CN111857351A

CN111857351A - Electroencephalogram dialing method

Info

Publication number: CN111857351A
Application number: CN202010742685.2A
Authority: CN
Inventors: 于扬; 叶泽祺; 周宗潭; 胡德文; 沈辉; 张艺博
Original assignee: National University of Defense Technology
Current assignee: National University of Defense Technology
Priority date: 2020-07-29
Filing date: 2020-07-29
Publication date: 2020-10-30

Abstract

The invention specifically discloses an electroencephalogram dialing method, which comprises the following steps: s1, sequentially coding the left-hand MI task, the right-hand MI task and the idle state to form a coding table, wherein the coding table comprises twelve sequences with three lengths, and the sequences in the coding table are in a one-to-one correspondence with operation instructions in the operation process of making a call; s2, the user carries out motor imagery task according to the coding table and the operation instruction of dialing, the electroencephalogram detection equipment obtains corresponding electroencephalogram according to the motor imagery task of the user and transmits the electroencephalogram to the computer; and S3, the computer performs classification processing according to the electroencephalogram signals acquired in the step S2 to obtain corresponding operation instructions, so that automatic dialing of the brain computer is realized. The invention realizes that the user can make a call only by idea, has great significance for the gunnery, the busy population and the like, improves the self-care ability of the patients with the motor dysfunction and reduces the burden of the families and the society.

Description

Electroencephalogram dialing method

Technical Field

The invention relates to the technical field of brain-computer interfaces, in particular to an electroencephalogram dialing method.

Background

The basic principle of Brain Computer Interface (BCI) technology is to collect brain nerve signals by using various modern brain activity observation means, extract user control intentions contained in the signals by using a signal processing and pattern recognition method, and convert the control intentions into instructions, thereby realizing direct control of external devices by ideas. Electroencephalograms (EEG) are currently the most popular means for monitoring brain signals, and have the advantages of being non-invasive, high in signal real-time performance, good in portability, simple and easy to use in equipment, relatively low in cost and the like. How to provide a device for directly communicating with the outside by means of ideas or autonomously controlling various rehabilitation disabled-assisting devices for patients with motor function impairment to improve the self-care ability of the life of the patients and reduce the burden on families and society of the patients is always the motivation of researchers related to BCI.

Motor Imagery (MI) technology is a brain-conscious task commonly used in spontaneous BCI technology today. When an individual is attempting to imagine the movement of one of the body's limbs, it causes changes in the sensory-motor cortex Rhythms (SMRs) of the brain. SMRs caused by motor imagery of different limb parts have different spatial-temporal distribution characteristics, and EEG thereof has high distinguishability. Therefore, the MI mode in the EEG can be decoded by a mode recognition method and converted into a control instruction, so that the external equipment can be controlled. The electroencephalogram signals generated by the MI task have the advantages of high autonomy, no dependence on external stimulation, good real-time performance and the like. Particularly, because the cerebral cortex of the left hand and the right hand is widely distributed, when imagination movement of the left hand and the right hand is executed, the phenomenon of the SMRs is obvious, the method for designing and distinguishing the left MI task and the right MI task is easy to realize, related algorithms are mature, and the signal decoding accuracy is high. Therefore, the motor imagery task of the left and right hands becomes the most widely used task in designing MI-BCI systems. But is limited by the limited number of task types recognizable by the current MI-BCI technology (generally, the motor imagery tasks of 2-3 types, left/right hands and feet), and it is difficult to directly output more instructions to realize more complex tasks.

Disclosure of Invention

The invention aims to provide an electroencephalogram dialing method, which converts the idea of a user into an operation instruction, realizes that the user only dials a call by the idea, liberates two hands from the dialed call, and has great practical significance for soldiers holding guns, busy people and the like, thereby improving the self-care ability of the patients with dyskinesia and reducing the burden of families and society.

In order to solve the technical problem, the invention provides an electroencephalogram dialing method, which comprises the following steps:

s1, sequentially coding three states of a left-hand MI task, a right-hand MI task and an idle state in the motor imagery task to form a coding table, wherein the coding table comprises twelve sequences with three lengths, and the sequences in the coding table correspond to operation instructions in the operation process of making a call one by one;

s2, the user carries out corresponding motor imagery tasks according to the coding table and the operation instructions of dialing, and the electroencephalogram detection equipment acquires corresponding electroencephalogram signals according to the motor imagery tasks of the user and transmits the electroencephalogram signals to the computer;

and S3, classifying and processing the electroencephalogram signals acquired in the step S2 by the computer to obtain operation instructions corresponding to the dialed call, so that automatic dialing of the brain computer is realized.

Preferably, the operation instructions in step S1 include start of dialing/answering and start of hanging up/exiting, and control over ten digits, which are 0, 1,2,3,4,5,6,7,8 and 9.

Preferably, the left-hand MI task, the right-hand MI task, and the idle state in step S1 are respectively marked as L, R and 0, and the specific encoding manner of the sequence in the encoding table is as follows: the first bit and the last bit in the sequence are coded by two states of L and R, and the middle bit is coded by three states of L, R and 0, so that a sequence and an operation instruction one-to-one corresponding coding table is formed.

Preferably, the electroencephalogram signals acquired through the electroencephalogram detection equipment in the step S3 may be displayed on a human-computer interaction interface.

Preferably, the human-computer interaction interface comprises an instruction feedback area, a motor imagery task result feedback area, a reference numeric keyboard and a coding prompt area thereof, a hang-up/exit key and a coding prompt area thereof, and a dialing/answering key and a coding prompt area thereof, and the computer can convert each three categories of task results in the motor imagery task result feedback area into operation instructions according to a coding table and display the operation instructions on the instruction feedback area, and simultaneously clear a register of the motor imagery task result feedback area.

Preferably, after each operation command corresponding to an electroencephalogram signal is acquired in step S3, the electroencephalogram detection apparatus acquires the next electroencephalogram signal at intervals of time in step S2.

Preferably, in the step S3, the computer intercepts an electroencephalogram signal with a window length of 2000ms every 100ms and transmits the electroencephalogram signal to the classifier for classification processing to obtain a status signal corresponding to the motor imagery task.

Preferably, the computer continuously intercepts 50 electroencephalogram signals with the window length of 2000ms, the electroencephalogram signals are all in an idle state O, and if the register of the motor imagery task result feedback area is empty, the last register of the instruction feedback area is deleted.

Preferably, the classifiers in step S3 include three types, i.e., a left-hand MI and right-hand MI classifier, a left-hand MI and idle-state classifier, and a right-hand MI and idle-state classifier.

Compared with the prior art, the electroencephalogram dialing method has the advantages that three states of the left hand MI, the right hand MI and the idle state are sequentially coded to form a coding table which corresponds to operation instructions in the telephone dialing operation process one by one, and the computer converts the idea of a user into the operation instructions according to the coding table, so that the user can release the two hands of the user from the telephone dialing only by the idea to dial the telephone, the electroencephalogram dialing method has a great practical significance for military personnel holding guns, busy crowds and the like, the life self-care capability of patients with dyskinesia is improved, and the burden of families and society is reduced.

Drawings

FIG. 1 is a flow chart of an electroencephalogram dialing method of the present invention,

FIG. 2 is a schematic diagram of a human-machine interface according to the present invention.

In the figure: 1. the system comprises an instruction feedback area, a motor imagery task result feedback area, a reference numeric keyboard and a coding prompt area thereof, a hang-up/quit key and a coding prompt area thereof, and a dialing/answering key and a coding prompt area thereof, wherein the instruction feedback area is 2.

Detailed Description

In order to make the technical solutions of the present invention better understood, the present invention is further described in detail below with reference to the accompanying drawings.

As shown in fig. 1 and 2, an electroencephalogram dialing method includes the following steps:

In the embodiment, firstly, three states of a left-hand MI task, a right-hand MI task and an idle state in the motor imagery movement are sequentially coded to form twelve coding tables with three lengths, and sequences in the coding tables correspond to operation instructions in the telephone dialing operation process one by one; then the user carries out the motor imagery task according to the coding table and the operation instruction of dialing the telephone; finally, the brain wave detection equipment acquires the corresponding electroencephalograms according to the motor imagination task of the user and transmits the electroencephalograms to the computer for classification and processing, so that the operation instruction corresponding to the call making is obtained, the user can make the call by means of idea, the invention can liberate the two hands of the user from making the call, has great practical significance for military personnel holding guns, busy people and the like, improves the self-care capability of the patients with motor dysfunction, and reduces the burden of families and society.

In this embodiment, the electroencephalogram detection device is integrated in the telephone, is connected with the electrode cap worn by the user, and is configured to receive electroencephalogram signals collected by the electrode cap, identify and classify the electroencephalogram signals, and convert the electroencephalogram signals into operation instructions corresponding to the code table for dialing, so that automatic dialing of electroencephalograms is achieved, and the electroencephalogram detection device includes an electroencephalogram collection device and a computer. In other embodiments, the electroencephalogram detection device can also be externally arranged on a telephone to perform corresponding processing on the acquired electroencephalogram signals.

As shown in fig. 1 and fig. 2, the operation instructions in step S1 include start of dialing/answering and start of hanging up/exiting, and control over ten digits, which are 0, 1,2,3,4,5,6,7,8 and 9; in the step S1, the left-hand MI task, the right-hand MI task, and the idle state are respectively marked as L, R and O, and the specific encoding manner of the sequence in the encoding table is as follows: the first bit and the last bit in the sequence are coded by two states of L and R, and the middle bit is coded by three states of L, R and O, so that a sequence and an operation instruction one-to-one corresponding coding table is formed.

In this embodiment, the operation commands are twelve, including start of dialing/answering and start of hanging up/exiting, and ten digits of 0, 1,2,3,4,5,6,7,8 and 9 are controlled. Firstly, a left-hand MI task, a right-hand MI task and an idle state in the motor imagery task are respectively recorded as L, R and O to be combined to form a plurality of sequences, as shown in table 1 (table 1 is a coding table corresponding to a dialing operation instruction and a sequence), wherein the first bit and the last bit in the sequences can only be coded by two states of L and R, and the middle bit is coded by three states of L, R and O, so that a coding table corresponding to the operation instruction one by one can be formed. On one hand, all functions of the dialing can be effectively coded and correspond to each other exactly one by one; on the other hand, the first bit of the sequence can only be coded in the L and R states in order to ensure that there is a signal triggering the start, i.e. when the computer recognizes the L or R signal, it means that the user wants to start making a call, while the last bit can only be coded in the L and R states in order to effectively distinguish from the gap time. In other embodiments, the corresponding code table between the dialing/answering key and the sequence may be adjusted according to the self condition, that is, the sequence corresponding to the dialing/answering key may be RRR, the sequence corresponding to the hang-up/quit key may be LLL, and the correspondence between other sequences and the operation command is similar.

Table 1: coding table for dialing operation instruction and sequence correspondence

Operation instruction	Sequence of	Operation instruction	Sequence of
				Start dialing/answering	LLL	0	LLR
1	L0L	2	L0R
				3	LRL	4	LRR
5	RLL	6	RLR
				7	R0L	8	R0R
9	RRL	Start hang-up/exit	RRR

As shown in fig. 1 and 2, the electroencephalogram signal acquired through the electroencephalogram detection apparatus in the step S3 may be displayed on a human-computer interaction interface.

As shown in fig. 2, the human-computer interaction interface includes an instruction feedback area 1, a motor imagery task result feedback area 2, a reference numeric keyboard and a coding prompt area 3 thereof, a hang-up/exit key and a coding prompt area 4 thereof, and a dial-up/answer key and a coding prompt area 5 thereof, and the computer can convert each three-bit category task result in the motor imagery task result feedback area 2 into an operation instruction according to a coding table and display the operation instruction on the instruction feedback area 1, and simultaneously clear a register of the motor imagery task result feedback area 2.

In the embodiment, the human-computer interaction interface is arranged, and the acquired electroencephalogram signal is displayed on the human-computer interaction interface, so that the electroencephalogram signal can be used for feeding back the identification signal of the computer and the translation condition of the operation instruction in real time, a user can know the real-time condition and correct the real-time condition in time, and the use experience of the user is greatly improved.

In one embodiment, after each operation command corresponding to the electroencephalogram signal is acquired in step S3, the electroencephalogram detecting apparatus in step S2 acquires the next electroencephalogram signal at a time interval. In this embodiment, in order to ensure that the attention of the user is kept focused and improve the quality of the signal acquired by the electroencephalogram detection equipment, the electroencephalogram signal is acquired at a time interval after each operation instruction is acquired.

In one embodiment, in the step S3, the computer intercepts a section of electroencephalogram signals with a window length of 2000ms every 100ms and transmits the electroencephalogram signals to the classifier for classification processing to obtain state signals corresponding to the motor imagery task, the computer intercepts 50 electroencephalograms with a window length of 2000ms continuously and all states are idle states O, and when the register of the motor imagery task result feedback area 2 is empty, the last register of the instruction feedback area 1 is deleted.

In this embodiment, in order to facilitate user error detection, when the computer continuously captures 50 electroencephalogram signals with a window length of 2000ms and all electroencephalogram signals are in an idle state O, and the register of the motor imagery task result feedback area 2 is empty, the last register of the instruction feedback area 1 is deleted, so that deletion of the last operation instruction of the instruction feedback area 1 is realized. In this embodiment, the computer intercepts an electroencephalogram signal with a window length of 2000ms every 100 ms.

In this embodiment, in order to distinguish between the idle state and the interval time, the first bit and the last bit of the sequence can only be encoded by using two states, i.e., a left-hand MI task or a right-hand MI task, and when the computer recognizes the left-hand MI task or the right-hand MI task as the first bit in the sequence, it is considered whether the middle bit is in the idle state, and if electroencephalogram signals within 2000ms of the computer recognition window length are all idle, the middle bit is considered as the idle state.

In this embodiment, the computer detects an electroencephalogram of a user by using electroencephalogram detection equipment, intercepts an electroencephalogram signal with a window length of 2000ms every 100ms, processes the intercepted electroencephalogram signal by using a BCI algorithm, performs feature extraction by using an Empirical Mode Decomposition (EMD) algorithm and a Common Spatial Pattern (CSP) algorithm, obtains a sample score by using a stepwise discriminant analysis (swada), divides the classifier into a left-handed MI classifier and a right-handed MI classifier, a left-handed MI classifier and an idle-state classifier, and a right-handed MI classifier and an idle-state classifier, obtains one of final classification results L, R or O in a voting form by using the three classifiers, and displays the final classification result on the motor task imagination result feedback area 2.

The electroencephalogram dialing method provided by the invention is described in detail above. The principles and embodiments of the present invention are explained herein using specific examples, which are presented only to assist in understanding the core concepts of the present invention. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.

Claims

1. An electroencephalogram dialing method is characterized by comprising the following steps:

2. The electroencephalogram dialing method according to claim 1, wherein the operation instructions in the step S1 include start of dialing/answering and start of hanging up/exiting, and control of ten arabic numerals, which are 0, 1,2,3,4,5,6,7,8 and 9.

3. The electroencephalogram dialing method according to claim 2, wherein in step S1, the left-hand MI task, the right-hand MI task, and the idle state are respectively marked as L, R and O, and the specific way of encoding the sequences in the encoding table is as follows: the first bit and the last bit in the sequence are coded by two states of L and R, and the middle bit is coded by three states of L, R and O, so that a sequence and an operation instruction one-to-one corresponding coding table is formed.

4. The electroencephalogram dialing method according to claim 3, wherein the electroencephalogram signal acquired through the electroencephalogram detection device in the step S3 is displayable on a human-computer interaction interface.

5. The electroencephalogram dialing method according to claim 4, wherein the human-computer interaction interface comprises an instruction feedback area (1), a motor imagery task result feedback area (2), a reference numeric keyboard and a coding prompt area (3) thereof, a hang-up/exit key and a coding prompt area (4) thereof, and a dialing/answering key and a coding prompt area (5) thereof, and the computer can convert each three-bit category task result in the motor imagery task result feedback area (2) into an operation instruction according to a coding table and display the operation instruction on the instruction feedback area (1), and simultaneously clear a register of the motor imagery task result feedback area (2).

6. The electroencephalogram dialing method according to claim 5, wherein, after each operation instruction corresponding to an electroencephalogram signal is acquired in step S3, the electroencephalogram detecting apparatus acquires the next electroencephalogram signal at intervals of time in step S2.

7. The electroencephalogram dialing method according to claim 6, wherein in the step S3, the computer intercepts an electroencephalogram signal with a window length of 2000ms every 100ms, and transmits the electroencephalogram signal to the classifier for classification processing so as to obtain a state signal corresponding to the motor imagery task.

8. The electroencephalogram dialing method according to claim 7, wherein the computer continuously intercepts 50 electroencephalogram signals with the window length of 2000ms, and all electroencephalogram signals are in an idle state O, and when the register of the motor imagery task result feedback area (2) is empty, the last register of the instruction feedback area (1) is deleted.

9. The electroencephalogram dialing method according to claim 8, wherein the classifiers in the step S3 include three types of left-hand MI and right-hand MI classifiers, left-hand MI and idle-state classifiers, and right-hand MI and idle-state classifiers.