CN101561850B

CN101561850B - System for comprehensively processing massive multi-channel nerve information and control method thereof

Info

Publication number: CN101561850B
Application number: CN2009103031072A
Authority: CN
Inventors: 骆清铭; 蒲江波; 李向宁
Original assignee: Huazhong University of Science and Technology
Current assignee: Huazhong University of Science and Technology
Priority date: 2009-06-10
Filing date: 2009-06-10
Publication date: 2011-12-28
Anticipated expiration: 2029-06-10
Also published as: CN101561850A

Abstract

The invention relates to the field of neuroelectricity physiological information data processing and provides a system for comprehensively processing massive multi-channel nerve information. The system comprises a script-driving control component, a data obtaining control component, a data analysis processing component and a data visualization processing component. Collection, processing and visualization processes of multi-channel nerve information undergo componentization and integration so as to establish a uniform service interface; and the processing process of the massive multi-channel nerve information can be automatically finished in terms of batches under the control of a script, and a visualization result can be provided. Simultaneously, the procedures for processing neuroelectricity physiological information data can be stored in a storage medium, thereby improving the repeatability and controllability of data processing. The invention also has the advantages that hardware control and the data processing are combined together; according to the processing result of neuroelectricity physiological information data, hardware operation is controlled, and an experiment proposal is dynamically revised; and due to the uniform interface, expansion processing and visualization algorithm can be dynamically carried out.

Description

System for comprehensively processing massive multi-channel nerve information and control method thereof

Technical field

The present invention relates to neuroelectricity physiologic information data processing field.Relate in particular to a kind of modularization total system and control method thereof of operating the multi-channel nerve electric physiological data.

Background technology

In the nerve information process field, the information that the researchist faces generally has: the size of data of (1) magnanimity; (2) Fu Za data definition; (3) feature such as data processing method flexibly.Once typical neuroelectricity Physiological Experiment can produce the experimental data greater than 2 G bytes, and this experimental data comprises the time electroneurographic signal that becomes of a plurality of passages usually.According to experiment purpose, the concrete definition of data is complicated and changeable often, example: the dynamic change of number of active lanes, the appointment of stimulation parameter etc.Simultaneously, the treatment scheme to experimental data embodies flexible and changeable feature.According to the space-time characterisation of experimental data, flow chart of data processing need be assigned algorithms of different.Algorithm must satisfy dynamically the demand that increases, deletes, revises.In addition, because neuroelectricity physiologic information data volume is huge, the researchist need assign nonidentity operation respectively to each lot data, and carries out subsequent analysis according to operation result, and this instant has visually proposed active demand to what data were handled PRELIMINARY RESULTS.In addition, the end result of scientific research need produce the figure as a result of printing quality, for example line chart, block diagram, pie chart etc.Therefore visualization result must satisfy research journal publication requirement (as size, gray scale, sharpness etc.).Yet existing assembly can't satisfy the processing requirements of above-mentioned experimental data.

In present neuroelectricity Physiological Experiment data handling procedure, data acquisition, storage, processing and visualization process generally are to be finished by discrete separately software, lack unified Workflow Management and control framework.And (1) data acquisition relies on particular hardware and operating system environment height; (2) reading and writing data relies on the concrete file layout height of manufacturer's definition, and lacks index and search to the data file description in storage administration; (3) data processing lacks customization, only can move existing Processing Algorithm according to named order, and the newly-increased and modification of algorithm logic is very difficult; (4) lack the robotization mechanism that a kind of batch processing meets the specified requirements data; (5) data visualization and data processing be by hard coded, can't increase new method for drafting or revise existent method, and visualization result and publication requires gap big, can't be directly used in publication.

The neuroelectricity physiological Study be one from (1) experimental design; (2) experimental data produces, analyzes, handles; (3) according to existing correction experimental design as a result; (4) obtain the entire flow of final experimental result etc.Existing each stage is finished by discrete software respectively, and this has proposed two problems: how (1) guarantees high-efficiency reliable exchanges data between each software; (2) how to preserve experiment and handle and analysis process, be convenient to the playback experiment step, the efficient that improves the CAE design and analyze from overall angle.Therefore, need a kind of modularization and extendible total system that is applicable to the neuroelectricity physiological Study.

Summary of the invention

At above-mentioned technical matters, the present invention proposes a kind of system for comprehensively processing massive multi-channel nerve information and control method thereof, the integration and the batch processing of the collection of realization nerve information, processing, visual three processes, and guarantee data stream and consistance and the reproducibility of controlling stream in the processing procedure.

For solving the problems of the technologies described above, a kind of system for comprehensively processing massive multi-channel nerve information that the present invention proposes is characterized in that, comprises that script drive controlling assembly, data are obtained to handle assembly, data analysis processing components, data visualization processing components;

Described script drive controlling assembly is used to provide user interactions control interface, and user instruction is accepted, the generation script order in described user interactions control interface; Described script drive controlling assembly is explained script command, calls described data and obtains manipulation assembly, data analysis processing components, data visualization processing components execution corresponding task, and to the user feedback result;

Described data are obtained the manipulation assembly, connect script drive controlling assembly, neuroelectricity Physiological Experiment system, data read-write system, be used to accept the data that script drive controlling assembly sends and obtain manipulation request, drive described neuroelectricity Physiological Experiment system acting, and receive the return data of described neuroelectricity Physiological Experiment system; Perhaps, control described data read-write system and carry out reading and writing data; Then, return result to script drive controlling assembly;

Described data analysis processing components, connect script drive controlling assembly, be used to accept the data analysis processing request that script drive controlling assembly is initiated, calling various data analysis Processing Algorithm handles the data analysis of neuroelectricity physiologic information, the analyzing and processing algorithm and the data thereof of history are kept in the data file, and return result to script drive controlling assembly;

Described data visualization processing components, be used to accept the visualization processing request that script drive controlling assembly is initiated, call various data visualization Processing Algorithm neuroelectricity physiologic information data are carried out visualization processing, the visualization processing algorithm and the data thereof of history are kept in the data file, and return result to script drive controlling assembly.

Preferably, described script drive controlling assembly comprises: Cmd Shell and script file interface unit, and the system directive resolution unit, the user instruction resolution unit, the task execution time planning unit, task status is preserved and monitoring unit, task executive's interface unit,

Described Cmd Shell and script file interface unit are used to receive user command, and the work planning instruction in the identification user command, and described work planning instruction is committed to described task execution time planning unit; Inoperative in user command planning instruction is committed to described system directive resolution unit; And receive the inoperative planning instruction that described system directive resolution unit is returned, be identified as user instruction and be committed to described user instruction resolution unit;

Described system directive resolution unit is used for discerning the system directive of inoperative planning instruction, described system directive generation system order set is called be committed to described task execution time planning unit; And its inoperative that can not discern planning instruction returned described Cmd Shell and script file interface unit;

Described user instruction resolution unit is used to accept the user instruction that described Cmd Shell and script file interface unit are submitted to, generates the user instruction system call and is committed to described task execution time planning unit;

Described task execution time planning unit is used for according to described work planning instruction the priority arrangement being carried out in described system directive system call, user instruction system call, and the generation task is carried out stream, and is committed to task status preservation and monitoring unit;

Described task status is preserved and monitoring unit, is used for management and calling task executive's interface,

Described task executive's interface is used to carry out described system directive system call, user instruction system call, controls described data and obtains and handle assembly, data analysis processing components, data visualization processing components and execute the task.

Preferably, described data are obtained and are handled assembly and comprise: request of data and data exchange interface unit, hardware requests Agent layer unit, hardware controls assembly, file request Agent layer unit, file protocol assembly, database component;

Described request of data and data exchange interface unit, be used to receive the data processing request that script drive controlling assembly is initiated, is the target type of judging described data processing request operation of neuroelectricity Physiological Experiment or data read-write operation? if neuroelectricity Physiological Experiment operation requests is then called described hardware requests Agent layer unit; If file request Agent layer unit is called in the data read-write operation request, finally to described script drive controlling assembly return results;

Described hardware requests Agent layer unit is used for the instruction according to described request of data and data exchange interface unit, calls described hardware controls assembly,

Described hardware controls assembly connects the neuroelectricity Physiological Experiment system that comprises described neuroelectricity physiological data harvester, neuroelectricity physiological stimulation output unit, be used for driving and comprise described neuroelectricity physiological data harvester, the action of neuroelectricity physiological stimulation output unit, return the result that comprises data and status word to described hardware requests Agent layer unit according to the output result of described neuroelectricity Physiological Experiment system;

Described file request Agent layer unit is used for the instruction according to described request of data and data exchange interface unit, calls database component or described file protocol assembly,

Described database component, be used for according to the data query of the SQL word string of the include file characteristic query of described file request Agent layer unit and/or write request, call database and carry out query statement or call described file protocol assembly, and return the file operation result who comprises data and status word to described request of data and data exchange interface unit;

Described file protocol assembly is used for directly calling file-storage device and carries out file and store to handle, and to himself caller return results.

Preferably, described data analysis processing components comprises that the request handled explains and feedback interface as a result data input layer assembly, Processing Algorithm trustship layer assembly, operation result output interface, Processing Algorithm performance element;

Described processing request explanation and feedback interface as a result, be used to receive the data analysis processing request that described script drive controlling assembly is initiated, call described data input layer assembly, Processing Algorithm trustship layer assembly is handled, and returns result to described script drive controlling assembly;

Described data input layer assembly is used to receive described processing request explanation and the data stream object that sends of feedback interface as a result, submits to the Processing Algorithm performance element to handle described data stream object,

Described Processing Algorithm trustship layer assembly is used to receive described processing request explanation and the instruction stream object that sends of feedback interface as a result, and call described Processing Algorithm performance element and handle,

Described Processing Algorithm performance element is used for carrying out calculation process according to described data stream object, instruction stream object, and calculation process finishes the back and returns the running status word to described algorithm trustship layer assembly, and the operation result data are by described operation result output interface output;

Described operation result output interface is accepted the output result from described Processing Algorithm performance element, and to described processing request explanation and as a result feedback interface return.

Preferably, described data visualization processing components comprises that visualization request and result transmit interface, the method for visualizing performance element,

Described visualization request and result transmit interface and are used to receive the data visualization request of initiating from script drive controlling assembly, call described method for visualizing performance element, and to script drive controlling assembly return results;

Described method for visualizing performance element is used for the drafting according to described data visualization request execution result figure, and/or with graphics memory in file system and/or database, and figure transmitted the interface return results to visualization request and result.

The invention also proposes a kind of control method of system for comprehensively processing massive multi-channel nerve information, may further comprise the steps:

Load described script drive controlling assembly;

Move described script drive controlling assembly;

Travel far and wide this from the human-computer interaction interface input command;

The described order line script of described script drive controlling analyzing component generates the work order sequence;

Described script drive controlling component priority order is carried out described work order sequence, and gives status object to every described work order;

Call described data when certain work order and obtain when handling assembly, described data are obtained and are handled assembly and call according to described work order, drive described neuroelectricity Physiological Experiment system acting, and receive the return data of described neuroelectricity Physiological Experiment system; Perhaps, control described data read-write system and carry out reading and writing data; Then, return result to script drive controlling assembly;

When described data analysis processing components is called in certain work order, described data analysis processing components is called various data analysis Processing Algorithm the data analysis of neuroelectricity physiologic information is handled, return result to script drive controlling assembly, and submit historical analyzing and processing algorithm and data thereof to;

When described data visualization processing components is called in certain work order, described data visualization processing components is called various data visualization Processing Algorithm neuroelectricity physiologic information data is carried out visualization processing, return result to script drive controlling assembly, and submit historical visualization processing algorithm and data thereof to;

Continue to carry out next bar work order, all instructions in the work order sequence are finished.

As preferred version, described loading script drive controlling assembly may further comprise the steps:

Load described Cmd Shell and script file interface unit,

Load described system directive resolution unit, user instruction resolution unit,

Load described task execution time planning unit,

Load described task status and preserve and monitoring unit,

Load described task executive's interface unit.

As preferred version, the described order line script of described script drive controlling analyzing component generates the work order sequence, may further comprise the steps equally:

Described Cmd Shell and script file interface unit receive user command,

Whether the identification user command is the work planning instruction,

If the work planning instruction then is committed to described task execution time planning unit with the work planning instruction;

If not the work planning instruction then is committed to described system directive resolution unit with inoperative planning instruction; Described system directive resolution unit receives and identifies inoperative planning instruction, judge whether to be system directive? if system directive is then called described system directive generation system order set and is committed to described task execution time planning unit;

If not system directive then returns the nonsystematic instruction to described Cmd Shell and script file interface unit; Cmd Shell and script file interface unit are identified as user instruction with the nonsystematic instruction and are committed to described user instruction resolution unit; Described user instruction resolution unit generates the user instruction system call with user instruction and is committed to described task execution time planning unit;

Described task execution time planning unit carries out the priority arrangement according to described work planning instruction to described system directive system call, user instruction system call, and the generation task is carried out stream, and is committed to task status preservation and monitoring unit;

Described task status is preserved and monitoring unit management and calling task executive's interface,

Described task executive's interface is carried out described system directive system call, user instruction system call, controls described data and obtains and handle assembly, data analysis processing components, data visualization processing components and execute the task.

Equally as preferred version, described data are obtained and are handled assembly and call according to described work order, drive described neuroelectricity Physiological Experiment system acting, and receive the return data of described neuroelectricity Physiological Experiment system; Perhaps, control described data read-write system and carry out reading and writing data; Then, return result, may further comprise the steps to script drive controlling assembly:

Described request of data and data exchange interface unit receive the data processing request that script drive controlling assembly is initiated,

The target type of judging described data processing request is system operation of neuroelectricity Physiological Experiment or data read-write operation,

If Electrophysiology experimental system operation requests; Then call described hardware requests Agent layer unit; Described hardware requests Agent layer unit is according to the instruction of described request of data and data exchange interface unit; Call described hardware controls assembly; Described hardware controls Component driver comprises that described Electrophysiology data acquisition unit, Electrophysiology stimulate the Electrophysiology experimental system action of output device and gather its output data; Return the result that comprises data and status word to described request of data and data exchange interface unit

Described request of data and data exchange interface unit be to described script drive controlling assembly return results status word, and submit detected data to;

If described file request Agent layer unit is then called in the data read-write operation request, described file request Agent layer unit calls described file protocol assembly or calls described database component according to the instruction of described request of data and data exchange interface unit:

---if call described file protocol assembly, then described file protocol component call file-storage device, described file-storage device is carried out file and is stored processing, return result to described file protocol assembly then, described file protocol assembly returns result to described file request Agent layer unit;

---if call described database component, then database component is according to the data query of the SQL word string of the include file characteristic query of described file request Agent layer unit and/or write request, call database and carry out query statement, and returning result to described file request Agent layer unit, described file request Agent layer unit returns result to described file request Agent layer unit;

Described request of data and data exchange interface unit are to described script drive controlling assembly return results.

Equally as preferred version, described data analysis processing components is called various data analysis Processing Algorithm the data analysis of neuroelectricity physiologic information is handled, return result to script drive controlling assembly, and submit historical analyzing and processing algorithm and data thereof to, may further comprise the steps:

Described processing request explanation and as a result feedback interface receive the data analysis that described script drive controlling assembly initiates and handle request, described data analysis is handled request analysis for comprising algorithm instruction stream object, data stream object;

Described data input layer assembly receives described processing request explanation and the data stream object that sends of feedback interface as a result,

Described Processing Algorithm trustship layer receives described processing request explanation and the algorithm instruction stream object that sends of feedback interface as a result,

The described Processing Algorithm performance element inquiry of described algorithm trustship course, can handle this algorithm instruction stream object? if, then described Processing Algorithm performance element calls the specific algorithm of this algorithm instruction stream object correspondence and handles from described data input layer assembly acquisition data stream object, and the operation result data are returned described operation result output interface; Described operation result output interface is accepted the output result from described Processing Algorithm performance element, and to described processing request explanation and as a result feedback interface return;

Otherwise, returning " algorithm does not exist " status word to described algorithm trustship layer assembly, described algorithm trustship layer assembly is to described processing request explanation and feedback interface return state word as a result;

Described processing request explanation and as a result feedback interface return result to described script drive controlling assembly.

Equally as preferred version, described data visualization processing components is called various data visualization Processing Algorithm neuroelectricity physiologic information data is carried out visualization processing, return result to script drive controlling assembly, and submit historical visualization processing algorithm and data thereof to, may further comprise the steps:

Described visualization request and result transmit interface and are used to receive the data visualization request of initiating from described script drive controlling assembly,

Call described method for visualizing performance element,

The drafting of described method for visualizing performance element execution result figure, and transmit interface to described visualization request and result and return execution result;

Described visualization request and result transmit interface to described script drive controlling assembly return results.

Based on technique scheme, the present invention has following overall performance characteristics:

(1) multi-channel nerve information acquisition, processing, visualization process are carried out the integrated fusion of modularization, set up unified service interface, allow various concrete grammars when operation, dynamically to be articulated on the three big service interfaces;

(2) native system has been set up the extendible nerve information processing instruction collection of a cover, is used for manipulation data, calls data processing algorithm, starts different method for visualizing and output.This instruction set comprises predefined system directive and user defined commands two parts.System directive comprises the steering order of command stream and built-in instruction.Built-in instruction comprises the call instruction of hardware operation instruction and built-in algorithms.By extendible instruction storage and call-by mechanism, the user can write algorithm voluntarily or articulate the external dynamic chained library, expansion custom instruction collection;

(3) provide one can be used for accepting and execution nerve information processing instruction collection for the processing interface of script driving; Handle neuroelectricity physiology hardware.Script drives and comprises two layers of meaning, and on the one hand, content for script is a series of nerve information processing instruction collection of carrying out successively, and nerve information is handled the interface and accepted these instructions, carries out and give user feedback; On the other hand, the input of script and execution can be interactively inputs successively, also can be to read in the script file that is stored as document form, explain automatically and carry out.

By implementing technique scheme, the present invention can reach following beneficial effect:

(1) overall process that nerve information is handled can be finished in a united frame;

(2) can be under the control of script, the magnanimity multi-channel nerve information processing process of finishing of batch robotization, and provide visualization result;

(3) flow process of neuroelectricity physiologic information data processing can be kept in the storage medium, improve the repeatability and the controllability of data processing;

(4) hardware controls is combined with data processing, can be according to the result of neuroelectricity physiologic information data, experimental program is revised in the control hardware operation dynamically;

(5) by unified interface, extension process and visualized algorithm dynamically.

Description of drawings

Fig. 1 is according to one embodiment of the present of invention, demonstrates the entire block diagram of example system;

Fig. 2 is according to one embodiment of the present of invention, demonstrates the internal operation block diagram of data acquisition interface layer;

Fig. 3 is according to one embodiment of the present of invention, demonstrates the internal operation block diagram of data-processing interface layer;

Fig. 4 is according to one embodiment of the present of invention, demonstrates the internal operation block diagram of data visualization interface layer;

Fig. 5 is according to one embodiment of the present of invention, demonstrates the internal operation block diagram that script drives user interface;

Fig. 6 is according to one embodiment of the present of invention, demonstrates the execution flow process of instructing in the system, and demonstrates a kind of example of the relation of instruction stream and data stream simultaneously;

Fig. 7 is according to one embodiment of the present of invention, demonstrates in the life cycle of a data object process flow diagram that various algorithms and method for visualizing combine with data object.

Specific embodiments

At first with reference to figure 1, block diagram shows an extendible system for comprehensively processing massive multi-channel nerve information 100 according to one embodiment of the present of invention.System 100 includes but not limited to: data are obtained and are handled assembly 101, data analysis processing components 102, data visualization processing components 103, script drive controlling assembly 104.In one embodiment, system 100 is the software systems that are made of one or more assemblies (as: data are obtained and handled assembly 101, data analysis processing components 102, data visualization processing components 103, script drive controlling assembly 104).In another embodiment, system 100 is the integrated software application systems that are made of the one or more application that comprise embodiments of the invention and/or software systems and electric physiology hardware system, as multichannel nerve electric Physiological Experiment system.

Data are obtained and handled assembly 101, data analysis processing components 102, data visualization processing components 103 is service interfaces of each corresponding with service, being unique interface of corresponding with service supplier assembly and system interaction, is the inside executive's interface of the nerve information processing instruction collection that comprises in the script.

Among Fig. 1, script drive controlling assembly 104 and data are obtained and are handled the instruction and data stream 105 that there is request of data in 101 of assemblies and returns.Comprise in the instruction and data stream 105, obtain to data from script drive controlling assembly 104 and handle the request of data instruction that assembly 101 initiates and/or upload data stream accordingly and data are obtained and handled the data download stream that returns after assembly 101 is handled.Request of data instruction includes but not limited to: to the request of the data stream that comprises in the data file, to the storage request of data stream, to the data acquisition request of neuroelectricity physiological signal acquisition hardware, to the stimulus waveform output request of neuroelectricity physiological stimulation hardware etc.

Fig. 2 example among embodiment, data are obtained the component relation block diagram of handling assembly 101 inside.It is mutual with hardware and storage medium that data are obtained the major function of handling assembly 101, finishes data and writes and read request.Request of data instruction 105 from script drive controlling assembly 104 is submitted to request of data and data exchange interface unit 201.The 201 pairs of data request instructions in request of data and data exchange interface unit are analyzed, and judge the target type of its request of data.If the request of data instruction has the hardware operation identifier, request of data explanation and 201 of data exchange interfaces should instruct and/or upload data stream and import hardware requests Agent layer unit 202 into as a hardware instruction target 211.

Hardware requests Agent layer unit 202 has two big functions.On the one hand, it is the level of abstraction of request of data instruction.The instruction of the request of data of hardware includes but not limited to: data collecting card begins to gather, stops to gather, electrophysiological stimulation device stimulus modality is selected, stimulation programs is uploaded, the appointment of stimulus data stream, begin to stimulate, stop to stimulate etc.These hardware requests instructions are included in the request of data instruction stream 105, and are encapsulated as the part of hardware instruction target 211.The hardware requests instruction in the target 211 is accepted in hardware requests Agent layer unit 202, and calls different hardware Control Component 204 according to type of hardware in the hardware operation identifier and model.In one embodiment, type of hardware comprises collecting device (as data acquisition hardware 207) and stimulates output device (as electro photoluminescence hardware 208).Ardware model number is meant concrete unit type, different unit type correspondences different hardware controls assemblies 204.On the other hand, hardware requests Agent layer unit 202 is the hosts that serve of each hardware controls assembly 204.Hardware controls assembly 204 is towards concrete hardware, and the hardware requests instruction 213 that hardware requests Agent layer unit 202 is transmitted converts the control word of corresponding hardware drive program interface to, handles particular hardware work.

In a specific embodiment, when carrying out hardware operation, instruction and data flows the another part that data stream is encapsulated as hardware instruction target 211 of uploading in 105, and submits to concrete hardware drive program by hardware controls assembly 204.

For data acquisition hardware 207, this hardware is not handled and is uploaded data stream, but can return a continuous data stream, this data stream will instruct 213 result to return as hardware requests, and convert a part of return data request explanation and the data exchange interface 201 of hardware instruction target 211 to, finally as the data stream of returning in the instruction and data stream 105.Simultaneously, the hardware controls assembly should return a status word according to the success or not of operation, this status word is also returned as the result of hardware requests instruction 213, belongs to the part of hardware instruction target 211, finally as the instruction executing state word that returns in the instruction and data stream 105.

In a specific embodiment, for electro photoluminescence hardware 208, this hardware will be uploaded data stream and be considered as stimulation programs, but there is not data stream to return, instruction executing state word return data request explanation and data exchange interface 201 are only arranged, and finally as the instruction executing state word that returns in the instruction and data stream 105.In this case, the data stream of returning in the instruction and data stream 105 is for empty.

If the request of data instruction does not have the hardware operation indications, then request of data explanation and data exchange interface 201 will instruct and/or upload the data stream packing and be submitted to file request Agent layer unit 203 for the part of storage system target 212.

File request Agent layer unit 203 is to explain the final exponent who transmits the relevant request of data instruction of next file from script drive controlling assembly 104.The request of data instruction that file is relevant can include but not limited to: the reading and writing data instruction of request of data instruction, the concrete memory location of include file and the file characteristic of the request of data instruction of the concrete memory location of include file, the SQL word string of include file characteristic query.For the request of data instruction of the concrete memory location of include file, file request Agent layer unit 203 sends file request 214 according to the file type identifier in the request of data instruction to different file protocol assemblies 206.File protocol assembly 206 directly to file storage facility 209(as demand file harddisk memory), and according to different file format protocol instrument of interpretation data, data stream is returned to file request Agent layer unit 203 as the result of file request 214, and as a part of return data request explanation and the data exchange interface 201 of storage system target 212, finally as the data stream of returning in the instruction and data stream 105.Simultaneously, file protocol assembly 206 should return a status word according to the success or not of file access, this status word is also returned as the result of file request 214, belongs to the part of storage system target 212, finally as the instruction executing state word that returns in the instruction and data stream 105.

Request of data instruction for the SQL word string of include file characteristic query, the query entries status word that file request Agent layer unit 203 is provided with in the database request 215 is true, and perform statement is set to this SQL word string, at last database request 215 submitted to database component 205.The instruction of the request of data of the SQL word string of include file characteristic query can include but not limited to: the inquiry of carrying out according to the file designation rule, according to the pairing characteristic information of file, the metamessage of data for example, the additional information that comprises data files such as experiment date, experimental implementation people, experiment type, experiment purpose, the employed algorithm types of data processing, these information are not data itself, but the attached situation that data are produced that they can be unique is described out.When inquiring about, database component 205 is submitted to database program 210 with the SQL request.If inquiry obtains the sky result, then database component 205 will be provided with inquiry status of fail word to the return results of database request 215.Accordingly, after file request Agent layer unit 203 receives this status of fail word, the status word that storage system target 212 is set is failure, and 212 data stream is set for empty, and status word and data stream are finally returned as the execution result of instruction and data stream 105.If inquiry obtains one or more file entries, database component 205 returns to file request Agent layer unit 203 with database query result and successful inquiring status word in the lump as the result of database request 215.After database query result comprises that File Agent requesting layer 203 gets access to the file entries title, manner of execution according to " the request of data instruction of the concrete memory location of include file ", initiate request to file protocol assembly 206, this will obtain two results: file request status word and file request data stream.In the lump as the request explanation of execution result return data and the data exchange interface 201 of storage system target 212, finally the execution result as instruction and data stream 105 returns File Agent requesting layer 203 with file request status word, data base querying solicited status word and file request data stream.

For the reading and writing data instruction of concrete memory location of include file and file characteristic, the interpolation entry state word that file request Agent layer unit 203 is provided with in the database request 215 is true, and perform statement is arranged to the file characteristic word string.At last database request 215 is submitted to database component 205.Whether database component 205 exists same directory entry at first according to the file characteristic keyword in the database request 215 to database program 210 inquiries.If exist, then database component 205 adds the status of fail word with the return results of database request 215 is provided with.Accordingly, after file request Agent layer unit 203 receives this status of fail word, the status word that storage system target 212 is set is failure, and 212 data stream is set for empty, and status word and data stream are finally returned as the execution result of instruction and data stream 105.If do not have same directory entry in the database, then database component 205 will store the entry file characteristic in the database into, and write executing state according to the operation result return data storehouse of database program 210 reality.When database component 205 sends request, file request Agent layer unit 203 will be included in the storage system target 212 upload data stream and file protocol assembly 206 is submitted in the concrete memory location of file.File protocol assembly 206 writes disk file with data stream, and backspace file writes executing state.File writes executing state and database writes request of status word return data and the data exchange interface unit 201 that executing state is used as storage system target 212, and finally the execution result as instruction and data stream 105 returns.

Fig. 3 example among embodiment, the component relation block diagram of the inside of data analysis processing components 102.The major function of data handling component is on the one hand to the application programming interface of various data processing algorithm exposure systems, these algorithms to be articulated in the workflow of nerve information processing; On the other hand, submit to the tabulation of existing algorithm to call to script drive controlling assembly 104 for the user.In the specific embodiment of the present invention, data and data processing algorithm are loosely-coupled, and this is defined as: the user can dynamically be articulated to various data processing algorithms on the data stream according to demand and handle.The processing history that the Processing Algorithm of while data stream warp is used as data is kept in the mark sheet of data file.

Being submitted to the request of processing from the processing request instruction 106 of script drive controlling assembly 104 explains and feedback interface 301 as a result.The request of handling is explained with feedback interface 301 as a result will be submitted to data input layer assembly 302 as internal data flow target 306 from the data stream of script drive controlling assembly 104, and instruction stream is submitted to Processing Algorithm trustship layer assembly 303 as built-in command stream target 307.

Data input layer assembly 302 can carry out necessary buffer memory to data, and provide request of data passage 309 to concrete Processing Algorithm performance element 305,309 pairs of Processing Algorithm performance elements 305 of request of data passage have shielded the details that data are obtained, only expose the essential attribute of data, for example: channel number, acquisition time axle, event information etc.Can obtain data by request of data passage 309 to data input layer assembly 302 at any time during 305 operations of Processing Algorithm performance element.In a specific embodiment, a Processing Algorithm performance element 305 may need a plurality of data stream.For the situation of a plurality of data stream, the number of data stream should be provided in script drive controlling assembly 104 by the user, handles to have included the address of pointing to desired data stream in the request instruction 106.Data input layer assembly 302 is not initiated new request of data.Need to prove that the responsibility of data input layer assembly 302 is to provide data stream to Processing Algorithm performance element 305, therefore be not limited among other the embodiment that data input layer assembly 302 also can be initiated new request of data.

Processing Algorithm trustship layer assembly 303 is Processing Algorithm performance elements 305 and the mutual key of system 100.In one embodiment, may have N concrete Processing Algorithm assembly 305, these algorithms are stored in the specified file system position according to predefined form.When system 100 started, Processing Algorithm trustship layer scanned the specified file system position, obtains the document location and the call format of each Processing Algorithm assembly 305.In other embodiment, can Processing Algorithm be registered to Processing Algorithm trustship layer assembly 303 by other modes.

After script drive controlling assembly 104 will be handled request and be submitted to the request handled and feedback interface 301 as a result, this interface will be handled algorithm in the request and instruct and further be submitted to Processing Algorithm trustship layer assembly 303.The registered algorithm routine of Processing Algorithm trustship layer assembly 303 inquiries, successful inquiring is then called this Processing Algorithm, and control is transferred to this algorithm, otherwise returns " algorithm does not exist " status word.

Processing Algorithm performance element 305 obtains the computational data of input by data Layer 302 after obtaining the control that Processing Algorithm trustship layer assembly 303 delivers; The Processing Algorithm end of run will return running status word (as: success, failure, failure code, failure cause) to Processing Algorithm trustship layer assembly 303, and operation result is by 304 outputs of operation result output interface.Operation result output interface 304 is accepted the output result 311 from Processing Algorithm on the one hand, and this result is carried out possible buffer memory (for example: can be buffered in the disk file, also can be buffered in the zone of internal memory appointment); On the other hand, resultant layer 304 is explained the pointer 308 that an operation result is provided with feedback interface 301 as a result to the request handled.The request of handling is explained with feedback interface 301 as a result Processing Algorithm performance element 305 running status words and operating result pointer 308 is returned script drive controlling assembly 104 as the operation result of processing request instruction 106 in the lump.

At certain concrete algorithm 305, data analysis processing components 102 can support its various different languages to realize.In one embodiment, algorithm 305 can use the C/C++ language compilation and be compiled into executable program, is finally called by Processing Algorithm trustship layer assembly 303.In another embodiment, algorithm 305 can use engineering computing language (as Matlab) to write, and moves in hosting environment, is called by Processing Algorithm trustship layer assembly 303 simultaneously.In other embodiment, various algorithm or the executable programs that utilize other script (as Python) to write of following the function definition of algorithm trustship layer can be registered in the Processing Algorithm trustship layer assembly 303 equally, and are carried out by the user by script drive controlling assembly 104.

Fig. 4 example among embodiment, the component relation block diagram of the inside of data visualization processing components 103.The major function of data visualization processing components 103 is on the one hand to the application programming interface of various data visualization algorithm exposure systems, these method for visualizing to be articulated in the workflow of nerve information processing; On the other hand, submit to the tabulation of existing method for visualizing to call to script drive controlling assembly 104 for the user.In the specific embodiment of the present invention, data processed result and data visualization method are loosely-coupled equally, and this is defined as: the user can dynamically be articulated to various data visualization methods on the data stream according to demand and handle.The processing history that the method for visualizing of while data stream warp is used as data is kept in the mark sheet of data file.

Be submitted to visualization request and the result transmits interface 401 from the data visualization request 107 of script drive controlling assembly 104.Data visualization request 107 comprises visual instruction and the visual data stream of needs.Each visual instruction corresponding a method for visualizing performance element 402.In one embodiment, may have N concrete method for visualizing performance element 402, these assemblies are stored in the specified file system position according to predefined form.When system 100 started, visualization request and result transmitted 401 pairs of specified file system positions of interface and scan, and obtain the document location and the call format of each method for visualizing performance element 402.In other embodiment, can method for visualizing 402 be registered to visualization request and the result transmits interface 401 by other modes.

After transmitting the order parameter of interface and data stream 403 and submit to method for visualizing performance element 402 from visualization request and result, method for visualizing performance element 402 is finished the drafting of figure as a result, can be under the file path of user's appointment with graphics memory, also can transmit interface and return a figure handle, be convenient to figure and raw data are stored in the database in the lump to visualization request and result.

In one embodiment, method for visualizing performance element 402 can be realized with different program languages.Data visualization processing components 103 can convert the request to method for visualizing performance element 402 to the band parameter call to the executable program of C/C++ language, and various engineering computing language (as: Matlab) and script (as: Python, Lua etc.).

Among Fig. 5 example among embodiment, the component relation block diagram of the inside of script drive controlling assembly 104.The major function of script drive controlling assembly 104 comprises: at first, script drive controlling assembly 104 is interfaces of system 100 and user interactions.All instructions that the user provides are all sent to script drive controlling assembly 104 with the form of script command, and script drive controlling assembly 104 fill orders also provide feedback to the user; Secondly, script drive controlling assembly 104 is keys of internal system Workflow Delegation.The nerve information treatment scheme has strict time sequencing, and the script interactive instruction can be according to the given instruction of user, and the working time and the processor distribution of planning instruction are supported the parallel computation of multiprocessor; The 3rd, at nerve information computing length consuming time, the long-time unattended characteristics of possibility, 104 pairs of execution process instruction monitoring of script drive controlling assembly, before each bar instruction of operation, all preserved ambient condition, the accident collapse of the failure of a certain instruction or system 100 does not influence the continuation of whole workflow to be carried out, and system 100 can recover to carry out from the interruptions of certain bar instruction.

Cmd Shell and script file interface 501 are shells of command interpretation and result's feedback, and it accepts the method for operation of two kinds of forms.(1) interactive mode.The user once imports an order, and system provides feedback immediately according to the implementation status of specific instructions; (2) script interpreter.Various neural processing commands are stored in the file system in the mode of script file, and the disposable submission of user provides feedback after system is complete.Instruction set comprises system directive resolution unit 502 and user instruction resolution unit 503.System directive resolution unit 502 is to realize Cmd Shell and embedded various instructions and the interface of script file interface 501 employed program languages in the specific embodiment.Among the embodiment, use Python to write Cmd Shell and script file interface 501, then the system directive collection then comprises all key words of Python.User instruction resolution unit 503 is to obtain the mapping of handling each specific procedure of definition in assembly 101, data analysis processing components 102, the data visualization processing components 103 in data.System directive resolution unit 502 is with command line interface 501 initialization when starting.During system start-up, transmit the various Processing Algorithm of interface 401 scannings by the Processing Algorithm trustship layer assembly 303 in the data analysis processing components 102 and the visualization request in the data visualization processing components 103 and result, finish the initialization of user instruction resolution unit 503.

The instruction calls of Cmd Shell and script file interface 501 (comprising system directive request 507 and user instruction request 509) is the parsing by system directive resolution unit 502 and user instruction resolution unit 503 at first, be transformed into system call, enter task execution time planning unit 504.In task execution time planning unit 504, system directive is called 510 execution priority and is higher than user instruction and calls 511 priority.There are work planning instruction stream 508 in 504 of Cmd Shell and script file interface 501 and task execution time planning units, comprise in the work planning instruction stream that execution precedence adjustment instruction, the workflow that Cmd Shell interface 501 is initiated begins, time-out and command for stopping.Task execution time planning unit 504 is carried out according to the priority arrangement instruction of instruction, and the instruction that priority is identical is not if exist the mutual exclusion status word, and then instruction is arranged on a plurality of processing units and carries out simultaneously.

Behind the instruction stream process task execution time planning unit 504, be converted to concrete execution stream 512, enter task status and preserve and monitoring unit 505.Each instruction of carrying out in the stream 512 has been endowed the unique state object in task status preservation and supervisory layers.The instruction object 513 that has status object is submitted to task executive's interface unit 506, and be recorded in simultaneously in the journal file, task executive's interface unit 506 calls data respectively and obtains and handle assembly 101, data analysis processing components 102, data visualization processing components 103 and finish the work.After data are obtained and are handled assembly 101, data analysis processing components 102, data visualization processing components 103 and carry out data stream that the back produces and status word and turn back to task executive's interface unit 506, be stored in the status object of every instruction correspondence, and in journal file, write down a status word.In the journal file, the instruction that unwritten instruction object is regarded as not being performed, only have instruction record not have the instruction object of executing state word to be regarded as the wrong instruction that meets accident in the implementation, the executing state word for the instruction record of failure be regarded as carrying out in because of condition satisfies or parameter error withdraws from voluntarily instruction.Task status preserve and monitoring unit 505 before instruction of each run, all daily record is checked, after the instruction of the status word of finding not run succeeded, will initiate the request that reschedule to task execution time planning unit 504, guarantee the execution once more of ordering.

Below in conjunction with Fig. 6 and an embodiment instruction execution flow of the present invention is described further.In the present invention, system 100 relies on the script of user's input to drive.The user imports directive script in script interactive interface 104.This instruction at first is converted to the built-in command sequence by system directive collection and user instruction collection, and the sequential process that these built-in command sequences are carried out successively according to section planning task time 504 is defined as instruction stream.The execution of instruction sequence will be accompanied by establishment, modification, the preservation process of data object.In the execution of instruction sequence, task status is preserved and monitor component 505 will be regularly to the executing state poll, and stores the executing state in each step.Instruction sequence carry out to finish, the execution result of data stream is taken out and the executing state word that adds self as instruction execution result.

The execution request that data object should instruct and creating, the Data Source that needs when data object is created can include but not limited to data acquisition hardware object 207, electro photoluminescence item of hardware 208, file storage object 209, database object 210.After data object is created, the execution request that should instruct and constantly change, the data that wherein comprise are revised.After instruction was complete, data object can be kept in file storage object 209 and the database object 210, as permanent reading and writing data; Also can be kept in the electro photoluminescence item of hardware 208, as the control signal of electrophysiological stimulation device.

In conjunction with Fig. 7 and an embodiment, the dynamic cohesive process of data stream and algorithm, method for visualizing is described further.The overall process that structure was created, handles, preserves and analysed to data object is defined as the life cycle of data object, and a plurality of data objects process that is created, revises, preserves and analyse structure parallel or serial can be considered a data stream.In a data object, when data object is created, automatically generating according to copy 0, is wherein being comprised all data of this data object.By directive script, arbitrary Processing Algorithm in the system can be combined with data trnascription.Data in the data trnascription are as the input of Processing Algorithm, and the output of Processing Algorithm will be created new data trnascription automatically.In a specific embodiment, data trnascription and data object are indiscriminate, and the establishment of data trnascription is equal to the establishment of new data-objects.In the opposing party's specific embodiment, data object inside can comprise a plurality of data trnascriptions, only just creates when the instruction explicitly requires to create new data object.The mutual relationship of the present invention and two or more data objects of unspecific qualification and data trnascription.By directive script, arbitrary Processing Algorithm in the system can be combined with data trnascription, the data in the data trnascription are as the input of method for visualizing.Method for visualizing will return the drawing handle and carry out drafting in screen or the file, also draw data return data object together can be stored with data trnascription.Among the present invention, the sequence of in store Processing Algorithm and method for visualizing in the data object, that is: for data trnascription 1, data object is with the relevant execution sequence information of the Processing Algorithm 1 of in store generation data trnascription 1.This is because data trnascription 1 can repeat by the operation once more of 1 pair of data copy 0 of Processing Algorithm to obtain, as long as preserve the corresponding information of each Processing Algorithm successively, just can reappear entire stream.For the method for visualizing sequence, data object will be preserved the relevant execution sequence information of method for visualizing 1 for data trnascription 0.As long as this is because by repeating the execution sequence of method for visualizing 1 on data trnascription 0, just can reappear the visualization result of being correlated with.The present invention has guaranteed the high performance reproducibility in data processing and the data visualization process by said method.

In view of above content, can see, realize some targets of the present invention and obtained other favourable outcomes.

When programming according to method described herein and technology, the present invention also comprises computing machine and electric physiology hardware itself.For purposes of illustration, operating system and other executable program components are illustrated as the frame that disperses herein, but persons skilled in the art can recognize, these programs and assembly and have the data processor of computing machine to carry out in the different memory modules of different moment main flows at computing machine.Although combining the exemplary computer system environment that comprises multi-purpose computer and neuroelectricity physiological data acquisition hardware 207, electro photoluminescence hardware 208 herein describes.Yet the present invention can operate with numerous other universal or special computing environment, and the computingasystem environment that provides in the example is not intended to usable range of the present invention or function are proposed any limitation.In addition, this computingasystem environment should not be interpreted as that any or its combination of assembly shown in the exemplary operation environment are had any dependence or requirement.The example that is applicable to known computing system of the present invention, environment and/or configuration includes but not limited to, personal computer, server computer, multiprocessing system, the system based on microprocessor, network PC, minicomputer, large scale computer, comprises distributed computing environment of arbitrary said system or equipment or the like.

The present invention can describe in the general context environmental of the computer executable instructions such as program module that one or more computing machines or other equipment are carried out.Usually, program module includes but not limited to, routine, program, object, assembly, data structure or the like, and they are carried out specific task or realize specific abstract data type.The present invention also can put into practice in distributed computing environment.When in distributed computing environment, task is carried out by the teleprocessing equipment that communication network connects, and program module can be arranged in the local and remote computer-readable storage medium that comprises storer.

Interface layer in the software architecture environment comprises software module, assembly, code section or other sequence of computer-executable instructions.Interface comprises: for example visit second module and represent first module to carry out first module of calculation task.In one example, first and second modules can be the application programming interface, The Component Object Model interface, interprocess communication interface and the extend markup language metadata Interchange Format interface that provide by operating system (as, when being used for Web service communication).Interface can be for example tightly coupled synchronous realization among .Net Frameworks, Java, the C++.Perhaps in addition, interface also can be the asynchronous realization of the loose coupling in for example Web service.Usually, interface comprises arbitrary combination of following feature: tightly coupled, loose coupling, synchronous and asynchronous.In addition, interface can the conformance with standard agreement, any combination of proprietary protocol or standard and proprietary protocol.Interface described herein can all be the part of individual interface, perhaps may be implemented as independent interface or any combination wherein.Interface can be carried out so that function to be provided on Local or Remote ground.In addition, interface can comprise than shown here or described more or less function.

Execution or realization the order shown here and method described are dispensable, unless otherwise.That is, the element of method can be carried out with any order, and unless otherwise, and method can comprise the more or less element that discloses than this place.Can think, before another element, carry out simultaneously or afterwards with it or realize that element-specific is within the scope of the present invention.

When introducing the element of the present invention or embodiment, article " ", " one ", " being somebody's turn to do " and " described " are meant and have one or more elements." comprise ", " comprising ", " having " be intended to inclusive, and means that except that listed element other elements can also be arranged.

Because various changes can made without departing from the scope of the invention, be included in therefore that above-mentioned instructions and all the elements illustrated in the accompanying drawings all should be interpreted as illustrative and nonrestrictive in the said goods and method.

It should be noted last that, above embodiment is only unrestricted in order to technical scheme of the present invention to be described, although the present invention is had been described in detail with reference to preferred embodiment, those of ordinary skill in the art is to be understood that, can make amendment or be equal to replacement technical scheme of the present invention, and not breaking away from the spirit and scope of technical solution of the present invention, it all should be encompassed in the middle of the claim scope of the present invention.

Claims

1. a system for comprehensively processing massive multi-channel nerve information is characterized in that, comprises that script drive controlling assembly, data are obtained to handle assembly, data analysis processing components, data visualization processing components;

2. system for comprehensively processing massive multi-channel nerve information according to claim 1, it is characterized in that, described script drive controlling assembly comprises: Cmd Shell and script file interface unit, the system directive resolution unit, the user instruction resolution unit, the task execution time planning unit, task status is preserved and monitoring unit, task executive's interface unit

3. system for comprehensively processing massive multi-channel nerve information according to claim 2, it is characterized in that, described data are obtained the manipulation assembly and are comprised: request of data and data exchange interface unit, hardware requests Agent layer unit, the hardware controls assembly, file request Agent layer unit, file protocol assembly, database component;

Described request of data and data exchange interface unit, be used to receive the data processing request that script drive controlling assembly is initiated, the target type of judging described data processing request is operation of neuroelectricity Physiological Experiment or data read-write operation, if neuroelectricity Physiological Experiment operation requests is then called described hardware requests Agent layer unit; If file request Agent layer unit is called in the data read-write operation request, finally to described script drive controlling assembly return results;

Described file protocol assembly is used for directly calling file read-write equipment and carries out file read-write and handle, and to himself caller return results.

4. system for comprehensively processing massive multi-channel nerve information according to claim 3, it is characterized in that, described data analysis processing components comprises that the request handled explains and feedback interface as a result, data input layer assembly, Processing Algorithm trustship layer assembly, the operation result output interface, the Processing Algorithm performance element;

5. system for comprehensively processing massive multi-channel nerve information according to claim 4 is characterized in that, described data visualization processing components comprises that visualization request and result transmit interface, the method for visualizing performance element,

6. the control method of a system for comprehensively processing massive multi-channel nerve information according to claim 5 is characterized in that may further comprise the steps:

Load described script drive controlling assembly;

Move described script drive controlling assembly;

7. the control method of system for comprehensively processing massive multi-channel nerve information according to claim 6 is characterized in that, described loading script drive controlling assembly may further comprise the steps:

Load described Cmd Shell and script file interface unit,

Load described task execution time planning unit,

Load described task status and preserve and monitoring unit,

Load described task executive's interface unit.

8. the control method of system for comprehensively processing massive multi-channel nerve information according to claim 7 is characterized in that, the described order line script of described script drive controlling analyzing component generates the work order sequence, may further comprise the steps:

Described Cmd Shell and script file interface unit receive user command,

Whether the identification user command is the work planning instruction,

If not the work planning instruction then is committed to described system directive resolution unit with inoperative planning instruction; Described system directive resolution unit receives and identifies inoperative planning instruction, judges whether to be system directive, if system directive is then called described system directive generation system order set and is committed to described task execution time planning unit;

9. the control method of system for comprehensively processing massive multi-channel nerve information according to claim 8, it is characterized in that, described data are obtained the manipulation assembly and are called according to described work order, drive described neuroelectricity Physiological Experiment system acting, and receive the return data of described neuroelectricity Physiological Experiment system; Perhaps, control described data read-write system and carry out reading and writing data; Then, return result, may further comprise the steps to script drive controlling assembly:

If call described file protocol assembly, then described file protocol component call file-storage device, described file-storage device is carried out file and is stored processing, return result to described file protocol assembly then, described file protocol assembly returns result to described file request Agent layer unit;

If call described database component, then database component is according to the data query of the SQL word string of the include file characteristic query of described file request Agent layer unit and/or write request, call database and carry out query statement, and returning result to described file request Agent layer unit, described file request Agent layer unit returns result to described file request Agent layer unit;

10. the control method of system for comprehensively processing massive multi-channel nerve information according to claim 9, it is characterized in that, described data analysis processing components is called various data analysis Processing Algorithm the data analysis of neuroelectricity physiologic information is handled, return result to script drive controlling assembly, and submit historical analyzing and processing algorithm and data thereof to, may further comprise the steps:

The described Processing Algorithm performance element inquiry of described algorithm trustship course, can handle this algorithm instruction stream object, if, then described Processing Algorithm performance element calls the specific algorithm of this algorithm instruction stream object correspondence and handles from described data input layer assembly acquisition data stream object, and the operation result data are returned described operation result output interface; Described operation result output interface is accepted the output result from described Processing Algorithm performance element, and to described processing request explanation and as a result feedback interface return;

11. the control method of system for comprehensively processing massive multi-channel nerve information according to claim 10, it is characterized in that, described data visualization processing components is called various data visualization Processing Algorithm neuroelectricity physiologic information data is carried out visualization processing, return result to script drive controlling assembly, and submit historical visualization processing algorithm and data thereof to, may further comprise the steps:

Call described method for visualizing performance element,