CN-113694374-B - EEG-based prediction system, method, device, processor and storage medium for predicting tDCS treatment obsessive-compulsive disorder efficacy

CN113694374BCN 113694374 BCN113694374 BCN 113694374BCN-113694374-B

Abstract

The invention relates to a prediction system for predicting the treatment obsessive-compulsive disorder effect of a tDCS based on an EEG, which comprises a data acquisition processing module, a data preprocessing module, a data spectrum analysis processing module and a treatment effect prediction processing module, wherein the data acquisition processing module is used for acquiring resting state EEG data of a subject, the data preprocessing module is used for carrying out filtering processing on the acquired EEG data and removing signals containing artifacts, the data spectrum analysis processing module is used for converting EEG time domain signals into frequency domain signals by using fast Fourier transformation and carrying out division processing according to different frequency bands, and the treatment effect prediction processing module is used for judging the treatment effect prediction condition of the tDCS according to the comparison relation between a characteristic curve of the subject and an optimal critical point. The invention also relates to a corresponding method, device, processor and storage medium thereof. The system, the method, the device, the processor and the storage medium thereof of the invention are adopted to initiate the curative effect prediction method for treating compulsive disorder by using the tDCS based on EEG, have high algorithm accuracy and are beneficial to solving the key problem of difficult treatment method selection in the clinical diagnosis and treatment process.

Inventors

WANG ZHEN
Cheng Jiayue
TANG YINGYING

Assignees

上海市精神卫生中心（上海市心理咨询培训中心）
上海市精神卫生中心（上海市心理咨询培训中心）

Dates

Publication Date: 20260421
Application Date: 20210831
Priority Date: 20210831

Claims (5)

1. A prediction system for predicting the efficacy of tDCS treatment for obsessive-compulsive disorder based on EEG, said system comprising: the data acquisition processing module is used for acquiring resting state EEG data of the subject in a specific environment; The data preprocessing module is connected with the data acquisition processing module and is used for carrying out filtering processing on the acquired EEG data and removing EEG signals containing artifacts; The data spectrum analysis processing module is connected with the data preprocessing module and is used for converting EEG time domain signals into frequency domain signals by using fast Fourier transform, dividing the frequency domain signals according to different frequency bands and obtaining the power of a region of interest by calculating an average value according to the international lead system named electrode, and The treatment effect prediction processing module is connected with the data spectrum analysis processing module and is used for extracting the average power of the interested region in different frequency bands, inputting the obtained average power into the treatment effect prediction processing module for data processing, and comparing the output value of the treatment effect prediction processing module with an optimal critical point so as to judge the treatment effect prediction condition of tDCS treatment, wherein the treatment effect prediction condition specifically comprises the following steps: Taking the power of different frequency bands of forehead She Dianji as input, taking the curative effect after tDCS treatment as output, constructing a prediction model, and adopting the effect of a subject characteristic curve and an area under the curve evaluation model; The electrode named by the international lead system specifically comprises: FP1, FPZ, FP2, AF3, AF4, AF7, and AF8 electrodes.
2. The prediction system for predicting the efficacy of tDCS treatment obsessive-compulsive disorder based on EEG according to claim 1, wherein the frequency domain signals in the data spectrum analysis processing module are divided according to the following frequency bands: The method comprises delta (1-4 Hz), theta (5-7 Hz), alpha (8-13 Hz), beta (14-30 Hz) and gamma (30-60 Hz) 5 frequency bands.
3. A predictive device for achieving an EEG-based prediction of the efficacy of tDCS treatment of obsessive-compulsive disorder, said device comprising: A processor configured to execute computer-executable instructions; a memory storing one or more computer-executable instructions which, when executed by the processor, perform the following steps of a predictive method for predicting the efficacy of a tDCS therapy for obsessive-compulsive disorder based on EEG: (1) Collecting resting EEG data of a subject under a specific sampling condition in a specific environment; (1.1) setting the data sampling rate as 1000Hz, setting the on-line reference electrode as CPz, setting the resistance reduction standard as below 10KΩ and setting the acquisition time as 5 minutes; (1.2) collecting electrode data of 64 conductive electroencephalogram caps corresponding to 10-20 international electroencephalogram systems in a state that a subject is calm; (2) Preprocessing the collected resting state EEG data to remove artifacts and obtaining EEG individual signals without artifacts; (2.1) removing 64 the electrooculogram electrode signal contained in the electrooculogram cap; (2.2) removing system power frequency interference by using high-pass filtering of 1Hz and low-pass filtering of 100Hz and notch of 48-52 Hz; (2.3) segmenting the resting EEG data for 2 seconds, manually checking the segments containing obvious artifacts therein and removing the segments, and performing interpolation calculation on the leads with high noise; (2.4) obtaining independent resting EEG data statistical components using independent component analysis and removing non-neural signals including blink, horizontal eye movement, electrocardiography, muscle artefacts therefrom to obtain an EEG-individual signal free of artefacts; (3) Converting EEG time domain signals into frequency domain signals distributed according to different frequency bands by using fast Fourier transformation, and obtaining power values of a region of interest; (3.1) converting the EEG time domain signal into a frequency domain signal divided by delta (1-4 Hz), theta (5-7 Hz), alpha (8-13 Hz), beta (14-30 Hz) and gamma (30-60 Hz) 5 frequency bands by using a fast Fourier transform; (3.2) taking forehead She Dianji as a region of interest, and obtaining average power of forehead She Dianji by calculating average values of FP1, FPZ, FP2, AF3, AF4, AF7, and AF8 electrodes; (4) Extracting the power value of the region of interest and inputting the power value into the curative effect prediction processing module to perform prediction processing of tDCS curative effect; (4.1) inputting the average power of the forehead She Dianji into the efficacy prediction processing module for data processing; (4.2) comparing the output value of the efficacy prediction processing module with an optimal critical point; And (4.3) judging the curative effect of the tDCS treatment according to the comparison result.
4. A prediction processor for predicting the efficacy of a treatment for obsessive-compulsive disorder based on an EEG, said processor being configured to execute computer-executable instructions which, when executed by said processor, perform the steps of a method for predicting the efficacy of a treatment for obsessive-compulsive disorder based on an EEG comprising: (1) Collecting resting EEG data of a subject under a specific sampling condition in a specific environment; (1.1) setting the data sampling rate as 1000Hz, setting the on-line reference electrode as CPz, setting the resistance reduction standard as below 10KΩ and setting the acquisition time as 5 minutes; (1.2) collecting electrode data of 64 conductive electroencephalogram caps corresponding to 10-20 international electroencephalogram systems in a state that a subject is calm; (2) Preprocessing the collected resting state EEG data to remove artifacts and obtaining EEG individual signals without artifacts; (2.1) removing 64 the electrooculogram electrode signal contained in the electrooculogram cap; (2.2) removing system power frequency interference by using high-pass filtering of 1Hz and low-pass filtering of 100Hz and notch of 48-52 Hz; (2.3) segmenting the resting EEG data for 2 seconds, manually checking the segments containing obvious artifacts therein and removing the segments, and performing interpolation calculation on the leads with high noise; (2.4) obtaining independent resting EEG data statistical components using independent component analysis and removing non-neural signals including blink, horizontal eye movement, electrocardiography, muscle artefacts therefrom to obtain an EEG-individual signal free of artefacts; (3) Converting EEG time domain signals into frequency domain signals distributed according to different frequency bands by using fast Fourier transformation, and obtaining power values of a region of interest; (3.1) converting the EEG time domain signal into a frequency domain signal divided by delta (1-4 Hz), theta (5-7 Hz), alpha (8-13 Hz), beta (14-30 Hz) and gamma (30-60 Hz) 5 frequency bands by using a fast Fourier transform; (3.2) taking forehead She Dianji as a region of interest, and obtaining average power of forehead She Dianji by calculating average values of FP1, FPZ, FP2, AF3, AF4, AF7, and AF8 electrodes; (4) Extracting the power value of the region of interest and inputting the power value into the curative effect prediction processing module to perform prediction processing of tDCS curative effect; (4.1) inputting the average power of the forehead She Dianji into the efficacy prediction processing module for data processing; (4.2) comparing the output value of the efficacy prediction processing module with an optimal critical point; And (4.3) judging the curative effect of the tDCS treatment according to the comparison result.
5. A computer readable storage medium having stored thereon a computer program executable by a processor to perform the following steps of a method for predicting efficacy of treatment of tDCS based on EEG prediction of compulsive disorder: (1) Collecting resting EEG data of a subject under a specific sampling condition in a specific environment; (1.1) setting the data sampling rate as 1000Hz, setting the on-line reference electrode as CPz, setting the resistance reduction standard as below 10KΩ and setting the acquisition time as 5 minutes; (1.2) collecting electrode data of 64 conductive electroencephalogram caps corresponding to 10-20 international electroencephalogram systems in a state that a subject is calm; (2) Preprocessing the collected resting state EEG data to remove artifacts and obtaining EEG individual signals without artifacts; (2.1) removing 64 the electrooculogram electrode signal contained in the electrooculogram cap; (2.2) removing system power frequency interference by using high-pass filtering of 1Hz and low-pass filtering of 100Hz and notch of 48-52 Hz; (2.3) segmenting the resting EEG data for 2 seconds, manually checking the segments containing obvious artifacts therein and removing the segments, and performing interpolation calculation on the leads with high noise; (2.4) obtaining independent resting EEG data statistical components using independent component analysis and removing non-neural signals including blink, horizontal eye movement, electrocardiography, muscle artefacts therefrom to obtain an EEG-individual signal free of artefacts; (3) Converting EEG time domain signals into frequency domain signals distributed according to different frequency bands by using fast Fourier transformation, and obtaining power values of a region of interest; (3.1) converting the EEG time domain signal into a frequency domain signal divided by delta (1-4 Hz), theta (5-7 Hz), alpha (8-13 Hz), beta (14-30 Hz) and gamma (30-60 Hz) 5 frequency bands by using a fast Fourier transform; (3.2) taking forehead She Dianji as a region of interest, and obtaining average power of forehead She Dianji by calculating average values of FP1, FPZ, FP2, AF3, AF4, AF7, and AF8 electrodes; (4) Extracting the power value of the region of interest and inputting the power value into the curative effect prediction processing module to perform prediction processing of tDCS curative effect; (4.1) inputting the average power of the forehead She Dianji into the efficacy prediction processing module for data processing; (4.2) comparing the output value of the efficacy prediction processing module with an optimal critical point; And (4.3) judging the curative effect of the tDCS treatment according to the comparison result.

Description

EEG-based prediction system, method, device, processor and storage medium for predicting tDCS treatment obsessive-compulsive disorder efficacy Technical Field The invention relates to the technical field of medical care, in particular to the technical field of nerve signal processing, and specifically relates to a prediction system, a method, a device, a processor and a computer readable storage medium thereof for predicting the treatment forcing effect of tDCS based on EEG. Background Obsessive-compulsive disorder is a disabling disease characterized by repeated and persistent compulsive thinking or compulsive behavior, and has a lifetime prevalence of 2.4% in our country. Obsessive compulsive disorder is early onset, usually occurs in early adulthood, and can cause extensive social function impairment. Currently, first-line treatments for obsessive-compulsive disorders mainly include drug therapies, which are mainly selective 5-hydroxytryptamine reuptake inhibitors, and psychological therapies, which are mainly cognitive behavioral therapies. However, nearly 40% of patients do not respond to these treatments, and treatment of obsessive-compulsive disorder rarely brings about complete relief of symptoms, and residual symptoms and periodic deterioration are also difficulties in treatment. Therefore, finding and optimizing a diagnosis and treatment method for obsessive-compulsive disorder is of great value. Transcranial direct current stimulation (tDCS) is one of the neuromodulation techniques of great interest in the field of mental disease in recent years, consisting of a direct current micro-electro-stimulator, a cathodic electrode and an anodic electrode. After the electrodes are placed on the surface of the brain, the stimulator outputs weak direct current of 1-2mA, and the current flows from the anode to the cathode to form a loop. Some of the current will be shunted by skin, skull, cerebrospinal fluid and some of the current can reach the brain parenchyma, stimulation of up to 2mA for 30 minutes is safe, improving the symptoms and cognitive function of the patient by altering the way the cortex is excitable. The research of the subject group shows that the effective rate of the tDCS for treating the compulsive disorder is about 30%, the treatment period of the tDCS is 2 weeks, and in consideration of the time consumption of the tDCS treatment, if a reliable index for predicting the curative effect of the tDCS before the treatment can be found, the treatment efficiency can be improved, and the additional pain and burden caused by ineffective treatment to a patient can be avoided. Currently, some researches explore the prediction index of tDCS for treating other mental diseases, for example, the excitability of forehead leaves at the left side and the back side of a base line can predict the curative effect of tDCS for treating depression, and the ratio of glutamic acid to creatine at the intersection of left side and temporal top can predict the curative effect of tDCS for treating the auditory hallucination of schizophrenia. However, no prediction index of the curative effect of tDCS treatment on obsessive-compulsive patients is currently studied. Electroencephalogram (EEG) is used as a common nerve function detection technology, can record the electrical activity of neurons, can record the instantaneous brain activity of the whole brain surface layer in a non-invasive way, has the advantage of high time resolution, and can provide a simple and convenient critical value detection method compared with magnetic resonance and other technologies. Patent application publication number CN109924973a discloses a "pre-epileptic EEG signal recognition method and cloud system based on GBDT model", and predicts whether a patient is in a pre-epileptic state by EEG signals and machine learning methods. However, there is no method for EEG to predict the efficacy of tDCS in treating obsessive-compulsive disorder. Disclosure of Invention It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art by providing a prediction system, method, device, processor and computer readable storage medium thereof, which enable to easily and quickly identify the efficacy of an applied tDCS treatment based on EEG prediction of tDCS treatment obsessive-compulsive disorder. To achieve the above object, the prediction system, method, apparatus, processor and computer readable storage medium thereof for predicting the efficacy of tDCS treatment obsessive-compulsive disorder based on EEG of the present invention are as follows: The prediction system for predicting the treatment obsessive-compulsive disorder effect of the tDCS based on the EEG is mainly characterized by comprising the following components: the data acquisition processing module is used for acquiring resting state EEG data of the subject in a specific environment; The data preprocessing module is connected with the data acquisition pr