medical image/image comprises physics and how much two class parameters.Physical token information includes, but are not limited to: one-tenth phase/scan mode (MR, CT, ultrasound wave or the like), scanning device model, image/image type, image/image reconstruction mode, image contrast and data extract time parameter (as TR, TE, the TI of MR, the KeV of CT etc.), or the like.Geometric landmarks information includes, but are not limited to: the geometric parameter of visual field (Field of View), picture element matrix, the resolution of pixel (Pixel/Voxel), scanning tomography thickness (Slice Thickness), volume coordinate orientation (Orientation) or the like.Part in the information of this group mark can obtain from the file header of DICOM, but needs the work of part post-processed.
2.9. image/image and area-of-interest (ROI) characteristic information
Because the visual field (Field of View) of image and the difference of resolution, it relatively is unpractical using the image based on the image pixel feature simply.The cardinal index of image through rotating revised deformable mold block eigenvalue and then can be used as image comparison and search is stored in the database.The type relevant with region of interest (ROI), anatomical position, size (volume), shape, contrast, absolute and relative signal intensity, signal intensity standard error also can be used as the criterion of contrast and search.
3. medical information search strategy
For a data management system that has been organized and has had fine sign, its information search itself is not the thing of a difficulty.For above-described medical information database, can adopt following three kinds of information search strategies.The user can adopt wherein any mode, or adopts their combination in any to carry out the medical information inquiry, thereby reaches best Query Result.Native system adopts following several ways of search and combination thereof:
3.1. search based on Word message
This is the general the most frequently used searching method of search engine.Comprise following but be not limited only to:
1. the search of key word, keyword sets or their combination: as patient name, disease name or the like.
2. the search of coding/code or their combination: as diagnostic code, sex, race or the like.
3. the search of quantized value or its scope: as age, date or the like.
3.2. search based on the human dissection positional information
This is the very useful method of medical information search field.Comprise following but be not limited only to:
1. the search of human organ classification.
2. the text search in anatomical position zone.
3. the space search of anatomical position.
3.3. search based on image/image feature information
This method is also referred to as the search based on image/image information.Comprise following but be not limited only to:
1. physics/set/the pathological parameter to close anatomical position and image/image minor structure compares and similar image/image search.
2. physics/set/the pathological parameter to close area-of-interest (ROI) and image/image minor structure compares and similar image/image search.
In sum, the present invention is summarized as:
The present invention relates to medicine technology field, is a kind of establishment and searching method thereof of multi-dimensional enhanced cross-marked medical information database specifically.The invention describes and a kind ofly can go to set up the method and the process of the organic connections of the medical information between them from the known information of existing HIS, RIS and PACS system, thereby set up a medical information database, thereby make the user can search for the medical information that stores in these systems easily with a new generation of strong inquiry means.Here the Given information of HIS or RIS system with processed and as the inquiry criterion.Because versatility of the present invention, same thinking also can be applied in other information management system relevant with image, image and literal with method.
A kind of establishment of multi-dimensional enhanced cross-marked medical information database and searching method thereof, described a kind of can be from existing HIS, the known information of RIS and PACS system, the method of describing in the native system in addition, go to set up the method and the process of the organic connections of the medical information between them, thereby set up a medical information database with a new generation of strong inquiry means, thereby make the user can search for the medical information that stores in these systems easily, because versatility of the present invention, same thinking and method also can be applied in other and image, in the image information management system relevant with literal; The various features that the invention describes with medical image/image is the management and the inquiry means in the medical image data storehouse of foundation, it provides the instrument of new search relevant medical information for the user, all these methods all are based upon on the basis of construction method of multi-dimensional enhanced cross-marked medical information database, extract all medical informations in this database, the feature that comprises the various information of image, image and literal, and as the criterion of inquiring about; Generation and the identification method of the PACS of the described enhancing intersection of first segment identification type have increased unexistent processing of conventional P ACS and sign, thereby have improved the search capability of system greatly; The establishment of described multi-dimensional enhanced cross-marked medical information database and searching method thereof, nine kinds of data structures of this novel PACS system multidimensional sign; The establishment of described multi-dimensional enhanced cross-marked medical information database and searching method thereof, three kinds of information query methods of this novel PACS system.
Claims (5)
1, a kind of establishment of multi-dimensional enhanced cross-marked medical information database and searching method thereof, it is characterized in that: the invention describes a kind of can be from existing HIS, the known information of RIS and PACS system, the method of describing in the native system in addition, go to set up the method and the process of the organic connections of the medical information between them, thereby set up a medical information database with a new generation of strong inquiry means, thereby make the user can search for the medical information that stores in these systems easily, because versatility of the present invention, same thinking and method also can be applied in other and image, in the image information management system relevant with literal.
2, the establishment of multi-dimensional enhanced cross-marked medical information database according to claim 1 and searching method thereof, it is characterized in that: the various features that the invention describes with medical image/image is the management and the inquiry means in the medical image data storehouse of foundation, it provides the instrument of new search relevant medical information for the user, all these methods all are based upon on the basis of construction method of multi-dimensional enhanced cross-marked medical information database, extract all medical informations in this database, comprise image, the feature of the various information of image and literal, and as the criterion of inquiring about.
3, the establishment of multi-dimensional enhanced cross-marked medical information database according to claim 1 and searching method thereof, generation and the identification method of the PACS of the described enhancing intersection of first segment identification type, it is characterized in that: increased unexistent processing of conventional P ACS and sign, thereby improved the search capability of system greatly.
4, the establishment of multi-dimensional enhanced cross-marked medical information database according to claim 1 and searching method thereof is characterized in that: nine kinds of data structures of this novel PACS system multidimensional sign.
5, the establishment of multi-dimensional enhanced cross-marked medical information database according to claim 1 and searching method thereof is characterized in that: three kinds of information query methods of this novel PACS system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2006100342644A CN1845096A (en) | 2006-03-15 | 2006-03-15 | Establishment of multi-dimensional enhanced cross-marked medical information database and search method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2006100342644A CN1845096A (en) | 2006-03-15 | 2006-03-15 | Establishment of multi-dimensional enhanced cross-marked medical information database and search method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1845096A true CN1845096A (en) | 2006-10-11 |
Family
ID=37064024
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2006100342644A Pending CN1845096A (en) | 2006-03-15 | 2006-03-15 | Establishment of multi-dimensional enhanced cross-marked medical information database and search method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1845096A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103116626A (en) * | 2013-01-31 | 2013-05-22 | 张粤新 | Matching method of medical image |
| CN110291535A (en) * | 2016-11-25 | 2019-09-27 | 康德斯弗洛股份有限公司 | For creating the method and system of medical image databases by convolutional neural networks |
-
2006
- 2006-03-15 CN CNA2006100342644A patent/CN1845096A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103116626A (en) * | 2013-01-31 | 2013-05-22 | 张粤新 | Matching method of medical image |
| CN110291535A (en) * | 2016-11-25 | 2019-09-27 | 康德斯弗洛股份有限公司 | For creating the method and system of medical image databases by convolutional neural networks |
| CN110291535B (en) * | 2016-11-25 | 2023-07-25 | 康德斯弗洛股份有限公司 | Method and system for creating medical image databases through convolutional neural networks |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2017365608B2 (en) | Method and a system for creating a medical image database by means of a convolutional neural network | |
| Gu et al. | Classification of bacterial and viral childhood pneumonia using deep learning in chest radiography | |
| US7489825B2 (en) | Method and apparatus for creating a multi-resolution framework for improving medical imaging workflow | |
| US20090228299A1 (en) | Methods and apparatus for context-sensitive telemedicine | |
| CN107492071A (en) | Medical image processing method and equipment | |
| Lowe et al. | Towards knowledge-based retrieval of medical images. The role of semantic indexing, image content representation and knowledge-based retrieval. | |
| Wu et al. | Development and evaluation of a deep learning algorithm for rib segmentation and fracture detection from multicenter chest CT images | |
| CN101040779A (en) | Method and system for virtual slice positioning in a 3d volume data set | |
| CN1920882A (en) | System and method for salient region feature based 3d multi modality registration of medical images | |
| Kim et al. | A new way for multidimensional medical data management: volume of interest (VOI)-based retrieval of medical images with visual and functional features | |
| CN1881254A (en) | Image processing device and image processing method | |
| AlZu'bi et al. | Transferable HMM probability matrices in multi‐orientation geometric medical volumes segmentation | |
| CN103116626A (en) | Matching method of medical image | |
| Wong et al. | Design methods and architectural issues of integrated medical image data base systems | |
| CN116469545A (en) | Method, device and equipment for carrying out auxiliary disease prediction by using medical image | |
| Rahman et al. | Medical image retrieval and registration: towards computer assisted diagnostic approach | |
| Althenayan et al. | Detection and classification of COVID-19 by radiological imaging modalities using deep learning techniques: a literature review | |
| Wong | Medical image databases | |
| CN1845096A (en) | Establishment of multi-dimensional enhanced cross-marked medical information database and search method thereof | |
| Huang et al. | Image-matching as a medical diagnostic support tool (DST) for brain diseases in children | |
| CN1820274A (en) | Cad (computer-aided decision) support for medical imaging using machine learning to adapt cad process with knowledge collected during routine use of cad system | |
| Li et al. | SOMA: Subject‐, object‐, and modality‐adapted precision atlas approach for automatic anatomy recognition and delineation in medical images | |
| Shi et al. | Deep learning in the diagnosis for cystic lesions of the jaws: a review of recent progress | |
| Kim et al. | Content-based dual-modality biomedical data retrieval using co-aligned functional and anatomical features | |
| CN118537567B (en) | Multi-mode medical image segmentation method and system based on SAM model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| DD01 | Delivery of document by public notice |
Addressee: Zhang Xiaoyue Document name: Notification that Application Deemed to be Withdrawn |