Integrated platform for rapid detection and targeted intervention of addiction susceptibility

Abstract

The invention relates to the cross field of mental health and artificial intelligence technology, in particular to an addiction susceptibility rapid detection and targeted intervention integrated platform which comprises an addiction susceptibility detection module, a data analysis and evaluation module, a targeted intervention module and a feedback and adjustment module. Through fusion of multi-dimensional data such as gene detection, physiological signal monitoring, behavior data acquisition and psychological assessment, rapid identification of individual addiction risks is realized, and based on a machine learning and reinforcement learning algorithm, a dynamic intervention scheme is matched for an individual, a refined intervention task is executed, and an intervention path is adjusted in real time. The platform can be widely applied to scenes of addiction prevention, addiction rehabilitation management, mental health screening and the like of teenagers, and has the advantages of high detection efficiency, high intervention precision and strong adaptability.

Description

Technical Field

[0001] The present invention relates to the intersection of mental health and artificial intelligence technology, and specifically to an addiction risk detection and intervention system, in particular an integrated platform for rapid detection and targeted intervention of addiction susceptibility. Background Art

[0002] Currently, research on addictive behavior identification and risk warning primarily focuses on behavioral data observation and subjective questionnaire assessments, lacking a mechanism for jointly analyzing multidimensional features such as genes, physiological signals, and psychological traits. On the one hand, traditional questionnaires are highly subjective and susceptible to emotional influences; on the other hand, single-dimensional data struggles to accurately reflect the physiological and psychological mechanisms of addiction, leading to delayed risk identification and a failure to effectively capture individual differences.

[0003] Most existing addiction intervention systems utilize fixed intervention templates, lacking targeted intervention mechanisms tailored to individual risk levels, addiction types, and behavioral patterns. This results in intervention pathways that are difficult to adapt to individual needs and limited effectiveness. Furthermore, current systems generally lack the ability to collect real-time feedback and adaptively adjust strategies after interventions, preventing closed-loop optimization of interventions. This results in generally low user engagement and response rates.

[0004] Currently, addiction risk warning and intervention systems are mostly modular and lack a unified platform-based integration mechanism. This is particularly true for multi-source data integration, model invocation, and task scheduling, which lack a unified coordination and control module. This results in weak inter-system collaboration and high response latency. Furthermore, existing systems lack adequate consideration for data security, permission management, and multi-device compatibility, limiting their widespread application across diverse scenarios and populations. Summary of the Invention

[0005] To achieve the above-mentioned purpose, the present invention provides the following technical solution: an integrated platform for rapid detection and targeted intervention of addiction susceptibility, including the following modules:

[0006] The addiction susceptibility detection module is used to quickly assess an individual's addiction susceptibility and obtain addiction susceptibility assessment results through biomarker testing, psychological assessment, and behavioral pattern analysis;

[0007] The data analysis and evaluation module is used to generate an individual addiction risk prediction model based on the evaluation results provided by the addiction susceptibility detection module and in combination with the data analysis algorithm;

[0008] A targeted intervention module, which provides individuals with tailored intervention plans based on an addiction risk prediction model. Interventions include cognitive behavioral therapy, medication intervention, and psychological support.

[0009] Feedback and adjustment module, used to monitor the intervention effect in real time and dynamically adjust the intervention plan based on the monitoring results;

[0010] The central control unit is connected to the above modules and is used to coordinate data flow, task scheduling and status monitoring, and provide a graphical interface and remote management interface.

[0011] Preferably, the addiction susceptibility detection module includes:

[0012] The gene detection submodule is used to extract and analyze user blood samples, detect genetic marker sites related to addiction through high-throughput sequencing technology, and form an individual genotype dataset;

[0013] The neurophysiological monitoring submodule is used to collect real-time electroencephalogram (EEG), heart rate variability, and galvanic skin response physiological signals to reflect an individual's emotional response and impulse control ability;

[0014] The behavioral data collection submodule is used to collect user behavior pattern data through wearable devices and smart terminals, including device usage frequency, Internet time, application type preferences, and circadian activity rhythm information;

[0015] The psychological measurement submodule is used to push standardized psychological questionnaires to users and collect psychological characteristic data such as personality traits, emotional state, impulsivity, and depression level.

[0016] Preferably, the data analysis and evaluation module includes:

[0017] The feature vector preprocessing submodule is used to normalize, fill in missing values, and perform principal component analysis dimensionality reduction on the addiction feature vector input by the detection module;

[0018] The addiction risk modeling submodule uses an addiction risk prediction model trained based on historical label data to assign a risk score to the input feature vector and output an individual's addiction probability value and influencing factor ranking;

[0019] The risk level classification submodule automatically divides users into low-risk, medium-risk and high-risk categories based on the scoring results and preset thresholds, and outputs risk level reports and decision-making basis.

[0020] Preferably, the targeted intervention module includes:

[0021] The intervention path planning submodule is used to automatically recommend intervention path templates based on the user's risk level, individual characteristics and behavioral preferences, and can be manually confirmed by a doctor;

[0022] The intervention task management submodule generates a daily intervention task list based on the selected path, including attention training, emotion regulation exercises, and task delay training. It can also adjust the task difficulty and frequency based on user feedback.

[0023] The immersive intervention execution sub-module presents intervention content through smart terminal devices. The content includes virtual cognitive training scenarios, meditation guidance content, and audio emotion training materials to improve intervention participation and effectiveness.

[0024] Preferably, the feedback and adjustment module includes:

[0025] The intervention feedback collection submodule is used to collect the user's behavioral feedback, completion status, emotional state changes and physiological index fluctuations during the intervention process in real time, and automatically generate daily feedback records;

[0026] The dynamic adjustment submodule is used to identify intervention fatigue, inefficiency, and negative emotions based on feedback records, triggering an adjustment mechanism to automatically adjust the intensity, rhythm, and content of the intervention;

[0027] The effectiveness evaluation and attribution module is used to periodically generate intervention effectiveness reports and analyze intervention response factors in combination with user portraits to support precise path optimization.

[0028] Preferably, the central control unit includes:

[0029] The data scheduling and management subunit is used to receive multi-source data collected by the addiction susceptibility detection module and push it to the data analysis and evaluation module for processing based on task priority and data type classification;

[0030] The model calling and reasoning subunit is used to call the preset intervention strategy model and personalized task generation engine after receiving the risk results output by the assessment module, and output the initial intervention path and task parameters;

[0031] The task distribution and execution monitoring subunit is used to distribute the task instructions generated by the intervention module to the user-side execution device, which includes a mobile terminal, wearable device, and virtual reality device, and monitor the execution status, response delay, and user interaction in real time;

[0032] The dynamic adjustment control subunit is used to compare the original path with the task execution trajectory based on the feedback data returned by the feedback and adjustment module, call the strategy adjustment model and regenerate the intervention task plan;

[0033] The permission and security management sub-unit is used to manage user permission verification, data access control and communication encryption mechanisms to ensure the security of platform operation and data privacy compliance.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] Achieve high-precision, multi-dimensional rapid identification of addiction susceptibility, significantly enhancing early warning capabilities: By integrating multiple detection submodules, including genetic testing, neurophysiological monitoring, behavioral data collection, and psychometrics, the system achieves comprehensive quantitative analysis of addiction susceptibility. The system combines high-throughput sequencing with physiological signal acquisition to construct precise individual profiles, enabling the platform to conduct highly sensitive and specific early risk identification before addictive behaviors occur, providing a scientific basis for subsequent intervention.

[0036] Building a dynamic, closed-loop intervention system based on individual profiles to improve intervention precision and response efficiency: This platform forms a closed-loop intervention system through risk level assessment, intelligent intervention path planning, immersive task execution, and real-time feedback mechanisms. During the intervention execution phase, the system automatically adjusts task frequency and content based on user feedback and physiological responses, effectively alleviating intervention fatigue, enhancing individual adaptability and sustainability, and improving overall intervention effectiveness and user compliance.

[0037] This platform provides a modular, intelligent, and integrated architecture suitable for addiction management needs across multiple scenarios. The platform utilizes a central control unit to centrally dispatch resources across multiple modules, offering high scalability and compatibility. It supports the identification and intervention of various addictive behaviors, including internet addiction, drug dependency, and compulsive behavior in adolescents. Deployed in the cloud and collaborating with edge devices, the platform ensures data processing efficiency and privacy compliance, providing reliable intelligent solutions for multiple application scenarios, including education, healthcare, and justice. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 Schematic diagram of the module flow provided for this application;

[0039] Figure 2 Schematic diagram of the addiction susceptibility detection module provided for this application;

[0040] Figure 3 Schematic diagram of the data analysis and evaluation module provided for this application;

[0041] Figure 4 Schematic diagram of the targeted intervention module provided for this application;

[0042] Figure 5 Schematic diagram of the feedback and adjustment module provided for this application;

[0043] Figure 6 Schematic diagram of the central control unit provided for this application. DETAILED DESCRIPTION

[0044] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0045] refer to Figures 1-6 The embodiment of the present invention provides an integrated platform for rapid detection and targeted intervention of addiction susceptibility, including the following modules:

[0046] The addiction susceptibility detection module is used to quickly assess an individual's susceptibility to addiction and obtain addiction susceptibility assessment results through biomarker testing, psychological assessment, and behavioral pattern analysis.

[0047] The Addiction Susceptibility Detection module integrates four submodules to acquire multi-source addiction-related data from four dimensions: genetics, physiology, behavior, and psychology. The genetic testing submodule collects user blood samples using EDTA anticoagulation. After DNA extraction, it uses a high-throughput sequencing platform to analyze addiction-related SNPs such as rs1800497, outputting a personalized genotype matrix. The neurophysiological monitoring submodule uses electrocardiogram (ECG) sensors and galvanic skin patches to monitor the user's brain waves and skin conductance responses in real time, uploading signal samples at a 1Hz frequency. The behavioral data collection submodule utilizes smart bracelets and mobile phone APIs to capture dynamic information such as user app usage, gait behavior, and circadian rhythms in real time. The psychological measurement submodule uses standardized scales to push data online, combining user response time and response latency to construct a psychological indicator database. All data is uniformly converted into a standard feature vector format for input into downstream analysis modules.

[0048] The data analysis and evaluation module is used to generate an individual addiction risk prediction model based on the evaluation results provided by the addiction susceptibility detection module and combined with the data analysis algorithm.

[0049] In the data analysis and evaluation module, the feature vector generated by the addiction susceptibility detection module is first normalized, missing value interpolated, and PCA dimension reduction is performed by the feature vector preprocessing submodule. Subsequently, the processed vector is input into the addiction risk modeling submodule, which loads a multi-classification model built based on the XGBoost algorithm and has been trained in a training set containing more than 10,000 historical samples. The model outputs the user's addiction probability distribution (low, medium, and high) and returns the ranking of the most contributing feature factors, such as "impulsivity score" and "rs4680 genotype". Finally, the risk level classification submodule compares the probability value with the system's preset threshold (low <0.3, medium 0.3-0.7, high >0.7), automatically generates a level label, and combines the risk source and key influencing factors to generate a structured assessment report.

[0050] The targeted intervention module is used to provide individuals with tailored intervention plans based on the addiction risk prediction model. Intervention methods include cognitive behavioral therapy, drug intervention, and psychological support.

[0051] In the targeted intervention module, the system calls the intervention path planning submodule to construct an intervention plan map based on the individual risk level generated by the assessment module. Taking medium-risk users as an example, the system recommends the "meditation training + task delay practice" intervention path, and the path content takes effect after confirmation by the doctor. The intervention task management submodule combines the user's work and rest data and behavioral preferences to push 3 training tasks on a daily basis, and automatically adjusts the task difficulty (meditation duration, intervention content structure) based on the user's completion status (click rate, task duration). The immersive intervention execution submodule calls the VR device to present interactive attention training scenes (such as "target search in an immersive forest") and enhances emotional self-regulation through audio training materials. The system records interactive behaviors and emotional change feedback during the training process to provide support for subsequent feedback modules.

[0052] The feedback and adjustment module is used to monitor the intervention effect in real time and dynamically adjust the intervention plan based on the monitoring results.

[0053] The feedback and adjustment module includes the following submodules:

[0054] Intervention feedback collection submodule: Through sensing components and interactive interfaces deployed in user terminals (such as mobile applications, wearable devices, and VR headsets), the system collects real-time behavioral feedback from users during intervention tasks. This includes behavioral indicators such as task completion rate, usage time, and interaction frequency. It also simultaneously collects fluctuations in physiological signals such as heart rate, skin electricity, and brain waves. The system also regularly pushes simplified emotional self-assessment questionnaires to supplement subjective emotional state data. All collected results are initially filtered and encoded by the edge processor before being uploaded to the platform data center.

[0055] The Dynamic Adjustment Submodule receives and analyzes data uploaded by the Intervention Feedback Collection Submodule, compares it with historical user data, and automatically identifies abnormalities such as intervention fatigue (e.g., decreased task completion rate), ineffective intervention (e.g., lack of significant mood improvement), or even negative reactions (e.g., increased stress levels). Once a negative trend is identified, the system automatically adjusts the subsequent intervention plan, such as reducing task frequency, changing task types, or pushing encouraging content, to ensure the dynamic adaptability of the intervention plan.

[0056] Effectiveness Evaluation and Attribution Module: This module conducts periodic evaluations of user intervention responses based on pre-set intervals (e.g., every seven days or after completing a phased intervention task). Using a multidimensional data fusion model, the platform comprehensively scores users' behavioral improvements, physiological stability, and emotional changes, generating an effectiveness evaluation report. Furthermore, by combining user profiles, task completion paths, and feedback data, it extracts labels for key factors influencing effectiveness (e.g., intervention content preferences, response timeframe, and emotional adaptation speed) for subsequent intervention strategy optimization and path attribution analysis.

[0057] The central control unit is connected to the above modules and is used to coordinate data flow, task scheduling and status monitoring, and provide a graphical interface and remote management interface.

[0058] The central control unit, serving as the platform's information hub and scheduling core, is primarily responsible for coordinating data flow, model invocation, task push, and execution monitoring among the five modules to ensure efficient closed-loop operation of the addiction detection and intervention process. The central control unit includes the following subunits:

[0059] The Data Scheduling and Management Subunit: Through a unified data access interface, this subunit receives standardized data from multiple sources within the Addiction Susceptibility Detection Module, including genotype data, physiological signals, behavioral data, and psychological questionnaire results. The system intelligently queues and labels data based on data type (structured / unstructured), timeliness, and processing priority. This data is then distributed to the Data Analysis and Evaluation Module for feature extraction and modeling analysis based on task flow dependency logic, ensuring maximum data utilization and efficient serial / parallel task execution.

[0060] The model invocation and reasoning subunit is automatically activated after the data analysis module outputs an individual risk assessment report. Based on the assessment results, the system automatically selects a matching intervention strategy model and a personalized intervention path recommendation engine. Taking into account parameters such as user profile, risk level, and intervention preferences, it generates preliminary intervention paths and task templates through a reasoning mechanism, annotating the execution conditions and adjustment parameters of task nodes for reference by subsequent intervention modules.

[0061] The Task Distribution and Execution Monitoring Subunit structures the generated intervention tasks into instruction packages recognizable by user terminals and automatically pushes them to various execution devices, including mobile applications, smart wristbands, and virtual reality devices. During the intervention execution process, this subunit monitors user task completion status, interaction frequency, system response latency, and other operational indicators in real time. It also records key behavioral nodes through a log upload mechanism for subsequent feedback and adjustment module processing.

[0062] The Dynamic Adjustment Control Subunit receives intervention feedback data from the Feedback and Adjustment Module, either periodically or in real time, and determines whether the current intervention task is offset from its target. Upon identifying a deviation threshold or an abnormal intervention response, this subunit automatically triggers a policy adjustment process, invoking a path adjustment model to replan the task flow. It also updates key parameters such as task pacing, content complexity, and interaction methods based on user status, enabling personalized, adaptive intervention iteration.

[0063] The Permissions and Security Management Sub-Unit builds a complete user permission tree and data access policy, responsible for login verification, multi-factor authentication, and sensitive data encryption and desensitization for all levels of accounts (such as users, doctors, and platform administrators). Furthermore, a TLS encrypted tunnel and access log tracking mechanism are implemented for data communication between all modules of the platform to ensure information security and compliance throughout the addiction detection and intervention process, meeting relevant data regulatory requirements in the healthcare sector.

[0064] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features and technical solutions therein may be combined with each other.

[0065] Obviously, the embodiments described above are only some embodiments of the present invention, rather than all embodiments. The preferred embodiments of the present invention are given in the accompanying drawings, but they do not limit the patent scope of the present invention. The present invention can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive. Although the present invention has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions described in the aforementioned specific embodiments, or to make equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the present invention specification and drawings, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present invention.

Claims

1. An integrated platform for rapid detection and targeted intervention of addiction susceptibility, characterized by: Includes the following modules: The addiction susceptibility detection module is used to quickly assess an individual's addiction susceptibility and obtain addiction susceptibility assessment results through biomarker testing, psychological assessment, and behavioral pattern analysis; The data analysis and evaluation module is used to generate an individual addiction risk prediction model based on the evaluation results provided by the addiction susceptibility detection module and in combination with the data analysis algorithm; A targeted intervention module, which provides individuals with tailored intervention plans based on an addiction risk prediction model. Interventions include cognitive behavioral therapy, medication intervention, and psychological support. Feedback and adjustment module, used to monitor the intervention effect in real time and dynamically adjust the intervention plan based on the monitoring results; The central control unit is connected to the above modules and is used to coordinate data flow, task scheduling and status monitoring, and provide a graphical interface and remote management interface.

2. The integrated platform for rapid detection and targeted intervention of addiction susceptibility according to claim 1, characterized in that: The addiction susceptibility detection module includes: The gene detection submodule is used to extract and analyze user blood samples, detect genetic marker sites related to addiction through high-throughput sequencing technology, and form an individual genotype dataset; The neurophysiological monitoring submodule is used to collect real-time electroencephalogram (EEG), heart rate variability, and galvanic skin response physiological signals to reflect an individual's emotional response and impulse control ability; The behavioral data collection submodule is used to collect user behavior pattern data through wearable devices and smart terminals, including device usage frequency, Internet time, application type preferences, and circadian activity rhythm information; The psychological measurement submodule is used to push standardized psychological questionnaires to users and collect psychological characteristic data such as personality traits, emotional state, impulsivity, and depression level.

3. The integrated platform for rapid detection and targeted intervention of addiction susceptibility according to claim 1, characterized in that: The data analysis and evaluation module includes: The feature vector preprocessing submodule is used to normalize, fill in missing values, and perform principal component analysis dimensionality reduction on the addiction feature vector input by the detection module; The addiction risk modeling submodule uses an addiction risk prediction model trained based on historical label data to assign a risk score to the input feature vector and output an individual's addiction probability value and influencing factor ranking; The risk level classification submodule automatically divides users into low-risk, medium-risk and high-risk categories based on the scoring results and preset thresholds, and outputs risk level reports and decision-making basis.

4. The integrated platform for rapid detection and targeted intervention of addiction susceptibility according to claim 1, characterized in that: The targeted intervention module includes: The intervention path planning submodule is used to automatically recommend intervention path templates based on the user's risk level, individual characteristics and behavioral preferences, and can be manually confirmed by a doctor; The intervention task management submodule generates a daily intervention task list based on the selected path, including attention training, emotion regulation exercises, and task delay training. It can also adjust the task difficulty and frequency based on user feedback. The immersive intervention execution sub-module presents intervention content through smart terminal devices. The content includes virtual cognitive training scenarios, meditation guidance content, and audio emotion training materials to improve intervention participation and effectiveness.

5. The integrated platform for rapid detection and targeted intervention of addiction susceptibility according to claim 1, characterized in that: The feedback and adjustment module includes: The intervention feedback collection submodule is used to collect the user's behavioral feedback, completion status, emotional state changes and physiological index fluctuations during the intervention process in real time, and automatically generate daily feedback records; The dynamic adjustment submodule is used to identify intervention fatigue, inefficiency, and negative emotions based on feedback records, triggering an adjustment mechanism to automatically adjust the intensity, rhythm, and content of the intervention; The effectiveness evaluation and attribution module is used to periodically generate intervention effectiveness reports and analyze intervention response factors in combination with user portraits to support precise path optimization.

6. The integrated platform for rapid detection and targeted intervention of addiction susceptibility according to claim 1, characterized in that: The central control unit comprises: The data scheduling and management subunit is used to receive multi-source data collected by the addiction susceptibility detection module and push it to the data analysis and evaluation module for processing based on task priority and data type classification; The model calling and reasoning subunit is used to call the preset intervention strategy model and personalized task generation engine after receiving the risk results output by the assessment module, and output the initial intervention path and task parameters; The task distribution and execution monitoring subunit is used to distribute the task instructions generated by the intervention module to the user-side execution device, which includes a mobile terminal, wearable device, and virtual reality device, and monitor the execution status, response delay, and user interaction in real time; The dynamic adjustment control subunit is used to compare the original path with the task execution trajectory based on the feedback data returned by the feedback and adjustment module, call the strategy adjustment model and regenerate the intervention task plan; The permission and security management sub-unit is used to manage user permission verification, data access control and communication encryption mechanisms to ensure the security of platform operation and data privacy compliance.

Images