WO2024103402A1 - 基于视觉诱发脑电和深度学习的戒毒后复吸风险评估方法 - Google Patents

基于视觉诱发脑电和深度学习的戒毒后复吸风险评估方法 Download PDF

Info

Publication number
WO2024103402A1
WO2024103402A1 PCT/CN2022/132939 CN2022132939W WO2024103402A1 WO 2024103402 A1 WO2024103402 A1 WO 2024103402A1 CN 2022132939 W CN2022132939 W CN 2022132939W WO 2024103402 A1 WO2024103402 A1 WO 2024103402A1
Authority
WO
WIPO (PCT)
Prior art keywords
drug
relapse
electroencephalogram
time
person
Prior art date
Application number
PCT/CN2022/132939
Other languages
English (en)
French (fr)
Inventor
孟志强
Original Assignee
中国科学院深圳先进技术研究院
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 中国科学院深圳先进技术研究院 filed Critical 中国科学院深圳先进技术研究院
Priority to PCT/CN2022/132939 priority Critical patent/WO2024103402A1/zh
Publication of WO2024103402A1 publication Critical patent/WO2024103402A1/zh

Links

Images

Classifications

    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
    • G16H10/20—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for electronic clinical trials or questionnaires

Definitions

  • the present disclosure relates to the technical field of drug control monitoring, and in particular to a method for assessing the risk of relapse after drug detoxification based on visually evoked electroencephalogram (EVO) deep learning.
  • EVO electroencephalogram
  • Drug addiction is a chronic relapsing brain disease characterized by compulsive drug seeking, drug use, and high relapse rates after withdrawal, of which the high relapse rate is the most difficult problem in drug rehabilitation.
  • drug detoxification treatment can be carried out for drug addicts through cold turkey method and low-toxic drug substitution therapy, there is still no effective solution for relapse after withdrawal.
  • drug addicts return to the community, and the speed of relapse varies from a few months to a few years, indicating that the effect of detoxification treatment is different for each person.
  • the current prediction system for relapse is based on survey scales or physiological indicators such as electrocardiograms.
  • Addiction itself is a brain disease.
  • Existing technologies for predicting relapse lack effective and objective brain-related indicators, and currently cannot predict the speed of relapse, that is, how long it will take to relapse after abstinence.
  • the above-mentioned prior art technologies for predicting relapse lack effective and objective indicators related to the brain, and currently cannot predict the speed of relapse.
  • the purpose of the present invention is to evaluate the effectiveness of withdrawal treatment and the risk of relapse after withdrawal, and to provide a risk assessment method based on objective physiological indicators.
  • the present invention proposes a method for assessing the risk of relapse after drug addiction treatment based on visual evoked electroencephalogram and deep learning, characterized in that the method comprises the following steps:
  • the person's electroencephalogram is performed to extract EEG features, and the drug rehabilitation effect is judged based on the EEG features and the relapse time is predicted;
  • the preset model is established based on a deep learning neural network model, and is trained using electroencephalogram samples of real drug addicts who have successfully quit drug addiction and those who relapse after drug addiction.
  • the EEG data of drug addicts under visual stimulation of drugs is used to predict the relapse time.
  • the above technical solution is a risk assessment method based on objective physiological indicators.
  • the predicted relapse time can provide an objective physiological indicator basis for community drug rehabilitation monitoring and early intervention.
  • the classified EEG data is real data based on the relapse time after drug rehabilitation, which is more objective than scale data and can more truly reflect the brain's response to drugs than physiological indicators such as electrocardiogram and body fluids.
  • the brain waves are collected by a 32-lead EEG acquisition device.
  • the electroencephalogram sample is composed of the electroencephalogram of the drug addict and the relapse time of the drug addict, and is obtained through the following steps:
  • the EEG data is filtered and pre-processed to reduce noise and generate an EEG graph.
  • the evaluation also includes judging the effect of drug rehabilitation and the time of relapse, specifically:
  • N1 and N2 are set values
  • the data of the fastest and slowest relapsers are used as critical values to judge the effect of drug rehabilitation and the time of relapse.
  • the effect of drug rehabilitation can be a qualitative evaluation based on the time of relapse.
  • the drug visual stimulation is achieved by presenting real drugs, or presenting drug-related pictures and videos through electronic devices.
  • the present invention proposes a device for assessing the risk of relapse after drug addiction treatment based on visual evoked electroencephalogram and deep learning, comprising a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute any of the above methods.
  • the present invention provides a computer-readable storage medium storing a computer program that can be loaded by a processor and execute any of the above methods.
  • the present invention proposes a drug relapse risk assessment system based on visual evoked electroencephalogram and deep learning, the system comprising the following modules:
  • An acquisition module is configured to provide drug visual stimulation to a person who has just completed drug rehabilitation, and acquire the EEG data of the person at that time;
  • the prediction module uses a preset model to extract EEG features from the person’s EEG, and based on the EEG features, determines the drug rehabilitation effect and predicts the time of relapse;
  • the preset model is established based on a deep learning neural network model, and is trained using electroencephalogram samples of real drug addicts who have successfully quit drug addiction and those who relapse after drug addiction.
  • the electroencephalogram sample is composed of the electroencephalogram of the drug addict and the relapse time of the drug addict, and is obtained by the following units:
  • the electroencephalogram generation unit is configured to filter and perform noise reduction preprocessing on the electroencephalogram data to generate an electroencephalogram.
  • the evaluation also includes judging the effect of drug rehabilitation and the time of relapse, specifically:
  • N1 and N2 are set values
  • the data of the fastest and slowest relapsers are used as critical values to determine the effectiveness of drug rehabilitation and the time when relapse occurs.
  • FIG1 is a schematic diagram of a method flow chart in one embodiment
  • FIG2 is a schematic diagram of a system structure in one embodiment
  • FIG3 is a schematic diagram of the structure of an electronic device provided by an embodiment.
  • FIG. 1 which shows a flow chart of a method for assessing the risk of relapse after drug addiction treatment based on visually evoked electroencephalogram and deep learning.
  • the risk assessment method for drug relapse after detoxification based on visual evoked EEG and deep learning is a risk assessment method based on objective physiological indicators, which aims to evaluate the effect of detoxification treatment and the risk of relapse after detoxification.
  • the steps include:
  • the preset model is established based on a deep learning neural network model, and is trained using EEG samples of real drug rehabilitation personnel who have successfully quit drug rehabilitation and those who have relapsed after drug rehabilitation.
  • the sample data used to train the deep learning neural network consists of the EEG of each sample drug addict and the relapse time of the drug addict. Since the EEG sample data based on real drug addicts who have successfully quit drugs and drug addicts who relapse after drug treatment are more objective than scale data, they can more truly reflect the brain's response to drugs than physiological indicators such as electrocardiogram and body fluids. Therefore, it is feasible to use the EEG data of drug addicts induced by drug vision to predict relapse, especially when the sample data is real data based on the relapse time after drug treatment, which can improve the reliability and accuracy of the prediction.
  • EEG sample data processing The EEG data of drug addicts who have just completed drug rehabilitation and are presented with drug-related visual stimuli are obtained, and these EEG data are filtered, pre-processed for noise reduction, and an EEG graph is generated.
  • the deep learning neural network can mine the spectral features of EEG waves, including the range of spectral value changes and curve changes. By limiting the duration of EEG acquisition, such as 10 minutes, 15 minutes, 20 minutes, 25 minutes or 30 minutes or even longer, EEG spectral features that effectively predict relapse time can be obtained. Furthermore, the trained deep learning neural network is inversely mapped back to the input EEG, and it can be known which part of the spectrum has an impact on the relapse time.
  • the data of the 30% of people who relapse the fastest (short relapse time) and the 30% of people who relapse the slowest (long relapse time) are taken as critical values to classify people who have completed drug rehabilitation and provide early warning prompts for community drug rehabilitation monitoring and early intervention.
  • the network structure, structural parameters, weight parameters, etc. of the model are recorded so as to facilitate migration and deployment of the model.
  • the number of leads of the EEG acquisition device may range from 2 to 256.
  • the ratio of the 30% of people who relapse the fastest to the 30% of people who relapse the slowest can also be 10%, 15%, 20%, 25% and so on.
  • a qualitative evaluation standard for drug rehabilitation effects is further established based on the time the drug addicts maintain abstinence, the time they relapse, and the time they lose contact.
  • risk levels are divided according to the time of relapse, and a qualitative risk level prompt is given after prediction, or a qualitative evaluation of the withdrawal effect is further given.
  • the probability of relapse is predicted and calculated based on the EEG characteristics of drug visual stimulation of drug addicts in combination with the duration of abstinence.
  • a post-detoxification relapse risk assessment system based on visual evoked EEG and deep learning may be implemented, which may include:
  • An acquisition module is configured to provide drug visual stimulation to a person who has just completed drug rehabilitation, and acquire the electroencephalogram of the person at that time;
  • the prediction module uses a preset model to extract EEG features from the person’s EEG, and based on the EEG features, determines the drug rehabilitation effect and predicts the time of relapse;
  • the preset model is established based on a deep learning neural network model, and is trained using electroencephalogram samples of real drug addicts who have successfully quit drug addiction and those who relapse after drug addiction.
  • the electroencephalogram sample is composed of the electroencephalogram of the drug addict and the relapse time of the drug addict, and is obtained by the following units:
  • the electroencephalogram generation unit is configured to filter and perform noise reduction preprocessing on the electroencephalogram data to generate an electroencephalogram.
  • This embodiment provides a post-drug relapse risk assessment system based on visual evoked EEG and deep learning, which can execute the embodiments of the above method. Its implementation principles and technical effects are similar and will not be repeated here.
  • Fig. 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. As shown in Fig. 3, a schematic diagram of the structure of an electronic device 300 suitable for implementing the embodiment of the present application is shown.
  • the electronic device 300 includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage part 308 into a random access memory (RAM) 303.
  • ROM read-only memory
  • RAM random access memory
  • various programs and data required for the operation of the device 300 are also stored.
  • the CPU 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304.
  • An input/output (I/O) interface 305 is also connected to the bus 304.
  • the following components are connected to the I/O interface 305: an input section 306 including a keyboard, a mouse, etc.; an output section 307 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet.
  • a drive 310 is also connected to the I/O interface 306 as needed.
  • a removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed, so that a computer program read therefrom is installed into the storage section 308 as needed.
  • an embodiment of the present disclosure includes a computer program product, which includes a computer program tangibly contained on a machine-readable medium, and the computer program includes a program code for executing the above-mentioned cascade hydropower scheduling model construction method.
  • the computer program can be downloaded and installed from a network through the communication part 309, and/or installed from a removable medium 311.
  • each box in the flow chart or block diagram can represent a module, a program segment or a part of the code, and the aforementioned module, program segment or a part of the code contains one or more executable instructions for realizing the specified logical function.
  • the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved.
  • each box in the block diagram and/or flow chart, and the combination of the boxes in the block diagram and/or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
  • the units or modules involved in the embodiments described in the present application may be implemented by software or hardware.
  • the units or modules described may also be arranged in a processor.
  • the names of these units or modules do not constitute limitations on the units or modules themselves in certain circumstances.
  • a typical implementation device is a computer.
  • the computer may be, for example, a personal computer, a laptop, a mobile phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device or a combination of any of these devices.
  • the present application further provides a storage medium, which may be the storage medium included in the aforementioned device in the above embodiment; or may be a storage medium that exists independently and is not assembled into the device.
  • the storage medium stores one or more programs, and the aforementioned programs are used by one or more processors to execute the cascade hydropower scheduling model construction method described in the present application.
  • Storage media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information.
  • Information can be computer-readable instructions, data structures, program modules or other data.
  • Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.
  • computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.
  • the present invention adopts effective and objective indicators related to the brain, based on the real data of relapse time after drug detoxification, combined with the EEG data induced by visual stimulation of drug-related pictures and videos during the withdrawal period, and uses a deep learning neural network to establish a model.
  • this model can correspond different EEG features to different relapse times, and use this model as a prediction of the relapse time of other drug addicts after withdrawal.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Development Economics (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Epidemiology (AREA)
  • Medical Informatics (AREA)
  • Primary Health Care (AREA)
  • Public Health (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Biomedical Technology (AREA)
  • Game Theory and Decision Science (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

本发明涉及一种基于视觉诱发脑电和深度学习的戒毒后复吸风险评估方法。所述方法包括下述步骤：对刚完成戒毒的人员进行毒品视觉刺激，获取该人员此时的脑电波图；利用预设的模型，对该人员进行脑电波图提取脑电特征，基于脑电特征判断戒毒效果并预测其复吸时间；其中，所述预设的模型，基于深度学习神经网络模型建立，利用真实戒毒成功的戒毒人员和戒毒后复吸人员的脑电波图样本训练获得。本发明基于戒毒后复吸时间的真实数据，结合戒断期间毒品视觉诱发的脑电数据，利用神经网络通过训练将脑电图特征与复吸时间相对应，再用此模型预测戒毒人员戒断后复吸时间，为社区戒毒监控及提早干预提供客观生理指标依据。

Description

基于视觉诱发脑电和深度学习的戒毒后复吸风险评估方法 技术领域
本公开涉及禁毒监测技术领域，尤其涉及一种基于视觉诱发脑电深度学习的戒毒后复吸风险评估方法。
背景技术
药物成瘾是一种慢性复发性脑疾病，其特征在于强迫性觅药、用药及戒断后的高复吸率，其中高复吸率是戒毒工作面中最棘手的问题。虽然目前可以通过冷火鸡法、低毒药物替代疗法等对吸毒人员进行药物脱毒治疗，但是对于戒断后的复吸仍然没有很有效的解决办法。吸毒人员在戒毒机构治疗满两年后回到社区，再次复吸的速度从几个月到几年并不相同，说明每个人戒断治疗的效果也不尽相同。
目前对于复吸的预测系统是基于调查量表或心电等生理指标，而成瘾本身是一种脑疾病，现有的预测复吸的技术缺乏大脑相关的有效的客观的指标，且目前还不能预测复吸的速度，即戒断后多久复吸。
发明内容
针对上述现有技术的预测复吸的技术缺乏大脑相关的有效的客观的指标，且目前还不能预测复吸的速度，本发明的目的在于评估戒断治疗的效果和戒断后的复吸风险，提供了一种基于客观生理指标的风险评估方法。
为了实现上述目的，本发明的技术方案如下。
第一方面，本发明提出一种基于视觉诱发脑电和深度学习的戒毒后复吸风险评估方法，其特征在于，所述方法包括下述步骤：
对刚完成戒毒的人员进行毒品视觉刺激，获取该人员此时的脑电波图；
利用预设的模型，对该人员进行脑电波图提取脑电特征，基于脑电特征判断戒毒效果并预测其复吸时间；
其中，所述预设的模型，基于深度学习神经网络模型建立，利用真实戒毒成功的戒毒人员和戒毒后复吸人员的脑电波图样本训练获得。
在上述技术方案中，通过采用训练好的深度学习神经网络模型，利用戒毒人员在毒品视觉刺激下的脑电数据，对复吸时间进行预测。上述技术方案是一种基于客观生理指标的风险评估方法，预测的复吸时间，可作为社区戒毒监控及提早干预提供客观生理指标依据。其中，所述分类的脑电数据为基于戒毒后复吸时间的真实数据，相比量表数据更加客观，相比心电、体液等生理指标更能真实的反映大脑对毒品的反映。在一种实施方式中，所述脑电波通过32导联脑电采集设备采集。
在上述技术方案中，所述脑电波图样本由戒毒人员的脑电波图和该戒毒人员的复吸时间构成，通过下述步骤获取：
获取已戒毒人员的复吸时间，同时获取该戒毒人员的刚戒完毒时进行毒品视觉刺激时的脑电波数据；
将脑电数据进行滤波、降噪预处理，生成脑电波图。
在上述技术方案中，所述评估还包括判断戒毒效果、复吸发生的时间，具体为：
将复吸时间按从短到长排序，将排序在前N1的人为最快复吸人员，将排序在后N2的人作为最慢复吸人员，N1、N2为设定值；
最快复吸人员以及最慢复吸人员的数据作为临界值，用于判断戒毒效果、复吸发生的时间。其中，戒毒效果可以是一种基于复吸时间的定性评价。
在上述技术方案中，所述毒品视觉刺激通过呈现毒品实物、或者通过电子设备呈现毒品相关的图片和视频实现。
第二方面，本发明提出一种基于视觉诱发脑电和深度学习的戒毒后复吸风险评估装置，包括存储器和处理器，所述存储器上存储有能够被处理器加载并执行上述任一种方法的计算机程序。
第三方面，本发明提出一种计算机可读存储介质，存储有能够被处理器加载并执行上述任一种方法的计算机程序。
第四方面，本发明提出一种基于视觉诱发脑电和深度学习的戒毒后复吸风险评估系统，所述系统包括下述模块：
获取模块，被配置用于对刚完成戒毒的人员进行毒品视觉刺激，获取该人员此时的脑电数据；
预测模块，利用预设的模型，对该人员进行脑电波图提取脑电特征，基于脑电特征判断戒毒效果并预测其复吸时间；
其中，所述预设的模型，基于深度学习神经网络模型建立，利用真实戒毒成功的戒毒人员和戒毒后复吸人员的脑电波图样本训练获得。
在上述技术方案中，所述脑电图样本由戒毒人员的脑电图和该戒 毒人员的复吸时间构成，通过下述单元配合获取：
信息单元，获取已戒毒人员的复吸时间，同时获取该戒毒人员的刚戒完毒时进行毒品视觉刺激时的脑电波数据；
脑电波图生成单元，被配置用于将脑电数据进行滤波、降噪预处理，生成脑电波图。
在上述技术方案中，所述评估还包括判断戒毒效果、复吸发生的时间，具体为：
将复吸时间按从短到长排序，将排序在前N1的人为最快复吸人员，将排序在后N2的人作为最慢复吸人员，N1、N2为设定值；
最快复吸人员以及最慢复吸人员的数据作为临界值，用于判断戒毒效果、复吸发生的时间。
附图说明
图1、一个实施例中的方法流程示意图；
图2、一个实施例中的系统结构示意图；
图3、一个实施例提供的一种电子设备的结构示意图。
具体实施方式
下面结合附图和实施例对本发明进行详细说明。
参照图1，其示出了一种基于视觉诱发脑电和深度学习的戒毒后复吸风险评估方法的流程示意图。
如图1所示，基基于视觉诱发脑电和深度学习的戒毒后复吸风险评估方法，是一种基于客观生理指标的风险评估方法，目的在于评估戒断治疗的效果和戒断后的复吸风险。在该方法中，步骤包括：
S100、对刚完成戒毒的人员进行毒品视觉刺激，获取该人员此时的脑电波图。
S200、利用预设的模型，对该人员进行脑电波图提取脑电特征，基于脑电特征判断戒毒效果并预测其复吸时间，为社区戒毒监控及提早干预提供客观生理指标依据。其中，所述预设的模型，基于深度学习神经网络模型建立，利用真实戒毒成功的戒毒人员和戒毒后复吸人员的脑电波图样本训练获得。
在上述方法中，用于训练深度学习神经网络的样本数据，每个样本戒毒人员的脑电波图和该戒毒人员的复吸时间构成。由于基于真实的戒毒成功的戒毒人员和戒毒后复吸人员的脑电波图样本数据，相比量表数据更加客观，相比心电、体液等生理指标更能真实的反映大脑对毒品的反映。因此使用戒毒人员在毒品视觉诱发的脑电波数据进行复吸预测是可行的，特别当样本数据为基于戒毒后复吸时间的真实数据时，能够提高预测的可靠性和准确性。
在采集样本时，包括：
(1)毒品相关图片视频呈现及脑电数据采集：通过电脑和显示器给戒毒即将完成即将回归社区的人员呈现毒品相关的图片和视频，通过量表调查戒毒人员对毒品的渴求程度；同时，通过32导联脑电采集设备采集脑电信号。
(2)社区回访：对回归社区的戒毒人员进行定期社区回访，统计失戒毒完成人员信息；通过社区回访数据，详细登记每个人复吸时间。
(3)脑电波样本数据处理：获取回归社区的戒毒人员在刚完成戒毒时，进行毒品相关视觉刺激呈现时的脑电波数据，将这些脑电波数据进行滤波，降噪预处理，生成脑电波图。
(4)利用脑电波图样本作为输入，输入到基于深度学习神经网络建立的模型，将复吸时间作为标签，对基于深度学习神经网络建立的模型进行训练，使模型学习到脑电波数据特征和复吸时间之间的关系，形成训练好的模型，该模型可用于对新的完成戒毒的人员进行脑电特征提取和分析，从而预测其复吸时间。深度学习神经网络可以挖掘脑电波的频谱特征，包括频谱值变化范围和曲线变化。通过限定脑电波获取的时长，比如10分钟，15分钟，20分钟，25分钟或者30分钟甚至更长，可以获取到有效预测复吸时间的脑电波频谱特征。进一步地，将训练好的深度学习神经网络逆映射回输入的脑电波，可获知是哪部分频谱对复吸时间产生了影响。
(5)根据复吸时间，进一步对戒毒人员的戒毒效果进行分类。
在一种实施方式中，取最快复吸的30％人员(复吸时间短)以及最慢复吸的30％人员(复吸时间长)的数据作为临界值，对戒毒完成人员进行分类，为社区戒毒监控及提早干预进行预警提示。
在一种实施方式中，通过记录模型的网络结构、结构参数、权重参数等，以便对模型进行迁移部署。
在上述实施方式中，脑电采集设备导联数可以从2-256不等。
上述实施方式中，最快复吸的30％人员和最慢复吸的30％人员中的比例还可以是10％、15％、20％、25％等。
在一种实施方式中，对社区回访中戒毒人员，基于戒毒人员保持戒断的时间，复吸的时间，失去联系的时间，进一步建立戒毒效果定性评价标准。
在一些实施方式中，根据复吸时间进行风险等级划分，预测后给出定性的风险等级提示。或者进一给出戒断效果的定性评估。
在一种实施方式中，基于戒毒人员的毒品视觉刺激的脑电特征，结合戒断时长，进行复吸概率预测计算。
如图2所示，根据上述方法，实现为一种基于视觉诱发脑电和深度学习的戒毒后复吸风险评估系统，可以包括：
获取模块，被配置用于对刚完成戒毒的人员进行毒品视觉刺激，获取该人员此时的脑电波图；
预测模块，利用预设的模型，对该人员进行脑电波图提取脑电特征，基于脑电特征判断戒毒效果并预测其复吸时间；
其中，所述预设的模型，基于深度学习神经网络模型建立，利用真实戒毒成功的戒毒人员和戒毒后复吸人员的脑电波图样本训练获得。
所述脑电图样本由戒毒人员的脑电图和该戒毒人员的复吸时间构成，通过下述单元配合获取：
信息单元，获取已戒毒人员的复吸时间，同时获取该戒毒人员的刚戒完毒时进行毒品视觉刺激时的脑电波数据；
脑电波图生成单元，被配置用于将脑电数据进行滤波、降噪预处理，生成脑电波图。
本实施例提供的一种基于视觉诱发脑电和深度学习的戒毒后复吸风险评估系统，可以执行上述方法的实施例，其实现原理和技术效果类似，在此不再赘述。
图3为本发明实施例提供的一种电子设备的结构示意图。如图3所示，示出了适于用来实现本申请实施例的电子设备300的结构示意图。
如图3所示，电子设备300包括中央处理单元(CPU)301，其可以根据存储在只读存储器(ROM)302中的程序或者从存储部分308加载到随机访问存储器(RAM)303中的程序而执行各种适当的动作和处理。在RAM 303中，还存储有设备300操作所需的各种程序和数据。CPU 301、ROM 302以及RAM 303通过总线304彼此相连。输入/输出(I/O)接口305也连接至总线304。
以下部件连接至I/O接口305：包括键盘、鼠标等的输入部分306；包括诸如阴极射线管(CRT)、液晶显示器(LCD)等以及扬声器等的输出部分307；包括硬盘等的存储部分308；以及包括诸如LAN卡、调制解调器等的网络接口卡的通信部分309。通信部分309经由诸如因特网的网络执行通信处理。驱动器310也根据需要连接至I/O接口306。可拆卸介质311，诸如磁盘、光盘、磁光盘、半导体存储器等等，根据需要安装在驱动器310上，以便于从其上读出的计算机程序根据需要被安装入存储部分308。
特别地，根据本公开的实施例，上文参考图1描述的过程可以被实现为计算机软件程序。例如，本公开的实施例包括一种计算机程序 产品，其包括有形地包含在机器可读介质上的计算机程序，计算机程序包含用于执行上述梯级水电调度模型构建方法的程序代码。在这样的实施例中，该计算机程序可以通过通信部分309从网络上被下载和安装，和/或从可拆卸介质311被安装。
附图中的流程图和框图，图示了按照本发明各种实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段、或代码的一部分，前述模块、程序段、或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个接连地表示的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或操作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
描述于本申请实施例中所涉及到的单元或模块可以通过软件的方式实现，也可以通过硬件的方式来实现。所描述的单元或模块也可以设置在处理器中。这些单元或模块的名称在某种情况下并不构成对该单元或模块本身的限定。
上述实施例阐明的系统、装置、模块或单元，具体可以由计算机芯片或实体实现，或者由具有某种功能的产品来实现。一种典型的实现设备为计算机。具体的，计算机例如可以为个人计算机、笔记本电 脑、行动电话、智能电话、个人数字助理、媒体播放器、导航设备、电子邮件设备、游戏控制台、平板计算机、可穿戴设备或者这些设备中的任何设备的组合。
作为另一方面，本申请还提供了一种存储介质，该存储介质可以是上述实施例中前述装置中所包含的存储介质；也可以是单独存在，未装配入设备中的存储介质。存储介质存储有一个或者一个以上程序，前述程序被一个或者一个以上的处理器用来执行描述于本申请的梯级水电调度模型构建方法。
存储介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括，但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带，磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质，可用于存储可以被计算设备访问的信息。按照本文中的界定，计算机可读介质不包括暂存电脑可读媒体(transitory media)，如调制的数据信号和载波。
综上，本发明通过采用与大脑先关的有效的客观的指标，基于戒毒后复吸时间的真实数据，结合戒断期间毒品相关图片视频视觉刺激诱发的脑电数据，利用深度学习神经网络建立模型，采用真实的训练 权重模型，此模型可以将不同的脑电特征对应到不同复吸时间，并以此模型作为预测其他戒毒人员戒断后复吸时间。
需要说明的是，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、商品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、商品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括要素的过程、方法、商品或者设备中还存在另外的相同要素。
本说明书中的各个实施例均采用递进的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其他实施例的不同之处。尤其，对于系统实施例而言，由于其基本相似于方法实施例，所以描述的比较简单，相关之处参见方法实施例的部分说明即可。

Claims (10)

  1. 一种基于视觉诱发脑电和深度学习的戒毒后复吸风险评估方法，其特征在于，所述方法包括下述步骤：
    对刚完成戒毒的人员进行毒品视觉刺激，获取该人员此时的脑电波图；
    利用预设的模型，对该人员进行脑电波图提取脑电特征，基于脑电特征判断戒毒效果并预测其复吸时间；
    其中，所述预设的模型，基于深度学习神经网络模型建立，利用真实戒毒成功的戒毒人员和戒毒后复吸人员的脑电波图样本训练获得。
  2. 根据权利要求1所述的方法，其特征在于，所述脑电波通过32导联脑电采集设备采集。
  3. 根据权利要求1所述的方法，其特征在于，所述脑电波图样本由戒毒人员的脑电波图和该戒毒人员的复吸时间构成，通过下述步骤获取：
    获取已戒毒人员的复吸时间，同时获取该戒毒人员的刚戒完毒时进行毒品视觉刺激时的脑电波数据；
    将脑电数据进行滤波、降噪预处理，生成脑电波图。
  4. 根据权利要求1所述的方法，其特征在于，所述评估还包括判断戒毒效果、复吸发生的时间，具体为：
    将复吸时间按从短到长排序，将排序在前N1的人为最快复吸人员，将排序在后N2的人作为最慢复吸人员，N1、N2为设定值；
    最快复吸人员以及最慢复吸人员的数据作为临界值，用于判断戒 毒效果、复吸发生的时间。
  5. 根据权利要求1所述的方法，其特征在于，所述毒品视觉刺激通过呈现毒品实物、或者通过电子设备呈现毒品相关的图片和视频实现。
  6. 一种基于视觉诱发脑电数据和深度学习的戒毒人员复吸评估装置，其特征在于：包括存储器和处理器，所述存储器上存储有能够被处理器加载并执行如权利要求1至5中任一种方法的计算机程序。
  7. 一种计算机可读存储介质，其特征在于：存储有能够被处理器加载并执行如权利要求1至5中任一种方法的计算机程序。
  8. 一种基于视觉诱发脑电数据和深度学习的戒毒人员复吸评估系统，其特征在于，所述系统包括下述模块：
    获取模块，被配置用于对刚完成戒毒的人员进行毒品视觉刺激，获取该人员此时的脑电波图；
    预测模块，利用预设的模型，对该人员进行脑电波图提取脑电特征，基于脑电特征判断戒毒效果并预测其复吸时间；
    其中，所述预设的模型，基于深度学习神经网络模型建立，利用真实戒毒成功的戒毒人员和戒毒后复吸人员的脑电波图样本训练获得。
  9. 根据权利要求8所述的系统，其特征在于，所述脑电图样本由戒毒人员的脑电图和该戒毒人员的复吸时间构成，通过下述单元配合获取：
    信息单元，获取已戒毒人员的复吸时间，同时获取该戒毒人员的 刚戒完毒时进行毒品视觉刺激时的脑电波数据；
    脑电波图生成单元，被配置用于将脑电数据进行滤波、降噪预处理，生成脑电波图。
  10. 根据权利要求8所述的系统，其特征在于，所述评估还包括判断戒毒效果、复吸发生的时间，具体为：
    将复吸时间按从短到长排序，将排序在前N1的人为最快复吸人员，将排序在后N2的人作为最慢复吸人员，N1、N2为设定值；
    最快复吸人员以及最慢复吸人员的数据作为临界值，用于判断戒毒效果、复吸发生的时间。
PCT/CN2022/132939 2022-11-18 2022-11-18 基于视觉诱发脑电和深度学习的戒毒后复吸风险评估方法 WO2024103402A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/132939 WO2024103402A1 (zh) 2022-11-18 2022-11-18 基于视觉诱发脑电和深度学习的戒毒后复吸风险评估方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/132939 WO2024103402A1 (zh) 2022-11-18 2022-11-18 基于视觉诱发脑电和深度学习的戒毒后复吸风险评估方法

Publications (1)

Publication Number Publication Date
WO2024103402A1 true WO2024103402A1 (zh) 2024-05-23

Family

ID=91083693

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/132939 WO2024103402A1 (zh) 2022-11-18 2022-11-18 基于视觉诱发脑电和深度学习的戒毒后复吸风险评估方法

Country Status (1)

Country Link
WO (1) WO2024103402A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110547793A (zh) * 2019-07-02 2019-12-10 上海大学 一种结合脑电与近红外的毒瘾评估方法
CN111724896A (zh) * 2020-06-11 2020-09-29 阿呆科技（北京）有限公司 一种基于多刺激图片或视频erp的毒瘾评估系统
CN111798942A (zh) * 2020-06-17 2020-10-20 西南大学 一种通过心理量表建立吸毒复吸时间预估模型的方法
CN112598184A (zh) * 2020-12-27 2021-04-02 上海达梦数据库有限公司 一种戒毒人员复吸风险预测的方法和装置
JP2022021944A (ja) * 2020-07-22 2022-02-03 ブレスコーポレーション株式会社 依存症回復支援システム

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110547793A (zh) * 2019-07-02 2019-12-10 上海大学 一种结合脑电与近红外的毒瘾评估方法
CN111724896A (zh) * 2020-06-11 2020-09-29 阿呆科技（北京）有限公司 一种基于多刺激图片或视频erp的毒瘾评估系统
CN111798942A (zh) * 2020-06-17 2020-10-20 西南大学 一种通过心理量表建立吸毒复吸时间预估模型的方法
JP2022021944A (ja) * 2020-07-22 2022-02-03 ブレスコーポレーション株式会社 依存症回復支援システム
CN112598184A (zh) * 2020-12-27 2021-04-02 上海达梦数据库有限公司 一种戒毒人员复吸风险预测的方法和装置

Similar Documents

Publication Publication Date Title
Moore et al. Applications for self‐administered mobile cognitive assessments in clinical research: A systematic review
Garcia-Ceja et al. Depresjon: a motor activity database of depression episodes in unipolar and bipolar patients
Jansen Quantitative analysis of electroencephalograms: is there chaos in the future?
Woodhead et al. Testing the unitary theory of language lateralization using functional transcranial Doppler sonography in adults
US20240212859A1 (en) Predicting addiction relapse and decision support tool
CN114190897B (zh) 睡眠分期模型的训练方法、睡眠分期方法及装置
Olokodana et al. Real-time automatic seizure detection using ordinary kriging method in an edge-IoMT computing paradigm
Harris et al. Digital Neuropsychology beyond Computerized Cognitive Assessment: Applications of Novel Digital Technologies
Kirschner et al. A P300‐based cognitive assessment battery
CN110473631B (zh) 基于真实世界研究的智能睡眠监测方法和系统
Cheung et al. Wrist accelerometry for monitoring dementia agitation behaviour in clinical settings: A scoping review
Fuchs et al. Tremor assessment using smartphone sensor data and fuzzy reasoning
Almahshi et al. Hypothyroidism prediction and detection using machine learning
Kim et al. Artificial intelligence-based patient monitoring system for medical support
WO2024103402A1 (zh) 基于视觉诱发脑电和深度学习的戒毒后复吸风险评估方法
Zontone et al. Low-complexity classification algorithm to identify drivers’ stress using electrodermal activity (EDA) measurements
Goldfine et al. Quantifying poststroke apathy with actimeters
Klapproth et al. Why do temporal generalization gradients change when people make decisions as quickly as possible?
US20230170097A1 (en) Methods and computing device related to sleep evaluation
Yang et al. Loneliness forecasting using multi-modal wearable and mobile sensing in everyday settings
Hossain et al. Measurement of cognitive load for writing tasks using galvanic skin response
Yuan et al. Detection and Quantification of Resting Tremor in Parkinson’s Disease Using Long‐Term Acceleration Data
Steele et al. Separable processes before, during, and after the N400 elicited by previously inferred and new information: Evidence from time–frequency decompositions
Dadhich et al. A flexible analytic wavelet transform and ensemble bagged tree model for electroencephalogram-based meditative mind-wandering detection
Georgieva-Tsaneva et al. Cardio-diagnostic assisting computer system

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22965582

Country of ref document: EP

Kind code of ref document: A1