Drug rehabilitation system combining virtual reality technology and psychological tutoring

Abstract

The invention relates to the technical field of drug addiction treatment, in particular to a drug addiction treatment system combining virtual reality technology and psychological guidance, which comprises a data acquisition unit, a drug addiction evaluation unit, a virtual equipment debugging unit, a VR scene generation unit, a man-machine interaction unit, a stage evaluation unit and a comprehensive evaluation and optimization unit, the drug taking record data and a preset drug rehabilitation strategy are processed through a virtual device debugging unit, virtual devices are debugged, and a plurality of virtual scenes corresponding to the preset drug rehabilitation strategy are selected so as to fit the life scene of a drug rehabilitation person, and the experience feeling of the drug rehabilitation person is enhanced; and replacing the face image of the drug addict in the video data corresponding to each virtual scene with the face image of the drug addict through a VR scene generation unit. Through coordination and cooperation of the units, a complete drug rehabilitation system combining virtual reality and psychological counseling is formulated, the problem that a traditional drug rehabilitation method is poor in drug rehabilitation effect is solved, and the relapse probability of drug rehabilitation personnel is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of drug rehabilitation, and in particular to a drug rehabilitation system combining virtual reality technology and psychological counseling. Background Art

[0002] Drug rehabilitation means that drug addicts quit the bad habits of smoking and injecting drugs and their addiction to drugs. The methods of drug rehabilitation for drug addicts generally include three stages: detoxification counseling, rehabilitation counseling, and counseling for re-entering society. At present, natural withdrawal methods, drugs and non-drug withdrawal methods are generally used to help drug addicts to quit drugs. Practice has proved that natural withdrawal methods require drug addicts to have strong willpower, the detoxification process is painful, and it is easy to take drugs again, and it will have adverse effects on the physical health of drug addicts; drug withdrawal methods can easily make drug addicts dependent on drugs, making them unable to leave drugs, and the detoxification effect is not good; non-drug withdrawal methods treat drug addicts through both psychological and physiological aspects, but the detoxification time is longer, and it is easy for drug addicts to relapse, the detoxification effect is not thorough, and it is impossible to fundamentally quit drug addiction.

[0003] Therefore, the present invention provides a drug rehabilitation system combining virtual reality technology and psychological counseling to solve the above problems. Summary of the invention

[0004] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a drug rehabilitation system that combines virtual reality technology and psychological counseling to solve the problem that traditional drug rehabilitation methods are not effective and drug addicts are prone to relapse.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] A drug rehabilitation system combining virtual reality technology and psychological counseling, including:

[0007] The data collection unit is used to obtain drug use record data, medical test data and facial feature information of drug addicts.

[0008] The drug addiction assessment unit assesses the drug addiction level of drug addicts based on drug use record data and medical test data, and matches the corresponding preset drug rehabilitation strategy according to the drug addiction level. The preset drug rehabilitation strategy includes three or more of the drug type, drug aversion stage, cognitive rehabilitation stage, psychotherapy stage, and regression adaptation stage.

[0009] The virtual device debugging unit debugs the virtual device based on drug abuse record data and preset drug rehabilitation strategies, and selects multiple virtual scenes corresponding to the preset drug rehabilitation strategies; wherein the virtual device includes a VR host, VR glasses, VR detection instruments, VR interactive handles, VR displays and VR servers.

[0010] The VR scene generation unit includes obtaining video data corresponding to each virtual scene from a database, replacing the facial feature information of the drug addict in the video data based on the facial feature information of the drug addict, so as to obtain simulated video data; and fusing each virtual scene and the corresponding simulated video data to generate a VR scene in which the drug addict participates.

[0011] The human-computer interaction unit is used to import the VR scene into the virtual device. After starting each virtual device, the VR scene is played in the order of the preset drug rehabilitation strategy, and various interactive operations in the VR scene are performed through the VR interactive handle, and a VR detection instrument is used to record multiple detection data of the drug rehabilitation personnel in real time.

[0012] The stage evaluation unit evaluates the scores of drug addicts according to the test data of this stage after the drug addicts reach the treatment time of the stage treatment, and judges whether the drug addicts are qualified according to the scores. If qualified, the VR scene of the next stage will be played. If unqualified, the extended time of treatment will be evaluated according to the test data, and the VR scene of this stage will continue to be played.

[0013] The comprehensive evaluation and optimization unit evaluates the comprehensive treatment effect of drug addicts based on the scores of each stage after the drug addicts complete the last stage of treatment; obtains feedback data from multiple drug addicts, and at least adjusts the preset drug rehabilitation strategy based on the feedback data and the comprehensive treatment effect.

[0014] Preferably, the drug addiction assessment unit includes: collecting various test questions for assessing the addiction status of drug addicts, and storing them in a database of addiction status assessment question bank; automatically screening out test questions of a set type and a set number from the question bank to form a drug addiction questionnaire, and obtaining the score of the drug addiction questionnaire of the drug addict; using a weighted average method to analyze the score of the drug addiction questionnaire, medical test data, and drug use record data to obtain a drug addiction score D, and the calculation formula of the drug addiction score D is:

[0015]

[0016] Among them, p i is the score of the drug addiction questionnaire, the medical test data, and the i-th data in the drug use record data, α i For p i The weight value corresponding to the data, the total number of drug addiction questionnaire scores, medical test data and drug use record data is n; the corresponding drug addiction level is determined according to the drug addiction score. The medical test data at least includes BMI index, vital capacity, heart rate, sit-and-reach, grip strength, reaction time, one-minute push-ups, shuttle run, and drug use record data includes drug use duration and drug use frequency.

[0017] Preferably, the virtual device debugging unit includes: filtering out all virtual scenes of the drug type from the virtual scene library according to the drug type; filtering out virtual scenes of the drug aversion stage and the cognitive rehabilitation stage from all the virtual scenes; judging whether the preset drug rehabilitation strategy includes the psychotherapy stage, and if so, filtering out virtual scenes of the psychotherapy stage; and judging whether the preset drug rehabilitation strategy includes the regression adaptation stage, and if so, filtering out virtual scenes of the regression adaptation stage.

[0018] Preferably, the virtual device debugging unit further comprises: determining the correlation degree c between each virtual scene and the experience of the drug addict according to the drug taking record data, and the calculation formula of the correlation degree c is:

[0019] c=β pl H pl +β pe H pe +β en H en +β la H la ,

[0020] Among them, H pl is the place association index, H pe is the character association index, H en is the environmental relevance index, H la is the language association index, β pl , β pe , β en , β la H pl , H pe , H en , H la Corresponding weight coefficients; screening a preset number of virtual scenes in each stage in order of relevance from high to low.

[0021] Preferably, the VR scene generation unit includes: using a feature extractor to extract facial feature information of drug addicts in each video frame of video data; performing feature mapping on the facial feature information of drug addicts and the facial feature information of drug addicts to obtain a simulated facial image whose similarity exceeds a set similarity value; obtaining the facial position of the drug addict in each video frame of the video data; replacing the facial image of the drug addict in the video frame with the simulated facial image to obtain a simulated video frame; and updating the video data according to each of the simulated video frames to obtain simulated video data.

[0022] Preferably, the VR scene generation unit includes: acquiring simulated video frame data that meets the requirements in the simulated video data according to the order of each virtual scene and the video requirements; fusing the video frame data of each virtual scene with the corresponding simulated video frame data in chronological order to obtain a VR scene that can be interactive and in which drug addicts can participate.

[0023] Preferably, the human-computer interaction unit includes: according to the prompt of the VR scene, the drug addict makes a selection through the VR interactive handle to play the next VR scene; wherein different selections will display different VR scenes.

[0024] Preferably, the stage evaluation unit includes: querying the corresponding preset standards according to the various detection data of this stage to obtain multiple initial scores; using the weighted average method to process each initial score and the corresponding weight to obtain the score of the drug addict in this stage; when the score is less than the corresponding preset score, determining the extension time according to the corresponding relationship between the score and the preset extension time.

[0025] Preferably, the stage evaluation unit also includes: determining the abnormal time corresponding to the abnormal detection data with a score less than the preset score, determining a first VR scene to which the drug addicts react strongly based on the abnormal time, and using the first VR scene as the playback scene during extended treatment; or, playing the VR scene of this stage in its entirety.

[0026] Preferably, the human-computer interaction unit also includes a voice interaction module, which uses a GPT model to recognize the voice of drug addicts, extracts the voice information features of drug addicts, matches corresponding functions from a preset function library according to the voice information features, and executes the operation of starting the function.

[0027] The beneficial effects of the present invention are:

[0028] 1. The present invention obtains drug taking record data, medical examination data and facial feature information of drug addicts through a data acquisition unit; processes drug taking record data and medical examination data of drug addicts through a drug addiction assessment unit, assesses the drug addiction level of drug addicts, and matches a preset drug rehabilitation strategy with a reasonable duration and guaranteed drug rehabilitation effect according to the drug addiction level; processes drug taking record data and preset drug rehabilitation strategy through a virtual device debugging unit, debugs each virtual device, selects multiple virtual scenes corresponding to the preset drug rehabilitation strategy, so as to fit the life scene of drug addicts and enhance the experience of drug addicts; obtains video data corresponding to each virtual scene from a database through a VR scene generation unit, and replaces the facial feature information of drug addicts in the video data based on the facial feature information of drug addicts obtained by the data acquisition unit, obtains simulated video data in which drug addicts participate, and compares each virtual scene data with the corresponding The simulated video data is fused to obtain VR scenes in which drug addicts participate, so that drug addicts feel immersive when experiencing these VR scenes, thereby increasing the treatment effect of drug addicts and reducing the chance of relapse; in the human-computer interaction unit, drug addicts can perform various operations in the VR scene through the VR interactive handle to realize the human-computer interaction function, and when each virtual device is started, the status of drug addicts is detected in real time through the VR detection instrument; the stage evaluation unit is used to evaluate whether the treatment effect of drug addicts at each stage is qualified, if qualified, the next stage of treatment is carried out, if unqualified, the treatment time of this stage is extended to ensure the treatment effect of drug addicts; the entire treatment stage of drug addicts is evaluated through the comprehensive evaluation and optimization unit, the overall treatment effect is evaluated, and the preset drug rehabilitation strategy is adjusted according to the feedback data of multiple drug addicts, so that the preset drug rehabilitation strategy is more complete and efficient. The present invention has formulated a complete drug rehabilitation system combining virtual reality and psychological counseling through the coordination and cooperation of the above-mentioned various units, which solves the problem of poor drug rehabilitation effect of traditional drug rehabilitation methods and reduces the chance of relapse of drug addicts.

[0029] 2. In the drug addiction assessment unit of the present invention, a large number of psychological addiction state assessment questions with rich types stored in the database are used to automatically screen out questions of set types and set numbers to form a drug addiction questionnaire, and the drug addiction questionnaire is different each time the questions are answered. Then, the score of the drug addiction questionnaire is determined based on the choice of the drug addict; and then a weighted assessment method is used to analyze the score of the drug addiction questionnaire, medical test data, and drug use record data to obtain a comprehensive assessment of the drug addiction score; finally, the corresponding drug addiction level is determined according to the drug addiction score. The drug addiction assessment unit of the present invention can comprehensively assess the psychological addiction state and physical state of the drug addict in the above manner to obtain a comprehensive assessment of the drug addiction level. Compared with the assessment of the psychological addiction state or the assessment of the physical state alone, the assessed drug addiction level is more accurate.

[0030] 3. In the virtual device debugging unit of the present invention, first, according to the drug type in the preset drug rehabilitation strategy, all virtual scenes corresponding to the drug type are screened out from the virtual scene library; secondly, virtual scenes of the drug aversion stage and the cognitive rehabilitation stage are screened out from the virtual scenes, and then it is determined whether there is a psychotherapy stage and / or a regression adaptation stage in the preset drug rehabilitation strategy, and whether to screen out the corresponding virtual scenes, so as to avoid the problem of the processor running too fast and falling into a processing bottleneck causing the equipment to heat up when screening virtual scenes at the same time; in addition, according to the drug use record data, the correlation between each virtual scene and the experience of the drug addict is determined, and a preset number of virtual scenes for each stage are screened in order of correlation from high to low, and the correlation is considered from four angles: place, person, environment, and language, so as to fit the actual life scenes of the drug addicts, enhance the experience of the drug addicts, and thus enhance the treatment effect.

[0031] 4. In the VR scene generation unit of the present invention, the facial feature information of the drug addict in each video frame of the video data is obtained by a feature extractor, and based on the facial feature information of the drug addict obtained by the data acquisition unit, the facial feature information of the drug addict and the facial feature information of the drug addict are feature mapped to obtain a simulated facial image whose similarity with the drug addict exceeds the set similarity value; then based on the facial image position of the drug addict in each video frame, the facial image of the drug addict in the video frame is replaced with a simulated facial image; then the modified video frame replaces the original video frame, and each video frame is fused to obtain the simulated video data in which the drug addict participates; after the simulated video data is fused with each virtual scene, the VR scene in which the drug addict participates is obtained. The VR scene generation unit of the present invention adopts the above-mentioned method to make the drug addict feel immersive when experiencing the VR scene, which can enhance the disgust effect of the drug disgust stage, enhance the cognitive effect of the cognitive rehabilitation stage, enhance the therapeutic effect of the psychotherapy stage, and the adaptation effect of the regression adaptation stage, thereby reducing the relapse rate of drug addicts. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 The figure is a schematic flow chart of a drug rehabilitation system combining virtual reality technology and psychological counseling according to the present invention. DETAILED DESCRIPTION

[0033] The following will refer to Figure 1 The embodiments of the present invention are described in detail. It should be understood by those skilled in the art that

[0034] These implementation modes are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0035] A drug rehabilitation system that combines virtual reality technology and psychological counseling, such as Figure 1As shown, it includes a data acquisition unit, a drug addiction assessment unit, a virtual device debugging unit, a VR scene generation unit, a human-computer interaction unit, a stage assessment unit and a comprehensive assessment and optimization unit. Among them, the data acquisition unit is used to obtain drug use record data, medical test data and facial feature information of drug addicts. Drug use record data includes but is not limited to drug use duration, drug use frequency, drug use method, drug use location, drug use environment, number of drug users around, etc. Medical test data includes but is not limited to BMI index, vital capacity, heart rate, sit-up, grip strength, reaction time, one-minute push-ups, shuttle run speed, blood pressure, etc. The facial feature information of drug addicts is the facial feature information extracted by the feature extractor based on the facial image of the drug addict; the feature extractor can be various extractors and extraction models that can extract facial features, for example, a feature extraction model constructed based on a convolutional neural network and a feature extraction model constructed based on machine learning.

[0036] The drug addiction assessment unit includes assessing the drug addiction level of drug addicts based on drug use record data and medical test data, matching the corresponding preset drug rehabilitation strategy based on the drug addiction level, and the preset drug rehabilitation strategy includes three or more of the drug type, drug aversion stage, cognitive rehabilitation stage, psychotherapy stage, and regression adaptation stage. Among them, the preset drug rehabilitation strategy must include drug type, drug aversion stage, and cognitive rehabilitation stage. Whether the preset drug rehabilitation strategy adds a psychotherapy stage and / or a regression adaptation stage is determined based on the drug addiction level; when the psychotherapy stage and / or the regression adaptation stage are not included, the preset drug rehabilitation strategy can be added with a psychotherapy stage and / or a regression adaptation stage according to the requirements of the drug addicts to improve the treatment effect. In the drug aversion stage, a strong contrast is formed by the drug addicts watching the drug addicts before and after the state to enhance the drug addicts' aversion and fear of drugs. In the cognitive rehabilitation stage, the cognitive ability of drug addicts is trained through training and human-computer interaction. The psychotherapy stage is used to treat the inferiority, distortion, or depression of drug addicts. The regression adaptation stage is based on the situations that drug addicts may encounter after returning to normal, so that drug addicts can integrate into their new life as soon as possible after returning to normal.

[0037] The virtual device debugging unit debugs the virtual device based on the drug abuse record data and the preset drug rehabilitation strategy, selects multiple virtual scenes corresponding to the preset drug rehabilitation strategy, and establishes connections between the virtual devices so that the virtual devices can coordinate and cooperate to jointly realize functions; wherein, the virtual device includes a VR host, VR glasses, VR detection instruments, VR interactive handles, VR displays and VR servers.

[0038] Preferably, a plurality of virtual scenes corresponding to the preset drug rehabilitation strategy can be obtained from a database, or a plurality of virtual scenes corresponding to the preset drug rehabilitation strategy can be obtained from a memory of a virtual device.

[0039] The VR scene generation unit includes: obtaining video data corresponding to each virtual scene from a database, replacing the facial feature information of the drug addict in the video data based on the facial feature information of the drug addict, so as to obtain simulated video data; and fusing each virtual scene and the corresponding simulated video data to generate a VR scene in which the drug addict participates.

[0040] Preferably, the database stores video data of various types of drugs and various scenes, and the database is constantly increasing to enrich the types of drugs and video types of various scenes.

[0041] The human-computer interaction unit is used to import the VR scene into the virtual device. After starting each virtual device, the VR scene is played in the order of the preset drug rehabilitation strategy, and various interactive operations in the VR scene are performed through the VR interactive handle to complete the tasks of the VR scene in the treatment process, etc., and a VR detection instrument is used to record multiple detection data of the drug addicts in real time.

[0042] The stage evaluation unit evaluates the scores of drug addicts according to the test data of this stage after the drug addicts reach the treatment time of the stage treatment, and judges whether the drug addicts are qualified according to the scores. If qualified, the VR scene of the next stage will be played. If unqualified, the extended time of treatment will be evaluated according to the test data, and the VR scene of this stage will continue to be played.

[0043] The comprehensive evaluation and optimization unit evaluates the comprehensive treatment effect of drug addicts based on the scores of each stage after the drug addicts complete the last stage of treatment; obtains feedback data from multiple drug addicts, and at least adjusts the preset drug rehabilitation strategy based on the feedback data and the comprehensive treatment effect.

[0044] Preferably, first feedback data that accounts for more than a set proportion of the feedback data of drug addicts is obtained, and the preset drug rehabilitation strategy is adjusted according to the problems in the first feedback data, including but not limited to refining the content of each stage, adding treatment stages for drug addicts, etc., or adjusting the preset drug rehabilitation strategy and the image quality in the VR scene generation unit to increase clarity, etc.

[0045] In this embodiment, the present invention obtains drug taking record data, medical examination data and facial feature information of drug addicts through a data acquisition unit; processes drug taking record data and medical examination data of drug addicts through a drug addiction assessment unit, assesses the drug addiction level of drug addicts, and matches a preset drug rehabilitation strategy with a reasonable duration and guaranteed drug rehabilitation effect according to the drug addiction level; processes drug taking record data and preset drug rehabilitation strategy through a virtual device debugging unit, debugs each virtual device, selects multiple virtual scenes corresponding to the preset drug rehabilitation strategy, so as to fit the life scenes of drug addicts and enhance the experience of drug addicts; obtains video data corresponding to each virtual scene from a database through a VR scene generation unit, and replaces the facial feature information of drug addicts in the video data based on the facial feature information of drug addicts obtained by the data acquisition unit, obtains simulated video data in which drug addicts participate, and replaces each virtual scene data and The corresponding simulated video data is fused to obtain VR scenes in which drug addicts participate, so that drug addicts feel immersive when experiencing these VR scenes, thereby increasing the treatment effect of drug addicts and reducing the chance of relapse; in the human-computer interaction unit, drug addicts can perform various operations in the VR scene through the VR interactive handle to realize the human-computer interaction function, and when each virtual device is started, the state of drug addicts is detected in real time through the VR detection instrument; the stage evaluation unit is used to evaluate whether the treatment effect of drug addicts at each stage is qualified, if qualified, the next stage of treatment is carried out, if unqualified, the treatment time of this stage is extended to ensure the treatment effect of drug addicts; the entire treatment stage of drug addicts is evaluated through the comprehensive evaluation and optimization unit, the overall treatment effect is evaluated, and the preset drug rehabilitation strategy is adjusted according to the feedback data of multiple drug addicts, so that the preset drug rehabilitation strategy is more complete and efficient. The present invention has formulated a complete drug rehabilitation system combining virtual reality and psychological counseling through the coordination and cooperation of the above-mentioned various units, which solves the problem of poor drug rehabilitation effect of traditional drug rehabilitation methods and reduces the chance of relapse of drug addicts.

[0046] In one embodiment of the present invention, the drug addiction assessment unit includes: collecting various test questions for assessing the addiction status of drug addicts, and storing them in a database of addiction status assessment question bank; automatically screening out test questions of a set type and a set number from the question bank to form a drug addiction questionnaire, and obtaining the score of the drug addiction questionnaire of the drug addict; using a weighted average method to analyze the score of the drug addiction questionnaire, medical test data, and drug use record data to obtain a drug addiction score D, and the calculation formula of the drug addiction score D is:

[0047]

[0048] Among them, p i is the score of the drug addiction questionnaire, the medical test data, and the i-th data in the drug use record data, α i For p iThe weight value corresponding to the data, the total number of drug addiction questionnaire scores, medical test data and drug use record data is n; the corresponding drug addiction level is determined according to the drug addiction score. The medical test data at least includes BMI index, vital capacity, heart rate, sit-and-reach, grip strength, reaction time, one-minute push-ups, shuttle run, and drug use record data includes drug use duration and drug use frequency.

[0049] Preferably, the answering method is online answering, and the number of test questions and the number of test question types in the generated drug addiction assessment questionnaire can be determined according to actual needs, for example, the number of test questions is 100 and the number of types is 15. The pictures in the drug addiction assessment questionnaire are color pictures. The automatic screening method is to randomly screen out test questions under the condition of setting the number of test questions and the type of test questions to form a set of online drug addiction questionnaires.

[0050] Preferably, in the calculation formula for the drug addiction score, the weight values ​​corresponding to the scores of the drug addiction questionnaire, the medical test data and the drug use record data are fixed values, which are weight values ​​determined by medical experts.

[0051] Preferably, a comparison table is established between the drug addiction score and the drug addiction level, and when the drug addiction score is obtained, the drug addiction level can be determined according to the comparison table. In addition, a strategy comparison table is established between the drug addiction level and the preset drug detoxification strategy, and the corresponding preset drug detoxification strategy can be found according to the drug type and drug level.

[0052] In this embodiment, a large number of psychological addiction state assessment questions with rich types stored in the database are used to automatically screen out questions of set types and a set number to form a drug addiction questionnaire, and the drug addiction questionnaire is different each time the questions are answered. Then, the score of the drug addiction questionnaire is determined based on the choice of the drug addict; and then a weighted assessment method is used to analyze the score of the drug addiction questionnaire, medical test data, and drug use record data to obtain a comprehensive assessment of the drug addiction score; finally, the corresponding drug addiction level is determined according to the drug addiction score. The drug addiction assessment unit of the present invention can comprehensively assess the psychological addiction state and physical state of the drug addict in the above manner to obtain a comprehensive assessment of the drug addiction level. Compared with the separate assessment of the psychological addiction state or the separate assessment of the physical state, the assessed drug addiction level is more accurate.

[0053] In one embodiment of the present invention, the virtual device debugging unit includes: filtering out all virtual scenes corresponding to the drug types from the virtual scene library according to the drug types; filtering out virtual scenes of the drug aversion stage and the cognitive rehabilitation stage from all virtual scenes; judging whether the preset drug rehabilitation strategy includes the psychotherapy stage, and if so, filtering out virtual scenes of the psychotherapy stage; and judging whether the preset drug rehabilitation strategy includes the regression adaptation stage, and if so, filtering out virtual scenes of the regression adaptation stage.

[0054] Furthermore, the virtual device debugging unit also includes: determining the correlation c between each virtual scene and the experience of the drug addict according to the drug taking record data, and the calculation formula of the correlation c is:

[0055] c=β pl H pl +β pe H pe +β en H en +β la H la ,

[0056] Among them, H pl is the place association index, H pe is the character association index, H en is the environmental relevance index, H la is the language association index, β pl , β pe , β en , β la H pl , H pe , H en , H la Corresponding weight coefficients; screening a preset number of virtual scenes in each stage in order of relevance from high to low.

[0057] Preferably, the place association index is the degree of relevance between the virtual scene and the place experienced by the drug addicts, wherein the place name, built-in items in the place, built-in color of the place and the shape of the place each account for 25%, and the place association index includes four values ​​of 0, 25%, 50% and 100%; the character association index is the degree of relevance between the virtual scene and the characters experienced by the drug addicts, mainly including the number of characters, the height of the characters, and the gender of the characters, each accounting for 33.3%; the environment association index is the degree of relevance between the virtual scene and the environment experienced by the drug addicts, which is divided into two values ​​of 50% and 100%. For example, the environment association index of common environments is 100%, and the common environments include hotels, KTVs, etc., and the environment association index of unfamiliar environments such as deserts and caves is 50%, which can be adjusted according to the actual situation of the drug addicts; the language association index is the similarity between the language in the virtual scene and the language frequently communicated with the drug addicts, wherein the language association index corresponding to the local spoken language and the standard native language is 100%, and other languages ​​are determined according to the familiarity of the drug addicts, and the language association index corresponding to the unfamiliar language is 0. The weight values ​​of the place association index, the person association index, the environment association index and the language association index may all be 25%, or may be adjusted according to actual conditions to obtain a more appropriate association degree.

[0058] In this embodiment, in the virtual device debugging unit of the present invention, first, according to the drug type in the preset drug rehabilitation strategy, all virtual scenes corresponding to the drug type are screened out from the virtual scene library; secondly, virtual scenes of the drug aversion stage and the cognitive rehabilitation stage are screened out from the virtual scenes, and then it is determined whether there is a psychotherapy stage and / or a regression adaptation stage in the preset drug rehabilitation strategy, and whether to screen out the corresponding virtual scenes, thereby avoiding the problem of the processor running too fast and falling into a processing bottleneck causing the equipment to heat up when screening virtual scenes at the same time; in addition, the correlation between each virtual scene and the experience of the drug addict is determined according to the drug use record data, and a preset number of virtual scenes for each stage are screened in order from high to low in correlation, and the correlation is considered from four perspectives: place, person, environment, and language, so as to fit the actual life scenes of the drug addicts, enhance the experience of the drug addicts, and thus enhance the treatment effect.

[0059] In one embodiment of the present invention, a VR scene generation unit includes: extracting facial feature information of a drug addict in each video frame of video data using a feature extractor; performing feature mapping on the facial feature information of a drug rehabilitation personnel and the facial feature information of a drug addict, modifying the facial feature information of the drug addict based on the facial feature information of the drug addict, modifying the data at the same position or adjacent position of the facial feature information of the drug addict to the data at the corresponding position of the facial feature information of the drug rehabilitation personnel, and then processing the modified facial feature information of the drug addict through a feature fusion device to obtain a simulated facial image whose similarity exceeds a set similarity value; obtaining the facial position of the drug addict in each video frame of the video data; replacing the facial image of the drug addict in the video frame with the simulated facial image to obtain a simulated video frame; updating the video data according to each of the simulated video frames, replacing the modified video frame data with the corresponding simulated video frame data, and obtaining the simulated video data after integration through a video converter.

[0060] Preferably, the similarity value is set to 70% or 80%, which can be adjusted according to the needs of drug addicts.

[0061] Preferably, the feature extractor can be any extractor or extraction model capable of extracting facial features, such as a feature extraction model based on a convolutional neural network or a feature extraction model based on machine learning. The feature fusion device is the reverse operation of the feature extractor, processing the feature information into an image, and the specific method is not limited.

[0062] In this embodiment, in the VR scene generation unit of the present invention, the facial feature information of the drug addict in each video frame of the video data is obtained by a feature extractor, and based on the facial feature information of the drug addict obtained by the data acquisition unit, the facial feature information of the drug addict and the facial feature information of the drug addict are feature mapped to obtain a simulated facial image whose similarity with the drug addict exceeds the set similarity value; then based on the facial image position of the drug addict in each video frame, the facial image of the drug addict in the video frame is replaced with a simulated facial image; then the modified video frame replaces the original video frame, and each video frame is fused to obtain simulated video data in which the drug addict participates; after the simulated video data is fused with each virtual scene, a VR scene in which the drug addict participates is obtained. The VR scene generation unit of the present invention adopts the above-mentioned method to make the drug addict feel immersive when experiencing the VR scene, which can enhance the disgust effect of the drug disgust stage, enhance the cognitive effect of the cognitive rehabilitation stage, enhance the therapeutic effect of the psychotherapy stage, and the adaptation effect of the regression adaptation stage, thereby reducing the relapse rate of drug addicts.

[0063] In an optional embodiment of the present invention, the VR scene generation unit includes: obtaining simulated video frame data that meets the requirements in the simulated video data according to the order of each virtual scene and the video requirements; fusing the video frame data of each virtual scene and the corresponding simulated video frame data in chronological order to obtain a VR scene that can be interactive and in which drug addicts can participate.

[0064] In an optional embodiment of the present invention, the VR scene generation unit includes: according to the video content requirements and time length in the virtual scene, cutting the simulated video data that meets the requirements from the simulated video data corresponding to the virtual scene, and then splicing the simulated video data that meets the requirements and the corresponding virtual scene by video splicing to obtain initial splicing data; splicing each initial splicing data according to the order of the scenes in the preset drug rehabilitation strategy to obtain an interactive VR scene in which drug addicts participate. Through the setting method of this embodiment, drug addicts can have an immersive feeling during treatment, as if they are experiencing it in person, which can enhance the disgust effect of the drug disgust stage, enhance the cognitive effect of the cognitive rehabilitation stage, enhance the therapeutic effect of the psychotherapy stage, and the adaptation effect of the regression adaptation stage, thereby improving the treatment effect of drug addicts and reducing the relapse rate of drug addicts.

[0065] In one embodiment of the present invention, the human-computer interaction unit includes: according to the prompt of the VR scene, the drug addict makes a selection through the VR interactive handle to play the next VR scene; wherein different selections will display different VR scenes.

[0066] In one embodiment of the present invention, the stage evaluation unit includes: querying the corresponding preset standards according to the various detection data of this stage to obtain multiple initial scores; using the weighted average method to process each initial score and the corresponding weight to obtain the score of the drug addict in this stage; when the score is less than the corresponding preset score, determining the extension time according to the corresponding relationship between the score and the preset extension time.

[0067] Furthermore, the stage evaluation unit also includes: determining the abnormal time corresponding to the abnormal detection data with a score less than a preset score, determining a first VR scene to which the drug addicts react strongly based on the abnormal time, and using the first VR scene as the playback scene during extended treatment; or, playing the VR scene of this stage in its entirety.

[0068] Preferably, the abnormal detection data is the detection data that causes the score of this stage to be less than a preset score.

[0069] Preferably, the preset score is generally 60 points, which can be adjusted according to actual conditions.

[0070] Preferably, the above reaction refers to the emotional fluctuations of the drug addicts that cause abnormal behavior in the detection data.

[0071] Preferably, a relationship table is established between the score and the preset extension time, and the corresponding preset extension time can be found out through the score.

[0072] In this embodiment, the present invention evaluates the treatment effect of this stage through a stage evaluation unit. When the treatment effect does not reach the expected effect, the treatment time of this stage is extended, thereby ensuring the overall treatment effect of the drug addicts and helping to reduce the relapse rate of drug addicts.

[0073] In one embodiment of the present invention, the human-computer interaction unit also includes a voice interaction module, which includes using a GPT model to recognize the voice of drug addicts, extracting the voice information features of drug addicts, matching corresponding functions from a preset function library according to keywords in the voice information features, and then the voice interaction module executes the operation of starting the function to achieve human-computer interaction. The GPT model is used to communicate with drug addicts, comfort and encourage drug addicts, switch music, pause the system, end the system early, and other functions to provide convenience for drug addicts.

[0074] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A drug rehabilitation system combining virtual reality technology and psychological counseling, characterized in that: include: A data collection unit, used to obtain drug use record data, medical test data and facial feature information of drug addicts; The drug addiction assessment unit assesses the drug addiction level of drug addicts based on drug use record data and medical test data, and matches the corresponding preset drug addiction strategy according to the drug addiction level. The preset drug addiction strategy includes three or more of the drug type, drug aversion stage, cognitive rehabilitation stage, psychotherapy stage, and regression adaptation stage; A virtual device debugging unit, which debugs the virtual device based on drug abuse record data and a preset drug rehabilitation strategy, and selects a plurality of virtual scenes corresponding to the preset drug rehabilitation strategy; wherein the virtual device includes a VR host, VR glasses, VR detection equipment, VR interactive handle, VR display and VR server; The VR scene generation unit includes obtaining video data corresponding to each virtual scene from a database, replacing facial feature information of drug addicts in the video data based on facial feature information of drug addicts, so as to obtain simulated video data; and fusing each virtual scene with the corresponding simulated video data to generate a VR scene in which drug addicts participate. A human-computer interaction unit is used to import the VR scene into the virtual device, and after starting each virtual device, play the VR scene in the order of the preset drug rehabilitation strategy, perform various interactive operations in the VR scene through the VR interactive handle, and use the VR detection instrument to record multiple detection data of the drug rehabilitation personnel in real time; The stage evaluation unit evaluates the scores of drug addicts according to the test data of this stage after the drug addicts reach the treatment time of the stage treatment, and judges whether the drug addicts are qualified according to the scores. If qualified, the VR scene of the next stage is played; if not qualified, the extended time of treatment is evaluated according to the test data, and the VR scene of this stage continues to be played; The comprehensive evaluation and optimization unit evaluates the comprehensive treatment effect of drug addicts based on the scores of each stage after the drug addicts complete the last stage of treatment; obtains feedback data from multiple drug addicts, and at least adjusts the preset drug rehabilitation strategy based on the feedback data and the comprehensive treatment effect.

2. The drug rehabilitation system according to claim 1, characterized in that: The drug addiction assessment unit includes: collecting various test questions for assessing the addiction status of drug addicts, and storing them in a question bank for assessing the addiction status in a database; automatically screening out test questions of a set type and a set number from the question bank to form a drug addiction questionnaire, and obtaining the score of the drug addiction questionnaire of the drug addict; using a weighted average method to analyze the score of the drug addiction questionnaire, medical test data, and drug use record data to obtain a drug addiction score D, and the calculation formula of the drug addiction score D is: Among them, p i is the score of the drug addiction questionnaire, the medical test data, and the i-th data in the drug use record data, α i For p i The weight value corresponding to the data, the total number of scores in the drug addiction questionnaire, medical test data and drug use record data is n; the corresponding drug addiction level is determined according to the drug addiction score; the medical test data at least includes BMI index, vital capacity, heart rate, sit-and-reach, grip strength, reaction time, one-minute push-ups, and shuttle run; the drug use record data includes drug use duration and drug use frequency.

3. The drug rehabilitation system according to claim 1, characterized in that: The virtual device debugging unit includes: filtering out all virtual scenes of the drug type from the virtual scene library according to the drug type; filtering out virtual scenes of the drug aversion stage and the cognitive rehabilitation stage from all the virtual scenes; judging whether the preset drug rehabilitation strategy includes the psychotherapy stage, and if so, filtering out virtual scenes of the psychotherapy stage; and judging whether the preset drug rehabilitation strategy includes the regression adaptation stage, and if so, filtering out virtual scenes of the regression adaptation stage.

4. The drug rehabilitation system according to claim 4, characterized in that: The virtual device debugging unit further includes: determining the correlation degree c between each virtual scene and the experience of the drug addict according to the drug taking record data, and the calculation formula of the correlation degree c is: c=β pl H pl +b pe H pe +b en H en +b la H la , Among them, H pl is the place association index, H pe is the character association index, H en is the environmental relevance index, H la is the language association index, β pl , β pe , β en , β la H pl , H pe , H en , H la Corresponding weight coefficients; screening a preset number of virtual scenes in each stage in order of relevance from high to low.

5. The drug rehabilitation system according to claim 1, characterized in that: The VR scene generation unit includes: using a feature extractor to extract facial feature information of drug addicts in each video frame of video data; performing feature mapping on the facial feature information of drug addicts and the facial feature information of drug addicts to obtain a simulated facial image whose similarity exceeds a set similarity value; obtaining the facial position of the drug addict in each video frame of the video data; replacing the facial image of the drug addict in the video frame with the simulated facial image to obtain a simulated video frame; and updating the video data according to each of the simulated video frames to obtain simulated video data.

6. The drug rehabilitation system according to claim 5, characterized in that: The VR scene generation unit includes: obtaining simulated video frame data that meets the requirements in the simulated video data according to the order of each virtual scene and the video requirements; fusing the video frame data of each virtual scene with the corresponding simulated video frame data in time sequence to obtain a VR scene that can be interactive and in which drug addicts can participate.

7. The drug rehabilitation system according to claim 6, characterized in that: The human-computer interaction unit includes: according to the prompt of the VR scene, the drug addict makes a selection through the VR interactive handle to play the next VR scene; different selections will display different VR scenes.

8. The drug rehabilitation system according to claim 1, characterized in that: The stage evaluation unit includes: querying the corresponding preset standards according to the various detection data of this stage to obtain multiple initial scores; using the weighted average method to process each initial score and the corresponding weight to obtain the score of the drug addict in this stage; when the score is less than the corresponding preset score, determining the extension time according to the corresponding relationship between the score and the preset extension time.

9. The drug rehabilitation system according to claim 8, characterized in that: The stage evaluation unit also includes: determining the abnormal time corresponding to the abnormal detection data with a score less than the preset score, determining a first VR scene to which the drug addicts react strongly based on the abnormal time, and using the first VR scene as the playback scene during extended treatment; or, completely playing the VR scene of this stage.

10. The drug rehabilitation system according to claim 1, characterized in that: The human-computer interaction unit also includes a voice interaction module, which uses a GPT model to recognize the voice of drug addicts, extracts voice information features of drug addicts, matches corresponding functions from a preset function library according to the voice information features, and executes the operation of starting the function.

Images