CN101876992B - Method for managing image data warehouse - Google Patents
Method for managing image data warehouse Download PDFInfo
- Publication number
- CN101876992B CN101876992B CN 200910237783 CN200910237783A CN101876992B CN 101876992 B CN101876992 B CN 101876992B CN 200910237783 CN200910237783 CN 200910237783 CN 200910237783 A CN200910237783 A CN 200910237783A CN 101876992 B CN101876992 B CN 101876992B
- Authority
- CN
- China
- Prior art keywords
- view data
- index file
- client
- image
- data
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 72
- 230000008569 process Effects 0.000 claims abstract description 28
- 238000013461 design Methods 0.000 claims description 14
- 230000001360 synchronised effect Effects 0.000 claims description 12
- 238000007726 management method Methods 0.000 claims description 8
- 238000003384 imaging method Methods 0.000 claims description 6
- 230000008520 organization Effects 0.000 claims description 3
- 230000000875 corresponding effect Effects 0.000 claims 5
- 230000015572 biosynthetic process Effects 0.000 claims 1
- 230000002596 correlated effect Effects 0.000 claims 1
- 230000004048 modification Effects 0.000 abstract 1
- 238000012986 modification Methods 0.000 abstract 1
- 238000011160 research Methods 0.000 description 21
- 201000010099 disease Diseases 0.000 description 12
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 12
- 238000010586 diagram Methods 0.000 description 7
- 208000024827 Alzheimer disease Diseases 0.000 description 6
- 238000013523 data management Methods 0.000 description 6
- 238000002474 experimental method Methods 0.000 description 6
- 239000012467 final product Substances 0.000 description 6
- 210000004556 brain Anatomy 0.000 description 5
- 238000013500 data storage Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 208000010877 cognitive disease Diseases 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000002059 diagnostic imaging Methods 0.000 description 3
- 238000013401 experimental design Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000002599 functional magnetic resonance imaging Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 208000027061 mild cognitive impairment Diseases 0.000 description 3
- 206010039966 Senile dementia Diseases 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000003542 behavioural effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000010191 image analysis Methods 0.000 description 2
- 230000007170 pathology Effects 0.000 description 2
- 238000003909 pattern recognition Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 206010003694 Atrophy Diseases 0.000 description 1
- 208000017667 Chronic Disease Diseases 0.000 description 1
- 238000005481 NMR spectroscopy Methods 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000037444 atrophy Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012984 biological imaging Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 210000001320 hippocampus Anatomy 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 230000002045 lasting effect Effects 0.000 description 1
- 238000002595 magnetic resonance imaging Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000002610 neuroimaging Methods 0.000 description 1
- 238000011112 process operation Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 201000000980 schizophrenia Diseases 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a method for managing an image warehouse, which carries out efficient management on the image warehouse by using text-based image data description information. The management process comprises the steps of creating and uploading image data index files, searching keywords and downloading image data; the method comprises image data uploading and downloading; and when a client user uploads the image data to the image warehouse, the user is made to have the unique modification operation authority to the warehouse in a way of locking. The whole uploading operation and the whole downloading operation are performed through a data description file, and the data description file records the detail information of all image data such as the subfield number, the storage path and various attributes. The method has the advantages that: the same image data description files are kept in a client and a server, so that the operation of searching the image data by the user can be performed at the client, the pressure of the network and the server is lightened and the efficiency of warehouse management is fundamentally improved.
Description
Technical field
The invention belongs to technical field of image processing, relate to method for managing image data warehouse.
Background technology
The raising of the development of computing technique and computer hardware performance has promoted the development of the research fields such as pattern-recognition, artificial intelligence, data mining and image processing greatly, so that machine learning can carry out at super large sample training collection, thereby obtain to have more characteristic of division and the recognition result of specificity and susceptibility.
Increase along with the machine learning training sample set, management to the sample set data warehouse also will be vital problem, it will directly affect the efficient of data storage and retrieval extraction, if there is not efficient data warehouse management method, with the huge bottleneck that has been follow-up machine learning research equipment for data warehouse undoubtedly.
According to the theories of vision of Ma Er, the external information 70%-80% that the people obtains is from human visual system (that is to say image), so one of mainly studying a question of pattern-recognition is understanding for picture material.For example in the research of computer face identification problem, often need to store a large amount of different crowds' (kind) in the view data warehouse, in the view data of different angles and illumination condition and different expression; In the research for medical image, for the research of certain institutional framework or certain pathology, also often need different medical imaging patterns (such as computer tomography, magnetic resonance imaging etc.), at different imaging parameters and at the image document of different developmental phases/course of disease.Therefore, for the research contents of image certainly will relate to huge view data warehouse and with the method for managing image data warehouse of data warehouse coupling.
Existing method for managing image data warehouse all lacks efficient data storage and retrieval obtain manner.People's face data warehouse (http://face.nist.gov/) such as American National Standard and technical institute's announcement, this data warehouse is that the mode with data message composing document name realizes Data classification and searches, because the attribute information of view data only is present in the filename, the method has limited the extensibility that the high efficiency of searching and data attribute are described; For another example by American National Biological imaging and bio-engineering research the unit's of grade senile dementia of initiating the neuroimaging project (
Http:// www.adni-info.org/), a large amount of medical images are stored in server end, and retrieving all is to finish by the computer program of server end, and this mode has increased the burden of server undoubtedly, when especially the external reference amount increased, the impact of this retrieval rate was with particularly evident.
Therefore total being summed up, existing method for managing image data warehouse has the following disadvantages at present:
1. can easily arbitrarily not expand for the description scheme of view data, make it can not satisfy scientific research based on view data to the multifarious demand of view data Expressive Features attribute;
2. imaged image data storage and numbering can not reflect the relevance between the view data, and the relation such as between different time points or the resulting view data of different images drainage pattern can not well be embodied;
3. the upload and download process of view data on server all finished at server end, and this design has increased the burden of server end, affected for the uploading of data warehouse, retrieval and speed of download;
4. when in the user need to comform the multilevel image data result for retrieval, selectively downloading partial results, there is not the method for well optionally downloading in batches.
Therefore, for the background demand of Computer Image Processing research, and the deficiency that has at present method for managing image data warehouse, we have proposed a kind of new method for managing image data warehouse.
Summary of the invention
View data warehouse recall precision is not high in the prior art, the storage/retrieval keyword expansion is bad in order to solve, and related indefinite problem between data and the data, the present invention seeks to propose the method for managing image data warehouse that a kind of text based can be expanded, can efficient storage/retrieval/download.
For reaching described purpose, the invention provides the management method in a kind of view data warehouse, to utilize text based view data descriptor, the imaged image data warehouse is carried out rapidly and efficiently management, and this management process comprises the uploading of generation, view data of view data view data description document, the keyword retrieval of view data and the download of view data; The method is divided into view data to be uploaded and the work of image data download two-way, and the step of realization is as follows:
Wherein view data is uploaded and is comprised:
Step su1: client terminal start-up view data upload operation is prepared to upload new image data, and is connected to the server that will operate;
Step su2: utilize lock to the operation that the view data warehouse of operated server locks, operate successfully if lock, so that synchronization only has a client to have the view data warehouse is carried out the authority that alterability operates; If it is unsuccessful to add latching operation, client does not have the authority of the view data warehouse being carried out the alterability operation;
Step su3: client is carried out the alterability operation with the authority of alterability operation to the view data warehouse, download index file from server end first by the client visitor, the time attribute of the index file of client indexes file and server is compared, if the time attribute of comparison is consistent, the view data index file of client is identical with the index file content of server, if the time attribute of comparison is inconsistent, then cover the index file of client with the index file of server end, be used for guaranteeing that all alterability operations are to carry out in up-to-date view data warehouse; The view data index file that forms in client at last;
Step su4: client is inputted first the title of new acquisition of image data, whether client can be retrieved this title and exist in the index file of client, obtain result for retrieval, judge whether relevant property of the view data of new collection and the existing view data of server end by the user of client; If there is not relevance, will directly accumulate being numbered RCCM (x to new acquisition of image data
1x
2x
3x
4... x
n+ 1) _ sMRI; If there is correlativity, for the existing view data that is associated of enumerating in the result for retrieval, the user of client selects the k time collection of this new acquisition of image data according to the title of new acquisition of image data, then after existing numbering, add tested times of collection suffix k, generate the RCCMx that is numbered of new acquisition of image data
1x
2x
3x
4... x
n_ sMRI_k, in the formula, RCCM represents data affiliated unit, x
1x
2x
3x
4... x
nThe expression data with existing is numbered, and wherein n and k are natural number, and sMRI is the abbreviation of image type in the view data, so the different images type of different field is somebody's turn to do the corresponding difference of abbreviation meeting;
Step su5: with the RCCM (x that new acquisition of image data is generated
1x
2x
3x
4... x
n+ 1) _ sMRI or RCCMx
1x
2x
3x
4... x
n_ sMRI_k is numbered the description document beginning of new acquisition of image data, and the generation user retrieves needed keyword and key assignments thereof to the view data warehouse, the antistop list registration represents the property value corresponding to classification of this feature according to a kind of classification of feature, the key assignments corresponding with keyword, the new image data description document of generation is text-only file;
Step su6: the client user puts into new acquisition of image data description document in the file at this new acquisition of image data place, the view data warehouse of then this document folder being uploaded onto the server together and holding;
Step su7: after the server end view data has new image data to upload, server end upgrades the index file in this view data warehouse accordingly according to this user instruction that upgrades the client at place, the renewal process of view data warehouse index file is exactly the description document that there is view data corresponding in the catalogue upgraded and this catalogue in all in time of user's appointment in this view data warehouse of retrieval, the new description document that obtains being updated, extract respectively again the new description document be updated and corresponding deposit the path and compare with the existing index file of this server end, if exist the view data numbering to repeat, then cover the corresponding contents of the existing index file of server end with the new description document that is updated and path thereof; If numbering does not repeat, then the new description document that is updated and path thereof are added in the existing index file of server end, thereby obtain the new index file that is updated in whole view data warehouse, so far, update image data warehouse index file is complete;
Step su8: the lock that arranges among the step su2 is opened, and client discharges the control authority to server end, so that other user has the renewal authority to this server data warehouse, thereby to the release of view data warehouse;
Wherein image data download comprises:
Step sd1: the user of client starts the image data download operation, and is connected to the server that will operate;
Step sd2: the user of client carries out synchronous operation to the view data index file of the view data index file of client and server end, generate up-to-date index file in client, so that the view data index file content of the image index file of client and server end is identical, thereby the retrieving that guarantees the user carries out in up-to-date view data warehouse;
Step sd3: the user is after client has obtained up-to-date index file, and keyword and key assignments that input will be retrieved are searched for, and is met the right view data directory listing of keyword key;
Step sd4: the user of client carries out preview to the image in the view data directory listing, take the result that determines search whether as own expected result;
Step sd5: in the view data directory listing that search obtains, adopt all view data that need of file check box final election, needed view data is carried out batch download.
Wherein, described lock and the operation of release is at view data warehouse entrance the Boolean variable that a called after " locks " to be set, when this Boolean variable was "True", this view data warehouse was temporarily locked and can not be accessed simultaneously by other user; When described alterability EO, view data is uploaded when finishing, and this Boolean variable is set to " vacation ", and then the view data warehouse is in " release " state.
Wherein, the view data index file of described client is identical from the index file content of server to be when having a plurality of different view data warehouse, and the view data index file of client is comprised of the index file in all view data warehouses.
Wherein, described view data description document is the user above-mentioned view data numbering and keyword: the hand-written one-tenth view data of key-value pair description document, or design table (schedule) allows the user fill in a form, and then allows computer programming language become the view data description document.
Wherein, the keyword of described needs represents that the different field acquisition of image data is different to the description of keyword according to the actual requirements, the user is according to keyword: the rules for writing of key assignments expand keyword: the key assignments key-value pair, in order to the descriptive power of expansion for the keyword of the acquisition of image data of newly arriving, thereby generation is for the description document of new acquisition of image data.
Wherein, described synchronous operation is exactly that the view data index file in each view data warehouse is downloaded to client, then generate a unified overall index file according to these index files, this overall index file is that each view data index file is spliced.
Wherein, all view data that need of described file check box final election, when perhaps the user of client needs a plurality of results among all results, the needed result who searches at this moment user's final election carries out disposable download.
Beneficial effect of the present invention:
A, the describing method of relevance between the view data in the view data description document.The view data of different batches collection often have time/and spatial point on relevance, the present invention realizes that in the mode of data being carried out multistage numbering the relevance between the different batches view data describes; The numbering form of the view data that generates is RCCMx
1x
2x
3x
4... x
n_ sMRI_k, in the formula, RCCM represents data affiliated unit, x
1x
2x
3x
4... x
nExpression one-level data number, n wherein, k is natural number, sMRI is the abbreviation of image type in the view data, so the different images type of different field should abbreviation can be corresponding different; If the x of two groups of data numbers
1x
2x
3x
4... x
nField is identical, and then showing has the one-level association between them, if the sMRI field is also identical, shows that then they have secondary related, and last field k is illustrated in the k secondary data collection under the identical situation of aforementioned all categories attribute;
B, view data is by the process of client upload to server end, and the user need not arrange/and recording image data is at the Actual path of server end storage, and this path will be identified by this inventive method in the process that view data warehouse index file generates automatically.After a certain view data warehouse was updated, this inventive method can the corresponding index file that upgrades this view data warehouse.This process be exactly retrieval in this warehouse all catalogues and this catalogue in the description document of correspondence image data.Extract respectively the path of depositing of all description document and correspondence, obtain the index file in whole view data warehouse.Certainly, in the practical operation, the index file of server end does not need whole view data warehouse is carried out whole process scanning in each the renewal, only need to travel through the related view data of this renewal operation and get final product, this partial sweep can be finished by " time " attribute that checks file, only scans 2 hours with the file of interior renewal such as being designed to.
C independently generates the view data index file in this warehouse for each view data warehouse, then generate unified image file in client.When the user retrieves the view data warehouse at every turn, need first the index file of local view data index file and server end to be carried out synchronous operation.The method all is to carry out in this locality in order to guarantee that the each image data retrieval of user operates, thereby has fundamentally alleviated the burden of server end, has guaranteed effectiveness of retrieval.Synchronous operation is exactly that the view data index file in each view data warehouse is downloaded to local client, then generate a unified overall index file according to these index files, this overall index file can be that the direct simple concatenation of each view data index file forms.
D, whole retrieving all is based on keyword to carry out, and keyword can upgrade and adjust at any time, especially for the view data of new interpolation, the user can reset keyword as required, thereby has guaranteed that image-data management method has good extensibility.For example, often relate to keyword shown in accompanying drawing 4 and accompanying drawing 6 based on the research of the neuropsychiatric disease of image, the user can add at any time new " keyword: key assignments " and to satisfying the feature that newly enters to organize the image data described demand certainly.
Description of drawings
Fig. 1 is the inventive method overall procedure schematic diagram.Wherein left frame represents to carry out the view data warehouse user of read/write operation, and su* represents each number of steps of whole view data upload procedure, and sd* represents each number of steps of whole image data download process.
Fig. 2 is the inventive method image data generating sequence number process flow diagram.
Fig. 3 is the schematic diagram of client view data index file of the present invention and server end view data index file synchronizing process.
Fig. 4 is the inventive method client view data index file organization structural representation.
Fig. 5 is that the inventive method result for retrieval shows with the tree-shaped form that contains check box, to make things convenient for the user batch of results needed is downloaded.
Fig. 6 is view data example description document.
Embodiment
For the ease of the understanding of the present invention, describe each related detailed problem in the technical solution of the present invention in detail below in conjunction with accompanying drawing.Be to be noted that described embodiment does not play any restriction effect to the present invention.
The below introduces embodiments of the invention, and the embodiment of the invention only is applied to the system of Medical Imaging data image data management for example.But the embodiment of the invention is not limited only to this, is applied to the other technologies field and does not repeat them here.
Be applied to the system of Medical Imaging data image data management, clinical application research for the medical image data view data, comprise diseases analysis (senile dementia, schizophrenia etc.), and behavioral study, principle such as the human thinking activity, the relation in IQ and brain district etc., they all are the statistical conclusions that draws at statistical significance according to the image data view data, therefore an important platform problem facing of these research institutes is exactly the management in mass data view data warehouse, comprises the storage of image data view data, retrieval, perhaps add up etc.
Such as Fig. 1 block diagram, whole invention system architecture and flow process have been described.
Step su* represents that the user carries out each step that view data is uploaded the operating process such as renewal to the view data warehouse, and sd* represents that the user downloads each step of required image process operation data from the retrieval of view data warehouse.Step su2 and su8 lock and the operation of release to operated image data base, this is can only have simultaneously a user to carry out the alterability operation in order to realize to same view data warehouse, and purpose is the uniqueness in order to guarantee that the view data storage is described.The view data warehouse is carried out alterability operation begin this warehouse is locked, after this operation is finished this warehouse is carried out release, thereby discharges the alterability operating right.
Step su1: the user starts the operation that view data uploads to the view data warehouse;
Step su2: utilize digital phase-locking that operated image data base is locked and the operation of release; , by at view data warehouse entrance the Boolean variable that a called after " locks " being set, when this Boolean variable was "True", this view data warehouse was temporarily locked and can not be accessed simultaneously by other user; When above-mentioned alterability EO, view data is uploaded when finishing, and this Boolean variable is set to " vacation ", is in " release " state, so that synchronization only has a user that the view data warehouse is carried out the alterability operation;
Step su3: synchronous operation allows the view data index file of the client that the user uses and the index file of server end keep synchronously, is used for guaranteeing that all alterability operations are to carry out in up-to-date view data warehouse.Synchronous operation is exactly that the up-to-date view data index file in each view data warehouse is downloaded to local client, then generate a unified overall index file according to these index files, this overall index file can be that the direct simple concatenation of each view data index file forms; Similarly, the view data warehouse of server end all will upgrade operation to the view data index file of this server end after each the renewal, and this will produce the source of client indexes file update among the step su3.The index file of server end does not need whole view data warehouse is carried out whole process scanning in each the renewal, only need to travel through the related view data of this renewal operation and get final product, this partial sweep can be finished by " time " attribute that checks file, only scans 2 hours with the file of interior renewal such as being designed to.This design philosophy can well keep the view data of client and server end synchronous all the time, also avoided the user when each connection layout is as data warehouse, to upgrade index file and waiting as long for of causing temporarily, especially when the view data warehouse was larger, this stand-by period was with very important.The last view data index file that forms in client has shown view data numbering, path as shown in Figure 4, and other information relevant with configuration file.Detailed operating process specifies in Fig. 3, also detailed introduction will be arranged in the chapters and sections below.
Step su4 carries out exclusive number to view data to be uploaded, this numbering should be considered uniqueness, also to consider the relevance between some view data, such as in order to carry out for the follow-up study of a certain chronic disease on continuous time point, the repeatedly image data acquiring that may relate to this subject, relevance between these view data will can embody from numbering, so that the user can be very clear to the view data attribute when using view data.This inventive method has adopted the mode of segment encoding, specifically has a detailed description in Fig. 2, also can further describe in ensuing branch is introduced.
Step su5 is the generation of view data description document, and it will produce the needed keyword of user and key assignments thereof, and this is the basis for the retrieval of view data warehouse.This class keywords is according to the nearly 10 years Experience Design of inventor place seminar, and has good extensibility.Consider the development along with science and technology, new neuropsychiatric disease and new parameter index all can emerge, and so only need simply rules for writing according to " keyword: key assignments " to expand and get final product.
Among the step su6, after having generated the view data numbering and view data description document that is associated with view data, the user just can upload to the view data warehouse together with view data, and, in theory, the user does not need the deposit position of being concerned about that view data is concrete when uploading, and specifically deposits the path because server end can the self-verifying view data when upgrading its index file.
The image data download process is relatively simple.The user at first needs through step sd2, the view data index file of the view data index file of client with server end is carried out synchronously, so that guarantee that user's retrieving is to carry out in up-to-date view data warehouse.Then keyword and the key assignments that will retrieve by step sd3 input are searched for, and carry out preview by step sd4, and final step sd5 downloads needed result.Download in order to make things convenient for the user that needed view data is carried out batch, inventive method has designed the file check box, can all results that need of final election download in batches.The Search Results schematic diagram is arranged among Fig. 5.
Fig. 2-Fig. 5 has explained respectively the exemplary operations process of several important steps among Fig. 1.
The user uploads view data and only need to simply fill in a view data description document to the view data warehouse and get final product; Same, needed view data is also downloaded in user's inquiry, also only needs to key in own desired keyword and get final product, and download before also the image that can easily preview retrieves.The method of medical image image data management of the present invention, the serial operating process such as generation, image data retrieval, view data preview and image data download of it is integrated uploading of view data, view data serial number gencration, view data description document;
The user does not need to consider the specifically memory location on server when uploading view data, can arbitrarily leave in the server, and this system can generate search index automatically according to uploading the file that view data carries, and makes things convenient for the user to retrieve efficiently;
The user is when uploading view data, this inventive method can be carried out according to subject's name of user's input local view data index file inquiry, examine the view data whether same subject is arranged in the view data warehouse, then itemize out the description document of these view data, remind this to upload the whether continuous acquisition (for example following up a case by regular visits to for certain disease patient's tracking) on the different time points of same tested place of view data, allow the user determine whether view data is carried out the relevance numbering; If so, system will generate the secondary numbering that the user can identify to this view data;
Special " remarks " subregion is arranged in the view data description document file, be designated " # describes the district ", this is in order to satisfy the multifarious demand of imaged image data, for example, some view data are to carry out for certain concrete experimental duties design, can be described in this part, system equally can be according to keyword retrieval;
The generation of view data description document can be undertaken by a shirtsleeve operation interface, and most contents all is directly to select by drop-down menu; Fill substance in the interface has carried out subregion according to the form of clinical scanning, and is simple and clear;
The keyword of image data retrieval can be keyword option (item), also can be key feature word (keyword); The latter mainly be for " # describes the district " carry out;
The view data search inquiry can be to carry out in the machine, also can be to carry out at the long-range FTP that has login (download) authority;
In order to realize that from the high-speed downloads in view data warehouse, the multithreading setting has been opened by system.Certainly, this needs the other side's view data warehouse server to allow to use multithreading to download;
System can show with tree-shaped catalogue the result who arrives according to keyword retrieval, and the user can live preview view data image and corresponding view data description document;
In the tree-shaped catalogue of Search Results, the user can be on certain file right mouse button select " individual layer downloads " or " recurrence download ", thereby realization is only downloaded son file in this grade catalogue or recurrence and is downloaded All Files and sub-folder content under this catalogue;
For the convenient view data of downloading in batches a plurality of files in the Search Results, system can arrange check box (checkbox) on all folder node, so that the user can download the content that all are chosen easily;
In order to improve view data keyword retrieval efficient, avoid retrieving for server retrieves all carrying out at long-range FTP at every turn, the user can be chosen on the local hard drive generated data and describe index file, owing to be text-only file all, so do not take too much hard drive space;
Research for medical image processing tends to relate to a large amount of iconography view data; especially for clinical practice; usually can adopt the imaged image data of large sample, comprise the tested of large group, perhaps the image data acquiring of a plurality of time points of single-subject.Therefore these software systems with regard to a managing image data warehouse of exigence solve the serial operating process such as storage, retrieval, preview and download that comprise view data.
The present invention proposes in order to address the above problem, and seminar is at research accumulation summary image-data management system out.This system mainly is divided into three parts: view data is uploaded, image data retrieval and image data download.
View data is uploaded
View data is uploaded the generation that part mainly solves the respective image data description document, because upload procedure does not need the user to consider the memory location of view data on server, the retrieving of view data mainly carries out according to index file exactly.Not only comprised the main information in conventional digital imagery and Interchange Format (hereinafter to be referred as the DICOM) header file in this index file, address name for example, contact method and some sweep parameters etc., and comprised for the more useful specialized information of medical image research, for example tested pathology and the course of disease, whether take medicine, IQ (hereinafter to be referred as IQ), lab design and tested behavior or the scoring of emotion scale etc., because these information play central role in based on the disease research of image or behavioural analysis, so this class view data description document often has more researching value than DICOM header file.
The generation of view data description document also comprises the generation of view data numbering (sequence number), and this sequence number and view data are one to one.In the medical image analysis of image data; usually can relate to for a certain tested tracking and following up a case by regular visits to; such as the research to Alzheimer disease (AD); from normal person (Normal); develop into mild cognitive impairment (MCI); arrive again AD; this process is understood the time of lasting for years usually; developing into AD from MCI approximately is the time in 3 years to 5 years; human brain has variation on many corresponding structures and the function in iconography during this; such as the variation of cortex thickness and complexity, expansion or the atrophy of hippocampus part, the activity change on the brain zone function or function connect variation.Therefore whole research process will comprise structure magnetic resonance (sMRI) repeatedly to the tested brain image scan that carries out, functional MRI (fMRI), and diffusion magnetic resonance (dMRI) etc.These view data are separate in iconography, but but are mutually additional and checking in the research of disease.So the view data numbering that we produce should represent the unique identification to certain view data, shows again the continuity between the view data.
We adopt syllogic that view data is numbered.For the data of new collection, whether we first first have duplication of name according to subject's name look-up, if there is no bear the same name, and then directly number add 1 the chief editor, are adding the numbering of imaging pattern as the second field; If exist to bear the same name, then data typing person will judge whether this data acquisition is relevant with data with existing according to the duplication of name result who enumerates, if uncorrelated, then catch up with and states the situation of not bearing the same name and equally process; If relevant then judge this new data whether for following up a case by regular visits to sequence according to data with existing, and be which time during sequence gathers gathers, this time numerical value is as the 3rd field number, and the second field number still is the imaging pattern of this collection.
Therefore, end is got up, if tested is Zhang San, his the structure nuclear magnetic resonance image view data numbering second time may be: RCCM00035_sMRI_2, first field is that (RCCM is for calculating medical center, Research Center for ComputationalMedicine for total independent image data bulk numbering.This prefix user can preset), the second field is that scintigram 2 illustrates system's image data generating sequence number process flow diagram.
The 3rd field is which time scan image data of this this mode of people.Some the time, when especially normal controls scanning is tested, if be Zhang San, perhaps Zhang San participates in sMRI scanning for the first time, but the purpose of front and back is not the contrast experiment who does on the different time points, therefore the first field number can't be identical, although be same person, is independently on the view data meaning.That is to say, our view data numbering is not distinguished with the people, but distinguishes with the independence of view data, this will be in ensuing image data retrieval uses the crucial meaning of performance.This also is the basic thought of whole system design.Be summed up, the generation of view data numbering can be according to the flow implementation of image data generating sequence number among Fig. 2.
View data description document agreement is the supplementary notes of user to view data or keyword section with the behavior description field of " # " beginning.This part can not followed the rules for writing of " keyword: key assignments ", arbitrarily writes according to user's request and description custom, the keyword commonly used of implication to be expressed should occur as far as possible yet.
The person fills in the view data description document in order to further facilitate the image data acquiring, and the readme.txt of this system User Interface adopts multiple solution to reduce workload.At first, system adopts drop-down menu in a large number, such as disease type, and sweep parameter etc., the user only needs to select to get final product from drop-down menu; In addition, system adopts interim memory style, such tested for same mass scanning, more information may all be identical, such as all being the blind person, perhaps experiment purpose design etc., so the user is in same mass scanning, basic only need to fill in one tested, other most information can generate automatically.In addition, fill in the corresponding simple declaration of Qu Douyou at each, pointed out should the zone this how to fill in.
Image data retrieval
The retrieval of view data is carried out for the correspondence image data description file, also substantially is divided into two types.The first is according to keyword option and key assignments search, such as the age option, and the age: 43 (by the concrete age), perhaps age: 35-50 (age-based interval); Another kind is that search described in keyword, such as the assignment decisions (Decision Making) in functional magnetic resonance imaging (hereinafter to be referred as the fMRI) experiment, related descriptive keyword may comprise " money ", " repayment " or " three months " etc., these keywords all can appear at keyword and describe the district, #description, be used for describing the original intention of this image data acquiring, they also are the fundamental basis of doing assignment decisions research for human brain.
In order to improve search efficiency, this inventive method can be so that whole search procedure be carried out in the view data description document of client, and this just need to set up local index to long-range view data description document (perhaps local view data description document of disperseing storage).Local index file synchronously will occur with the descriptor format of Fig. 4.
View data upload and downloading process must be first synchronous the data description file of the data description file of client and server end.When uploading data, first the data warehouse is locked, so that the operation of this property upgraded has the authority of monopolizing, other user can't change data warehouse simultaneously.After upgrading complete (upload or revise) data warehouse, regenerate new data description file at server end again, this index file for next time for this database upload or during down operation.Similarly, the user is from the data warehouse downloading data time, after the log-on data warehouse management procedures is connected to data server, at first synchronous the data description file of the data terminal of the data description file of client and server end, and then retrieve the operations such as download.Sum up, client view data index file and server images data directory file are synchronous by process shown in Figure 3.
This index file generates corresponding new version by each server after view data is updated, view data is upgraded and included that new view data is uploaded, view data moves catalogue and the operation such as view data deletion; The user is after being connected to server, and meeting this new version of automatic synchronization also adds on the own local search index.This Design Mode both guaranteed server and the good view data of client synchronously more new relation, also avoided the user in time that the client awaits index file generates.
The number of searching key word and field setting can freely arrange according to user's request.
Fig. 4 client indexes file organization structural representation.This document comprises data number, store path, and for the descriptor of these data, comprise coordination responsible official and the acquisition time of this secondary data data acquisition, subject's name, sex, contact method etc.
Result for retrieval after keyword is selected is shown to the user with tree-shaped catalogue, and the user can carry out preview to DICOM image and corresponding view data description document, is positioned at simultaneously respectively different windows, checks and confirms whether it is the view data that oneself needs.And the preview of image also has different aids, such as the adjusting of gray scale window and convergent-divergent etc.
Image data download
Image data download designs mainly for the data management of remote image data server images.Download partial design multithreading download downloading mode, the user can freely select Thread Count according to the network bandwidth of oneself and the requirement for restriction of image data server, in order to maximize speed of download.In addition, on catalogue was downloaded, system was provided with " choosing " (checkbox) check box in each catalogue,
Fig. 5, result for retrieval show with the tree-shaped form that contains check box, download in batches with convenient.End user can carry out batch to all check boxes of having chosen and download, and this meets the demand of batch contrast experiment in the medical image analysis of image data fully.Its effect as shown in Figure 5.
In addition, download for single catalogue, system be provided with also that the individual layer catalogue is downloaded and this catalogue with and the recurrence of all sub-directories download.
One of main innovation of this invention system is design and the classification of image data structure.We are divided into five classes to retrieving information: the scan task brief introduction, and tested general information, tested clinical information, scanning machine parameter and experimental design are described, and each specifying information that comprises is as follows.
The scan task brief introduction.The overall coordination responsible official, time, address and the radiation doctor that comprise this time scanning.
Tested general information.Comprise the person's of being scanned name, sex, contact method and left/right handedness.
The general information of tested and whole experiment, these information have in the DICOM of routine file.What describe in ensuing several block diagram is exactly that the DICOM file can not satisfy these demands with the tight clinical information of clinical research.It is view data example description document among Fig. 6.
Tested clinical information.Comprise whether disease and kinds of Diseases are arranged, IQ, age, education degree, and the correlation parameter information of various diseases.
The scanning machine parameter.Comprise clinical scanning parameter corresponding to various mode scannings.
Experimental design is described.The design original intention that comprises this time experiment comprises experimental design (allow the tested picture of seeing, perhaps answer etc.), and the basic problem that will study.
The above; only be the embodiment among the present invention; but protection scope of the present invention is not limited to this; anyly be familiar with the people of this technology in the disclosed technical scope of the present invention; can understand conversion or the replacement expected; all should be encompassed in of the present invention comprising within the scope, therefore, protection scope of the present invention should be as the criterion with the protection domain of claims.
Claims (7)
1. the management method in a view data warehouse, it is characterized in that, to utilize text based view data descriptor, the imaged image data warehouse is managed, the method is divided into view data and uploads step and image data download step, and described view data uploads step and the image data download step is as follows:
Wherein view data is uploaded step and is comprised:
Step su1: client terminal start-up view data upload operation is prepared to upload new image data, and is connected to the server end that will operate;
Step su2: utilize lock to the operation that the view data warehouse of operated server end locks, operate successfully if lock, so that synchronization only has a client to have the view data warehouse is carried out the authority that alterability operates; If it is unsuccessful to add latching operation, client does not have the authority of the view data warehouse being carried out the alterability operation;
Step su3: the image index file update process of client is as follows: download the image index file from server end first by client, compare the existing image index file of client with from the time attribute that server end is downloaded the image index file obtain, if the time attribute of comparison is consistent, the image index file of client and the image index file content of server end are identical; If the time attribute of comparison is inconsistent, then cover the image index file of client with the image index file of server end, form the image index file in client, be used for guaranteeing that all alterability operations are to carry out in up-to-date view data warehouse; Client image index file organization structure comprises view data numbering, path, and the information relevant with configuration file;
Step su4: client is for new image data, first whether the name look-up according to the subject has the step of duplication of name as follows: the subject's name according to the data typing person input of client carries out local image index file polling, examine the view data whether same subject is arranged in the view data warehouse, if there is not same subject's view data, then directly new image data is accumulated numbering as the first field, add the view data imaging pattern as the second field of view data numbering; If same subject's view data is arranged, then the data typing person of client will be according to the existing view data description document of this subject, judge whether the new image data that this data acquisition obtains is correlated with existing view data, if uncorrelated, then catching up with to state does not have same subject's view data situation equally to process; If relevant then judge this new data whether for following up a case by regular visits to sequence according to data with existing, and be which time during sequence gathers gathers, this time numerical value is as the 3rd field of view data numbering, and view data the second field number still is the imaging pattern of this collection; Therefore, the view data of formation is numbered:
RCCM (x
1x
2x
3x
4... x
n+ 1) _ sMRI or RCCMx
1x
2x
3x
4... x
n_ sMRI_k,
Wherein RCCM is data affiliated unit, (x
1x
2x
3x
4... x
n+ 1) or (x
1x
2x
3x
4... x
n) expression one-level data number, the first field RCCM (x
1x
2x
3x
4... x
n+ 1) or RCCMx
1x
2x
3x
4... x
nBe total independent image data bulk numbering; SMRI is the view data imaging pattern, as the second field of view data numbering; If there is the 3rd field in view data numbering, then be the k time the subject to be carried out sMRI to scan, wherein n and k are natural number;
Step su5: with the RCCM (x that new image data is generated
1x
2x
3x
4... x
n+ 1) _ sMRI or RCCMx
1x
2x
3x
4... x
n_ sMRI_k numbering is as the first row of the description document beginning of new image data, and described description document is text-only file; Data typing person fills in the user by operation interface needed keyword and key assignments thereof is retrieved in the view data warehouse, the antistop list registration is according to a kind of classification of feature, the key assignments corresponding with keyword represents the property value that the classification of this feature is corresponding, thereby generates new view data description document;
Step su6: the client user puts into the new image data description document file at new image data place, the view data warehouse of then this document folder being uploaded onto the server together and holding;
Step su7: after the user uploads new image data toward server end, the image index file of server end update server end; The renewal process of image index file is exactly that in time of user's appointment in this view data warehouse of server-side retrieval all exist corresponding view data description document in the catalogue upgraded and this catalogue, the new description document that obtains being updated, extract respectively again the new description document be updated and corresponding deposit the path and compare with the existing image index file of this server end, if exist the view data numbering to repeat, then cover the corresponding contents of the existing image index file of server end with the new description document that is updated and path thereof; If numbering does not repeat, then the new description document that is updated and path thereof are added in the existing image index file of server end, thereby obtain the new images index file that is updated in whole view data warehouse at server end, so far, the update image index file is complete;
Step su8: the lock that arranges among the step su2 is opened, and client discharges the authority to the alterability operation of server end, thereby to the release of view data warehouse, so that other user has the authority of alterability operation to this servers' data warehouse;
Wherein the image data download step comprises:
Step sd1: the user of client starts the image data download operation, and is connected to the server end that will operate;
Step sd2: the user of client carries out synchronous operation to the image index file of the image index file of client and server end, generate up-to-date image index file in client, so that the image index file content of the image index file of client and server end is identical, thereby the retrieving that guarantees the user carries out in up-to-date view data warehouse;
Step sd3: the user is after client has obtained up-to-date image index file, and the keyword that input will be retrieved and key assignments are directly searched in the image index file of client, is met the right view data directory listing of keyword key;
Step sd4: the user of client carries out preview to the image in the view data directory listing, take the result that determines search whether as own expected result;
Step sd5: in the view data directory listing that search obtains, adopt all view data that need of file check box final election, needed view data is carried out batch download.
2. described method for managing image data warehouse according to claim 1, it is characterized in that, lock and the operation of release is at view data warehouse entrance the Boolean variable that a called after " locks " to be set, when this Boolean variable was "True", this view data warehouse was temporarily locked and can not be accessed simultaneously by other user; When the alterability EO, view data is uploaded when finishing, and this Boolean variable is set to " vacation ", and then the view data warehouse is in " release " state.
3. described method for managing image data warehouse according to claim 1, it is characterized in that, the image index file of described client is identical from the image index file content of server end to be: when having a plurality of different view data warehouse, the image index file of client is comprised of the image index file in all view data warehouses.
4. described method for managing image data warehouse according to claim 1, it is characterized in that, described view data description document is that the user numbers above-mentioned view data and keyword: key-value pair is write into the view data description document, or design table (schedule) allows the user fill in a form, and then allows computer programming language image data generating description document.
5. described method for managing image data warehouse according to claim 1, it is characterized in that, the keyword of described needs is expression different field acquisition of image data, different to the description of keyword according to the actual requirements, the user expands " keyword: key assignments " key-value pair according to the rules for writing of " keyword: key assignments ", in order to the descriptive power of expansion for the keyword of new image data, thereby generation is for the description document of new image data.
6. described method for managing image data warehouse according to claim 1, it is characterized in that, described synchronous operation is exactly that the image index file in the view data warehouse on each server is downloaded to client, then according to a unified overview image index file of these image index file generateds, this overview image index file is that each image index file is spliced.
7. described method for managing image data warehouse according to claim 1, it is characterized in that, all view data that need of described file check box final election, when the user of client needs a plurality of results among all results, at this moment the needed result who searches in user's final election carries out disposable download.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200910237783 CN101876992B (en) | 2009-11-17 | 2009-11-17 | Method for managing image data warehouse |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200910237783 CN101876992B (en) | 2009-11-17 | 2009-11-17 | Method for managing image data warehouse |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101876992A CN101876992A (en) | 2010-11-03 |
| CN101876992B true CN101876992B (en) | 2013-03-06 |
Family
ID=43019550
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200910237783 Active CN101876992B (en) | 2009-11-17 | 2009-11-17 | Method for managing image data warehouse |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101876992B (en) |
Families Citing this family (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102467520A (en) * | 2010-11-10 | 2012-05-23 | 金蝶软件（中国）有限公司 | Data storage method and device |
| CN102495848B (en) * | 2011-11-17 | 2013-06-19 | 深圳市赛格导航科技股份有限公司 | Method for processing massive GPS (global positioning system) data and system |
| CN103793701A (en) * | 2014-02-27 | 2014-05-14 | 彭大维 | Electric transmission line image automatic recognition system based on neural network technology |
| CN104053016B (en) * | 2014-06-24 | 2018-09-28 | 深圳市江波龙电子有限公司 | A kind of image preview method and device |
| US10198445B2 (en) | 2014-06-30 | 2019-02-05 | Google Llc | Automated archiving of user generated media files |
| CN105719105A (en) * | 2014-12-03 | 2016-06-29 | 镇江雅迅软件有限责任公司 | Inventory quick lookup method based on keywords |
| CN105205122A (en) * | 2015-09-11 | 2015-12-30 | 烽火通信科技股份有限公司 | Telecommunication service retrieval method based on data synchronization |
| CN105243101B (en) * | 2015-09-17 | 2019-05-10 | 浪潮(北京)电子信息产业有限公司 | The method and system that multi-controller concurrently accesses same compressed data file |
| CN106294860A (en) * | 2016-08-23 | 2017-01-04 | 浪潮电子信息产业股份有限公司 | The system of a kind of real time indexing data syn-chronization and its implementation |
| CN106776720A (en) * | 2016-11-18 | 2017-05-31 | 北京奇虎科技有限公司 | A kind of document handling method and device |
| CN107066521A (en) * | 2017-02-16 | 2017-08-18 | 国网上海市电力公司 | Transmission line equipment association picture crawling method based on text keyword |
| CN108052516A (en) * | 2017-10-19 | 2018-05-18 | 福建中金在线信息科技有限公司 | The display methods of picture, device and mobile terminal in multidate information |
| CN107993051A (en) * | 2017-11-29 | 2018-05-04 | 珠海市艾森科技有限公司 | Material document management method and its management system |
| CN107861858B (en) * | 2017-12-04 | 2021-02-19 | 网易（杭州）网络有限公司 | File detection method and device, storage medium and processor |
| CN108259992B (en) * | 2018-04-25 | 2021-02-23 | 浙江博尚电子有限公司 | Network television set top box video transmission method |
| CN110688672A (en) * | 2019-09-17 | 2020-01-14 | 上海易点时空网络有限公司 | Method and system for ensuring file downloading safety of data warehouse |
| CN112073681A (en) * | 2020-07-14 | 2020-12-11 | 中国电力科学研究院有限公司 | Processing method and system for routing inspection, positioning and shooting images of overhead power line unmanned aerial vehicle |
| CN112420147B (en) * | 2020-11-18 | 2022-03-04 | 孙应实 | System and method for real-time communication between image AI result and structured report desktop |
| CN113241153A (en) * | 2021-05-08 | 2021-08-10 | 上海联影医疗科技股份有限公司 | Image loading method and device, electronic equipment and storage medium |
| CN115391289B (en) * | 2022-08-15 | 2023-08-11 | 广州市玄武无线科技股份有限公司 | Method and device for generating directory structure containing file information |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1434402A (en) * | 2002-01-21 | 2003-08-06 | 中国科学院研究生院 | Method and system of URL based long distance image files description service |
-
2009
- 2009-11-17 CN CN 200910237783 patent/CN101876992B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1434402A (en) * | 2002-01-21 | 2003-08-06 | 中国科学院研究生院 | Method and system of URL based long distance image files description service |
Non-Patent Citations (1)
| Title |
|---|
| JP特开2003-288344A 2003.10.10 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101876992A (en) | 2010-11-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101876992B (en) | Method for managing image data warehouse | |
| US8874552B2 (en) | Automated generation of ontologies | |
| US8856169B2 (en) | Multi-modality, multi-resource, information integration environment | |
| CN110415831A (en) | A kind of medical treatment big data cloud service analysis platform | |
| US20050187794A1 (en) | Electronic medical record registry including data replication | |
| US20050015381A1 (en) | Database management system | |
| KR20120011583A (en) | System for generating electron form in unified medical information system and management method thereof | |
| US20160070751A1 (en) | Database management system | |
| CN101271489A (en) | Medicine case information management system | |
| CN106919608A (en) | Medical data processing method, device and platform | |
| Kiong et al. | Health ontology system | |
| CN103246691B (en) | Document file management system and its method | |
| Jin et al. | Pai-wsit: An AI service platform with support for storing and sharing whole-slide images with metadata and annotations | |
| Kundu et al. | Building a graph database for storing heterogeneous healthcare data | |
| Madiraju et al. | A mobile agent approach for global database constraint checking | |
| JP2004102818A (en) | Retrieval support method and retrieval support device | |
| CN114510491A (en) | Dynamic follow-up quantity table design method and system | |
| CN106326626B (en) | A kind of health sufferer information input system | |
| Brunette et al. | Global goods software for the immunization cold chain | |
| Ziegler et al. | PAL: toward a recommendation system for manuscripts | |
| Liepins et al. | A browser based image bank, useful tool or expensive toy? | |
| Mak et al. | Enabling and integrating ETD repositories through Linked Data | |
| WO2006005715A1 (en) | System for interrogating heterogeneous databases and method for interrogation | |
| Vogeler et al. | Data Exchange in Practice: Towards a Prosopographical API. | |
| Haras et al. | Patient data synchronization process in a continuity of care environment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |