CN115227504A - Automatic lifting sickbed system based on electroencephalogram and electrooculogram signals - Google Patents
Automatic lifting sickbed system based on electroencephalogram and electrooculogram signals Download PDFInfo
- Publication number
- CN115227504A CN115227504A CN202210842339.0A CN202210842339A CN115227504A CN 115227504 A CN115227504 A CN 115227504A CN 202210842339 A CN202210842339 A CN 202210842339A CN 115227504 A CN115227504 A CN 115227504A
- Authority
- CN
- China
- Prior art keywords
- electroencephalogram
- electro
- data set
- data
- oculogram
- 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.)
- Granted
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61G—TRANSPORT, PERSONAL CONVEYANCES, OR ACCOMMODATION SPECIALLY ADAPTED FOR PATIENTS OR DISABLED PERSONS; OPERATING TABLES OR CHAIRS; CHAIRS FOR DENTISTRY; FUNERAL DEVICES
- A61G7/00—Beds specially adapted for nursing; Devices for lifting patients or disabled persons
- A61G7/002—Beds specially adapted for nursing; Devices for lifting patients or disabled persons having adjustable mattress frame
- A61G7/018—Control or drive mechanisms
-
- 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
- A61B5/398—Electrooculography [EOG], e.g. detecting nystagmus; Electroretinography [ERG]
-
- 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
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/74—Details of notification to user or communication with user or patient ; user input means
- A61B5/746—Alarms related to a physiological condition, e.g. details of setting alarm thresholds or avoiding false alarms
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61G—TRANSPORT, PERSONAL CONVEYANCES, OR ACCOMMODATION SPECIALLY ADAPTED FOR PATIENTS OR DISABLED PERSONS; OPERATING TABLES OR CHAIRS; CHAIRS FOR DENTISTRY; FUNERAL DEVICES
- A61G7/00—Beds specially adapted for nursing; Devices for lifting patients or disabled persons
- A61G7/002—Beds specially adapted for nursing; Devices for lifting patients or disabled persons having adjustable mattress frame
- A61G7/012—Beds specially adapted for nursing; Devices for lifting patients or disabled persons having adjustable mattress frame raising or lowering of the whole mattress frame
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
- G06F3/015—Input arrangements based on nervous system activity detection, e.g. brain waves [EEG] detection, electromyograms [EMG] detection, electrodermal response detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61G—TRANSPORT, PERSONAL CONVEYANCES, OR ACCOMMODATION SPECIALLY ADAPTED FOR PATIENTS OR DISABLED PERSONS; OPERATING TABLES OR CHAIRS; CHAIRS FOR DENTISTRY; FUNERAL DEVICES
- A61G2203/00—General characteristics of devices
- A61G2203/10—General characteristics of devices characterised by specific control means, e.g. for adjustment or steering
- A61G2203/18—General characteristics of devices characterised by specific control means, e.g. for adjustment or steering by patient's head, eyes, facial muscles or voice
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- Animal Behavior & Ethology (AREA)
- Molecular Biology (AREA)
- Medical Informatics (AREA)
- Heart & Thoracic Surgery (AREA)
- Pathology (AREA)
- Surgery (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physiology (AREA)
- Artificial Intelligence (AREA)
- Psychiatry (AREA)
- General Physics & Mathematics (AREA)
- Signal Processing (AREA)
- Nursing (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Ophthalmology & Optometry (AREA)
- Computing Systems (AREA)
- Genetics & Genomics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Psychology (AREA)
- Dermatology (AREA)
- Neurology (AREA)
- Neurosurgery (AREA)
- Human Computer Interaction (AREA)
Abstract
The invention discloses an automatic lifting sickbed system based on electroencephalogram and electro-oculogram signals, which belongs to the field of rehabilitation and medical treatment, and comprises a brain wave sensor, a sickbed control system and a sickbed body, wherein the sickbed body has lifting, active calling and alarming functions; the brain wave sensor is connected with the hospital bed control system and used for collecting the electroencephalogram and electro-oculogram fusion signals of the patient and transmitting the electroencephalogram and electro-oculogram fusion signals to the hospital bed control system; the sickbed control system sends a control instruction according to the electroencephalogram and electrooculogram fusion signal of a patient, the control instruction is used for controlling the sickbed body to rise and level, actively calling and alarming, and the sickbed body responds according to the control instruction; the patient can be lifted, actively called and intelligently alarmed in a bed-lying state by monitoring the electroencephalogram signals in real time.
Description
Technical Field
The invention belongs to the field of rehabilitation and medical treatment, and particularly relates to an automatic lifting sickbed system based on electroencephalogram and electrooculogram signals.
Background
For patients lying in bed for a long time, particularly patients with inconvenient limbs, the state of the sickbed needs to be adjusted with the help of nursing staff, the labor intensity of the nursing staff is increased, and the patients cannot independently adjust the state of the sickbed and call doctors and the like. Although the existing sickbed can automatically realize the functions of lifting, alarming and the like through a simple button, the sickbed is only suitable for patients who can move simply, and the sickbed does not have any mobility for patients who are inconvenient to lie in bed completely, and the common intelligent sickbed cannot meet the requirements.
With the continuous development of scientific technology, scientific research means of human beings on the brain are diversified, wherein a brain-computer interface method adopts electroencephalogram and an eye electrical signal to form corresponding relations with the movement of eyes and eye accessories, and a body-independent sickbed system is designed by utilizing the corresponding relations, so that a bedridden patient can get up, actively call and intelligently alarm, and the brain-computer interface method becomes a research hotspot of rehabilitation medical engineering.
Disclosure of Invention
In order to solve the technical problem, the invention provides an automatic lifting sickbed system based on electroencephalogram signals, which can realize lifting, active calling and intelligent alarming of a patient in a bed state by monitoring the electroencephalogram signals in real time.
The invention adopts the following technical scheme:
an automatic lifting sickbed system based on electroencephalogram and electrooculogram signals comprises an electroencephalogram sensor, a sickbed control system and a sickbed body, wherein the sickbed body has lifting, active calling and alarming functions;
the brain wave sensor is connected with the hospital bed control system and used for collecting the electroencephalogram and electrooculogram fusion signals of the patient and transmitting the electroencephalogram and electrooculogram fusion signals to the hospital bed control system; the sickbed control system sends a control instruction according to the electroencephalogram and electrooculogram fusion signal of the patient, the control instruction is used for controlling the sickbed body to rise and level, actively calling and alarming, and the sickbed body responds according to the control instruction;
the hospital bed control system comprises:
the brain electric eye electrical fusion signal noise filtering module is used for carrying out noise filtering processing on the brain electric eye electrical fusion signal;
the electroencephalogram electro-fusion signal segmentation window module is used for segmenting the electroencephalogram electro-fusion signal subjected to noise filtering into different groups with consistent length;
and the brain electric eye electric fusion signal pattern recognition module is used for recognizing the characteristic type of each group of brain electric eye electric fusion signals and sending an instruction according to the characteristic type.
Preferably, the brain wave sensor is wirelessly connected to a hospital bed control system.
Preferably, the hospital bed control system sends a control instruction according to the electroencephalogram and electrooculogram fused signal of the patient, and the hospital bed control system comprises:
s1, carrying out noise filtering processing on the electroencephalogram and electro-oculogram fusion signal to obtain electroencephalogram and electro-oculogram data subjected to noise filtering;
s2, dividing the electroencephalogram electro-ocular data after noise filtering into groups to obtain electroencephalogram electro-ocular data groups, and calculating waveform characteristics of each electroencephalogram electro-ocular data group, wherein the waveform characteristics of the electroencephalogram electro-ocular data groups comprise an average value, an absolute average value, a mean square error and a standard deviation;
s3, classifying the waveform characteristics of each electro-oculogram data set relative to the waveform characteristics of the sample library data set by adopting a K-nearest neighbor algorithm to obtain the characteristic type of each electro-oculogram data set; and sending a control command according to the type of the characteristic of each electro-oculogram data set.
As the optimization of the invention, the characteristic types of each electroencephalogram and electrooculogram data set comprise one-time blinking, two-time blinking, eyeball rotation and electroencephalogram abnormity, and the corresponding control commands are respectively lifting up and leveling up of a sickbed, active calling and alarming.
Preferably, step S3 includes:
calculating the distance between the waveform characteristics of the electroencephalogram electro-ocular data set to be tested and each electroencephalogram electro-ocular data set in the sample library data set to obtain a distance set D = { D = 1,1 ，d 1,2 ，…，d 1,m1 ，d 2,1 ，d 2,2 ，…，d 2,m2 ，…，d 4,1 ，d 4,2 ，…，d 4,m4 M1 represents the number of electroencephalogram and electro-oculogram data sets belonging to a one-time blinking type in the sample library data set, m2 represents the number of electroencephalogram and electro-oculogram data sets belonging to a two-time blinking type in the sample library data set, m3 represents the number of electroencephalogram and electro-oculogram data sets belonging to an eyeball rotation type in the sample library data set, m4 represents the number of electroencephalogram and electro-oculogram data sets belonging to an electroencephalogram abnormal type in the sample library data set, and d 1,m1 Representing the distance, d, between the waveform characteristics of the electroencephalogram electro-ocular data set to be tested and the waveform characteristics of the electroencephalogram electro-ocular data set of the m1 st blink type 2,m2 Representing the distance, d, between the waveform characteristics of the electroencephalogram electro-ocular data set to be tested and the waveform characteristics of the electroencephalogram electro-ocular data set of the m2 th blink type 3,m3 Represents the distance between the waveform characteristics of the electroencephalogram electro-ocular data set to be tested and the waveform characteristics of the electroencephalogram electro-ocular data set of the m 3-th eyeball rotation type, d 4,m4 The waveform characteristics of the electroencephalogram electro-ocular data group to be tested and the m4 th electroencephalogram abnormityThe distance between waveform features of common types of electroencephalogram electro-ocular data sets;
and (3) performing ascending order arrangement on the numerical values in the distance set D, taking the first K distances, judging the number of the four characteristic types in the sample library data set corresponding to the first K distances, and taking the characteristic type with the largest number as the characteristic type of the electroencephalogram and electrooculogram data set to be tested.
Preferably, the method for constructing the sample library data set comprises:
s31, acquiring electroencephalogram electro-oculogram signals of different patients under the conditions of primary blink, secondary blink, eyeball rotation and electroencephalogram abnormity through an electroencephalogram sensor, carrying out noise filtering on the electroencephalogram electro-oculogram signals, dividing the electroencephalogram electro-oculogram signals into groups, extracting waveform characteristics of each data group, and obtaining a database with p groups of total data;
s32, randomly generating m sample library data sets, r training library data sets and S testing library data sets from the data base with the total data volume of p groups; the sample library data set, the training library data set and the testing library data set respectively contain four types of data of blink once, blink twice, eyeball rotation and electroencephalogram abnormity and are not repeated;
s33, repeating the step S32, initializing a plurality of data sets, wherein each data set consists of a randomly initialized sample base data set, a training base data set, a test base data set and a K value parameter of a K nearest neighbor algorithm;
calculating a distance set between the waveform characteristics of each training library data set and the waveform characteristics of each sample library data set, performing characteristic classification on one training library data set by adopting a K nearest neighbor algorithm, constructing a target function by using the recognition rate and the misjudgment rate, and selecting an optimal data set;
s34, taking the sample library data group in the optimal data set obtained in the step S33 as an optimal sample library data group, calculating a distance set between the waveform characteristics of the test library data group in the optimal data set and the waveform characteristics of each sample library data group, performing characteristic classification on one test library data group by adopting a K-nearest neighbor algorithm, constructing a target function by using the recognition rate and the misjudgment rate, and judging whether the preset requirements are met;
if yes, the sample library data group in the optimal data set obtained in the step S33 is taken as a final optimal sample library data group;
if not, returning to the step S32, and reinitializing a plurality of data sets.
Preferably, in step S33, a genetic algorithm is used to select, cross and mutate data groups in the initialized data sets, and the obtained new data set is used to replace the original data set.
Compared with the prior art, the invention has the advantages that:
1) The sickbed system provided by the invention realizes automatic lifting, active calling and intelligent alarming by monitoring the electroencephalogram and electro-oculogram signals of a patient, adopts a K-nearest neighbor method to identify the characteristic types of the electroencephalogram and electro-oculogram signals, has high operation speed, and ensures timely correspondence of the sickbed system;
2) The invention provides a sample library construction method based on an embedded AI algorithm, which ensures the accuracy of the sample library and improves the accuracy of the characteristic type identification of the electroencephalogram electro-ocular signal.
Drawings
FIG. 1 is an overall schematic diagram of an automatic lifting sickbed system based on electroencephalogram and electrooculogram signals according to an embodiment of the invention;
fig. 2 is a data processing diagram of a patient bed control system according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a flow of recognizing a brain-electric-eye electrical fusion signal pattern according to an embodiment of the present invention;
fig. 4 is a schematic diagram illustrating a sample library construction process based on an embedded AI algorithm according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail below by way of examples with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby. Some of the block diagrams shown in the figures are functional entities, which do not necessarily have to correspond to physically or logically separate entities, and which may be implemented in software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor devices and/or microcontroller devices.
The flow charts shown in the drawings are merely illustrative and do not necessarily include all of the steps. For example, some steps may be decomposed, and some steps may be combined or partially combined, so that the actual execution sequence may be changed according to the actual situation.
Fig. 1 is an overall schematic view of an automatic lifting hospital bed system based on electroencephalogram electro-oculogram signals according to an embodiment of the present invention, the automatic lifting hospital bed system includes a brain wave sensor, a hospital bed control system and a hospital bed body, the hospital bed body has lifting, leveling, active calling and alarming functions, and can be realized by a motion module or an alarm module for realizing different functions, and the four functions are controlled by sending control instructions through the hospital bed control system.
The brain wave sensor is used for collecting and transmitting brain electrical signals and eye electrical signals, is connected with the hospital bed control system, and is used for collecting brain electrical eye electrical fusion signals of a patient and transmitting the brain electrical eye electrical fusion signals to the hospital bed control system; in the signal acquisition process, the electroencephalogram signal and the electro-oculogram signal are fused together and output. In this embodiment, the brain wave sensor is wirelessly connected to the hospital bed control system, for example, using wireless communication methods such as bluetooth and wifi.
The sickbed control system is used for carrying out noise filtering, feature extraction and mode identification on the electroencephalogram and electrooculogram fusion signals of the patient and converting the electroencephalogram and electrooculogram fusion signals into sickbed control instructions, namely the control instructions are sent according to the electroencephalogram and electrooculogram fusion signals of the patient and used for controlling the sickbed body to rise, level, actively call and alarm, and the sickbed body responds according to the control instructions. The hospital bed control system comprises:
the brain electric eye electrical fusion signal noise filtering module is used for carrying out noise filtering processing on the brain electric eye electrical fusion signal;
and the electroencephalogram electro-ocular fusion signal segmentation window module is used for segmenting the electroencephalogram electro-ocular fusion signal after noise filtering processing into different groups with consistent length, for example, the length n of the data fetching segment is a group of electroencephalogram electro-ocular data groups.
And the electro-oculogram fused signal pattern recognition module is used for extracting the waveform characteristics of the electro-oculogram fused signals, recognizing the type of the characteristic of each group of electro-oculogram fused signals and sending instructions according to the type of the characteristic. In this embodiment, as shown in fig. 2, the characteristic types of the electroencephalogram and electrooculogram data sets include single blink, double blink, eyeball rotation and electroencephalogram abnormality, and the corresponding control commands are respectively a sickbed lifting, a sickbed leveling, an active calling and an alarm. The electroencephalogram abnormality refers to an electroencephalogram electro-ocular signal acquired when a patient is in acute cerebral infarction, cerebral blood supply insufficiency, epilepsy and the like; the waveform characteristics include mean, absolute mean, sample mean square error, sample standard deviation, and the like.
As shown in fig. 3, a K-nearest neighbor algorithm is adopted to classify and judge each electroencephalogram electro-ocular data set with respect to a sample library, so as to obtain a feature type of each electroencephalogram electro-ocular data set; and according to the characteristic types of the electroencephalogram and electrooculogram data sets, the hospital bed control system sends out hospital bed control instructions.
In one embodiment of the present invention, for a certain electroencephalogram electro-ocular data set, firstly, the waveform characteristics are calculated; and then a distance function, such as a euclidean distance, calculating a distance set between the waveform characteristics of the electroencephalogram electro-oculogram data set and the waveform characteristics of each electroencephalogram electro-oculogram data set in the sample library data set, and recording as the distance set D = { D = 1,1 ，d 1,2 ，…，d 1,m1 ，d 2,1 ，d 2,2 ，…，d 2,m2 ，…，d 4,1 ，d 4,2 ，…，d 4,m4 Wherein m1 represents the number of electroencephalogram and electro-oculogram data groups belonging to a blink once type in the sample library data group, m2 represents the number of electroencephalogram and electro-oculogram data groups belonging to a blink twice type in the sample library data group, m3 represents the number of electroencephalogram and electro-oculogram data groups belonging to a eyeball rotation type in the sample library data group, m4 represents the number of electroencephalogram and electro-oculogram data groups belonging to an electroencephalogram abnormality type in the sample library data group, and d 1,m1 Represents the distance between the waveform characteristics of the electroencephalogram electro-ocular data set to be tested and the waveform characteristics of the electroencephalogram electro-ocular data set of the m1 st blink type, d 2,m2 Representing the distance, d, between the waveform characteristics of the electroencephalogram electro-ocular data set to be tested and the waveform characteristics of the electroencephalogram electro-ocular data set of the m2 th blink type 3,m3 Representing the distance, d, between the waveform characteristics of the electroencephalogram electro-ocular data set to be tested and the waveform characteristics of the electroencephalogram electro-ocular data set of the m3 th eyeball rotation type 4,m4 Representing the distance between the waveform characteristics of the electroencephalogram electro-ocular data set to be tested and the waveform characteristics of the electroencephalogram electro-ocular data set of the m4 th electroencephalogram abnormal type; finally, the values in the distance set D are sorted in ascending order, and a new set of distances is obtained, for example, the new set of distances is D' = { D = 2,1 ，d 1,1 ，d 2,4 ，…，d 2,m2 ，d 3,1 ，d 4,1 ，…，d 4,m4 Get the first K distances { d } 2,1 ，d 1,1 ，d 2,4 ，…，d 2,m2 And judging that the number of the second blink types in the sample library data set corresponding to the first K distances is the largest, so that the electroencephalogram electro-oculogram data set belongs to the second blink type.
In a specific implementation of the invention, the embedded AI algorithm is adopted to construct the sample library, so as to quickly find out the optimal sample library data set, use the optimal sample library data set for the characteristic classification of the electroencephalogram and electrooculogram signals and further improve the signal identification rate.
As shown in fig. 4, the sample library construction process based on the embedded AI algorithm is as follows:
1. and (3) constructing a constraint condition:
1.1 To increase the diversity of the sample library and avoid falling into local extreme values, a sample library data set is formed by randomly generating a primary blink data set, a secondary blink data set, an eyeball rotation data set and an electroencephalogram abnormity data set.
s.t.rand (m 1 blink data set)
rand (m 2 double blink data set)
rand (m 3 eyeball rotation data set)
rand (m 4 electroencephalogram abnormal data set)
1.2 Adopting a K nearest neighbor algorithm and a distance function, calculating the waveform characteristics of the electroencephalogram and electrooculogram data set, and judging the characteristic classification. The K value parameter in the K nearest neighbor algorithm takes a smaller value or a larger value, which causes large deviation of the feature recognition rate and the feature misjudgment rate, and adopts a mode of randomly generating the K value to obtain a proper K value, thereby avoiding falling into a local extreme value.
s.t.rand(K)
1.3 For the same reason, to avoid getting trapped in local extrema, training library data sets are randomly generated.
s.t.rand (r training database data set)
1.4 K value should be less than the number of sample library data sets.
s.t.K<m
1.5 The number of the sample library data sets and the training library data sets is less than that of the electroencephalography data sets.
s.t.m<p
r<p
m+r<p
m1+m2+m3+m4＝m
2. Constructing a multi-objective function:
and adopting the feature recognition rate and the feature error judgment rate to represent the feature classification accuracy and the error judgment rate of the electroencephalogram and electrooculogram data set. Aiming at the characteristic classification result of the electroencephalogram and electrooculogram data set, the larger the characteristic identification rate is, the better the characteristic misjudgment rate is, and the better the characteristic misjudgment rate is. Therefore, the maximum feature recognition rate and the minimum feature misjudgment rate are used as the multi-objective function.
Max { feature recognition rate }
Min { feature misjudgment rate }
3. Optimizing a sample library dataset
And (5) constructing a model by using the sample library, namely, the linear programming problem. The solving process of the sample library construction model is actually the solving process of a linear programming problem, and an embedded AI algorithm is adopted for calculating to quickly and accurately solve the sample library construction model.
Taking a basic genetic algorithm as an example, updating the evolutionary cycle process of the current optimal chromosome after population selection, a crossover algorithm and a mutation algorithm, and the method comprises the following basic steps:
3.1 Randomly generating a sample library data set (primary blink data, secondary blink data, eyeball rotation data and electroencephalogram abnormal data), a K value and a training library data set, wherein the rest electroencephalogram data sets are test library data sets (the rest data sets except the sample library data set and the training library data set in the p electroencephalogram electro-oculogram data sets), forming a data set and carrying out feasibility check;
3.2 Repeating the step of 3.1), and randomly generating POP _ SIZE data sets, wherein each data set comprises a sample library data set, a K value, a training library data set and a testing library data set;
3.3 In a data set, calculating a distance set between waveform characteristics of a training library data set and waveform characteristics of each sample library data set, performing characteristic classification on the training library data set by adopting a K nearest neighbor algorithm, and performing characteristic classification on each training library data set in the data set by analogy to calculate characteristic recognition rate and characteristic misjudgment rate of the data set;
3.4 The POP _ SIZE data sets are arranged in an ascending order or a descending order according to the feature recognition rate and the feature misjudgment rate of the data sets, the corresponding data set contents (sample library data sets and K values) are exchanged, and the current best data set is stored;
3.5 Adopting roulette selection, crossover algorithm and variation algorithm to update the data set to generate a new POP _ SIZE data set;
3.6 Repeating the steps 3.3) to 3.4) to generate the current best data set, and storing a better data set compared with the best data set of the previous generation;
3.7 Repeating the steps 3.5) to 3.6) until the termination condition is met, outputting the current best data set as an optimal solution, and taking a sample database data group in the current best data set as an optimal sample database;
3.8 In the current best data set, calculating a distance set between the waveform characteristics of a test library data set and the waveform characteristics of each sample library data set, performing characteristic classification on one test library data set by adopting a K nearest neighbor algorithm, and so on, performing characteristic classification on each test library data set in the data set, and calculating the characteristic recognition rate and the characteristic misjudgment rate of the test library data sets. If the target is reached, the characteristic recognition rate of the test library data set is more than 85%, the characteristic misjudgment rate is less than 8%, and the optimal sample library data set in the current best data set is feasible. If the target is not achieved, repeating the steps from 3.1) to 3.8), and retraining the sample library data set until the optimal sample library meeting the target is output.
In the embodiment, an optimal sample library data set is obtained, and in the feature classification of the electroencephalogram and electrooculogram data set, the feature recognition rate is further improved to be more than 85%, and the feature misjudgment rate is reduced to be less than 8 per thousand.
When the automatic lifting sickbed system is used, firstly, a patient electroencephalogram electro-oculogram fused signal with a certain length is collected through the electroencephalogram sensor to be subjected to noise filtering processing, waveform characteristics are extracted, the waveform characteristics are compared with an optimized sample library data set, the waveform characteristics to be recognized are classified by adopting a K-nearest neighbor algorithm, a control instruction is sent to a sickbed body according to a classification result, and the sickbed body responds according to the control instruction. The sickbed control scheme is as follows: the sickbed is lifted corresponding to one blink, the sickbed is leveled corresponding to the second blink, the eyeball rotation corresponds to active calling, and the electroencephalogram abnormity corresponds to intelligent alarm.
The sickbed system provided by the invention realizes automatic lifting, active calling and intelligent alarming by monitoring the electroencephalogram and electro-oculogram signals of a patient, adopts a K-nearest neighbor method to identify the type of the electroencephalogram and electro-oculogram signals, has high operation speed, and ensures timely correspondence of the sickbed system; the optimized sample library ensures the accuracy of the type identification of the electroencephalogram electro-oculogram signals.
The foregoing lists merely illustrate specific embodiments of the invention. It is obvious that the invention is not limited to the above embodiments, but that many variations are possible. All modifications which can be derived or suggested by a person skilled in the art from the disclosure of the present invention are to be considered within the scope of the invention.
Claims (7)
1. An automatic lifting sickbed system based on electroencephalogram and electrooculogram signals is characterized by comprising an electroencephalogram sensor, a sickbed control system and a sickbed body, wherein the sickbed body has lifting, active calling and alarming functions;
the brain wave sensor is connected with the hospital bed control system and used for collecting the electroencephalogram and electro-oculogram fusion signals of the patient and transmitting the electroencephalogram and electro-oculogram fusion signals to the hospital bed control system; the sickbed control system sends a control instruction according to the electroencephalogram and electrooculogram fusion signal of the patient, the control instruction is used for controlling the sickbed body to rise and level, actively calling and alarming, and the sickbed body responds according to the control instruction;
the hospital bed control system comprises:
the brain electric eye electrical fusion signal noise filtering module is used for carrying out noise filtering processing on the brain electric eye electrical fusion signal;
the electroencephalogram electro-fusion signal segmentation window module is used for segmenting the electroencephalogram electro-fusion signal subjected to noise filtering into different groups with consistent length;
and the brain electrical eye electrical fusion signal mode identification module is used for identifying the type of the characteristic of each group of brain electrical eye electrical fusion signals and sending instructions according to the type of the characteristic.
2. The electroencephalogram electro-oculogram signal based automatic lifting hospital bed system according to claim 1, wherein the electroencephalogram sensor is wirelessly connected with a hospital bed control system.
3. The automatic lifting sickbed system based on the electroencephalogram electro-oculogram signal as claimed in claim 1, wherein the sickbed control system sends a control instruction according to the electroencephalogram electro-oculogram fusion signal of the patient, and comprises:
s1, noise filtering is carried out on the electroencephalogram electro-oculogram fusion signal, and electroencephalogram electro-oculogram data after noise filtering are obtained;
s2, dividing the electroencephalogram electro-ocular data after noise filtering into groups to obtain electroencephalogram electro-ocular data groups, and calculating waveform characteristics of each electroencephalogram electro-ocular data group, wherein the waveform characteristics of the electroencephalogram electro-ocular data groups comprise an average value, an absolute average value, a mean square error and a standard deviation;
s3, classifying the waveform characteristics of each electro-oculogram data set relative to the waveform characteristics of the sample library data set by adopting a K-nearest neighbor algorithm to obtain the characteristic type of each electro-oculogram data set; and sending a control command according to the characteristic type of each electroencephalogram electro-ocular data set.
4. The electroencephalogram electro-oculogram signal based automatic lifting hospital bed system according to claim 3, it is characterized in that the preparation method is characterized in that, the characteristic types of each brain electricity and eye data set comprise one-time blinking, two-time blinking, eyeball rotation and brain electricity abnormity, and the corresponding control instructions are respectively lifting and leveling of the sickbed, active calling and alarming.
5. The EEG-ELECTRO-OPTOMIC SIGNAL-BASED AUTOMATIC RISE sickbed system according to claim 3, wherein said step S3 comprises:
calculating the distance between the waveform characteristics of the electroencephalogram electro-ocular data set to be tested and each electroencephalogram electro-ocular data set in the sample library data set to obtain a distance set D = { D = 1,1 ，d 1,2 ，…，d 1,m1 ，d 2,1 ，d 2,2 ，…，d 2,m2 ，…，d 4,1 ，d 4,2 ，…，d 4,m4 M1 represents the number of electroencephalogram and electro-oculogram data sets belonging to a one-time blinking type in the sample library data set, m2 represents the number of electroencephalogram and electro-oculogram data sets belonging to a two-time blinking type in the sample library data set, m3 represents the number of electroencephalogram and electro-oculogram data sets belonging to an eyeball rotation type in the sample library data set, m4 represents the number of electroencephalogram and electro-oculogram data sets belonging to an electroencephalogram abnormal type in the sample library data set, and d 1,m1 Representing the distance, d, between the waveform characteristics of the electroencephalogram electro-ocular data set to be tested and the waveform characteristics of the electroencephalogram electro-ocular data set of the m1 st blink type 2,m2 Representing the distance, d, between the waveform characteristics of the electroencephalogram electro-ocular data set to be tested and the waveform characteristics of the electroencephalogram electro-ocular data set of the m2 th blink type 3,m3 Representing the electroencephalogram and ocular counts to be testedAccording to the distance between the waveform characteristics of the group and the waveform characteristics of the m3 th eyeball rotation type electroencephalogram electro-ocular data group, d 4,m4 Representing the distance between the waveform characteristics of the electroencephalogram electro-ocular data set to be tested and the waveform characteristics of the electroencephalogram electro-ocular data set of the m4 th electroencephalogram abnormal type;
and (3) carrying out ascending sequence arrangement on the numerical values in the distance set D, taking the first K distances, judging the number of the four characteristic types in the sample library data group corresponding to the first K distances, and taking the characteristic type with the largest number as the characteristic type of the electroencephalogram and electrooculogram data group to be tested.
6. The automatic lifting sickbed system based on electroencephalogram and electrooculogram signals according to claim 3 or 5, the method for constructing the sample database data set is characterized by comprising the following steps:
s31, acquiring electroencephalogram electro-oculogram signals of different patients under the conditions of primary blink, secondary blink, eyeball rotation and electroencephalogram abnormity through an electroencephalogram sensor, carrying out noise filtering on the electroencephalogram electro-oculogram signals, dividing the electroencephalogram electro-oculogram signals into groups, extracting waveform characteristics of each data group, and obtaining a database with p groups of total data;
s32, randomly generating m sample base data sets, r training base data sets and S testing base data sets from the data base with the total data volume of p groups; the sample library data set, the training library data set and the testing library data set respectively contain four types of data of blink once, blink twice, eyeball rotation and electroencephalogram abnormity and are not repeated;
s33, repeating the step S32, initializing a plurality of data sets, wherein each data set consists of a sample base data group, a training base data group, a testing base data group and a K value parameter of a K neighbor algorithm which are initialized randomly;
calculating a distance set between the waveform characteristics of each training base data set and the waveform characteristics of each sample base data set, performing characteristic classification on one training base data set by adopting a K nearest neighbor algorithm, constructing a target function according to the recognition rate and the misjudgment rate, and selecting an optimal data set;
s34, taking the sample library data group in the optimal data set obtained in the step S33 as an optimal sample library data group, calculating a distance set between the waveform characteristics of the test library data group in the optimal data set and the waveform characteristics of each sample library data group, performing characteristic classification on one test library data group by adopting a K-nearest neighbor algorithm, constructing a target function according to the recognition rate and the misjudgment rate, and judging whether the preset requirements are met;
if yes, taking the sample database data group in the optimal data set obtained in the step S33 as a final optimal sample database data group;
if not, returning to the step S32, and reinitializing a plurality of data sets.
7. The electroencephalogram electro-ocular signal based automatic lifting hospital bed system according to claim 6, wherein in the step S33, the method further comprises the steps of selecting, intersecting and mutating data groups in the initialized data sets by adopting a genetic algorithm, comparing the obtained new data set with the original data set, and storing a more optimal data set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210842339.0A CN115227504B (en) | 2022-07-18 | 2022-07-18 | Automatic lifting sickbed system based on electroencephalogram-eye electric signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210842339.0A CN115227504B (en) | 2022-07-18 | 2022-07-18 | Automatic lifting sickbed system based on electroencephalogram-eye electric signals |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115227504A true CN115227504A (en) | 2022-10-25 |
| CN115227504B CN115227504B (en) | 2023-05-26 |
Family
ID=83672650
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210842339.0A Active CN115227504B (en) | 2022-07-18 | 2022-07-18 | Automatic lifting sickbed system based on electroencephalogram-eye electric signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115227504B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116849942A (en) * | 2023-07-28 | 2023-10-10 | 中国医学科学院生物医学工程研究所 | Brain-control intelligent lifting and turning-over multifunctional medical care bed |
Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6674908B1 (en) * | 2002-05-04 | 2004-01-06 | Edward Lasar Aronov | Method of compression of binary data with a random number generator |
| US20040024298A1 (en) * | 2002-08-05 | 2004-02-05 | Infraredx, Inc. | Spectroscopic unwanted signal filters for discrimination of vulnerable plaque and method therefor |
| US20090137924A1 (en) * | 2007-08-27 | 2009-05-28 | Microsoft Corporation | Method and system for meshing human and computer competencies for object categorization |
| US20100063948A1 (en) * | 2008-09-10 | 2010-03-11 | Digital Infuzion, Inc. | Machine learning methods and systems for identifying patterns in data |
| CN104063645A (en) * | 2014-07-01 | 2014-09-24 | 清华大学深圳研究生院 | Identity recognition method based on ECG dynamic self-updating samples |
| CN106200984A (en) * | 2016-07-21 | 2016-12-07 | 天津大学 | Mental imagery brain-computer interface model modelling approach |
| CN106709572A (en) * | 2015-11-16 | 2017-05-24 | 阿里巴巴集团控股有限公司 | Data processing method and equipment |
| CN109165615A (en) * | 2018-08-31 | 2019-01-08 | 中国人民解放军军事科学院军事医学研究院 | A kind of parameter training algorithm under multi-categorizer single channel mode towards EEG signals |
| CN109284004A (en) * | 2018-10-29 | 2019-01-29 | 中国矿业大学 | A kind of intelligent nursing system based on brain-computer interface |
| CN110162182A (en) * | 2019-05-28 | 2019-08-23 | 深圳市宏智力科技有限公司 | Brain electric control module device and its method for controlling controlled plant |
| CN110969108A (en) * | 2019-11-25 | 2020-04-07 | 杭州电子科技大学 | Limb action recognition method based on autonomic motor imagery electroencephalogram |
| CN112370017A (en) * | 2020-11-09 | 2021-02-19 | 腾讯科技（深圳）有限公司 | Training method and device of electroencephalogram classification model and electronic equipment |
| CN112890834A (en) * | 2021-03-01 | 2021-06-04 | 福州大学 | Attention-recognition-oriented machine learning-based eye electrical signal classifier |
| CN113095511A (en) * | 2021-04-16 | 2021-07-09 | 广东电网有限责任公司 | Method and device for judging in-place operation of automatic master station |
| CN113269217A (en) * | 2020-12-14 | 2021-08-17 | 北方信息控制研究院集团有限公司 | Radar target classification method based on Fisher criterion |
| CN113269201A (en) * | 2021-04-25 | 2021-08-17 | 浙江师范大学 | Hyperspectral image band selection method and system based on potential feature fusion |
| CN113842118A (en) * | 2021-12-01 | 2021-12-28 | 浙江大学 | Epileptic seizure real-time detection monitoring system for epileptic video electroencephalogram examination |
| CN113850192A (en) * | 2021-09-26 | 2021-12-28 | 杭州电子科技大学 | Blink artifact detection method based on EMD-CSP electroencephalogram feature learning and intelligent fusion |
| CN216168407U (en) * | 2021-10-20 | 2022-04-05 | 科大讯飞华南有限公司 | Hospital bed control system based on brain waves |
-
2022
- 2022-07-18 CN CN202210842339.0A patent/CN115227504B/en active Active
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6674908B1 (en) * | 2002-05-04 | 2004-01-06 | Edward Lasar Aronov | Method of compression of binary data with a random number generator |
| US20040024298A1 (en) * | 2002-08-05 | 2004-02-05 | Infraredx, Inc. | Spectroscopic unwanted signal filters for discrimination of vulnerable plaque and method therefor |
| US20090137924A1 (en) * | 2007-08-27 | 2009-05-28 | Microsoft Corporation | Method and system for meshing human and computer competencies for object categorization |
| US20100063948A1 (en) * | 2008-09-10 | 2010-03-11 | Digital Infuzion, Inc. | Machine learning methods and systems for identifying patterns in data |
| CN104063645A (en) * | 2014-07-01 | 2014-09-24 | 清华大学深圳研究生院 | Identity recognition method based on ECG dynamic self-updating samples |
| CN106709572A (en) * | 2015-11-16 | 2017-05-24 | 阿里巴巴集团控股有限公司 | Data processing method and equipment |
| CN106200984A (en) * | 2016-07-21 | 2016-12-07 | 天津大学 | Mental imagery brain-computer interface model modelling approach |
| CN109165615A (en) * | 2018-08-31 | 2019-01-08 | 中国人民解放军军事科学院军事医学研究院 | A kind of parameter training algorithm under multi-categorizer single channel mode towards EEG signals |
| CN109284004A (en) * | 2018-10-29 | 2019-01-29 | 中国矿业大学 | A kind of intelligent nursing system based on brain-computer interface |
| CN110162182A (en) * | 2019-05-28 | 2019-08-23 | 深圳市宏智力科技有限公司 | Brain electric control module device and its method for controlling controlled plant |
| CN110969108A (en) * | 2019-11-25 | 2020-04-07 | 杭州电子科技大学 | Limb action recognition method based on autonomic motor imagery electroencephalogram |
| CN112370017A (en) * | 2020-11-09 | 2021-02-19 | 腾讯科技（深圳）有限公司 | Training method and device of electroencephalogram classification model and electronic equipment |
| CN113269217A (en) * | 2020-12-14 | 2021-08-17 | 北方信息控制研究院集团有限公司 | Radar target classification method based on Fisher criterion |
| CN112890834A (en) * | 2021-03-01 | 2021-06-04 | 福州大学 | Attention-recognition-oriented machine learning-based eye electrical signal classifier |
| CN113095511A (en) * | 2021-04-16 | 2021-07-09 | 广东电网有限责任公司 | Method and device for judging in-place operation of automatic master station |
| CN113269201A (en) * | 2021-04-25 | 2021-08-17 | 浙江师范大学 | Hyperspectral image band selection method and system based on potential feature fusion |
| CN113850192A (en) * | 2021-09-26 | 2021-12-28 | 杭州电子科技大学 | Blink artifact detection method based on EMD-CSP electroencephalogram feature learning and intelligent fusion |
| CN216168407U (en) * | 2021-10-20 | 2022-04-05 | 科大讯飞华南有限公司 | Hospital bed control system based on brain waves |
| CN113842118A (en) * | 2021-12-01 | 2021-12-28 | 浙江大学 | Epileptic seizure real-time detection monitoring system for epileptic video electroencephalogram examination |
Non-Patent Citations (4)
| Title |
|---|
| ALI, AMJAD, 等: "A k-Nearest Neighbours Based Ensemble via Optimal Model Selection for Regression" * |
| 张英凯: "基于机器学习的重症监护病患死亡率预测" * |
| 曹宁: "基于静态图像的人脸表情识别算法研究" * |
| 蔡建程,等: "离心风机振动噪声及压力脉动实验研究" * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116849942A (en) * | 2023-07-28 | 2023-10-10 | 中国医学科学院生物医学工程研究所 | Brain-control intelligent lifting and turning-over multifunctional medical care bed |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115227504B (en) | 2023-05-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117457147B (en) | Personalized nursing planning method and system for rehabilitation patient | |
| Barsocchi | Position recognition to support bedsores prevention | |
| EP1864246A1 (en) | Spatio-temporal self organising map | |
| KR20210085867A (en) | Apparatus and method for estimating user's blood pressure | |
| CN109284004A (en) | A kind of intelligent nursing system based on brain-computer interface | |
| CN115227504B (en) | Automatic lifting sickbed system based on electroencephalogram-eye electric signals | |
| CN117854739A (en) | Intelligent internal medicine nursing monitoring system | |
| CN109949438A (en) | Abnormal driving monitoring model method for building up, device and storage medium | |
| CN114849063B (en) | Extracorporeal charger, program-controlled system, and computer-readable storage medium | |
| CN117877734A (en) | Mental disorder patient risk level judging system | |
| Turgunov et al. | Comparative analysis of the results of EMG signal classification based on machine learning algorithms | |
| Khorshidtalab et al. | Evaluation of time-domain features for motor imagery movements using FCM and SVM | |
| Pogorelc et al. | Home-based health monitoring of the elderly through gait recognition | |
| JP2014239789A (en) | Sleep stage estimating device, method, and program | |
| CN113380397A (en) | Medical system for providing treatment recommendations | |
| CN118038548A (en) | Abnormal behavior detection method, device, electronic equipment and storage medium | |
| Morresi et al. | Machine learning algorithms for the activity monitoring of elders by home sensor network | |
| EP4294259A1 (en) | Systems, methods, and media for decoding observed spike counts for spiking cells | |
| CN115501442A (en) | Sleep regulation system based on pulse and electroencephalogram signals | |
| KR101375673B1 (en) | Method for warning of epileptic seizure using excitatory-inhibitory model based on the chaos neuron and electronic device supporting the same | |
| Wang et al. | Epileptic seizures prediction based on unsupervised learning for feature extraction | |
| Suganthi et al. | Pattern recognition for EMG based forearm orientation and contraction in myoelectric prosthetic hand | |
| CN118658603B (en) | Medical care monitoring method and system based on computer vision and various sensors | |
| Madhavan | Design and Evaluation of a Brain Signal-based Monitoring System for Differently-Abled People | |
| Stephen et al. | Automatic autonomous heart monitoring device using machine learning for COVID patients |
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 |