CN114521868B - Pulse signal classification device and wearable equipment - Google Patents
Pulse signal classification device and wearable equipment Download PDFInfo
- Publication number
- CN114521868B CN114521868B CN202210172964.9A CN202210172964A CN114521868B CN 114521868 B CN114521868 B CN 114521868B CN 202210172964 A CN202210172964 A CN 202210172964A CN 114521868 B CN114521868 B CN 114521868B
- Authority
- CN
- China
- Prior art keywords
- pulse
- action potential
- classification
- signals
- average value
- 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
- 230000036982 action potential Effects 0.000 claims abstract description 151
- 238000000034 method Methods 0.000 claims abstract description 29
- 238000012545 processing Methods 0.000 claims abstract description 27
- 238000001514 detection method Methods 0.000 claims abstract description 18
- 230000006870 function Effects 0.000 claims description 102
- 230000000875 corresponding effect Effects 0.000 claims description 59
- 230000008569 process Effects 0.000 claims description 17
- 239000012634 fragment Substances 0.000 claims description 16
- 230000006386 memory function Effects 0.000 claims description 11
- 230000001276 controlling effect Effects 0.000 claims description 8
- 210000004556 brain Anatomy 0.000 claims description 6
- 238000012544 monitoring process Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 3
- 230000001960 triggered effect Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 14
- 230000005540 biological transmission Effects 0.000 description 5
- 238000004590 computer program Methods 0.000 description 5
- 210000002569 neuron Anatomy 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000000835 fiber Substances 0.000 description 2
- 230000001537 neural effect Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000001902 propagating effect Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 210000000653 nervous system Anatomy 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/369—Electroencephalography [EEG]
- A61B5/372—Analysis of electroencephalograms
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/369—Electroencephalography [EEG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
Abstract
The present disclosure relates to a pulse signal classification device, a wearable apparatus, the device comprising: the pulse detection module stores the received bioelectric signal in a buffer memory area, generates a pulse existence signal which is obtained by intercepting the bioelectric signal and corresponds to an input action potential segment of the pulse peak under the condition that the pulse peak exists in the bioelectric signal, and sends the pulse existence signal to the pulse alignment module; the pulse alignment module performs alignment processing on the received pulse existence signals and sends the obtained processed pulse existence signals to the pulse clustering module; the pulse clustering module performs clustering determination on the received processed pulse existence signals and then performs pulse classification identification corresponding to the processed pulse existence signals. The method can be used for rapidly classifying and processing the input action potential of the bioelectricity signals acquired by the sensor in a highly parallel and event-triggered working mode in real time, and can be used for remarkably improving the real-time property of classification and reducing the running power consumption.
Description
Technical Field
The disclosure relates to the technical field of signal processing, in particular to a pulse signal classification device and a wearable device.
Background
Brain-computer interface (BCI) is a technology for realizing information interaction and function integration between a nervous system and external devices (such as a computer, a robot, etc.) by establishing a direct connection path between a human Brain nerve and the external devices. An important step in brain-computer interfaces is the acquisition and transmission of neural activity in real time through various sensors. The extracellular recording technology in neuroscience is a technology of implanting electrodes into extracellular tissues of brain neurons to record the activities of individual neurons, and is a common method for researching brain operations. While each neuron of an organism tends to produce a pulse of a specific shape, also known as an action potential. The process of classifying neural activity of different neurons is called impulse classification. In practice, however, the data collected in the sensor electrodes typically includes a plurality of surrounding neuron electrical activity as well as ambient noise. In the related art, an analog signal acquired through a sensor electrode is converted into a digital form through analog-to-digital conversion and transmitted to an external computing unit for recording and offline processing. Although high-accuracy signal processing analysis is performed by using a computationally intensive algorithm, the method has a large processing delay and limits application in wearable equipment with relatively high real-time performance. How to quickly realize pulse classification and reduce delay is a technical problem to be solved.
Disclosure of Invention
In view of this, the present disclosure proposes a pulse signal classification device, a wearable device
According to an aspect of the present disclosure, there is provided a pulse signal classifying apparatus, the apparatus including:
the pulse detection module is used for storing the received continuous bioelectric signals sent by the sensor into a buffer memory area, generating a pulse existence signal under the condition that the pulse peak exists in the bioelectric signals, and sending the pulse existence signal to the pulse alignment module, wherein the pulse existence signal comprises an input action potential segment corresponding to the pulse peak obtained by intercepting the bioelectric signals;
the pulse alignment module is used for performing alignment processing on the received pulse existence signals to obtain processed pulse existence signals and sending the processed pulse existence signals to the pulse clustering module;
the pulse clustering module is used for clustering the received processed pulse existence signals, determining pulse classification identifiers corresponding to the processed pulse existence signals according to clustering results, and the pulse classification identifiers are used for indicating action potential classifications corresponding to the input action potential fragments.
In one possible implementation, the pulse clustering module includes: the system comprises a plurality of classification storage function cores, a feature comparison function core and a classification merging function core, wherein action potential classifications corresponding to the classification storage function cores are different;
the feature comparison function core is used for calculating an input action potential average value vector according to the received input action potential fragments in the processed pulse existence signal, determining a pulse classification identifier corresponding to the processed pulse existence signal according to the difference between the input action potential average value vector and action potential feature average values of different categories, and sending the input action potential fragments to the corresponding classification storage function core;
each classification storage function core is used for counting and storing received input action potential fragments to obtain a counting result, calculating a new action potential characteristic average value according to the stored input action potential fragments under the condition that the counting result meets a characteristic value updating condition, and sending the new action potential characteristic average value to the characteristic comparison function core and the classification merging function core to update as a new action potential characteristic average value of a corresponding action potential classification;
the classification and merger function core is configured to perform merger processing of at least two merger function cores under the condition that at least two merger function cores satisfying a classification update condition are determined from a plurality of classification and storage function cores according to a difference between current action potential feature averages of each classification and storage function core.
In one possible implementation manner, determining, according to a difference between the input action potential average value vector and action potential feature averages of different categories, a pulse classification identifier corresponding to the processed pulse presence signal includes:
calculating the arithmetic square root of the difference between the input action potential average value vector and the action potential characteristic average value of each classification to obtain a plurality of first arithmetic square roots corresponding to the input action potential segments;
determining an action potential class corresponding to the minimum first arithmetic square root as an action potential class corresponding to the input action potential segment when the minimum first arithmetic square root of the plurality of first arithmetic square roots is less than or equal to a preset comparison threshold;
and determining a pulse classification identifier corresponding to the processed pulse existence signal according to the action potential classification corresponding to the input action potential segment.
In one possible implementation manner, in a case where at least two combinable class storage function cores satisfying a class update condition are determined from a plurality of class storage function cores according to a difference between current action potential feature averages of each class storage function core, performing a combining process of the at least two combinable class storage function cores, including:
when a new action potential characteristic average value is received, calculating the arithmetic square root of the difference between the new action potential characteristic average value and the action potential characteristic average values of other classifications, and obtaining a plurality of second arithmetic square roots corresponding to the new action potential characteristic average value;
determining a first classification memory function core corresponding to the new action potential characteristic average value and a second classification memory function core corresponding to the minimum second arithmetic square root as combinable classification memory function cores when the minimum second arithmetic square root of the plurality of second arithmetic square roots is less than or equal to a combining threshold;
and controlling the first classified storage function core to send the stored input action potential fragments to the second classified storage function core, and controlling the first classified storage function core to stop working.
In one possible implementation manner, the alignment process includes a normalization process, and the input action potential segments in each of the processed pulse existence signals have a uniform characteristic value, where the characteristic value includes any one of the following: amplitude, energy, maximum slope.
In one possible implementation, the apparatus includes a many-core brain architecture chip, the pulse detection module and the pulse alignment module are external interface modules of the chip, and the pulse clustering module is a plurality of functional cores of the chip.
In one possible implementation, the apparatus further includes:
and the at least one sensor is used for monitoring organisms, acquiring bioelectric signals and sending the detected real-time bioelectric signals to the pulse detection module.
According to another aspect of the present disclosure, there is provided a wearable device, comprising:
the pulse signal classification device with the sensor.
According to another aspect of the present disclosure, there is provided a wearable device, comprising:
the at least one sensor is used for monitoring organisms, acquiring bioelectric signals and sending the detected real-time bioelectric signals to the pulse detection module;
the pulse signal classifying device.
The pulse signal classifying device and the wearable equipment provided by the disclosure can rapidly classify and process the bioelectricity signals collected by the sensor in real time in a highly parallel and event-triggered working mode, and can remarkably improve the classified instantaneity and reduce the running power consumption.
Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments, features and aspects of the present disclosure and together with the description, serve to explain the principles of the disclosure.
Fig. 1 and 2 are block diagrams showing a structure of a pulse signal classifying apparatus according to an embodiment of the present disclosure.
Fig. 3 shows a block diagram of the structure of a pulse clustering module according to an embodiment of the disclosure.
Detailed Description
Various exemplary embodiments, features and aspects of the disclosure will be described in detail below with reference to the drawings. In the drawings, like reference numbers indicate identical or functionally similar elements. Although various aspects of the embodiments are illustrated in the accompanying drawings, the drawings are not necessarily drawn to scale unless specifically indicated.
The word "exemplary" is used herein to mean "serving as an example, embodiment, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.
In addition, numerous specific details are set forth in the following detailed description in order to provide a better understanding of the present disclosure. It will be understood by those skilled in the art that the present disclosure may be practiced without some of these specific details. In some instances, methods, means, elements, and circuits well known to those skilled in the art have not been described in detail in order not to obscure the present disclosure.
Fig. 1 and 2 are block diagrams showing a structure of a pulse signal classifying apparatus according to an embodiment of the present disclosure. As shown in fig. 1, the apparatus includes: a pulse detection module 10, a pulse alignment module 20, and a pulse clustering module 30.
The pulse detection module 10 stores the received continuous bioelectric signals sent by the sensor in a buffer area, generates a pulse existence signal under the condition that pulse peaks exist in the bioelectric signals, and sends the pulse existence signal to the pulse alignment module 20, wherein the pulse existence signal comprises input action potential fragments corresponding to the pulse peaks obtained by intercepting the bioelectric signals.
In this embodiment, the buffer of the pulse detection module 10 may have a set length, and the pulse detection module 10 may perform pulse spike detection once each time the buffer is full, and empty the buffer after the detection is completed. In some embodiments, the input action potential segment may be an action potential vector comprising a pulse spike action. In this way, the positions of the pulse peaks in the input action potential segments can be the same, so that deviation caused by misalignment of the pulse peak positions in the subsequent pulse clustering module 30 clustering process is avoided.
The pulse alignment module 20 performs alignment processing on the received pulse presence signal to obtain a processed pulse presence signal, and sends the processed pulse presence signal to the pulse clustering module 30.
In one possible implementation, the alignment process may include a normalization process, and the input action potential segments in each of the processed pulse presence signals have a uniform characteristic value, where the characteristic value may include any one of the following: amplitude, energy, maximum slope. In this way, the input action potential segments have uniform characteristic values, so that the pulse clustering module 30 can cluster through the characteristic values of the input action potential segments to obtain the pulse classification identifiers corresponding to the pulse existence signals.
The pulse clustering module 30 is configured to cluster the received processed pulse presence signals, and determine, according to a clustering result, a pulse classification identifier corresponding to the processed pulse presence signals, where the pulse classification identifier is used to indicate an action potential classification corresponding to the input action potential segment.
In this embodiment, the pulse classification identifiers of different action potential classifications are different, and the pulse classification identifiers corresponding to different action potential classifications can be set by using symbols such as numerals and letters, or other identifiers, which is not limited in this disclosure.
In one possible implementation, as shown in fig. 2, the pulse clustering module 30 may include a plurality of functional cores 300, and the functional cores 300 may be set as a class storage functional core, a feature comparison functional core, and a class merge functional core, as desired. Wherein each functional core 300 includes a storage unit for data storage, a calculation unit for data operation, and a route for inter-core communication.
In this implementation manner, the apparatus may be implemented based on a many-core brain architecture chip, where the pulse detection module and the pulse alignment module are external interface modules of the chip, and the pulse clustering module is a plurality of functional cores of the chip. The functional cores can be triggered and processed in parallel according to the events, so that the real-time and synchronous performance of pulse classification, action potential characteristic average value updating and similar classification merging is ensured, and the accuracy and the real-time performance of classification are improved while the efficiency is ensured.
Fig. 3 shows a block diagram of the structure of a pulse clustering module according to an embodiment of the disclosure. In one possible implementation, as shown in fig. 3, the multiple function cores 300 may be respectively set according to a setting, so that the pulse clustering module 30 may include multiple classification storage function cores 301, a feature comparison function core 302, and a classification merge function core 303. As shown in fig. 3, the action potential classifications corresponding to the classification memory function cores 301 are different, and the action potential classifications corresponding to the classification memory function cores 301 are respectively classified into classification 1, classification 2, classification 3 and …, n being a positive integer.
The feature comparison function core 302 is configured to calculate an input action potential average value vector according to the received input action potential segment in the processed pulse presence signal, determine a pulse classification identifier corresponding to the processed pulse presence signal according to a difference between the input action potential average value vector and action potential feature averages of different classes, and send the input action potential segment to the corresponding classification storage function core 301.
In one possible implementation manner, determining the pulse classification identifier corresponding to the processed pulse existing signal according to the difference between the input action potential average value vector and the action potential characteristic average value of different classifications may include: calculating the arithmetic square root of the difference between the input action potential average value vector and the action potential characteristic average value of each category to obtain a plurality of first arithmetic square roots corresponding to the input action potential segments; determining an action potential class corresponding to the minimum first arithmetic square root as an action potential class corresponding to the input action potential segment when the minimum first arithmetic square root of the plurality of first arithmetic square roots is less than or equal to a preset comparison threshold; and determining a pulse classification identifier corresponding to the processed pulse existence signal according to the action potential classification corresponding to the input action potential segment.
For example, assuming that there are n action potential classifications, the first arithmetic square root of the difference between the input action potential average value vector of one input action potential segment and the action potential feature average value of the action potential classification of the i-th class is A1i, and if the minimum value of a11, a12, a13 … A1n is a13 and a13 is smaller than the preset comparison threshold value, it is possible to determine that the action potential classification corresponding to the input action potential segment is the 3-th class action potential classification. And then according to the corresponding relation between the action potential classification and the pulse classification mark, the mark of the pulse classification corresponding to the input action potential segment is determined to be the mark of the type 3 action potential classification, namely, the mark of the pulse classification corresponding to the processed pulse existence signal comprising the input action potential segment is determined to be the mark of the type 3 action potential classification.
Each of the classification and storage function cores 301 is configured to count and store received input action potential segments to obtain a count result, calculate a new action potential feature average value according to the stored input action potential segments when the count result meets a feature value update condition, and send the new action potential feature average value to the feature comparison function core 302 and the classification and combination function core 303 as a new action potential feature average value of a corresponding action potential classification for updating.
In this implementation manner, the feature value update condition may be set according to actual needs, for example, the feature value update condition may be that the count result is greater than or equal to a preset count value. The smaller the preset count value is, the higher the update frequency of the action potential characteristic average value is, and the higher the accuracy of pulse classification is. The greater the preset count value, the lower the frequency of updating the action potential characteristic average value.
For example, assume that the preset count value is 2 n (n is a positive integer greater than or equal to zero), each class store function core 301 stores each accumulated 1 (n is equal to 0) or 2 of the received and stored input action potential segments n When (n.noteq.0), the average value is calculated by averaging all the currently stored input action potential segments, and the obtained average value is used as a new action potential characteristic average value. Feature comparison function core 302 and class merge function core 303 receive new action potential features from one or more class store function cores 301After the average value, it is determined that the action potential feature average value of the action potential class corresponding to the class storage function core 301 that sent the new action potential feature average value is updated to the currently received value, and the update is completed.
The classification merge function core 303 is configured to perform a merging process of at least two combinable classification storage function cores when at least two combinable classification storage function cores satisfying a classification update condition are determined from the plurality of classification storage function cores 301 according to a difference between current action potential feature averages of each classification storage function core.
In this implementation, the classification merge function core 303 may make a determination as to whether or not there is a mergeable classification storage function core in the case where it is detected that the merge process start condition is satisfied. Wherein the merging process start condition may include at least one of: the method comprises the steps of receiving a new action potential characteristic average value (the condition can ensure the timely progress of merging processing), enabling the time length between the current time and the time when the merging and classifying storage function core judgment is performed last time to reach a time length threshold value (the condition can ensure the time length threshold value of each interval of merging processing), enabling the current time to be the preset time of the set merging and classifying storage function core judgment (the condition can ensure the timing progress of the merging processing), and the like.
In one possible implementation manner, in a case where the merging process starting condition is that a new action potential feature average value is received, and in a case where at least two combinable class storage function cores satisfying the class update condition are determined from the plurality of class storage function cores 301 according to a difference between current action potential feature average values of each class storage function core, the merging process of the at least two combinable class storage function cores may include:
when a new action potential characteristic average value is received, calculating the arithmetic square root of the difference between the new action potential characteristic average value and the action potential characteristic average values of other classifications, and obtaining a plurality of second arithmetic square roots corresponding to the new action potential characteristic average value; determining a first classification memory function core corresponding to the new action potential characteristic average value and a second classification memory function core corresponding to the minimum second arithmetic square root as combinable classification memory function cores when the minimum second arithmetic square root of the plurality of second arithmetic square roots is less than or equal to a combining threshold; and controlling the first classified storage function core to send the stored input action potential fragments to the second classified storage function core, and controlling the first classified storage function core to stop working.
In this implementation manner, controlling the first class storage function core to send the stored input action potential segment to the second class storage function core, and controlling the first class storage function core to suspend operation may include: after determining two combinable classified storage function cores capable of performing combination processing, the classified combination function core 303 sends a pause work instruction to the first classified storage function core, so that the first classified storage function core can firstly send the stored input action potential fragments to the second classified storage function core based on the pause work instruction, and then empty the stored input action potential fragments to pause work.
For example, assuming that the new action potential feature average value W3 is a new action potential feature average value of the class 3 action potential class, if the second arithmetic square root of the difference between the calculated action potential feature average values of W3 and the class i action potential class is a23-i, and the minimum value a23-4 of a23-1, a23-2, a23-4 … a23-n is smaller than the merge threshold, it may be determined that the class 3 action potential class is similar to the class 4 action potential class, and the class storage function core corresponding to the class 3 action potential class and the class 4 action potential class may be determined as the mergeable class function core, and the mergeable processing may be performed. The class-3 action potential class-corresponding class-storing function core may be determined as the first class-storing function core, and the class-4 action potential class-corresponding class-storing function core may be determined as the second class-storing function core. Alternatively, the class storage function core corresponding to the class 3 action potential class may be determined as the second class storage function core, and the class storage function core corresponding to the class 4 action potential class may be determined as the first class storage function core.
The disclosure also provides another wearable device comprising the pulse signal classifying device and at least one sensor. At least one sensor for monitoring the living being, acquiring bioelectric signals, and transmitting the detected real-time bioelectric signals to the pulse detection module 10.
In one possible implementation manner, the pulse signal classifying device may further include:
at least one sensor for monitoring the living being, acquiring bioelectric signals, and transmitting the detected real-time bioelectric signals to the pulse detection module 10. The disclosure also provides a wearable device comprising the pulse signal classification device with the sensor.
In this embodiment, the classification storage function core, the feature comparison function core and the classification merging function core all perform computation under a specific trigger condition, and can work in parallel without mutual influence, so that the instantaneity of the classification system can be remarkably improved and the running power consumption can be reduced, and the method is suitable for wearable equipment. Moreover, the device is based on a brain-like many-core architecture, and can directly run in the same architecture with a subsequent processing system (such as a decision processing system based on a neural network) for processing and operating pulse signals, so that the classification and processing of brain-computer signals can be rapidly carried out end to end.
It should be noted that, although the pulse signal classifying device and the wearable apparatus are described above by taking the above embodiments as examples, those skilled in the art can understand that the present disclosure should not be limited thereto. In fact, the user can flexibly set each module according to personal preference and/or actual application scene, so long as the technical scheme of the disclosure is met.
In some embodiments, functions or modules included in an apparatus provided by the embodiments of the present disclosure may be used to perform a method described in the foregoing method embodiments, and specific implementations thereof may refer to descriptions of the foregoing method embodiments, which are not repeated herein for brevity.
The present disclosure may be a system, method, and/or computer program product. The computer program product may include a computer readable storage medium having computer readable program instructions embodied thereon for causing a processor to implement aspects of the present disclosure.
The computer readable storage medium may be a tangible device that can hold and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: portable computer disks, hard disks, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static Random Access Memory (SRAM), portable compact disk read-only memory (CD-ROM), digital Versatile Disks (DVD), memory sticks, floppy disks, mechanical coding devices, punch cards or in-groove structures such as punch cards or grooves having instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media, as used herein, are not to be construed as transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., optical pulses through fiber optic cables), or electrical signals transmitted through wires.
The computer readable program instructions described herein may be downloaded from a computer readable storage medium to a respective computing/processing device or to an external computer or external storage device over a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmissions, wireless transmissions, routers, firewalls, switches, gateway computers and/or edge servers. The network interface card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium in the respective computing/processing device.
Computer program instructions for performing the operations of the present disclosure can be assembly instructions, instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, c++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program instructions may be executed entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, aspects of the present disclosure are implemented by personalizing electronic circuitry, such as programmable logic circuitry, field Programmable Gate Arrays (FPGAs), or Programmable Logic Arrays (PLAs), with state information of computer readable program instructions, which can execute the computer readable program instructions.
Various aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.
These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable medium having the instructions stored therein includes an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the technical improvements in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (8)
1. A pulse signal classifying apparatus, said apparatus comprising:
the pulse detection module is used for storing the received continuous bioelectric signals sent by the sensor into a buffer memory area, generating a pulse existence signal under the condition that the pulse peak exists in the bioelectric signals, and sending the pulse existence signal to the pulse alignment module, wherein the pulse existence signal comprises an input action potential segment corresponding to the pulse peak obtained by intercepting the bioelectric signals;
the pulse alignment module is used for performing alignment processing on the received real-time pulse existence signals to obtain processed pulse existence signals and sending the processed pulse existence signals to the pulse clustering module;
the pulse clustering module is used for clustering the received processed pulse existence signals and determining pulse classification identifiers corresponding to the processed pulse existence signals according to clustering results, wherein the pulse classification identifiers are used for indicating action potential classifications corresponding to the input action potential fragments;
wherein, the pulse clustering module includes: the system comprises a plurality of classification storage function cores, a feature comparison function core and a classification merging function core, wherein action potential classifications corresponding to the classification storage function cores are different;
the feature comparison function core is used for calculating an input action potential average value vector according to the received input action potential fragments in the processed pulse existence signal, determining a pulse classification identifier corresponding to the processed pulse existence signal according to the difference between the input action potential average value vector and action potential feature average values of different categories, and sending the input action potential fragments to the corresponding classification storage function core;
each classification storage function core is used for counting and storing received input action potential fragments to obtain a counting result, calculating a new action potential characteristic average value according to the stored input action potential fragments under the condition that the counting result meets a characteristic value updating condition, and sending the new action potential characteristic average value to the characteristic comparison function core and the classification merging function core to update as a new action potential characteristic average value of a corresponding action potential classification;
the classification and merger function core is configured to perform merger processing of at least two merger function cores under the condition that at least two merger function cores satisfying a classification update condition are determined from a plurality of classification and storage function cores according to a difference between current action potential feature averages of each classification and storage function core.
2. The apparatus of claim 1, wherein determining the pulse classification identifier corresponding to the processed pulse presence signal based on a difference between the input action potential average vector and action potential feature averages of different classes comprises:
calculating the arithmetic square root of the difference between the input action potential average value vector and the action potential characteristic average value of each classification to obtain a plurality of first arithmetic square roots corresponding to the input action potential segments;
determining an action potential class corresponding to the minimum first arithmetic square root as an action potential class corresponding to the input action potential segment when the minimum first arithmetic square root of the plurality of first arithmetic square roots is less than or equal to a preset comparison threshold;
and determining a pulse classification identifier corresponding to the processed pulse existence signal according to the action potential classification corresponding to the input action potential segment.
3. The apparatus according to claim 1, wherein in a case where at least two combinable class-storing function cores satisfying a class update condition are determined from a plurality of class-storing function cores based on a difference between current action potential feature averages of each class-storing function core, performing a merging process of the at least two combinable class-storing function cores, comprises:
when a new action potential characteristic average value is received, calculating the arithmetic square root of the difference between the new action potential characteristic average value and the action potential characteristic average values of other classifications, and obtaining a plurality of second arithmetic square roots corresponding to the new action potential characteristic average value;
determining a first classification memory function core corresponding to the new action potential characteristic average value and a second classification memory function core corresponding to the minimum second arithmetic square root as combinable classification memory function cores when the minimum second arithmetic square root of the plurality of second arithmetic square roots is less than or equal to a combining threshold;
and controlling the first classified storage function core to send the stored input action potential fragments to the second classified storage function core, and controlling the first classified storage function core to stop working.
4. A device according to any one of claims 1 to 3, wherein the alignment process comprises a normalization process, and the input action potential segments in each of the processed pulse presence signals have a uniform characteristic value, the characteristic value comprising any one of: amplitude, energy, maximum slope.
5. A device according to any one of claims 1 to 3, wherein the device comprises a many-core brain architecture chip, the pulse detection module and the pulse alignment module are external interface modules of the chip, and the pulse clustering module is a plurality of functional cores of the chip.
6. A device according to any one of claims 1 to 3, characterized in that the device further comprises:
and the at least one sensor is used for monitoring organisms, acquiring bioelectric signals and sending the detected real-time bioelectric signals to the pulse detection module.
7. A wearable device, comprising:
the pulse signal classifying apparatus according to any one of claims 1 to 6.
8. A wearable device, comprising:
the at least one sensor is used for monitoring organisms, acquiring bioelectric signals and sending the detected real-time bioelectric signals to the pulse detection module;
the pulse signal classifying device according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210172964.9A CN114521868B (en) | 2022-02-24 | 2022-02-24 | Pulse signal classification device and wearable equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210172964.9A CN114521868B (en) | 2022-02-24 | 2022-02-24 | Pulse signal classification device and wearable equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114521868A CN114521868A (en) | 2022-05-24 |
| CN114521868B true CN114521868B (en) | 2024-04-09 |
Family
ID=81624947
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210172964.9A Active CN114521868B (en) | 2022-02-24 | 2022-02-24 | Pulse signal classification device and wearable equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114521868B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107837078A (en) * | 2016-09-20 | 2018-03-27 | 三星电子株式会社 | Apparatus and method and wearable device for biometrics infomation detection |
| CN112034426A (en) * | 2020-08-27 | 2020-12-04 | 上海朱光亚战略科技研究院 | Radar signal processing method, apparatus, computer device, and storage medium |
| CN113313919A (en) * | 2021-05-26 | 2021-08-27 | 中国工商银行股份有限公司 | Alarm method and device using multilayer feedforward network model and electronic equipment |
| CN114052744A (en) * | 2021-11-02 | 2022-02-18 | 西安电子科技大学重庆集成电路创新研究院 | Electrocardiosignal classification method based on pulse neural network |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120041293A1 (en) * | 2008-12-23 | 2012-02-16 | Commissariat A L'energie Atomique Et Aux Ene Alt | Methods and devices for processing pulse signals, and in particular neural action potential signals |
-
2022
- 2022-02-24 CN CN202210172964.9A patent/CN114521868B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107837078A (en) * | 2016-09-20 | 2018-03-27 | 三星电子株式会社 | Apparatus and method and wearable device for biometrics infomation detection |
| CN112034426A (en) * | 2020-08-27 | 2020-12-04 | 上海朱光亚战略科技研究院 | Radar signal processing method, apparatus, computer device, and storage medium |
| CN113313919A (en) * | 2021-05-26 | 2021-08-27 | 中国工商银行股份有限公司 | Alarm method and device using multilayer feedforward network model and electronic equipment |
| CN114052744A (en) * | 2021-11-02 | 2022-02-18 | 西安电子科技大学重庆集成电路创新研究院 | Electrocardiosignal classification method based on pulse neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114521868A (en) | 2022-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11232133B2 (en) | System for detecting and characterizing seasons | |
| Sikder et al. | Human activity recognition using multichannel convolutional neural network | |
| EP3692901A1 (en) | Automatic recognition and classification method for electrocardiogram heartbeat based on artificial intelligence | |
| US20200258005A1 (en) | Unsupervised method for classifying seasonal patterns | |
| Kamboh et al. | Computationally efficient neural feature extraction for spike sorting in implantable high-density recording systems | |
| JP7248102B2 (en) | Information processing device, personal identification device, information processing method and storage medium | |
| CN112949710A (en) | Image clustering method and device | |
| US11291413B2 (en) | Systems and methods for linear-time clustering for bounded, repeatable, rare events in physiological signals | |
| Santhanam et al. | An extensible infrastructure for fully automated spike sorting during online experiments | |
| WO2016060620A1 (en) | Frame based spike detection module | |
| CN114521868B (en) | Pulse signal classification device and wearable equipment | |
| JP2019041811A (en) | Determination device, determination method, and determination program | |
| Lewis et al. | Mixed effects random forests for personalised predictions of clinical depression severity | |
| CN113672782A (en) | Internet of things equipment importance matching method based on data query | |
| Zviagintsev et al. | Algorithms and architectures for low power spike detection and alignment | |
| CN116407140A (en) | Pulse signal detection method, device, equipment and storage medium | |
| CN111709480A (en) | Method and device for identifying image category | |
| JP2016030096A (en) | Living body electric noise identification system and living body electric noise removal system | |
| WO2017031581A1 (en) | System and method for detecting spikes in noisy signals | |
| WO2019130840A1 (en) | Signal-processing device, analysis system, signal-processing method, and signal-processing program | |
| CN114255502B (en) | Face image generation method and device, face recognition method, equipment and medium | |
| JP6706996B2 (en) | Biological signal processing device, abnormality determination method and program | |
| WO2021176529A1 (en) | Learning method, learning system, device, learning apparatus, and program | |
| AU2017234374B2 (en) | Activity pulse | |
| CN116616722B (en) | Brain function abnormality detection method and device based on dynamic cerebral cortex function connection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |