Smoking quitter skill learning and smoking intervention system

Abstract

The present application provides a kind of smoker smoking cessation skill learning and smoking cessation intervention system, belong to medical care information technology field, method is applied to mobile terminal, system is realized accurate smoking cessation intervention by multi-end cooperation.Wearing equipment end real-time acquisition user physiological index, sleep quality, motion data, psychological data and carbon monoxide concentration;Server is based on deep learning framework and constructs neural network prediction model, analyzes the collected data, predicts the probability of user smoking in future period;Mobile phone application end pushes hierarchical intervention content according to prediction result: when probability exceeds threshold, push visual smoking hazards science popularization;When probability is medium, push acupoint pressing etc. instant smoking control prompt;When probability is low, push interest diversion guide.The present application integrates multi-source data through AI model, constructs "data acquisition-risk prediction-hierarchical intervention" closed loop system, dynamically generates personalized smoking cessation strategy, and synchronously provides skill learning module and smoking cessation process management function.

Description

Technical Field

[0001] The present invention relates to the field of medical care information technology, and in particular to a system for learning and intervening in smoking cessation skills for smokers and a method for learning and intervening in smoking cessation skills for smokers. Background Art

[0002] Smoking is a major health hazard. Tobacco dependence is a chronic addictive disease, and quitting smoking is an extremely challenging process for smokers, requiring continuous motivation, effective supervision, and personalized guidance. Traditional smoking cessation methods have limited effectiveness, lack of real-time support, and difficulty in long-term adherence. For example, nicotine replacement therapy relies on external medication, which can easily lead to drug dependence and has wide individual differences in adaptability. Psychological counseling and intervention services are difficult to guarantee in terms of frequency and continuity, lacking a real-time dynamic adjustment mechanism. With the popularization of smartphones and the development of artificial intelligence technology, leveraging the powerful functions of smartphones and AI algorithms to assist in smoking cessation has become a new possibility.

[0003] Existing smartphones have relatively simple functions in assisting people to quit smoking. They rely solely on users to actively input smoking behavior data and lack objective monitoring methods. Most of them provide simple smoking cessation reminders or information push notifications, lacking comprehensive monitoring and personalized intervention of smokers' behavior, physiological and psychological states, making it difficult to provide accurate and effective smoking cessation support.

[0004] Therefore, how to achieve multi-dimensional data collection and improve the effectiveness of smoking cessation support has become a technical problem that needs to be solved urgently. Summary of the Invention

[0005] The present invention provides a system for smokers to learn smoking cessation skills and provide smoking cessation intervention, which is used to solve the defects of the existing technology in terms of single data collection dimension and insufficient accuracy of smoking cessation assistance intervention, realize multi-dimensional data collection, and improve the effectiveness of smoking cessation support.

[0006] The present invention provides a smoking cessation skill learning and smoking cessation intervention system for smokers, comprising:

[0007] The wearable device is used to collect the user's physiological data, sleep data, exercise data, psychological data and carbon monoxide concentration data; the physiological data includes blood pressure, blood oxygen and heart rate; the sleep data includes sleep duration, deep sleep time and light sleep time;

[0008] The server is configured to use a pre-trained LSTM neural network model to predict the probability of the user smoking within a preset future time period based on the user's historical smoking history, the user's real-time environmental data, and physiological data;

[0009] The mobile phone application is used to push multimedia content introducing the dangers of smoking, or push prompt information for controlling smoking urges, or push reminders of methods to divert smoking urges based on the smoking probability predicted by the server.

[0010] According to a smoking cessation skill learning and smoking cessation intervention system provided by the present invention, the mobile phone application terminal is also used for:

[0011] The user's location data and voice data are collected in real time.

[0012] According to a smoking cessation skills learning and smoking cessation intervention system provided by the present invention, the mobile phone application terminal includes a smoking hazard motivation intervention module, which is used to:

[0013] Display health indicators related to smoking and the impact of smoking on said health indicators;

[0014] Based on the user's physical examination data, health indicators with abnormal values ​​are marked.

[0015] According to a smoking cessation skill learning and smoking cessation intervention system provided by the present invention, the mobile application terminal further includes a smoking cessation process assistance module, which is used to:

[0016] A smoking log template is provided; the smoking log template includes smoking time, smoking location type, number of cigarettes smoked and smoking triggers.

[0017] Based on the smoking log template and the user's input, the user's smoking log is generated.

[0018] According to a smoking cessation skill learning and smoking cessation intervention system provided by the present invention, the mobile phone application terminal also includes a smoking cessation method guidance module, which is used to:

[0019] The guidance content on smoking cessation methods is presented in multimedia form.

[0020] According to a smoking cessation skills learning and smoking cessation intervention system provided by the present invention, the mobile application terminal further includes an incentive and social module, which is used to:

[0021] issuing dynamic rewards based on the user's smoking cessation behavior data;

[0022] Push user cases to the user whose smoking cessation success rate is greater than a preset threshold and who are in the same smoking cessation stage as the user.

[0023] The present invention also provides a method for smokers to learn smoking cessation skills and to intervene in smoking cessation, which uses any of the above-mentioned systems for smokers to learn smoking cessation skills and to intervene in smoking cessation, and the method for smokers to learn smoking cessation skills and to intervene in smoking cessation comprises the following steps:

[0024] The wearable device collects the user's physiological data, sleep data, exercise data, psychological data and carbon monoxide concentration data; the physiological data includes blood pressure, blood oxygen and heart rate; the sleep data includes sleep duration, deep sleep time and light sleep time;

[0025] The server uses a pre-trained LSTM neural network model to predict the probability of the user smoking within a preset time period in the future based on the user's historical smoking records, the user's real-time environmental data, and physiological data;

[0026] The mobile application pushes multimedia content introducing the dangers of smoking, or pushes prompt information for controlling smoking urges, or pushes reminders of methods for diverting smoking urges based on the smoking probability predicted by the server.

[0027] According to a method for learning and intervening in smoking cessation skills for smokers provided by the present invention, the method further comprises:

[0028] The server periodically generates a smoking cessation achievement report based on the user's historical smoking records and smoking cessation behavior data;

[0029] The server pushes the smoking cessation achievement report to the mobile phone application;

[0030] The mobile phone application stores and displays the smoking cessation achievement report.

[0031] According to a method for learning smoking cessation skills and intervening in smoking cessation provided by the present invention, the mobile application pushes multimedia content introducing the dangers of smoking, or pushes a no-smoking reminder, or pushes a reminder of a smoking cessation alternative activity based on the smoking probability predicted by the server, including:

[0032] Determining the type of intervention strategy, intervention intensity, and information push timing based on the smoking probability predicted by the server; the intervention strategy types include pushing multimedia content introducing the dangers of smoking, pushing prompts for controlling smoking cravings, and pushing reminders for methods to divert smoking cravings;

[0033] Based on the intervention strategy type and the intervention intensity, intervention is performed at the information push timing.

[0034] According to a method for learning and intervening in smoking cessation skills for smokers provided by the present invention, the method further comprises:

[0035] After receiving the user's response to the intervention, the mobile application feeds the response back to the server;

[0036] The server optimizes the LSTM neural network model based on the user's response.

[0037] The present invention provides a system for learning smoking cessation skills and providing intervention for smokers, which includes a wearable device end for collecting the user's physiological data, sleep data, exercise data, psychological data and carbon monoxide concentration data; the physiological data includes blood pressure, blood oxygen and heart rate; the sleep data includes sleep duration, deep sleep time and light sleep time; a server end for predicting the probability of the user smoking in a preset time period in the future using a pre-trained LSTM neural network model based on the user's historical smoking records, the user's real-time environmental data and physiological data; and a mobile phone application end for pushing multimedia content introducing the hazards of smoking, or pushing prompt information for controlling smoking cravings, or pushing reminders of methods for diverting smoking cravings based on the smoking probability predicted by the server end. The present invention collects multimodal data through wearable devices, integrates the user's physiological data, sleep data, exercise data, psychological data, and carbon monoxide concentration data, and provides a reliable data basis for smoking cessation assistance; the LSTM neural network model can capture long-distance dependencies and selectively retain the user's historical smoking cessation information. The LSTM neural network model accurately predicts the probability of the user smoking in the future, and then conducts precise auxiliary intervention in advance, thereby improving the success rate of smoking cessation and user compliance. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 This is a schematic diagram of the structure of the smoking cessation skill learning and smoking cessation intervention system provided by the present invention;

[0040] Figure 2 This is one of the flow charts of the method for learning smoking cessation skills and intervening in smoking cessation provided by the present invention;

[0041] Figure 3 This is the second flow chart of the method for smokers to learn smoking cessation skills and intervene in smoking cessation provided by the present invention;

[0042] Figure 4 It is a flow chart of the prediction model provided by the present invention. DETAILED DESCRIPTION

[0043] In order to make the objects, technical solutions and advantages of the present application clearer, the following will clearly and completely describe the technical solutions in the present application with reference to the drawings in the present application. Obviously, the described embodiments are only some, but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall into the scope of the present application.

[0044] It should be noted that in the description of the embodiments of the present application, the terms "comprise", "contain" or any other variants thereof are intended to cover the non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment. Without more limitation, the element defined by the statement "comprises a" does not exclude the presence of another identical element in the process, method, article or equipment comprising the element. The terms "upper", "lower" and the like indicate the orientation or positional relationship shown in the drawings, and are only used to facilitate the description of the present application and simplify the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. Unless otherwise specified and limited, the terms "mount", "connect", "connect" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be connected inside two elements. For those of ordinary skill in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0045] The terms "first", "second", and the like in the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are generally a class, and do not limit the number of objects, for example, the first object can be one or more. In addition, "and / or" means at least one of the connected objects, and the character " / ", generally means that the front and rear associated objects are in a "or" relationship.

[0046] Figure 1 is a structural schematic diagram of a smoker smoking cessation skill learning and smoking cessation intervention system provided by the present application, as shown in Figure 1 The smoker smoking cessation skill learning and smoking cessation intervention system comprises the following:

[0047] The wearable device is used to collect the user's physiological data, sleep data, exercise data, psychological data and carbon monoxide concentration data; the physiological data includes blood pressure, blood oxygen and heart rate; the sleep data includes sleep duration, deep sleep time and light sleep time;

[0048] The server is configured to use a pre-trained LSTM neural network model to predict the probability of the user smoking within a preset future time period based on the user's historical smoking history, the user's real-time environmental data, and physiological data;

[0049] The mobile phone application is used to push multimedia content introducing the dangers of smoking, or push prompt information for controlling smoking urges, or push reminders of methods to divert smoking urges based on the smoking probability predicted by the server.

[0050] It should be noted that Figure 1 A schematic diagram of an example architecture is shown. For descriptive purposes, the architecture depicted is only an example of a suitable environment and does not limit the scope of use or functionality of the present application. The system should not be interpreted as Figure 1 No dependencies or requirements are present on any one component or combination of components shown.

[0051] In order to further enrich the data collection dimensions, the mobile application terminal is optionally further used to:

[0052] The user's location data and voice data are collected in real time.

[0053] like Figure 1 As shown, the wearable device side can include but is not limited to watches, bracelets, glasses and other devices; the wearable device side uses sensors to collect the wearer's physiological data (including but not limited to blood pressure data, blood oxygen data, heart rate data), sleep data (including but not limited to sleep duration, deep sleep time, light sleep time), exercise data, psychological data (including but not limited to stress values), and carbon monoxide concentration data. At the same time, the mobile phone application side also includes a multi-source data acquisition module, which uses mobile phone sensors (including but not limited to cameras, GPS / Beidou, microphones, accelerometers) to collect data, including environmental data (including but not limited to time, activity venue type), voice data, and behavioral data (smoking action recognition). At the same time, it receives the collected data returned by the wearable device side through the interface connected to the wearable device. In addition, it can also receive data manually entered by the user, such as the user's tobacco craving intensity, mood log, smoking log, and other data.

[0054] The smoking cessation skills learning and smoking cessation intervention system provided by the embodiment of the present invention collects the user's exercise status, sleep status, and activity venue type, and connects with wearable devices to collect physiological data, sleep data, exercise data, psychological data, and carbon monoxide concentration data, etc., enriching the data dimensions and providing targeted smoking cessation services for smokers based on this data.

[0055] In the embodiment of the present invention, carbon monoxide concentration data is collected to calculate the nicotine concentration. Nicotine concentration includes environmental nicotine concentration and human nicotine concentration. The environmental nicotine concentration is the nicotine concentration in the air, which is obtained by inversely monitoring the carbon monoxide concentration in the air. Specifically:

[0056] Ambient nicotine concentration (ppm) = [CO]_{ppm} × K × MW_{nicotine} / MW_{CO} × CF;

[0057] Where [CO]_{ppm} represents the actual CO concentration (ppm) measured by the wearable device sensor; K represents the tobacco release coefficient (default 0.08, range 0.06-0.12); MW_{Nicotine} represents the molecular weight of nicotine (162.23 g / mol); MW_{CO} represents the molecular weight of CO (28 g / mol); and CF represents the environmental calibration factor (ventilated space = 0.3, confined space = 1.2).

[0058] Human nicotine concentration is the nicotine concentration in the human body, which is obtained by reverse calculation based on the carbon monoxide concentration exhaled by the human body.

[0059] During the specific implementation process, based on the preset detection frequency, the mobile phone application or wearable device will periodically remind the user to measure the exhaled carbon monoxide concentration. After the user confirms the carbon monoxide concentration measurement, the user is prompted to blow for 5-10 seconds and the blowing countdown is displayed. When the countdown ends, the blowing is completed, and the carbon monoxide concentration exhaled by the user is detected, and the nicotine concentration in the user's body is calculated.

[0060] When humans metabolize nicotine, the proportion of CO bound to hemoglobin is strongly correlated with nicotine intake: {exhaled CO (ppm)} ≈ 0.92 {Blood carboxyhemoglobin (COHb\%)}; The relationship between COHb% and nicotine dose is as follows: COHb\% = 0.16 Nicotine intake (mg); the nicotine concentration in the body (ppm) is calculated based on the following formula:

[0061] Nicotine concentration in the body (ppm) = (exhaled CO - ambient CO) ÷ (0.92 0.16) × α;

[0062] Where exhaled CO represents the average value of the user's final 10-second breath (with anti-interference processing); ambient CO represents the simultaneously measured air CO concentration (with background subtraction); and α represents the individual metabolic factor (default 1.0, smoker = 1.2-1.5).

[0063] In one embodiment, a test reminder is pushed one hour after waking up each morning to avoid the morning CO peak. Specifically, the wearable device vibrates and displays a test reminder, such as "Please blow into the sensor." After the user confirms the start of the blow, a countdown animation is displayed, such as a progress bar with a 10-second countdown and a bar chart showing real-time airflow intensity feedback. After successful data collection, the user is notified of the completion of data collection through flashing lights and vibrations, and the corresponding calculated nicotine concentration data is displayed. If data collection fails, the user is prompted that insufficient airflow is present and needs to repeat the blow test. Optionally, after the wearable device collects the user's nicotine concentration, it is uploaded to and stored by the mobile application, which then generates weekly / monthly nicotine concentration curves to motivate the user.

[0064] Figure 4 It is a flow chart of the prediction model provided by the present invention, such as Figure 4 As shown, in this embodiment of the present invention, the mobile application uploads its own collected data, as well as data collected by the wearable device, to the server for prediction and calculation. The server's smoking cessation AI analysis and prediction engine uses historical smoking records, real-time environmental data, and physiological data uploaded by the mobile application to predict the probability that the user will crave a cigarette in the future. For example, if real-time environmental data indicates that the user is located in a bar and physiological data indicates an elevated heart rate, the pre-trained LSTM neural network model predicts an 85% probability that the user will smoke in the next five minutes.

[0065] In one embodiment, the neural network prediction model includes an input layer, a hidden layer, and an output layer. The input layer receives ten feature data including heart rate (HR), systolic blood pressure (SBP), diastolic blood pressure (DBP), stress index (Stress), ambient nicotine concentration (AirNic), body nicotine concentration (BodyNic), emotion, scene, location, and time period. Among them, emotion, scene, location, and time period need to be encoded. The specific encoding method used is not limited here. For example, one-hot encoding can be used. The hidden layer includes at least two fully connected layers and uses the ReLU activation function. The output layer contains a neuron that uses the sigmoid activation function to output the probability of smoking between 0 and 1. Specifically:

[0066] The input features of the neural network prediction model include: heart rate x1, systolic blood pressure x2, diastolic blood pressure x3, stress index x4, ambient nicotine concentration x5, internal nicotine concentration x6, emotion x7 (categorical variable, M emotions encoded as M-dimensional vectors), scene x8 (categorical variable, P scenes encoded as P-dimensional vectors), location x9 (categorical variable, Q locations encoded as Q-dimensional vectors), and time period x10 (categorical variable, R time periods encoded as R-dimensional variables). If one-hot encoding is used, the actual input vector includes the continuous part features X = [x1, x2, x3, x4, x5, x6] and the classification part encoding vector X_class = [x7_onehot,x8_onehot, x9_onehot, x10_onehot]. The total input vector X_total = [X, X_class] has a dimension of D = 6 + M + P + Q + R. That is, the input layer contains D neurons.

[0067] Hidden layer 1 contains H1 neurons (e.g. 128) and uses the ReLU activation function: Z1 = W1 X_total +b1, A1 = ReLU(Z1) = max(0, Z1);

[0068] Hidden layer 2 contains H2 neurons (e.g. 64) and uses the ReLU activation function: Z2 = W2 A1 + b2, A2 = ReLU(Z2);

[0069] The output layer contains 1 neuron, and uses the sigmoid activation function to map the prediction result to (0,1): Z3 = W3 A2 + b3, smoking probability Y_hat = sigmoid(Z3) = 1 / (1 + exp(-Z3)).

[0070] Optionally, the neural network prediction model uses binary cross entropy as the loss function:

[0071] L = - [ Y log(Y_hat) + (1-Y) log(1-Y_hat) ];

[0072] Among them, Y is the true label, that is, the actual smoking probability value. In the specific implementation process, the user's smoking probability is normalized to a value between 0 and 1 through real-time user reports or experimental settings.

[0073] Optionally, the Adam optimizer is used to accelerate the model learning process.

[0074] Optionally, the smoking cessation AI analysis and prediction engine is also used to identify smoking cessation stages. Specifically, using a pre-trained SVM classifier, based on seven consecutive days of user behavioral data, it identifies the user's smoking cessation stage, such as the preparation stage (preparing to quit smoking), the action stage (implementing smoking cessation behaviors), or the maintenance stage (the stage of successful smoking cessation and preventing relapse).

[0075] Optionally, the smoking cessation AI analysis and prediction engine is also used to generate smoking cessation evaluations. Specifically, it uses AI algorithms such as time series analysis and prediction, natural language processing, classification, and clustering to evaluate smoking cessation achievements for the day, week, and month.

[0076] Optionally, the smoking cessation AI analysis and prediction engine can also be used to provide smoking cessation guidance. Specifically, it uses deep learning algorithms to provide reasonable smoking cessation guidance.

[0077] Optionally, the smoking cessation AI analysis and prediction engine is also used for smoking cessation optimization. Specifically, it uses deep learning algorithms to evaluate and improve smoking cessation deficiencies.

[0078] In the embodiment of the present invention, multimedia content introducing the dangers of smoking, or prompting information on controlling smoking urges, or reminders on methods to divert smoking urges are pushed to the user based on the smoking probability predicted by the server.

[0079] During implementation, intervention strategies are determined based on preset thresholds. Specifically, if the predicted probability is greater than a first threshold, multimedia content explaining the dangers of smoking, such as a lung pathology animation, is pushed to the user. If the predicted probability is less than the first threshold but greater than a second threshold, prompts for controlling smoking cravings are pushed to the user, such as mindful breathing and acupressure techniques. If the predicted probability is less than the second threshold, prompts for methods to divert smoking cravings are pushed to the user, such as exercise challenges, meditation courses, and other interest-based courses.

[0080] The present invention provides a smoking cessation skills learning and intervention system for smokers, including a wearable device for collecting a user's physiological data, sleep data, exercise data, psychological data, and carbon monoxide concentration data; the physiological data includes blood pressure, blood oxygen, and heart rate; the sleep data includes sleep duration, deep sleep time, and light sleep time; a server for predicting the user's probability of smoking within a preset future time period using a pretrained LSTM neural network model based on the user's historical smoking history, the user's real-time environmental data, and physiological data; and a mobile phone application for pushing multimedia content introducing the dangers of smoking, or pushing information on controlling smoking cravings, or pushing reminders on methods for diverting smoking cravings, based on the smoking probability predicted by the server. The present invention collects multimodal data through a wearable device and integrates the user's physiological data, sleep data, exercise data, psychological data, and nicotine concentration data to provide a reliable data foundation for smoking cessation assistance. The LSTM neural network model can capture long-range dependencies and selectively retain the user's historical smoking cessation information. The LSTM neural network model accurately predicts the user's probability of smoking within a future time period, thereby providing precise assistance intervention in advance, improving the success rate of smoking cessation and user compliance.

[0081] In an optional embodiment, the mobile phone application terminal includes a smoking hazard motivation intervention module, and the smoking hazard motivation intervention module is used to:

[0082] Display health indicators related to smoking and the impact of smoking on said health indicators;

[0083] Based on the user's physical examination data, health indicators with abnormal values ​​are marked.

[0084] In an embodiment of the present invention, after the user completes the physical examination, the smoking hazard motivation intervention module updates the physical examination data in real time, displays health indicators related to smoking, and explains the impact of smoking on the corresponding health indicators. It marks out health indicators with abnormal values ​​such as blood pressure, blood oxygen, blood lipids, heart, and endocrine in the physical examination. Upward or downward arrows can be used to indicate that the health indicators are higher or lower than the normal value, and abnormal indicators can be marked by highlighting / different colors.

[0085] Optionally, the smoking hazard motivation intervention module is further configured to push content related to the hazards of smoking. Specifically, the module utilizes "No Smoking Day," lectures by authoritative experts, real stories, successful quitting cases, and other means to spread the message about the hazards of smoking.

[0086] The smoking cessation skills learning and smoking cessation intervention system provided in the embodiment of the present invention reminds users of the impact of smoking on health through the smoking hazard motivation intervention module, thereby further encouraging users to quit smoking and providing users with accurate and effective smoking cessation support.

[0087] In an optional embodiment, the mobile phone application further includes a smoking cessation process assistance module, which is used to:

[0088] A smoking log template is provided; the smoking log template includes smoking time, smoking location type, number of cigarettes smoked and smoking triggers.

[0089] Based on the smoking log template and the user's input, the user's smoking log is generated.

[0090] In this embodiment of the present invention, the smoking cessation assistance module provides a smoking log template. Users can fill in each smoking record based on the smoking log template, thereby providing data for subsequent prediction of the user's smoking probability. Specifically, according to the standardized smoking log template, daily smoking records are filled in, including but not limited to smoking time, type of venue, number of cigarettes smoked, degree of craving before smoking, smoking triggers, etc., to generate a smoking log record.

[0091] Optionally, the smoking cessation process auxiliary module is further used to record the user's smoking cessation cycle, specifically, to record information such as the user's smoking cessation start time, smoking cessation cycle, and smoking cessation plan.

[0092] Optionally, the smoking cessation assistance module is also used to assist users in establishing a favorable smoking cessation environment. For example, it may prompt users to stay away from smoking environments, such as public smoking areas, restaurants, bars, KTVs, and groups of people smoking; and it may prompt users to remove smoking items, such as ashtrays, lighters, and cigarette cases.

[0093] Optionally, the smoking cessation assistance module can also be used to collaborate with other users to supervise smokers who have quit. Specifically, based on the designated smoking cessation supervisor and effective supervision strategy, the smoker's family and friends can access the smoker's smoking history, health indicators, physiological data, psychological data, etc. through linked accounts, thereby assisting in supervision and psychological counseling of the smoker.

[0094] Optionally, the smoking cessation process assistance module is further configured to prompt the user to learn smoking cessation techniques. Specifically, based on a set frequency and time, the module prompts and assists the user in learning smoking cessation techniques.

[0095] Optionally, the smoking cessation process assistance module is also used to prepare alternative activities. For example, through mobile phone prompts and reminders, the user is prompted to take alternative activities (such as drinking ice water or taking deep breaths) to keep the brain busy, distract and reduce the urge to smoke.

[0096] The smoking cessation skill learning and smoking cessation intervention system provided by the embodiment of the present invention assists in generating a smoking log through a smoking cessation process auxiliary module, thereby providing a data basis for subsequent prediction of smoking probability.

[0097] In an optional embodiment, the mobile phone application further includes a smoking cessation guidance module, which is used to:

[0098] The guidance content on smoking cessation methods is presented in multimedia form.

[0099] Specifically, the content presents smoking cessation guidance from authoritative organizations and experts in the form of short videos, animations, audio, images, and text. This includes tips on avoiding public smoking environments and individuals, changing life and work habits and situations that are closely tied to smoking, how to escape temporary smoking situations, delaying the urge to smoke, and unbinding the urge to smoke.

[0100] Optionally, the mobile app also includes a smoking cessation support resource module, which provides smoking cessation support resources and filters out content that triggers smoking cravings based on the smoker's preferences. The server's smoking cessation AI analysis and prediction engine intelligently recommends suitable, high-quality content, such as three-minute breathing exercises, three-minute Ba Duan Jin (Eight-Section Brocade), three-minute Tai Chi, music, crosstalk, and talk shows. Suitable content is then redirected to the corresponding app for display.

[0101] The smoking cessation skill learning and smoking cessation intervention system provided by the embodiment of the present invention guides users on specific smoking cessation methods through a smoking cessation method guidance module, thereby providing users with rich and scientific smoking cessation method guidance.

[0102] In an optional embodiment, the mobile application terminal further includes an incentive and social module, and the incentive and social module is used to:

[0103] issuing dynamic rewards based on the user's smoking cessation behavior data;

[0104] Push user cases to the user whose smoking cessation success rate is greater than a preset threshold and who are in the same smoking cessation stage as the user.

[0105] In an embodiment of the present invention, based on the user's smoking cessation behavior data, the user's smoking cessation milestones, such as 24 hours of smoke-free, are determined, and dynamic rewards, such as virtual president, points redemption, etc., are then issued.

[0106] In the embodiment of the present invention, based on similar smoking cessation stages / demographic characteristics, user cases with high success rates are pushed to motivate users to quit smoking.

[0107] The smoking cessation skill learning and smoking cessation intervention system provided in the embodiment of the present invention motivates users to quit smoking by issuing dynamic rewards and pushing high success rate user cases, thereby further improving the user's smoking cessation success rate.

[0108] In an optional embodiment, the mobile application terminal further includes a dynamic real-time smoking cessation intelligent intervention module, including:

[0109] Real-time intervention library: including but not limited to deep breathing guided videos, cognitive behavioral therapy (CBT) micro-courses, alternative behavior suggestions (such as drinking ice water), emergency contact calls, etc.;

[0110] Intervention methods: Provide targeted smoking cessation measures to smokers in real time based on their location, physical and mental condition through mobile phones or wearable devices; make full use of holographic means (sound, image, video, vibration, information interaction, etc.) of mobile phones or wearable devices (watches, bracelets, glasses) to intervene and provide guidance;

[0111] Strategy matching rules: If the probability of smoking is greater than a preset threshold (such as 80%), an instant stress-relieving game will be pushed; if it is detected that the user is at a dinner party, a prompt message to control the urge to smoke will pop up on the mobile phone or wearable device; if it is identified that the user is in the "action period" and the user's hand-held smoking action is detected, the mobile phone camera will be activated to guide the behavior interruption.

[0112] Figure 3 This is the second flow chart of the method for smokers to learn smoking cessation skills and to intervene in smoking cessation provided by the present invention. Figure 3 As shown, in an optional embodiment, the mobile app also includes a smoking cessation behavior learning and evaluation module, which uses LSTM or Transformer models to analyze user behavior data and improve AI smoking cessation measures. Specifically, it conducts periodic smoking cessation behavior assessments, such as hourly assessments; extracts and generates appropriate smoking cessation measures for smokers; and updates smoking cessation measures based on smoking cessation results to address imperfections in smoking cessation plans.

[0113] In an optional embodiment, the server also includes a smoking cessation knowledge base for storing knowledge content related to smoking cessation, classifying and labeling it, regularly updating the content of the smoking cessation knowledge base, incorporating the latest research results, eliminating outdated methods, and continuously improving the knowledge base in iterations. Specifically, the knowledge base content includes basic cognitive knowledge, such as popular science on the hazards of smoking and scientific principles of smoking cessation; practical method knowledge, such as personalized smoking cessation plans, behavioral intervention techniques, and relapse prevention strategies; resource support knowledge, medical resource navigation, psychological support channels, auxiliary content, and tool recommendations. The knowledge base content also includes knowledge base classification, which classifies and labels the smoking cessation knowledge base according to male smoking cessation, female smoking cessation, young and middle-aged smoking cessation, elderly smoking cessation, and youth prevention.

[0114] The following describes the smoker quitting smoking skills learning and smoking cessation intervention method provided in the embodiment of the present application. The smoker quitting smoking skills learning and smoking cessation intervention method described below is applied to any of the smoker quitting smoking skills learning and smoking cessation intervention systems described above.

[0115] Figure 2is one of the process schematic diagram of the smoking cessation skill learning and smoking cessation intervention method provided by the present application, as shown in Figure 2 The smoking cessation skill learning and smoking cessation intervention method can include but is not limited to:

[0116] S210, the wearable device end collects physiological data, sleep data, motion data, psychological data and carbon monoxide concentration data of the user; the physiological data includes blood pressure, blood oxygen and heart rate; the sleep data includes sleep duration, deep sleep time and light sleep time;

[0117] S220, the server predicts the probability of the user smoking in a future preset period based on the historical smoking record of the user, real-time environmental data and physiological data of the user, and uses a pre-trained LSTM neural network model;

[0118] S230, the mobile phone application end pushes multimedia content introducing the harm of smoking, or pushes control smoking desire prompt information, or pushes method reminder for diverting smoking desire based on the smoking probability predicted by the server.

[0119] The smoking cessation skill learning and smoking cessation intervention method provided by the embodiment of the present application, the wearable device end collects physiological data, sleep data, motion data, psychological data and carbon monoxide concentration data of the user; the physiological data includes blood pressure, blood oxygen and heart rate; the sleep data includes sleep duration, deep sleep time and light sleep time; the server predicts the probability of the user smoking in a future preset period based on the historical smoking record of the user, real-time environmental data and physiological data of the user, and uses a pre-trained LSTM neural network model; the mobile phone application end pushes multimedia content introducing the harm of smoking, or pushes control smoking desire prompt information, or pushes method reminder for diverting smoking desire based on the smoking probability predicted by the server. The embodiment of the present application collects multi-modal data through the wearable device, integrates physiological data, sleep data, motion data, psychological data and nicotine concentration data of the user, provides a reliable data basis for smoking cessation assistance; the LSTM neural network model can capture long-distance dependence relationship, selectively retain historical smoking cessation information of the user, accurately predict the probability of the user smoking in a future period of time through the LSTM neural network model, and then perform accurate auxiliary intervention in advance, improve the success rate of smoking cessation and user compliance.

[0120] In an optional embodiment, the smoking cessation skill learning and smoking cessation intervention method further includes:

[0121] The server generates a smoking cessation achievement report based on the historical smoking record and smoking behavior data of the user;

[0122] The server pushes the smoking cessation achievement report to the mobile phone application end;

[0123] The mobile phone application stores and displays the smoking cessation achievement report.

[0124] In an embodiment of the present invention, the AI ​​analysis and prediction engine on the server performs AI calculations regularly. For example, AI calculations are performed at midnight every day to evaluate the smoking cessation achievements of each smoker who quit the previous day. The server pushes the smoking cessation achievements to the mobile phone application system, and the mobile phone / wearable device presents a smoking cessation achievement report. For example, "Congratulations! Today is your third No Smoking Day! You reported 5 strong urges today, but successfully dealt with 4 of them (success rate 80%), which is an improvement over yesterday (success rate 60%)! Stress increases around 3 pm, so it is recommended that you try deep breathing exercises at this time tomorrow. Keep it up!"

[0125] The smoking cessation skills learning and smoking cessation intervention method provided by the embodiment of the present invention regularly generates smoking cessation achievement reports based on the user's smoking log records and behavior data, intuitively reflecting the user's smoking cessation achievements and effectively promoting the user to quit smoking.

[0126] In an optional embodiment, the mobile application pushes multimedia content introducing the dangers of smoking, or pushes prompt information for controlling smoking urges, or pushes reminders of methods for diverting smoking urges based on the smoking probability predicted by the server, including:

[0127] Determining the type of intervention strategy, intervention intensity, and information push timing based on the smoking probability predicted by the server; the intervention strategy types include pushing multimedia content introducing the dangers of smoking, pushing prompts for controlling smoking cravings, and pushing reminders for methods to divert smoking cravings;

[0128] Based on the intervention strategy type and the intervention intensity, intervention is performed at the information push timing.

[0129] The smoking cessation skill learning and smoking cessation intervention method provided by the embodiment of the present invention adopts different intervention strategies according to different smoking probabilities, thereby performing precise auxiliary intervention in advance and improving the success rate of smoking cessation and user compliance.

[0130] In an optional embodiment, the method for smokers to learn smoking cessation skills and intervene in smoking cessation further includes:

[0131] After receiving the user's response to the intervention, the mobile application feeds the response back to the server;

[0132] The server optimizes the LSTM neural network model based on the user's response.

[0133] In an embodiment of the present invention, the LSTM neural network model is optimized based on the user's response to the intervention strategy (such as ignoring or completing), thereby continuously optimizing the prediction accuracy.

[0134] For example, if a user enters a bar using their phone's GPS / Beidou positioning and detects the sound of a lighter through their phone's microphone, the server, using historical data, calculates that the user's probability of craving a cigarette has risen to 85%, exceeding a set threshold. It then immediately pushes a smoking cessation alternative activity reminder, such as a 15-second deep breathing animation, and retrieves information from the knowledge base about the health benefits of successfully quitting smoking. If the user chooses to perform the deep breathing action, the user's effective response is recorded and awarded corresponding points.

[0135] In an optional embodiment, the server can also issue risk warnings based on the user's historical physiological data, allowing for early intervention. For example, if it detects that a user's sleep quality has deteriorated for three consecutive days and the user's work schedule is marked as "project deadline," the AI ​​will generate a high-risk warning, push a "stress management CBT course" in advance, and recommend contacting a smoking cessation counselor.

[0136] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0137] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A smoking cessation skills learning and smoking cessation intervention system for smokers, characterized in that: include: Wearable device side is used to collect users' physiological data, sleep data, exercise data, psychological data and carbon monoxide concentration data; The physiological data includes blood pressure, blood oxygen and heart rate; the sleep data includes sleep duration, deep sleep time and light sleep time; The server is used to predict the probability of the user smoking within a preset time period in the future using a pre-trained LSTM neural network model based on the user's historical smoking records, the user's real-time environmental data and physiological data; the pre-trained LSTM neural network model includes an input layer, two hidden layers and an output layer, and the feature data received by the input layer include ten feature data: heart rate, systolic blood pressure, diastolic blood pressure, stress index, ambient nicotine concentration, body nicotine concentration, emotion, scene, location and time period; the first hidden layer of the pre-trained LSTM neural network model contains H1 neurons and uses the ReLU activation function: Z1=W1 X_total+b1, A1=ReLU(Z1), where A1 is the output of the first hidden layer, ReLU() is the ReLU activation function, and X_total is the total input vector; the second hidden layer of the pre-trained LSTM neural network model contains H2 neurons and uses the ReLU activation function: Z2=W2 A1+b2, A2=ReLU(Z2), where A2 is the output of the first hidden layer; the output layer of the pre-trained LSTM neural network model contains 1 neuron, and the sigmoid activation function is used to map the prediction result to (0,1): Z3=W3 A2+b3, smoking probability Y_hat=sigmoid(Z3)=1 / (1+exp(-Z3)); the loss function L of the pre-trained LSTM neural network model is as follows: L=-[Y log(Y_hat)+(1-Y) log(1-Y_hat)], where Y is the true smoking probability value; The mobile phone application is used to push multimedia content introducing the dangers of smoking, or push reminders on controlling smoking urges, or push reminders on methods to divert smoking urges, based on the smoking probability predicted by the server; The wearable device is further configured to calculate a nicotine concentration based on the carbon monoxide concentration data; the nicotine concentration includes an ambient nicotine concentration and a human nicotine concentration; the ambient nicotine concentration is the nicotine concentration in the air, which is calculated based on the carbon monoxide concentration in the air; the human nicotine concentration is the nicotine concentration in the human body, which is calculated based on the carbon monoxide concentration exhaled by the user; Ambient nicotine concentration = [CO]_{ppm}×K×MW_{nicotine} / MW_{CO}×CF; Where [CO]_{ppm} represents the CO concentration measured by the wearable device sensor; K represents the tobacco release coefficient; MW_{nicotine} represents the molecular weight of nicotine; MW_{CO} represents the molecular weight of CO; CF represents the environmental calibration factor, with CF = 0.3 for ventilated space and CF = 1.2 for confined space. Nicotine concentration in the body = (exhaled CO2-ambient CO2) ÷ (0.92 0.16) × α; Among them, exhaled CO represents the average value of the user's 10-second exhalation; ambient CO represents the CO concentration of the air measured simultaneously; α represents the individual metabolic factor; The mobile phone application is also used to receive and store the nicotine concentration uploaded by the wearable device, and then generate a nicotine concentration curve.

2. The smoking cessation skills learning and smoking cessation intervention system for smokers according to claim 1, characterized in that: The mobile application terminal is also used for: The user's location data and voice data are collected in real time.

3. The smoking cessation skills learning and smoking cessation intervention system for smokers according to claim 1, characterized in that: The mobile phone application terminal includes a smoking hazard motivation intervention module, which is used to: Display health indicators related to smoking and the impact of smoking on said health indicators; Based on the user's physical examination data, health indicators with abnormal values ​​are marked.

4. The smoking cessation skills learning and smoking cessation intervention system for smokers according to claim 1, characterized in that: The mobile application terminal also includes a smoking cessation process auxiliary module, which is used to: Providing a smoking log template; the smoking log template includes smoking time, type of smoking location, number of cigarettes smoked and smoking triggers; Based on the smoking log template and the user's input, the user's smoking log is generated.

5. The smoking cessation skills learning and smoking cessation intervention system for smokers according to claim 1, characterized in that: The mobile phone application terminal also includes a smoking cessation method guidance module, which is used to: The guidance content on smoking cessation methods is presented in multimedia form.

6. The smoking cessation skills learning and smoking cessation intervention system for smokers according to claim 1, characterized in that: The mobile application terminal also includes an incentive and social module, which is used to: issuing dynamic rewards based on the user's smoking cessation behavior data; Push user cases to the user whose smoking cessation success rate is greater than a preset threshold and who are in the same smoking cessation stage as the user.

7. A method for smokers to learn smoking cessation skills and to intervene in smoking cessation, characterized in that: The system for learning and intervening in smoking cessation skills for smokers according to claim 1, wherein the method for learning and intervening in smoking cessation skills for smokers comprises: The wearable device collects the user's physiological data, sleep data, exercise data, psychological data and carbon monoxide concentration data; the physiological data includes blood pressure, blood oxygen and heart rate; the sleep data includes sleep duration, deep sleep time and light sleep time; The server uses a pre-trained LSTM neural network model to predict the probability of the user smoking within a preset time period in the future based on the user's historical smoking records, the user's real-time environmental data, and physiological data; The mobile application pushes multimedia content introducing the dangers of smoking, or pushes prompt information for controlling smoking urges, or pushes reminders of methods for diverting smoking urges based on the smoking probability predicted by the server.

8. The method for learning smoking cessation skills and intervening in smoking cessation according to claim 7, characterized in that: The smoker's smoking cessation skills learning and smoking cessation intervention method also includes: The server periodically generates a smoking cessation achievement report based on the user's historical smoking records and smoking cessation behavior data; The server pushes the smoking cessation achievement report to the mobile phone application; The mobile phone application stores and displays the smoking cessation achievement report.

9. The method for learning smoking cessation skills and intervening in smoking cessation according to claim 7, characterized in that: The mobile application pushes multimedia content introducing the dangers of smoking, or pushes prompt information for controlling smoking urges, or pushes reminders of methods for diverting smoking urges based on the smoking probability predicted by the server, including: Determining the type of intervention strategy, intervention intensity, and information push timing based on the smoking probability predicted by the server; the intervention strategy types include pushing multimedia content introducing the dangers of smoking, pushing prompts for controlling smoking cravings, and pushing reminders for methods to divert smoking cravings; Based on the intervention strategy type and the intervention intensity, intervention is performed at the information push timing.

10. The method for learning smoking cessation skills and intervening in smoking cessation according to claim 9, characterized in that: The smoker's smoking cessation skills learning and smoking cessation intervention method also includes: After receiving the user's response to the intervention, the mobile application feeds the response back to the server; The server optimizes the LSTM neural network model based on the user's response.

Images