CN117694906B - Early warning method, device and system for cerebral apoplexy attack - Google Patents
Early warning method, device and system for cerebral apoplexy attack Download PDFInfo
- Publication number
- CN117694906B CN117694906B CN202410073352.3A CN202410073352A CN117694906B CN 117694906 B CN117694906 B CN 117694906B CN 202410073352 A CN202410073352 A CN 202410073352A CN 117694906 B CN117694906 B CN 117694906B
- Authority
- CN
- China
- Prior art keywords
- electroencephalogram
- signals
- time
- cerebral apoplexy
- airspace
- 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
- 208000006011 Stroke Diseases 0.000 title claims abstract description 69
- 230000002490 cerebral effect Effects 0.000 title claims abstract description 65
- 206010008190 Cerebrovascular accident Diseases 0.000 title claims abstract description 61
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000012545 processing Methods 0.000 claims abstract description 15
- 238000000605 extraction Methods 0.000 claims abstract description 13
- 238000007781 pre-processing Methods 0.000 claims abstract description 12
- 230000002159 abnormal effect Effects 0.000 claims abstract description 5
- 210000004556 brain Anatomy 0.000 claims description 71
- 238000004422 calculation algorithm Methods 0.000 claims description 58
- 239000011159 matrix material Substances 0.000 claims description 46
- 230000004927 fusion Effects 0.000 claims description 12
- 238000009826 distribution Methods 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 5
- 239000013598 vector Substances 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 239000000203 mixture Substances 0.000 claims description 3
- 208000010125 myocardial infarction Diseases 0.000 claims 1
- 230000009471 action Effects 0.000 abstract description 19
- 238000004458 analytical method Methods 0.000 abstract description 12
- 230000008713 feedback mechanism Effects 0.000 abstract description 4
- 238000012544 monitoring process Methods 0.000 abstract description 3
- 230000002123 temporal effect Effects 0.000 abstract description 3
- 238000004364 calculation method Methods 0.000 abstract description 2
- 238000003745 diagnosis Methods 0.000 abstract description 2
- 238000011156 evaluation Methods 0.000 abstract 1
- 238000012880 independent component analysis Methods 0.000 description 21
- 125000004122 cyclic group Chemical group 0.000 description 10
- 230000003993 interaction Effects 0.000 description 10
- 238000012937 correction Methods 0.000 description 9
- 238000004590 computer program Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 230000005856 abnormality Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000001994 activation Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 206010008118 cerebral infarction Diseases 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 230000004075 alteration Effects 0.000 description 2
- 208000026106 cerebrovascular disease Diseases 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 201000006474 Brain Ischemia Diseases 0.000 description 1
- 206010008111 Cerebral haemorrhage Diseases 0.000 description 1
- 206010008120 Cerebral ischaemia Diseases 0.000 description 1
- 206010008132 Cerebral thrombosis Diseases 0.000 description 1
- 201000001429 Intracranial Thrombosis Diseases 0.000 description 1
- 208000032851 Subarachnoid Hemorrhage Diseases 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 208000035474 group of disease Diseases 0.000 description 1
- 208000037907 haemorrhagic injury Diseases 0.000 description 1
- 230000002008 hemorrhagic effect Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000000302 ischemic effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 231100000862 numbness Toxicity 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 210000004761 scalp Anatomy 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000012549 training Methods 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]
-
- 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
- A61B5/72—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
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
-
- 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
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
-
- 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
- A61B5/7271—Specific aspects of physiological measurement analysis
-
- 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
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
-
- 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
- A61B5/742—Details of notification to user or communication with user or patient; User input means using visual displays
-
- 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
- A61B5/7455—Details of notification to user or communication with user or patient; User input means characterised by tactile indication, e.g. vibration or electrical stimulation
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- 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
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Medical Informatics (AREA)
- Heart & Thoracic Surgery (AREA)
- Pathology (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Psychiatry (AREA)
- Data Mining & Analysis (AREA)
- Physiology (AREA)
- Signal Processing (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Psychology (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses a method, a device and a system for early warning of cerebral apoplexy attacks, and relates to the technical field of cerebral apoplexy attack early warning. According to the early warning method for cerebral apoplexy attacks, the spatial and temporal characteristics of the cerebral electrical signals are combined through the steps of acquisition, preprocessing, feature extraction and analysis, the cerebral electrical signals of the user are comprehensively analyzed from multiple angles, information of different stages is integrated, the prediction precision of the cerebral apoplexy attacks is improved, the data processing and feature extraction enable the system to accurately identify abnormal changes in the cerebral electrical signals, a real-time model is built by a digital twin principle, the whole early warning system is deployed on a cloud platform, the calculation power of real-time data processing, model updating and feedback mechanisms is increased, the framework allows continuous monitoring of the cerebral electrical signals and action states of the user, the model is updated at any time, and data are corrected to adapt to individual changes, so that the comprehensive application system for daily diagnosis and evaluation of individuals is formed, and the time window of cerebral apoplexy attacks early warning is rebuilt.
Description
Technical Field
The invention relates to the technical field of cerebral apoplexy attack pre-warning, in particular to a method, a device and a system for pre-warning cerebral apoplexy attack.
Background
Cerebral apoplexy is a group of diseases which take cerebral ischemia and hemorrhagic injury symptoms as main clinical manifestations, has extremely high fatality rate and disability rate, and is mainly divided into two major categories of hemorrhagic cerebral apoplexy (cerebral hemorrhage or subarachnoid hemorrhage) and ischemic cerebral apoplexy (from cerebral infarction and cerebral thrombosis), wherein cerebral infarction is the most common. Cerebral apoplexy is urgent and has high death rate. According to the report 2020 for preventing and treating cerebral apoplexy, cerebral apoplexy is the primary cause of death and disability of adults in China, and the burden of cerebral apoplexy diseases in China has an increased situation. Cerebral apoplexy is used as a controllable disease, and the early screening and intervention has obvious prognosis effect.
The prior art can be used for early warning cerebral apoplexy, and can realize early warning at the first time after limb numbness, but a method for early warning cerebral apoplexy through brain waves is not available yet, and more timely early warning is difficult to carry out, so that the problem of high disability rate and mortality rate of patients is caused.
Precision medicine is widely known and has expanded applications.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a method, a device and a system for early warning of cerebral apoplexy, which solve the problems that the prior art does not have a method for early warning cerebral apoplexy through brain waves, is difficult to perform early warning in time, and leads to high disability rate and death rate of patients.
In order to achieve the above purpose, the invention is realized by the following technical scheme: the early warning method for cerebral apoplexy comprises the following steps: an electroencephalogram signal is obtained by using an electrode helmet, preprocessing is carried out on the collected electroencephalogram signal based on an ICA algorithm, interference and artifacts are removed, an electroencephalogram signal matrix is obtained, and key characteristics of an original signal are reserved; the brain electrical signal matrix enters a CSP algorithm, the difference between left brain signals and right brain signals is extracted, and airspace characteristics are extracted and used for calculating brain electrical signals to obtain spatial distribution; the airspace features enter an LSTM algorithm, modeling is carried out on the evolution of the electroencephalogram signals in time, a threshold matrix is established on the fluctuation of the electroencephalogram signals, the model learns the historical data of the electroencephalogram signals, and the future possible attack situation is predicted; and feeding back the predicted result to the AI glove, analyzing the user state by the AI glove based on the temperature, pressure, humidity and position sensors, comparing the feedback of the digital signal and the analog signal with the predicted electroencephalogram signal, performing cyclic interaction and matching correction of actions and the electroencephalogram signal, and controlling a program by virtue of a terminal.
Further, the steps of obtaining the electroencephalogram signal matrix and retaining key features of the original signal are as follows: the electrode helmet acquires an electroencephalogram signal, and the acquired multichannel electroencephalogram signal is decomposed into components which are independent in time and fixed in space; the ICA algorithm finds a decomposition matrix W such that u=wx, where U is an electroencephalogram matrix representing the time course of ICA component activation and X is an electroencephalogram initial matrix acquired by the electrode helmet.
Further, the steps of extracting the difference between the left brain signal and the right brain signal and extracting the airspace characteristics are as follows: acquiring an electroencephalogram signal in a normal state and an electroencephalogram signal in a cerebral apoplexy state; acquiring a spatial filter based on the brain electrical signals in a normal state and the brain electrical signals in a cerebral apoplexy state through a set optimization function J (W); and the airspace filter processes the electroencephalogram signal matrix U to obtain airspace characteristics F.
Further, the obtained spatial characteristic F is f= [ F 1,f2,f3,...fm]T, where F m represents a matrix obtained by processing the electroencephalogram signal matrix U by the mth spatial filter.
Further, the airspace features enter an LSTM algorithm, the evolution of the electroencephalogram signals in time is modeled, the model learns the historical data of the electroencephalogram signals, and the step of predicting the future possible attack situation is as follows: the airspace characteristics are formed into a time sequence; the LSTM model learns the mode of the historical data, and understands the change trend of the brain electrical signal in time, wherein the historical data carries the label of the cerebral apoplexy attack situation; inputting a time sequence formed by airspace characteristics as an input sequence of an LSTM model; the LSTM model outputs a predicted value for representing the probability of future stroke attacks.
Further, the spatial feature composition time sequence is x= [ X 1,x2,x3,...xn]T ], where X n represents an input feature vector of the nth time step.
Further, the steps of performing the cyclic interaction and the cooperation correction of the actions and the brain waves based on the analysis of the user state by the AI glove and comparing with the predicted brain wave signals are as follows: the predicted value output by the LSTM model is transmitted to a digital twin cloud and is connected with AI gloves worn by a user; cloud loading, namely presetting an action instruction, a scene database and a target classification parameter database; the AI glove monitors the current motion state of the user through the sensor, including the position, humidity, temperature and pressure of the hand; and (3) interactively comparing the predicted value with the actual motion state monitored by the AI glove, correcting the data, and comparing the corrected data with the detection result before 24 hours, if the comparison between the predicted result and the actual motion state shows that the abnormality occurs, namely the brain electrical signal predicts the occurrence of the potential cerebral apoplexy, and notifying the user and the medical staff.
The early warning device for cerebral apoplexy comprises an electroencephalogram signal matrix acquisition module, an airspace feature extraction module, a prediction model construction module and a feedback module, wherein: the electroencephalogram signal matrix acquisition module is used for acquiring an electroencephalogram signal by using the electrode helmet, preprocessing the acquired electroencephalogram signal based on an ICA algorithm, removing interference and artifacts, acquiring an electroencephalogram signal matrix, and retaining key characteristics of an original signal; the airspace feature extraction module is used for enabling the electroencephalogram signal matrix to enter a CSP algorithm, extracting the difference between left and right brain signals, extracting airspace features and calculating the electroencephalogram signals to obtain spatial distribution; the prediction model construction module is used for enabling airspace characteristics to enter an LSTM algorithm, modeling evolution of an electroencephalogram signal in time, and predicting possible future attack conditions by learning historical data of the electroencephalogram signal; the feedback module is used for feeding back the predicted result to the AI glove, comparing the predicted electroencephalogram signal with the predicted electroencephalogram signal based on the analysis of the user state by the AI glove, and carrying out the cyclic interaction and the coordination correction of the action and the electroencephalogram.
A early warning system for cerebral apoplexy is at a start, including algorithm fusion module, feedback module and last Yun Rong cloud module, wherein: the algorithm fusion module is used for cleaning, extracting and interacting the acquired electroencephalogram signals in a multi-algorithm fusion mode; the feedback module is used for synchronously collecting brain wave signals in the execution process of the AI glove actions and feeding back to the calculation force center again, so that the action and brain wave cyclic interaction and matching correction are realized; the upper Yun Rong cloud module is used for fusing algorithms and carrying out upper cloud fusion through a digital twin principle building architecture.
Further, the algorithm of the algorithm fusion module comprises an ICA algorithm, a CSP algorithm and an LSTM algorithm, wherein the ICA algorithm is used for preprocessing the acquired brain electrical signals, removing interference and artifacts to obtain brain electrical signal matrixes, retaining key features of original signals, the CSP algorithm is used for extracting differences between left brain signals and right brain signals, extracting airspace features and calculating brain electrical signals to obtain spatial distribution, the LSTM algorithm is used for modeling evolution of the brain electrical signals in time, learning historical data of the brain electrical signals by a model, and predicting possible future attack conditions.
The invention has the following beneficial effects:
(1) According to the early warning method for the cerebral apoplexy attack, through the steps of acquisition, preprocessing, feature extraction and analysis, and by combining with the spatial and temporal features of the brain electrical signals, the method can comprehensively analyze the brain electrical signals of the user from multiple angles and integrate the information of different stages, so that the prediction precision of the cerebral apoplexy attack is improved, the system can accurately identify abnormal changes in the brain electrical signals through careful data processing and feature extraction, the false alarm rate is reduced, and the early warning reliability is improved.
(2) According to the early warning method for cerebral apoplexy, a real-time model is established through a digital twin principle, the whole early warning system is deployed on a cloud platform, a real-time data processing, model updating and feedback mechanism is realized, the framework allows the system to continuously monitor the electroencephalogram signals and action states of users, the model is updated at any time to adapt to individual changes, and meanwhile, cloud deployment also provides possibility for summarizing and analyzing multi-user data, so that the early warning model and algorithm are further optimized, and the early warning accuracy is improved.
(3) According to the early warning method for cerebral apoplexy, various algorithms (ICA, CSP and LSTM) are fused together, and the early warning system can comprehensively analyze brain electrical signals from different angles, and clean, extract and interact data. The advantages of each algorithm are fully exerted, so that the accuracy and reliability of prediction are improved, the ICA is used for eliminating interference, the CSP is used for extracting airspace characteristics, the LSTM modeling time sequence is used for constructing a powerful prediction model together, the application of the upper Yun Rong cloud module enables the whole system to be deployed and managed in the cloud, the relation between the electroencephalogram signals and the motion of a user can be simulated in real time by the system through the digital twin principle, the model is updated at any time to adapt to individual changes, and the framework allows the early warning system to be optimized continuously, so that the timeliness of prediction is improved.
Of course, it is not necessary for any one product to practice the invention to achieve all of the advantages set forth above at the same time.
Drawings
FIG. 1 is a flow chart of the method for early warning of stroke episodes according to the present invention.
Fig. 2 is a flowchart of the steps for obtaining an electroencephalogram signal matrix and retaining key features of an original signal in the early warning method for cerebral apoplexy attacks.
Fig. 3 is a flowchart showing steps of extracting differences between left and right brain signals and extracting airspace characteristics according to the early warning method for cerebral apoplexy attacks of the present invention.
Fig. 4 is a flowchart showing the steps of the method for early warning of stroke attacks according to the present invention for predicting possible future attacks.
Detailed Description
The embodiment of the application realizes the early warning of the cerebral apoplexy through the brain wave by the early warning method, the device and the system for cerebral apoplexy attack, and improves the reliability of early warning.
The problems in the embodiment of the application have the following general ideas:
Preprocessing the acquired brain electrical signals by using an Independent Component Analysis (ICA) algorithm to remove interference and artifacts, simultaneously retaining key characteristics of original signals, constructing an brain electrical signal matrix, retaining time sequence details of the brain electrical signals, extracting differences between left and right brain signals by using the brain electrical signal matrix through a Common SPATIAL PATTERNS (CSP) algorithm to capture relevant information of cerebral apoplexy, and using the characteristics to calculate spatial distribution of the brain electrical signals.
Features processed by CSP are sent into a long and short time memory network (LSTM) to model the time sequence change trend of the electroencephalogram signals, the LSTM model learns the history data of the electroencephalogram signals, the predicted result is connected to AI gloves worn by a user through a digital twin model to analyze the current motion state of the user, and compared with the predicted electroencephalogram signal attack situation, if abnormality is predicted, the AI gloves inform the user through vibration, an indicator lamp and other modes, and the assistance of the gloves to motion can be possibly adjusted, so that potential risks are reduced.
And a digital twin model is established, the relation between the electroencephalogram signal and the motion of a user is simulated, and the whole system is deployed on a cloud platform, so that a real-time data processing, model updating and feedback mechanism is realized, and the continuous optimization and early warning precision improvement of the system are ensured.
Referring to fig. 1, the embodiment of the invention provides a technical scheme: the early warning method for cerebral apoplexy comprises the following steps: using bioelectric electroencephalogram signal acquisition equipment such as an electrode helmet, acquiring electroencephalogram signals through an active dry electrode, a 32 wireless EEG channel and a signal amplifier, preprocessing the acquired electroencephalogram signals based on an ICA algorithm, removing interference and artifacts, acquiring an electroencephalogram signal matrix, and retaining key characteristics of original signals; the brain electrical signal matrix enters a CSP algorithm, the difference between left brain signals and right brain signals is extracted, and airspace characteristics are extracted and used for calculating brain electrical signals to obtain spatial distribution; the airspace features enter an LSTM algorithm, modeling is carried out on the evolution of the electroencephalogram signals in time, the model learns the historical data of the electroencephalogram signals, and the possible future attack situation is predicted; and feeding back the predicted result to the AI glove, comparing the user state with the predicted electroencephalogram signal based on the analysis of the AI glove, and performing cyclic interaction and matching correction of actions and the electroencephalogram.
Specifically, as shown in fig. 2, the steps of obtaining an electroencephalogram signal matrix and retaining key features of an original signal are as follows: the electrode helmet acquires an electroencephalogram signal, and the acquired multichannel electroencephalogram signal is decomposed into components which are independent in time and fixed in space; the ICA algorithm finds a decomposition matrix W that decomposes the multi-channel scalp data into a linear sum of components that are independent in time and fixed in space, such that u=wx, where U is an electroencephalogram matrix, representing the time course of ICA component activation, and X is an electroencephalogram initial matrix acquired by the electrode helmet.
In this embodiment, the electroencephalogram signals are often interfered by the aspects of environment, electrodes and the like, and the mixed electroencephalogram signals may reduce the accuracy of predicting cerebral apoplexy, and by decomposing the electroencephalogram signals and retaining key characteristics, the ICA algorithm can effectively remove the interference and the artifacts, thereby improving the quality of the signals.
By preserving key features, i.e. preserving details of the original signal, it can be ensured that important information in the electroencephalogram signal is not lost, the predictive model can better identify and capture seizure-related signal patterns, the electroencephalogram signal is typically multi-channel, and each channel may contain a large number of data points, by decomposing the electroencephalogram signal into separate components, the dimensionality of the data can be reduced while preserving the key features, making subsequent analysis more efficient.
The decomposed independent component U represents the time activation process of the ICA component, and the information is very useful for analyzing the change of the brain electrical signal in time sequence, and is helpful for building a time model to predict the possible attack situation of the cerebral apoplexy.
Specifically, as shown in fig. 3, the difference between the left and right brain signals is extracted, and the steps of extracting airspace features are as follows: acquiring an electroencephalogram signal in a normal state and an electroencephalogram signal in a cerebral apoplexy state; acquiring a spatial filter based on the brain electrical signals in a normal state and the brain electrical signals in a cerebral apoplexy state through a set optimization function J (W); and the airspace filter processes the electroencephalogram signal matrix U to obtain airspace characteristics F.
The obtained spatial characteristic F is f= [ F 1,f2,f3,...fm]T ], wherein F m represents a matrix obtained by processing the electroencephalogram signal matrix U by the mth spatial filter.
In this embodiment, by acquiring the electroencephalogram signals in the normal state and the cerebral apoplexy state, the difference between the electroencephalogram signals in the two states can be captured, and the airspace filter obtained by the optimization function J (W) can be adjusted according to the characteristics of the electroencephalogram signals of each individual, so that the feature extraction process can adapt to the difference between the individuals, the individuation capability of the prediction model is improved, and the airspace characteristics related to the normal state and the cerebral apoplexy state can be extracted through the processing of the airspace filter.
The extracted spatial features F are the result of the electroencephalogram signal matrix U processed by the spatial filter, and compared with the original electroencephalogram signal, the features can be more representative, and meanwhile, the dimension of data can be reduced, so that the subsequent analysis is more efficient.
Specifically, as shown in fig. 4, the airspace feature enters an LSTM algorithm, the evolution of an electroencephalogram signal in time is modeled, a threshold matrix is established for fluctuation of the electroencephalogram signal, the model learns the historical data of the electroencephalogram signal, and the steps for predicting possible future attack conditions are as follows: the airspace characteristics are formed into a time sequence; the LSTM model learns the mode of the historical data, understands the change trend of the brain electrical signal in time, and the historical data carries the label of the cerebral apoplexy attack situation; inputting a time sequence formed by airspace characteristics as an input sequence of an LSTM model; the LSTM model outputs a predicted value for representing the probability of future stroke attacks.
The spatial signature constitutes a time series of x= [ X 1,x2,x3,...xn]T ], where X n represents the input signature vector for the nth time step.
In this embodiment, the spatial features are formed into a time sequence according to a time sequence, which is helpful for capturing the evolution situation of the electroencephalogram at different time points, the LSTM model can understand the time trend of the electroencephalogram by learning the mode of the historical data, and the model can establish the connection between the electroencephalogram and the seizure situation through the labels because the historical data carries the labels of the seizure situation of the cerebral apoplexy.
The LSTM model takes the current characteristics and the past information into consideration at each time point when a time sequence composed of airspace characteristics is input, so that the evolution of an electroencephalogram signal in the time dimension is captured, the probability of the future possible cerebral apoplexy can be output after training, and whether the future possible cerebral apoplexy is possible to occur can be predicted according to the current state of the electroencephalogram signal.
Specifically, the AI glove is based on analysis of the user state by temperature, pressure, humidity and position sensors, compares with the predicted electroencephalogram signal through feedback of digital signals and analog signals, performs cyclic interaction and matching correction of actions and the electroencephalogram, and performs the following steps by means of a terminal control program: the predicted value output by the LSTM model is transmitted to a digital twin cloud and is connected with AI gloves worn by a user; and a framework is built by using a digital twin principle, the three algorithms are fused in a time dimension, and the data are completely cleaned, extracted and interacted. And comparing the historical data with the actual motion state monitored by the AI glove, and if the comparison between the monitoring result and the historical data shows that the abnormality occurs, namely the electroencephalogram signal predicts the occurrence of the potential cerebral apoplexy, notifying the user to seek medical advice in time.
In this embodiment, the predicted value output by the LSTM model is combined with the actual motion state of the user, so that comprehensive analysis can be performed in two aspects: the prediction of the brain electrical signals and the actual actions of the users are carried out, the actual motion states of the users are monitored through the AI glove sensors, and the actual action information of the individuals can be obtained, so that the system can carry out personalized adaptation according to the characteristics of each user.
If abnormality is detected, namely the brain wave signal predicts that the brain stroke is likely to occur, the system can inform the user in real time in a vibration mode, an indicator lamp mode and the like, the user can seek medical attention in time, further diagnosis is carried out through imaging, and the time window for early warning of the brain stroke is increased.
The early warning device for cerebral apoplexy comprises an electroencephalogram signal matrix acquisition module, an airspace feature extraction module, a prediction model construction module and a feedback module, wherein: the electroencephalogram signal matrix acquisition module is used for acquiring an electroencephalogram signal by using the electrode helmet, preprocessing the acquired electroencephalogram signal based on an ICA algorithm, removing interference and artifacts, acquiring an electroencephalogram signal matrix, and retaining key characteristics of an original signal; the airspace feature extraction module is used for enabling the brain signal matrix to enter a CSP algorithm, extracting the difference between left and right brain signals, extracting airspace features and calculating brain signals to obtain spatial distribution; the prediction model construction module is used for enabling airspace characteristics to enter an LSTM algorithm, modeling the evolution of the electroencephalogram signals in time, and predicting possible future attack conditions by learning historical data of the electroencephalogram signals; the feedback module is used for feeding back the predicted result to the AI glove, comparing the predicted electroencephalogram signal with the predicted electroencephalogram signal based on the analysis of the user state by the AI glove, and carrying out the cyclic interaction and the cooperation correction of the action and the electroencephalogram.
A early warning system for cerebral apoplexy is at a start, including algorithm fusion module, feedback module and last Yun Rong cloud module, wherein: the algorithm fusion module is used for cleaning, extracting and interacting the acquired electroencephalogram signals in a multi-algorithm fusion mode; the feedback module is used for synchronously collecting brain wave signals in the execution process of the AI glove actions and feeding back to the calculation center again, so that the cyclic interaction and the coordination correction of the actions and the brain waves are realized; the upper Yun Rong cloud module is used for carrying out cloud loading and cloud loading on the fused algorithm through a digital twin principle building architecture.
Specifically, the algorithm of the algorithm fusion module comprises an ICA algorithm, a CSP algorithm and an LSTM algorithm, wherein the ICA algorithm is used for preprocessing the acquired electroencephalogram signals, removing interference and artifacts, obtaining an electroencephalogram signal matrix, retaining key characteristics of original signals, the CSP algorithm is used for extracting differences between left and right brain signals, extracting airspace characteristics and calculating the electroencephalogram signals to obtain spatial distribution, the LSTM algorithm is used for modeling evolution of the electroencephalogram signals in time, and the model learns historical data of the electroencephalogram signals and predicts possible future attack conditions.
In the embodiment, by fusing a plurality of algorithms (ICA, CSP and LSTM), the early warning system can comprehensively analyze the electroencephalogram signals, clean, extract and interact data from different angles. The advantages of each algorithm are fully exerted, so that the accuracy and reliability of prediction are improved, ICA (independent component analysis) eliminates interference, CSP (space vector analysis) extracts airspace characteristics, LSTM (linear model) models time sequences, and a powerful prediction model is built together.
The design of the feedback module enables the early warning system to realize the cyclic interaction of actions and brain waves, the AI glove synchronously collects the brain electrical signals of the user, the actual action state is compared with the predicted brain electrical signals, the cyclic interaction can correct the predicted result in real time, and the early warning system is ensured to be more accurate and reliable.
The application of the Yun Rong cloud module enables the whole system to be deployed and managed in the cloud, the system can simulate the relationship between the electroencephalogram signal and the motion of a user in real time through the digital twin principle, and the model is updated at any time to adapt to individual changes.
In summary, the present application has at least the following effects:
By means of the method, the electroencephalogram signals of the user can be comprehensively analyzed from multiple angles through the steps of acquisition, preprocessing, feature extraction and analysis and by combining the spatial and temporal features of the electroencephalogram signals, information of different phases is integrated, accordingly, the prediction accuracy of cerebral apoplexy attacks is improved, the abnormal changes in the electroencephalogram signals can be accurately identified through careful data processing and feature extraction, the false alarm rate is reduced, and the reliability of early warning is improved.
The brain electrical signal, the motion state and the prediction model of the individual are fused, personalized cerebral apoplexy early warning is realized, the user can perceive potential dangerous situations through monitoring and real-time feedback of the AI glove, the system can be adjusted according to the actual action situations of the user, targeted assistance and correction are provided, and the user experience of the whole early warning system is improved.
The real-time model is built through the digital twin principle, the whole early warning system is deployed on the cloud platform, a real-time data processing, model updating and feedback mechanism is realized, the framework allows the system to continuously monitor the electroencephalogram signals and the action states of users, the model is updated at any time to adapt to individual changes, meanwhile, cloud deployment also provides possibility for summarizing and analyzing multi-user data, and therefore the early warning model and algorithm are further optimized, and early warning accuracy is improved.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of systems, apparatuses (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, 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 specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (5)
1. The early warning device for cerebral apoplexy is characterized by comprising an electroencephalogram signal matrix acquisition module, an airspace feature extraction module, a prediction model construction module and a feedback module, wherein:
the electroencephalogram signal matrix acquisition module is used for acquiring an electroencephalogram signal by using the electrode helmet, preprocessing the acquired electroencephalogram signal based on an ICA algorithm, removing interference and artifacts, acquiring an electroencephalogram signal matrix, and retaining key characteristics of an original signal;
The airspace feature extraction module is used for enabling the electroencephalogram signal matrix to enter a CSP algorithm, extracting the difference between left and right brain signals, extracting airspace features and calculating the electroencephalogram signals to obtain spatial distribution;
The prediction model construction module is used for enabling the space characteristics to enter an LSTM model, modeling the evolution of the electroencephalogram signals in time, and the model learns the historical data of the electroencephalogram signals to capture the evolution of the electroencephalogram signals in the time dimension; specifically, the airspace features are formed into a time sequence, an LSTM model learns the mode of historical data, the change trend of the electroencephalogram signals in time is understood, the historical data carries a cerebral apoplexy onset condition label, the time sequence formed by the airspace features is input and used as an input sequence of the LSTM model, the LSTM model outputs a predicted value for representing the probability of future cerebral apoplexy onset, and the airspace feature composition time sequence is that WhereinRepresent the firstInputting feature vectors of the time steps;
The feedback module is used for comparing the predicted value output by the LSTM model with the actual motion state monitored by the AI glove, if the comparison between the predicted value and the actual motion state is abnormal, namely the brain electrical signal predicts the occurrence of the potential brain stroke attack, informing the user to seek medical attention in time, specifically, transmitting the predicted value output by the LSTM model to the digital twin model, and connecting the predicted value and the AI glove worn by the user; AI gloves monitor the actual motion state of the user, including the position and speed of the hand, through sensors.
2. The apparatus of claim 1, wherein the obtaining the electroencephalogram signal matrix and retaining key features of the original signal comprises:
The electrode helmet acquires an electroencephalogram signal, and the acquired multichannel electroencephalogram signal is decomposed into components which are independent in time and fixed in space;
ICA algorithm finds the decomposition matrix So thatWhereinFor an electroencephalogram signal matrix, representing the time course of the activation of the ICA component,Is an electroencephalogram signal initial matrix acquired by the electrode helmet.
3. The apparatus according to claim 2, wherein the extracting of the difference between the left and right brain signals, extracting spatial features, comprises:
acquiring an electroencephalogram signal in a normal state and an electroencephalogram signal in a cerebral apoplexy state;
By means of a set optimization function Acquiring a spatial filter based on the brain electrical signals in the normal state and the brain electrical signals in the cerebral apoplexy state;
brain-electrical signal matrix of airspace filter Processing to obtain airspace characteristics。
4. A warning device as claimed in claim 3, wherein the spatial signature is obtainedIs thatWhereinRepresent the firstBrain-electrical signal matrix of individual airspace filterA processed matrix.
5. A early warning system for cerebral apoplexy is at a heart attack, its characterized in that includes algorithm fusion module, feedback module and last Yun Rong cloud module, wherein:
The algorithm fusion module is used for preprocessing the acquired electroencephalogram signals based on an ICA algorithm, removing interference and artifacts, obtaining an electroencephalogram signal matrix and retaining key characteristics of original signals; the brain electrical signal matrix enters a CSP algorithm, the difference between left brain signals and right brain signals is extracted, airspace characteristics are extracted, and the brain electrical signals are calculated to obtain spatial distribution; the method comprises the steps that space features enter an LSTM model, modeling is carried out on the evolution of the electroencephalogram signals in time, and the LSTM model learns the historical data of the electroencephalogram signals so as to capture the evolution of the electroencephalogram signals in the time dimension; specifically, the airspace features are formed into a time sequence, an LSTM model learns the mode of historical data, the change trend of the electroencephalogram signals in time is understood, the historical data carries a cerebral apoplexy onset condition label, the time sequence formed by the airspace features is input and used as an input sequence of the LSTM model, the LSTM model outputs a predicted value for representing the probability of future cerebral apoplexy onset, and the airspace feature composition time sequence is that WhereinRepresent the firstInputting feature vectors of the time steps;
the feedback module is used for comparing the predicted value output by the LSTM model with the actual motion state monitored by the AI glove, if the comparison between the predicted value and the actual motion state is abnormal, namely the brain electrical signal predicts the occurrence of the potential brain stroke attack, informing the user to seek medical attention in time, specifically, transmitting the predicted value output by the LSTM model to the digital twin model, and connecting the predicted value and the AI glove worn by the user; AI gloves monitor the actual motion state of the user, including the position and speed of the hand, through sensors;
The upper Yun Rong cloud module is used for building a framework through a digital twin principle to enable a fused algorithm to carry out cloud-up fusion, and the relation between the electroencephalogram signals and the movements of the user is simulated in real time through the digital twin principle, so that the model is updated at any time to adapt to individual changes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410073352.3A CN117694906B (en) | 2024-01-18 | 2024-01-18 | Early warning method, device and system for cerebral apoplexy attack |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410073352.3A CN117694906B (en) | 2024-01-18 | 2024-01-18 | Early warning method, device and system for cerebral apoplexy attack |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117694906A CN117694906A (en) | 2024-03-15 |
| CN117694906B true CN117694906B (en) | 2024-10-25 |
Family
ID=90157270
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410073352.3A Active CN117694906B (en) | 2024-01-18 | 2024-01-18 | Early warning method, device and system for cerebral apoplexy attack |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117694906B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118216937B (en) * | 2024-04-30 | 2024-11-12 | 北京瑞蜜达国际生物科技有限公司 | A method and system for analyzing and processing electromyographic signals based on big data |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112489793A (en) * | 2020-12-16 | 2021-03-12 | 郑州航空工业管理学院 | Early warning system for stroke risk patient |
| CN214283479U (en) * | 2020-12-17 | 2021-09-28 | 济宁三合服饰有限公司 | Multifunctional woven outdoor sports glove |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101836773A (en) * | 2009-03-20 | 2010-09-22 | 新奥特硅谷视频技术有限责任公司 | Gloves for monitoring hand temperature and humidity of testifier on court |
| CN109480780B (en) * | 2018-11-14 | 2021-08-24 | 重庆三峡医药高等专科学校 | Evaluation method and system of stroke early warning system |
| CN113545789B (en) * | 2021-08-24 | 2025-02-11 | 南京邮电大学 | A method for constructing an electroencephalogram (EEG) signal analysis model based on a CSP algorithm and a PSD algorithm, an electroencephalogram (EEG) signal analysis method and a system |
| CN117158994A (en) * | 2023-08-24 | 2023-12-05 | 齐鲁工业大学(山东省科学院) | Brain electrical signal classification method and system for cerebral apoplexy patient based on motor imagery |
-
2024
- 2024-01-18 CN CN202410073352.3A patent/CN117694906B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112489793A (en) * | 2020-12-16 | 2021-03-12 | 郑州航空工业管理学院 | Early warning system for stroke risk patient |
| CN214283479U (en) * | 2020-12-17 | 2021-09-28 | 济宁三合服饰有限公司 | Multifunctional woven outdoor sports glove |
Non-Patent Citations (1)
| Title |
|---|
| "基于BMI多通道FES脑卒中上肢康复技术的研究";孙文斌;《中国优秀硕士学位论文全文数据库 医药卫生科技辑》;20220215;第E070-237页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117694906A (en) | 2024-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhao et al. | EEG-Based Seizure detection using linear graph convolution network with focal loss | |
| US11834052B2 (en) | Estimator generation apparatus, monitoring apparatus, estimator generation method, and computer-readable storage medium storing estimator generation program | |
| CN117694906B (en) | Early warning method, device and system for cerebral apoplexy attack | |
| WO2009149126A2 (en) | Method, system, and computer-accessible medium for classification of at least one ictal state | |
| CN117877734A (en) | Mental disorder patient risk level judging system | |
| Tatara et al. | Interpreting ECG data by integrating statistical and artificial intelligence tools | |
| Golande et al. | Smart hospital for heart disease prediction using IoT | |
| CN116994755A (en) | Medical observation management system | |
| CN118797245B (en) | On-line monitoring system and method for hemodialysis process of patient | |
| CN114711790B (en) | Newborn electroconvulsive type determination method, newborn electroconvulsive type determination device, newborn electroconvulsive type determination equipment and storage medium | |
| Yu et al. | An extended computer aided diagnosis system for robust BCI applications | |
| CN119092088A (en) | A multi-center medical equipment big data cloud platform based on cloud-edge-end architecture | |
| WO2022120277A1 (en) | Systems and methods for detection of subject activity by processing video and other signals using artificial intelligence | |
| Xu et al. | Lightweight eeg classification model based on eeg-sensor with few channels | |
| CN118352064A (en) | Multichannel nerve monitoring system based on machine learning | |
| CN117224080B (en) | Human body data monitoring method and device for big data | |
| Zhang et al. | Learning robust representations of tonic-clonic seizures with cyclic transformer | |
| Lu et al. | SlimSeiz: Efficient Channel-Adaptive Seizure Prediction Using a Mamba-Enhanced Network | |
| CN117877747A (en) | A method, device and storage medium for classifying epilepsy pathology data | |
| US11324457B2 (en) | Explainable artificial intelligence framework for electrocardiography analysis | |
| Zavala-Yoé et al. | Real time acquisition and processing of massive electro-encephalographic signals for modeling by nonlinear statistics | |
| Pippa et al. | Automatic estimation of the optimal AR order for epilepsy analysis using EEG signals | |
| US20250209822A1 (en) | Method and apparatus for feedbacking human-machine collaboration state based on virtual-real integration, and electronic device | |
| Ni et al. | Dynamic multivariate multiscale entropy based analysis on brain death diagnosis | |
| CN117860201B (en) | Feature classification method and system based on deep learning analysis model |
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 |