Virtual platform-based lung cancer patient psychological health assessment intervention system

Abstract

The invention discloses a lung cancer patient psychological health assessment intervention system based on a virtual platform, and belongs to the field of psychological health. A lung cancer patient psychological health assessment intervention system based on a virtual platform comprises a virtual environment creating module, a psychological state monitoring module, a psychological health assessment module and a psychological health intervention module. The problem that an existing mental health assessment and intervention mode is limited is solved, an immersive environment is created through the virtual reality technology, a real lung cancer patient treatment scene is restored, a basis is provided for accurate assessment, psychological state assessment is more accurate by comprehensively obtaining multi-dimensional data such as physiology, voice and actions of a patient in real time, and the accuracy of assessment is improved. The emotion type and the psychological problem severity are quickly and accurately recognized through the constructed psychological health assessment model, a scientific basis is provided for subsequent intervention, a personalized scheme is formulated according to an assessment result, and the psychological stress of a patient can be effectively relieved by combining virtual role interaction.

Description

Technical Field

[0001] The present invention relates to the field of mental health technology, and in particular to a mental health assessment and intervention system for lung cancer patients based on a virtual platform. Background Art

[0002] Lung cancer is one of the most morbid and fatal malignancies worldwide. Patients often face immense psychological stress after diagnosis. From fear and anxiety during diagnosis, to physical pain and concerns about treatment effectiveness during treatment, to fear of recurrence and frustration with a decline in quality of life during recovery, these psychological burdens persist throughout the disease course. Studies have shown that a high proportion of lung cancer patients suffer from psychological disorders, such as depression and anxiety, which severely impact treatment compliance, quality of life, and survival outcomes. Currently, the assessment of the mental health status of lung cancer patients relies primarily on psychological scales and clinicians' subjective judgment. While these scales have some scientific validity, they are susceptible to subjective factors (such as the patient's current emotional state and misunderstanding of the problem), which limits the accuracy of the assessment results. Therefore, these scales do not meet existing needs. Therefore, we propose a virtual platform-based mental health assessment and intervention system for lung cancer patients. Summary of the Invention

[0003] The purpose of the present invention is to provide a psychological health assessment and intervention system for lung cancer patients based on a virtual platform. By creating an immersive virtual environment through virtual reality technology and combining a variety of advanced monitoring technologies and machine learning algorithms, it can achieve real-time, dynamic monitoring and accurate assessment of the psychological state of lung cancer patients, and provide personalized psychological intervention plans, thereby solving the problems raised in the above-mentioned background technology.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a virtual platform-based mental health assessment and intervention system for lung cancer patients, the system comprising: a virtual environment creation module configured to use virtual reality technology to create an immersive virtual environment to simulate the hospital environment, surgical scenes, and rehabilitation scenes of lung cancer patients during treatment, thereby inducing real emotional responses from patients; The psychological state monitoring module is configured to integrate multiple biosensors to monitor the patient's physiological response data in real time in a virtual environment. At the same time, it combines voice recognition technology and motion capture technology to analyze the patient's voice tone and speech speed changes, as well as record the patient's body movements and facial expressions to obtain the patient's comprehensive psychological state data; Among them, denoising and feature extraction are performed on physiological response data, speech enhancement and speech feature extraction are performed on speech data, and data smoothing, coordinate transformation and motion feature extraction are performed on motion data; During speech enhancement, the speech frame is evaluated from multiple dimensions, including envelope and signal strength, to comprehensively assess the noise level in the speech frame. The abnormality level of the time-frequency unit is then determined to locate the time-frequency unit that requires noise reduction, achieving targeted noise reduction. A mental health assessment module is configured to establish a mental health assessment model based on a machine learning algorithm, automatically identify the patient's emotional type and the severity of psychological problems, and generate a mental health assessment report; The mental health intervention module is configured to develop personalized psychological intervention plans based on mental health assessment reports and patient information, and evaluate patient behavior changes by comparing psychological scale scores before and after intervention. The psychological intervention plan includes psychological counseling dialogues, relaxation training and cognitive behavioral therapy in a virtual environment, and interacts with patients through virtual characters to guide patients to adjust their mindsets and relieve psychological stress.

[0005] Furthermore, the mental state monitoring module includes: The data acquisition module is configured to uniformly access and manage multiple biosensors, collect data on the patient's physiological responses in the virtual environment, collect voice data using speech recognition technology, and analyze changes in the patient's voice intonation and speech speed, and collect motion data using motion capture technology to record changes in the patient's body movements and facial expressions; Among them, the biosensors include heart rate sensors, skin conductance sensors and brain wave sensors. At the same time, corresponding sampling frequencies are set according to different biosensors, and the sampling times of all biosensors are synchronized; a data processing module configured to perform denoising and feature extraction on the collected physiological response data, perform speech enhancement and speech feature extraction on the collected speech data, and perform data smoothing, coordinate conversion, and motion feature extraction on the collected motion data; The data fusion module is configured to fuse the processed physiological response data, voice data and action data using a weighted average method and based on the importance and relevance of different data to obtain comprehensive psychological state data.

[0006] Furthermore, the data processing module is specifically: For physiological response data: Denoising: Each physiological response data corresponds to a parameter table, including data type, noise reduction algorithm, parameter name and parameter value. According to the collected physiological response data type, the corresponding noise reduction algorithm and its parameters are queried from the stored parameter table. Based on the queried noise reduction algorithm and parameters, the corresponding function in the algorithm library is called to perform noise reduction on the physiological response data; Feature extraction: Extract heart rate and heart rate variability features through standard deviation and root mean square deviation, extract skin conductance level and skin galvanic response features, perform spectrum analysis, and extract power spectral density in different frequency bands; For voice data: Speech enhancement: Through noise suppression algorithms, the impact of background noise on speech data is reduced, improving speech clarity; Speech feature extraction: extract intonation features and speech features and normalize them to the same range; For action data: Data smoothing: Smoothing motion data to remove noise and jitter through moving average filtering or Kalman filtering. Coordinate transformation: transforming motion data from the original coordinate system to a unified virtual environment coordinate system to ensure data consistency; Motion feature extraction: extract the amplitude, frequency, speed and acceleration features of limb movements, extract the features of body posture, and extract facial expression features.

[0007] Furthermore, the speech enhancement is specifically: A framing module is configured to use a speech signal in the collected speech data as a speech signal to be enhanced, perform frame processing on the speech signal to be enhanced, and obtain a plurality of speech frames; and sort the plurality of speech frames based on a time series to obtain a speech frame sequence; Randomly select a speech frame as the first speech frame; a determination module, configured to obtain a first evaluation index value and a second evaluation index value of a first speech frame, and determine a noise evaluation value of the first speech frame based on the first evaluation index value and the second evaluation index value of the first speech frame; a comparison module, configured to compare the noise evaluation value of the first speech frame with a preset noise evaluation threshold, and when determining that the noise evaluation value of the first speech frame is greater than or equal to the preset noise evaluation threshold, use the first speech frame as a target noise reduction frame; Performing a short-time Fourier transform on the target noise reduction frame to determine a sound spectrogram corresponding to the target noise reduction frame; obtaining a time-frequency unit of the sound spectrogram to obtain a plurality of time-frequency units; Anomaly identification module, used to: Obtain the energy value of each time-frequency unit within the time-frequency range and determine the energy distribution of each time-frequency unit; Take any time-frequency unit as the first time-frequency unit; Determine the target area with the first time-frequency unit as the center; calculate the standard deviation of the energy value in the target area as the standard deviation of the energy distribution corresponding to the first time-frequency unit; Traverse all time-frequency units and determine the standard deviation of energy distribution corresponding to each time-frequency unit; Randomly select a time-frequency unit in the target area except the first time-frequency unit as the second time-frequency unit; Calculating a ratio between a standard deviation of energy distribution of a first time-frequency unit and a standard deviation of energy distribution corresponding to a second time-frequency unit in the target area to obtain a first ratio; Calculating the Euclidean distance between the first time-frequency unit and the second time-frequency unit; Calculating the product of the first ratio and the Euclidean distance, and using the product as the difference degree value between the first time-frequency unit and the second time-frequency unit; Traversing other time-frequency units except the first time-frequency unit in the target area, and obtaining a difference degree value between the first time-frequency unit and each other time-frequency unit; Sum and average the difference values ​​between the first time-frequency unit and each other time-frequency unit to obtain the abnormality value of the first time-frequency unit; Traverse all time-frequency units and determine the abnormality value of each time-frequency unit; Enhancement modules for: Comparing the abnormality level of each time-frequency unit with a preset abnormality level threshold, taking the time-frequency unit corresponding to the abnormality level greater than or equal to the preset abnormality level threshold as a noise time-frequency unit, and obtaining a plurality of noise time-frequency units; performing noise reduction and suppression on the plurality of noise time-frequency units based on a noise reduction algorithm; Traverse all speech frames to complete the enhancement of the speech signal to be enhanced.

[0008] Furthermore, analyze the patient's voice tone and speech speed changes, and record the patient's body movements and facial expressions, specifically: Intonation analysis: By comparing with the normal range of intonation, the abnormality of the patient's intonation changes can be determined; Speech rate change analysis: Analyze the changes in speech rate and judge the patient's emotional state based on the absolute value and change trend of speech rate; Body movement analysis: Analyze the patient's body movement characteristics based on the amplitude, frequency, and speed of the body movements. By analyzing the pattern of body movements, the patient's behavioral intention in the virtual environment can be determined. Expression analysis: Based on the expression recognition results, the duration and frequency of the patient's expressions in different expression categories are counted.

[0009] Furthermore, the mental health assessment module includes: A data collection module is configured to collect a large amount of historical data as training samples and label the training samples with emotion types and severity of psychological problems; The model building module is configured to combine the random forest algorithm and the linear regression algorithm to construct a hybrid mental health assessment model, and the labeled training samples are used to train and optimize the constructed mental health assessment model; The report generation module is configured to input the comprehensive psychological state data into the trained mental health assessment model, output the patient's emotional type and the severity of psychological problems through the mental health assessment model, generate a corresponding mental health assessment report based on the output results and display it visually.

[0010] Furthermore, the mental health intervention module includes: The report analysis module is configured to interpret the mental health assessment report, extract the patient's anxiety level, depression level and stress sources, and collect and organize the patient's information to understand the background of the patient's psychological problems. The psychological status information in the mental health assessment report is combined with the patient's basic information to comprehensively analyze the causes and development trends of the patient's psychological problems; A program development module is configured to clarify the main goal of the intervention based on the comprehensive analysis results, select an appropriate method combination from the three main methods of psychological counseling dialogue, relaxation training, and cognitive behavioral therapy in a virtual environment based on the main intervention goal, and then customize the specific content according to the selected psychological intervention program. For psychological counseling dialogue, the dialogue script is written according to the patient's problem situation; for relaxation training, relaxation techniques suitable for the patient are selected; and for cognitive behavioral therapy, the type of cognitive distortion that needs to be corrected and the behavior change plan are determined; A virtual interaction module is configured to create a suitable virtual character in a virtual environment based on the patient information collected and organized above, and interact with the patient through the virtual character; The effectiveness evaluation module is configured to collect various data during the intervention process, including conversation records, relaxation training process records, and cognitive behavioral therapy implementation records. Based on the collected data, the effectiveness of the psychological intervention program is evaluated. By comparing the patient's psychological scale scores before and after the intervention, it is determined whether anxiety and depression have been alleviated, and the patient's behavioral changes are evaluated. Based on the intervention effectiveness evaluation report, the psychological intervention program is adjusted; If the intervention is effective, continue with the original plan; If the intervention effect is not good, adjust the combination of intervention methods, modify the intervention content or reset the intervention goals.

[0011] Furthermore, the effect evaluation module includes: The collection module is used to collect conversation records, relaxation training process records and cognitive behavioral therapy implementation records during the intervention process; A feature extraction module is used to extract features from the conversation records, relaxation training process records, and cognitive behavioral therapy implementation records; A first acquisition module is used to acquire a preset evaluation template set, wherein the evaluation template set includes a plurality of evaluation templates; A first evaluation module is configured to evaluate the conversation records, relaxation training process records, and cognitive behavioral therapy implementation records multiple times based on the multiple evaluation templates to obtain multiple evaluation values; A second acquisition module is used to obtain the template weight of the evaluation template; A second evaluation module is used to determine the evaluation value of the psychological intervention program based on the evaluation value of each evaluation template and the weight value of the corresponding template; in, It represents the evaluation value of the psychological intervention program; Indicates the total number of evaluation templates; Indicates the The first evaluation template was obtained after multiple evaluations of the conversation records, relaxation training process records and cognitive behavioral therapy implementation records. An assessment value; Indicates the The total number of evaluation values ​​obtained after multiple evaluations of the conversation records, relaxation training process records, and cognitive behavioral therapy implementation records using the evaluation templates; represents an intermediate variable; Indicates the The weight value of each evaluation template; Indicates the preset template weight threshold; when When , it means that the evaluation value obtained by evaluating the conversation record, relaxation training process record and cognitive behavioral therapy implementation record based on the evaluation template is not credible; A query module, configured to query a preset evaluation value-psychological scale score based on the evaluation value of the psychological intervention program, and determine a target psychological scale score corresponding to the evaluation value of the psychological intervention program; Judgment module, used to: Compare the target psychological scale scores with the patient's pre-intervention psychological scale scores; If the target psychological scale score is lower than the patient's psychological scale score before intervention, it means that the patient's anxiety and depression have been alleviated and the original plan should be continued; If the target psychological scale score is greater than or equal to the patient's psychological scale score before intervention, it means that the patient's anxiety and depression are facing worsening, and it is necessary to adjust the combination of intervention methods, modify the intervention content, or reset the intervention goals.

[0012] Furthermore, the virtual characters interact with the patients, specifically: In a virtual environment, psychological counseling dialogues are conducted with patients through virtual characters: Based on customized psychological counseling dialogue scripts, we guide patients to express their inner feelings, help them understand the causes of their emotions, and provide support and suggestions; In a virtual environment, patients are guided through relaxation training by virtual characters: Based on customized relaxation training content, the virtual character demonstrates relaxation movements and guides patients to perform breathing regulation or muscle relaxation exercises; In a virtual environment, virtual characters are used to help patients conduct cognitive behavioral therapy: Following a customized cognitive behavioral therapy plan, the avatar guides the patient in identifying and correcting irrational cognitions.

[0013] Furthermore, the virtual environment creation module includes: The scene management module is configured to store, classify and manage all created virtual environments, specifically: Establish a clear directory structure and add rich metadata tags for each virtual environment; Provide powerful search function and visual preview function; Control the access and operation permissions of different roles to the virtual environment; It also provides a standardized process for importing new virtual environments and passing the created virtual environments to the personalized configuration module for subsequent interactive configuration. The personalized configuration module is configured to select target scenarios from the scenario management module and perform personalized configuration and dynamic adjustment during runtime based on the patient's condition, treatment stage and evaluation purpose.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The present invention uses virtual reality technology to create an immersive environment, restore the real treatment scene of lung cancer patients, effectively induce emotional responses, and provide a basis for accurate assessment. By comprehensively and in real time obtaining multi-dimensional data such as patient physiology, voice, and movement, the psychological state assessment is made more accurate. Based on the machine learning algorithm, a mental health assessment model is constructed, which can quickly and accurately identify the emotion type and the severity of psychological problems and generate detailed reports, providing a scientific basis for subsequent intervention. According to the assessment results, personalized plans are formulated, combined with virtual character interaction, psychological counseling, relaxation training and cognitive behavioral therapy are carried out to effectively relieve the patient's psychological stress. Compared with traditional methods, the system can induce the patient's emotional response more realistically, comprehensively obtain psychological state data, and provide personalized intervention, thereby effectively alleviating the patient's psychological stress. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a structural diagram of the virtual platform-based mental health assessment and intervention system for lung cancer patients of the present invention. DETAILED DESCRIPTION

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments 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 are within the scope of protection of the present invention.

[0017] To address the technical issue that the current assessment of the mental health status of lung cancer patients mainly relies on psychological scales and the subjective judgment of clinicians, which, although scientifically sound, are easily affected by the patient's subjective factors, thus limiting the accuracy of the assessment results, please refer to Figure 1 , this embodiment provides the following technical solutions: A mental health assessment and intervention system for lung cancer patients based on a virtual platform, the system comprising: a virtual environment creation module configured to use virtual reality technology to create an immersive virtual environment to simulate the hospital environment, surgical scenes, and rehabilitation scenes of lung cancer patients during treatment, thereby inducing real emotional responses from patients; The psychological state monitoring module is configured to integrate multiple biosensors to monitor the patient's physiological response data in real time in a virtual environment. At the same time, it combines voice recognition technology and motion capture technology to analyze the patient's voice tone and speech speed changes, as well as record the patient's body movements and facial expressions to obtain the patient's comprehensive psychological state data; Among them, denoising and feature extraction are performed on physiological response data, speech enhancement and speech feature extraction are performed on speech data, and data smoothing, coordinate transformation and motion feature extraction are performed on motion data; During speech enhancement, the speech frame is evaluated from multiple dimensions, including envelope and signal strength, to comprehensively assess the noise level in the speech frame. The abnormality level of the time-frequency unit is then determined to locate the time-frequency unit that requires noise reduction, achieving targeted noise reduction. A mental health assessment module is configured to establish a mental health assessment model based on a machine learning algorithm, automatically identify the patient's emotional type and the severity of psychological problems, and generate a mental health assessment report; The mental health intervention module is configured to develop personalized psychological intervention plans based on mental health assessment reports and patient information, and evaluate patient behavior changes by comparing psychological scale scores before and after intervention. The psychological intervention plan includes psychological counseling dialogues, relaxation training and cognitive behavioral therapy in a virtual environment, and interacts with patients through virtual characters to guide patients to adjust their mindsets and relieve psychological stress.

[0018] The technical effects of the above technical scheme are as follows: the virtual environment creation module creates an immersive virtual environment through virtual reality technology, simulating real scenes during the patient's treatment process, which can more naturally induce the patient's real emotional response. Compared with traditional questionnaires or simple interviews, it can more deeply explore the patient's inner psychological state. The psychological state monitoring module combines a variety of biosensors, voice recognition technology and motion capture technology to comprehensively obtain the patient's comprehensive psychological state data from multiple dimensions such as physiology, language and behavior, providing a richer and more accurate basis for mental health assessment, avoiding the deviation that may be caused by a single assessment method. The mental health assessment module establishes an assessment model based on a machine learning algorithm, which can automatically identify the patient's emotional type and the severity of psychological problems, and generate an accurate mental health assessment report. The mental health intervention module formulates a personalized psychological intervention plan based on the assessment report and patient information, including psychological counseling dialogues, relaxation training and cognitive behavioral therapy in a virtual environment, etc., which can provide the most suitable intervention measures for the individual differences of different patients and improve the intervention effect.

[0019] Mental state monitoring module, including: The data acquisition module is configured to uniformly access and manage multiple biosensors, collect data on the patient's physiological responses in the virtual environment, collect voice data using speech recognition technology, and analyze changes in the patient's voice intonation and speech speed, and collect motion data using motion capture technology to record changes in the patient's body movements and facial expressions; Among them, the biosensors include heart rate sensors, skin conductance sensors and brain wave sensors. At the same time, corresponding sampling frequencies are set according to different biosensors, and the sampling times of all biosensors are synchronized; a data processing module configured to perform denoising and feature extraction on the collected physiological response data, perform speech enhancement and speech feature extraction on the collected speech data, and perform data smoothing, coordinate conversion, and motion feature extraction on the collected motion data; The data fusion module is configured to fuse the processed physiological response data, voice data and action data using a weighted average method and based on the importance and relevance of different data to obtain comprehensive psychological state data.

[0020] The technical effect of the above technical solution is: through multi-dimensional data collection and efficient data processing, it ensures that the mental state monitoring module can provide complete and accurate mental state data, providing a solid foundation for the system's mental health assessment module and mental health intervention module, and the fusion processing of multi-source data can effectively reduce the errors or deviations that may be caused by a single data source, and improve the reliability and stability of the system's mental state assessment.

[0021] Data processing module, specifically: For physiological response data: Denoising: Each physiological response data corresponds to a parameter table, including data type, noise reduction algorithm, parameter name and parameter value. According to the collected physiological response data type, the corresponding noise reduction algorithm and its parameters are queried from the stored parameter table. Based on the queried noise reduction algorithm and parameters, the corresponding function in the algorithm library is called to perform noise reduction on the physiological response data; In this embodiment, as shown in Table 1, the parameter table includes the following fields: Data type: identifies the type of physiological screening data; Noise reduction algorithm: specify the corresponding noise reduction algorithm name (such as "low-pass filter", "wavelet transform", etc.); Parameter name: such as "cutoff frequency", "filter order", "wavelet basis function", etc.; Parameter value: a specific numerical value or string, such as "10Hz", "3rd order", "Daubechies", etc. Feature extraction: Extract heart rate and heart rate variability features through standard deviation and root mean square deviation, extract skin conductance level and skin galvanic response features, perform spectrum analysis, and extract power spectral density in different frequency bands; For voice data: Speech enhancement: Use noise suppression algorithms (such as spectral subtraction and Wiener filtering) to reduce the impact of background noise on speech data and improve speech clarity. Speech feature extraction: extract intonation features and speech features and normalize them to the same range; For action data: Data smoothing: Smoothing motion data to remove noise and jitter through moving average filtering or Kalman filtering. Coordinate transformation: transforming motion data from the original coordinate system to a unified virtual environment coordinate system to ensure data consistency; Motion feature extraction: extract the amplitude, frequency, speed and acceleration features of limb movements, extract the features of body posture, and extract facial expression features.

[0022] The technical effects of the above technical solution are: through denoising, speech enhancement and data smoothing, the system can more accurately capture the patient's true psychological state and reduce misjudgments caused by data quality issues. By accurately extracting speech and movement features, the system can interact with the patient in a more natural and intuitive way. Based on the above high-quality data processing and accurate feature extraction, the data processing module can provide more accurate input data for the mental health assessment module, enabling the system to more accurately identify the patient's emotional type and the severity of psychological problems, thereby providing a more reliable basis for personalized intervention.

[0023] Voice enhancement, specifically: A framing module is configured to use a speech signal in the collected speech data as a speech signal to be enhanced, perform frame processing on the speech signal to be enhanced, and obtain a plurality of speech frames; and sort the plurality of speech frames based on a time series to obtain a speech frame sequence; Randomly select a speech frame as the first speech frame; a determination module, configured to obtain a first evaluation index value and a second evaluation index value of a first speech frame, and determine a noise evaluation value of the first speech frame based on the first evaluation index value and the second evaluation index value of the first speech frame; a comparison module, configured to compare the noise evaluation value of the first speech frame with a preset noise evaluation threshold, and when determining that the noise evaluation value of the first speech frame is greater than or equal to the preset noise evaluation threshold, use the first speech frame as a target noise reduction frame; Performing a short-time Fourier transform on the target noise reduction frame to determine a sound spectrogram corresponding to the target noise reduction frame; obtaining a time-frequency unit of the sound spectrogram to obtain a plurality of time-frequency units; Anomaly identification module, used to: Obtain the energy value of each time-frequency unit within the time-frequency range and determine the energy distribution of each time-frequency unit; Take any time-frequency unit as the first time-frequency unit; Determine the target area with the first time-frequency unit as the center; calculate the standard deviation of the energy value in the target area as the standard deviation of the energy distribution corresponding to the first time-frequency unit; Traverse all time-frequency units and determine the standard deviation of energy distribution corresponding to each time-frequency unit; Randomly select a time-frequency unit in the target area except the first time-frequency unit as the second time-frequency unit; Calculating a ratio between a standard deviation of energy distribution of a first time-frequency unit and a standard deviation of energy distribution corresponding to a second time-frequency unit in the target area to obtain a first ratio; Calculating the Euclidean distance between the first time-frequency unit and the second time-frequency unit; Calculating the product of the first ratio and the Euclidean distance, and using the product as the difference degree value between the first time-frequency unit and the second time-frequency unit; Traversing other time-frequency units except the first time-frequency unit in the target area, and obtaining a difference degree value between the first time-frequency unit and each other time-frequency unit; Sum and average the difference values ​​between the first time-frequency unit and each other time-frequency unit to obtain the abnormality value of the first time-frequency unit; Traverse all time-frequency units and determine the abnormality value of each time-frequency unit; Enhancement modules for: Comparing the abnormality level of each time-frequency unit with a preset abnormality level threshold, taking the time-frequency unit corresponding to the abnormality level greater than or equal to the preset abnormality level threshold as a noise time-frequency unit, and obtaining a plurality of noise time-frequency units; performing noise reduction and suppression on the plurality of noise time-frequency units based on a noise reduction algorithm; Traverse all speech frames to complete the enhancement of the speech signal to be enhanced.

[0024] In this embodiment, the first evaluation index value of the first speech frame is determined based on the dynamic value range, specifically: obtaining the maximum value and the minimum value in the dynamic value range; and taking the difference between the maximum value and the minimum value as the first evaluation index value of the first speech frame.

[0025] In this embodiment, the determination module includes: A first evaluation module is configured to obtain an envelope of the first speech frame, determine a dynamic value range of the envelope of the first speech frame, and determine a first evaluation index value of the first speech frame based on the dynamic value range; a second evaluation module, configured to obtain the signal strength in the first speech frame, determine a signal strength sequence corresponding to the first speech frame, and calculate a mean of the signal strengths in the signal strength sequence as a first mean; Calculating the difference between each signal strength value in the signal strength sequence and the first mean value to obtain a plurality of differences, and taking the sum of the absolute values ​​of the plurality of differences as the second evaluation index value of the first speech frame; The noise evaluation determination module is configured to determine a noise evaluation value of the first speech frame based on a first evaluation index value and a second evaluation index value of the first speech frame.

[0026] In this embodiment, based on the first evaluation index value and the second evaluation index value of the first speech frame, the noise evaluation value of the first speech frame is determined, specifically: the weight values ​​of the first evaluation index and the second evaluation index are preset respectively. and , the first evaluation index value × +Second evaluation index value× The calculation result of is taken as the noise evaluation value of the first speech frame; .

[0027] In this embodiment, the preset noise evaluation threshold is obtained by pre-training calculation based on experimental data.

[0028] In this embodiment, the time-frequency unit is the basic structural unit of the sound signal in the time-frequency domain (spectrogram).

[0029] In this embodiment, the noise reduction algorithm includes but is not limited to spectral subtraction, Wiener filtering, and wavelet threshold noise reduction.

[0030] The working principle and beneficial effects of the above technical solution are as follows: the framing module performs frame processing on the collected speech signal (speech signal to be enhanced) to obtain multiple speech frames, and sorts them in time series to form a speech frame sequence; framing helps to analyze the speech characteristics of different time periods; the determination module first obtains the envelope of any first speech frame and determines the dynamic value range of the envelope, thereby obtaining the first evaluation index value; measures the characteristics of the speech frame from the perspective of the envelope; then obtains the signal strength in the first speech frame, determines the signal strength sequence, calculates its mean (first mean), and then calculates the signal strength of the first speech frame. The second evaluation index value is obtained by calculating the sum of the absolute values ​​of the differences between each value in the signal strength sequence and the first mean value; the two evaluation index values ​​describe the speech frame from different aspects (envelope and signal strength); finally, the noise evaluation value of the first speech frame is determined based on the two evaluation index values; the comparison module compares the noise evaluation value of the first speech frame with the preset noise evaluation threshold, and if it is greater than or equal to the threshold, it is determined as the target noise reduction frame; for the target noise reduction frame, a short-time Fourier transform is performed to obtain a sound spectrogram, and then a number of time-frequency units are obtained; the abnormality recognition module first determines the noise of each time-frequency unit in time- Energy value distribution within the frequency range; for any first time-frequency unit, the target area is determined with it as the center, and the standard deviation of the energy value in the area is calculated to obtain the corresponding energy distribution standard deviation; after traversing all time-frequency units to determine the energy distribution standard deviation of each time-frequency unit, the ratio (first ratio) and Euclidean distance of the energy distribution standard deviation of the first time-frequency unit and other time-frequency units (such as the second time-frequency unit) in the target area are calculated, and the product of the two is used as the difference degree value, and then the other time-frequency units except the first time-frequency unit in the target area are traversed to obtain the mean of the difference degree values ​​between the first time-frequency unit and each other time-frequency unit as the abnormality degree value of the first time-frequency unit, and finally all time-frequency units are traversed to determine the abnormality degree value of each time-frequency unit; the enhancement module converts each time-frequency unit into an abnormal value. The abnormality value of the time-frequency unit is compared with the preset abnormality threshold to determine the noise time-frequency unit, and then these noise time-frequency units are subjected to noise reduction suppression based on the noise reduction algorithm. Finally, all speech frames are traversed to complete the enhancement of the speech signal to be enhanced; the speech frame is evaluated from multiple dimensions such as envelope and signal strength, which can more comprehensively evaluate the noise situation in the speech frame instead of relying on a single feature for judgment, thereby improving the accuracy of noise assessment; by determining the abnormality value of the time-frequency unit, the time-frequency unit that needs noise reduction can be accurately located, thereby achieving targeted noise reduction suppression, avoiding excessive processing of the normal speech part, and better preserving the original characteristics of the speech while effectively reducing the noise; the frame processing combined with time-frequency analysis takes into account the temporal segmentation characteristics of the speech and utilizes the characteristics of the time-frequency unit in the time-frequency domain. It can adapt to the changes of the speech signal at different times and frequencies, and improves the adaptability of the speech enhancement technology to different types of speech signals and noise environments.

[0031] Analyze the patient's voice tone and speech speed changes, and record the patient's body movements and facial expressions, specifically: Intonation analysis: By comparing with the normal range of intonation, the abnormality of the patient's intonation changes can be determined; Speech rate change analysis: Analyze the changes in speech rate and judge the patient's emotional state based on the absolute value and change trend of speech rate; Body movement analysis: Analyze the patient's body movement characteristics based on the amplitude, frequency, and speed of the body movements. By analyzing the pattern of body movements, the patient's behavioral intention in the virtual environment can be determined. Expression analysis: Based on the expression recognition results, the duration and frequency of the patient's expressions in different expression categories are counted.

[0032] The technical effect of the above technical solution is: by real-time analysis of voice intonation, speaking speed, body movements and facial expressions, the system can dynamically monitor the patient's emotional state and behavioral intentions, so that the system can promptly detect fluctuations in the patient's emotions and provide support for timely intervention.

[0033] Mental health assessment modules, including: A data collection module is configured to collect a large amount of historical data as training samples and label the training samples with emotion types and severity of psychological problems; The model building module is configured to combine the random forest algorithm and the linear regression algorithm to construct a hybrid mental health assessment model, and the labeled training samples are used to train and optimize the constructed mental health assessment model; Among them, the random forest algorithm can handle complex nonlinear relationships and has strong anti-overfitting capabilities; the linear regression algorithm can provide more intuitive explanation and interpretability; The report generation module is configured to input the comprehensive psychological state data into the trained mental health assessment model, output the patient's emotional type and the severity of psychological problems through the mental health assessment model, generate a corresponding mental health assessment report based on the output results and display it visually.

[0034] The technical effects of the above technical solution are as follows: the data collection module collects a large amount of historical data as training samples, providing a rich and accurately labeled data basis for model training. The model construction module combines the random forest algorithm and the linear regression algorithm, and combines the advantages of the two algorithms to construct a mental health assessment model, which can more comprehensively handle the complex relationships in the data and improve the accuracy and reliability of the assessment. The report generation module inputs the comprehensive psychological state data into the trained mental health assessment model, and can output the patient's emotional type (such as anxiety, depression, anger, etc.) and the severity of psychological problems (such as mild, moderate, severe), providing an important basis for personalized mental health intervention. The mental health assessment module realizes an automated process from data collection, model training to report generation, which greatly improves the efficiency of psychological assessment. Compared with traditional manual assessment methods, this module can complete the assessment of a large number of patients in a short time, saving manpower and time costs.

[0035] Mental health intervention modules, including: The report analysis module is configured to interpret the mental health assessment report, extract the patient's anxiety level, depression level and stress sources, and collect and organize the patient's information to understand the background of the patient's psychological problems. The psychological status information in the mental health assessment report is combined with the patient's basic information to comprehensively analyze the causes and development trends of the patient's psychological problems; a program development module configured to clarify the primary goal of the intervention based on the comprehensive analysis results, select an appropriate method combination from the three main methods of psychological counseling dialogue, relaxation training, and cognitive behavioral therapy in a virtual environment based on the primary intervention goal, and then customize the specific content based on the selected psychological intervention program. This includes, for psychological counseling dialogue, writing a dialogue script based on the patient's problem situation; for relaxation training, selecting relaxation techniques suitable for the patient, such as progressive muscle relaxation or breathing relaxation; and for cognitive behavioral therapy, determining the type of cognitive distortion that needs to be corrected and the behavior change plan; The virtual interaction module is configured to create a suitable virtual character in a virtual environment based on the patient information collected and organized above, and interact with the patient through the virtual character, specifically: In a virtual environment, psychological counseling dialogues are conducted with patients through virtual characters: Based on customized psychological counseling dialogue scripts, we guide patients to express their inner feelings, help them understand the causes of their emotions, and provide support and suggestions; In a virtual environment, patients are guided through relaxation training by virtual characters: Based on customized relaxation training content, the virtual character demonstrates relaxation movements and guides patients to perform breathing regulation or muscle relaxation exercises; In a virtual environment, virtual characters are used to help patients conduct cognitive behavioral therapy: Following a customized cognitive behavioral therapy plan, the avatar guides the patient in identifying and correcting irrational cognitions; The effectiveness evaluation module is configured to collect various data during the intervention process, including conversation records, relaxation training process records, and cognitive behavioral therapy implementation records. Based on the collected data, the effectiveness of the psychological intervention program is evaluated. By comparing the patient's psychological scale scores before and after the intervention, it is determined whether anxiety and depression have been alleviated, and the patient's behavioral changes are evaluated. Based on the intervention effectiveness evaluation report, the psychological intervention program is adjusted; If the intervention is effective, continue with the original plan; If the intervention effect is not good, adjust the combination of intervention methods, modify the intervention content or reset the intervention goals.

[0036] The technical effects of the above technical scheme are as follows: the report analysis module can provide accurate input for the program formulation module by interpreting the mental health assessment report, ensuring that the psychological intervention program fully meets the patient's personalized needs. The program formulation module clarifies the main goals of the intervention based on the comprehensive analysis results. This goal-oriented intervention strategy can ensure that the intervention measures are more targeted and improve the intervention effect. The virtual interaction module is based on the collected and organized patient information. It creates suitable virtual characters in the virtual environment to interact with the patient, which can provide patients with a more natural and comfortable intervention environment and enhance patient participation and compliance. By collecting various data during the intervention process, we can fully understand the implementation of the psychological intervention program, so that we can timely optimize the intervention measures according to the patient's real-time feedback and changes in psychological state to ensure the maximization of the intervention effect.

[0037] The effect evaluation module includes: The collection module is used to collect conversation records, relaxation training process records and cognitive behavioral therapy implementation records during the intervention process; A feature extraction module is used to extract features from the conversation records, relaxation training process records, and cognitive behavioral therapy implementation records; A first acquisition module is used to acquire a preset evaluation template set, wherein the evaluation template set includes a plurality of evaluation templates; A first evaluation module is configured to evaluate the conversation records, relaxation training process records, and cognitive behavioral therapy implementation records multiple times based on the multiple evaluation templates to obtain multiple evaluation values; A second acquisition module is used to obtain the template weight of the evaluation template; A second evaluation module is used to determine the evaluation value of the psychological intervention program based on the evaluation value of each evaluation template and the weight value of the corresponding template; in, It represents the evaluation value of the psychological intervention program; Indicates the total number of evaluation templates; Indicates the The first evaluation template was obtained after multiple evaluations of the conversation records, relaxation training process records and cognitive behavioral therapy implementation records. An assessment value; Indicates the The total number of evaluation values ​​obtained after multiple evaluations of the conversation records, relaxation training process records, and cognitive behavioral therapy implementation records using the evaluation templates; represents an intermediate variable; Indicates the The weight value of each evaluation template; Indicates the preset template weight threshold; when When , it means that the evaluation value obtained by evaluating the conversation record, relaxation training process record and cognitive behavioral therapy implementation record based on the evaluation template is not credible; A query module, configured to query a preset evaluation value-psychological scale score based on the evaluation value of the psychological intervention program, and determine a target psychological scale score corresponding to the evaluation value of the psychological intervention program; Judgment module, used to: Compare the target psychological scale scores with the patient's pre-intervention psychological scale scores; If the target psychological scale score is lower than the patient's psychological scale score before intervention, it means that the patient's anxiety and depression have been alleviated and the original plan should be continued; If the target psychological scale score is greater than or equal to the patient's psychological scale score before intervention, it means that the patient's anxiety and depression are facing worsening, and it is necessary to adjust the combination of intervention methods, modify the intervention content, or reset the intervention goals.

[0038] In this embodiment, the preset evaluation value-psychological scale is a control mapping table obtained by training based on experimental data.

[0039] The working principle and beneficial effects of the above technical solution are: by collecting multiple intervention process records (conversations, relaxation training, cognitive behavioral therapy, etc.), it can comprehensively reflect all aspects of mental health intervention for lung cancer patients, avoiding the one-sidedness of evaluation caused by a single data source; using multiple evaluation templates to evaluate the intervention process, the intervention effect can be considered from different dimensions, making the evaluation more detailed and comprehensive; considering the weight of the evaluation template, the importance of different evaluation templates can be reasonably set according to actual needs and experience, which can provide a reliable basis for further optimizing the psychological plan for mental health intervention for lung cancer patients.

[0040] Virtual environment creation module, including: The scene management module is configured to store, classify and manage all created virtual environments, specifically: Establish a clear directory structure and add rich metadata tags for each virtual environment; Provides powerful search function (by name, tag, type, creation date, etc.) and visual preview function (thumbnail, 3D preview); Control the access and operation permissions of different roles (modelers, doctor reviewers, system administrators) to the virtual environment; It also provides a standardized process for importing new virtual environments and passing the created virtual environments to the personalized configuration module for subsequent interactive configuration. The personalized configuration module is configured to select target scenarios from the scenario management module and perform personalized configuration and dynamic adjustment during runtime based on the patient's condition, treatment stage and evaluation purpose.

[0041] The technical effects of the above technical solution are as follows: the scene management module can achieve efficient classification and management of virtual environments by establishing a clear directory structure and adding rich metadata tags for each virtual environment. It provides search functions by multiple dimensions such as name, tag, type, creation date, etc., which can help users quickly find the target virtual environment. The visual preview function allows users to intuitively understand the layout and details of each scene when selecting a virtual environment. By controlling the access and operation permissions of different roles to the virtual environment, the system security and data integrity can be ensured. The personalized configuration module can select the most appropriate target scene from the scene management module according to the patient's specific situation, treatment stage and evaluation purpose, and perform further personalized configuration to ensure that the virtual environment fully meets the individual needs of the patient, thereby improving the effect of intervention and patient acceptance.

[0042] Working principle: By using virtual reality technology to create an immersive virtual environment and simulate real scenes during the patient's treatment process, it can more naturally induce the patient's real emotional response, thereby deeply exploring the patient's inner psychological state. Combined with a variety of biosensors, voice recognition technology and motion capture technology, it can comprehensively obtain the patient's comprehensive psychological state data from multiple dimensions such as physiology, language and behavior, providing a richer and more accurate basis for mental health assessment. An assessment model is established based on machine learning algorithms, which can automatically identify the patient's emotional type and the severity of psychological problems and generate a mental health assessment report. Based on the assessment report and patient information, a personalized psychological intervention plan can be formulated, thereby providing the most suitable intervention measures for the individual differences of different patients and improving the intervention effect.

[0043] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0044] While the embodiments of the present invention have been shown and described, it will be apparent to those skilled in the art that various changes, modifications, substitutions, and alterations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A mental health assessment and intervention system for lung cancer patients based on a virtual platform, characterized by: The system comprises: a virtual environment creation module configured to use virtual reality technology to create an immersive virtual environment to simulate the hospital environment, surgical scenes, and rehabilitation scenes of lung cancer patients during treatment, thereby inducing real emotional responses from patients; The psychological state monitoring module is configured to integrate multiple biosensors to monitor the patient's physiological response data in real time in a virtual environment. At the same time, it combines voice recognition technology and motion capture technology to analyze the patient's voice tone and speech speed changes, as well as record the patient's body movements and facial expressions to obtain the patient's comprehensive psychological state data; Among them, denoising and feature extraction are performed on physiological response data, speech enhancement and speech feature extraction are performed on speech data, and data smoothing, coordinate transformation and motion feature extraction are performed on motion data; During speech enhancement, the speech frame is evaluated from multiple dimensions, including envelope and signal strength, to comprehensively assess the noise level in the speech frame. The abnormality level of the time-frequency unit is then determined to locate the time-frequency unit that requires noise reduction, achieving targeted noise reduction. A mental health assessment module is configured to establish a mental health assessment model based on a machine learning algorithm, automatically identify the patient's emotional type and the severity of psychological problems, and generate a mental health assessment report; The mental health intervention module is configured to develop personalized psychological intervention plans based on mental health assessment reports and patient information, and evaluate patient behavior changes by comparing psychological scale scores before and after intervention. The psychological intervention plan includes psychological counseling dialogues, relaxation training and cognitive behavioral therapy in a virtual environment, and interacts with patients through virtual characters to guide patients to adjust their mindsets and relieve psychological stress.

2. The virtual platform-based mental health assessment and intervention system for lung cancer patients according to claim 1 is characterized in that: The mental state monitoring module includes: The data acquisition module is configured to uniformly access and manage multiple biosensors, collect data on the patient's physiological responses in the virtual environment, collect voice data using speech recognition technology, and analyze changes in the patient's voice intonation and speech speed, and collect motion data using motion capture technology to record changes in the patient's body movements and facial expressions; Among them, the biosensors include heart rate sensors, skin conductance sensors and brain wave sensors. At the same time, corresponding sampling frequencies are set according to different biosensors, and the sampling times of all biosensors are synchronized; a data processing module configured to perform denoising and feature extraction on the collected physiological response data, perform speech enhancement and speech feature extraction on the collected speech data, and perform data smoothing, coordinate conversion, and motion feature extraction on the collected motion data; The data fusion module is configured to fuse the processed physiological response data, voice data and action data using a weighted average method and based on the importance and relevance of different data to obtain comprehensive psychological state data.

3. The virtual platform-based mental health assessment and intervention system for lung cancer patients according to claim 2 is characterized in that: The data processing module is specifically: For physiological response data: Denoising: Each physiological response data corresponds to a parameter table, including data type, noise reduction algorithm, parameter name and parameter value. According to the collected physiological response data type, the corresponding noise reduction algorithm and its parameters are queried from the stored parameter table. Based on the queried noise reduction algorithm and parameters, the corresponding function in the algorithm library is called to perform noise reduction on the physiological response data; Feature extraction: Extract heart rate and heart rate variability features through standard deviation and root mean square deviation, extract skin conductance level and skin galvanic response features, perform spectrum analysis, and extract power spectral density in different frequency bands; For voice data: Speech enhancement: Through noise suppression algorithms, the impact of background noise on speech data is reduced, improving speech clarity; Speech feature extraction: extract intonation features and speech features and normalize them to the same range; For action data: Data smoothing: Smoothing motion data to remove noise and jitter through moving average filtering or Kalman filtering. Coordinate transformation: transforming motion data from the original coordinate system to a unified virtual environment coordinate system to ensure data consistency; Motion feature extraction: extract the amplitude, frequency, speed and acceleration features of limb movements, extract the features of body posture, and extract facial expression features.

4. The virtual platform-based mental health assessment and intervention system for lung cancer patients according to claim 3 is characterized in that: The speech enhancement is specifically: A framing module is configured to use a speech signal in the collected speech data as a speech signal to be enhanced, perform frame processing on the speech signal to be enhanced, and obtain a plurality of speech frames; and sort the plurality of speech frames based on a time series to obtain a speech frame sequence; Randomly select a speech frame as the first speech frame; a determination module, configured to obtain a first evaluation index value and a second evaluation index value of a first speech frame, and determine a noise evaluation value of the first speech frame based on the first evaluation index value and the second evaluation index value of the first speech frame; a comparison module, configured to compare the noise evaluation value of the first speech frame with a preset noise evaluation threshold, and when determining that the noise evaluation value of the first speech frame is greater than or equal to the preset noise evaluation threshold, use the first speech frame as a target noise reduction frame; Performing a short-time Fourier transform on the target noise reduction frame to determine a sound spectrogram corresponding to the target noise reduction frame; obtaining a time-frequency unit of the sound spectrogram to obtain a plurality of time-frequency units; Anomaly identification module, used to: Obtain the energy value of each time-frequency unit within the time-frequency range and determine the energy distribution of each time-frequency unit; Take any time-frequency unit as the first time-frequency unit; Determine the target area with the first time-frequency unit as the center; Calculate the standard deviation of the energy value in the target area as the standard deviation of the energy distribution corresponding to the first time-frequency unit; Traverse all time-frequency units and determine the standard deviation of energy distribution corresponding to each time-frequency unit; Randomly select a time-frequency unit in the target area except the first time-frequency unit as the second time-frequency unit; Calculating a ratio between a standard deviation of energy distribution of a first time-frequency unit and a standard deviation of energy distribution corresponding to a second time-frequency unit in the target area to obtain a first ratio; Calculating the Euclidean distance between the first time-frequency unit and the second time-frequency unit; Calculating the product of the first ratio and the Euclidean distance, and using the product as the difference degree value between the first time-frequency unit and the second time-frequency unit; Traversing other time-frequency units except the first time-frequency unit in the target area, and obtaining a difference degree value between the first time-frequency unit and each other time-frequency unit; Sum and average the difference values ​​between the first time-frequency unit and each other time-frequency unit to obtain the abnormality value of the first time-frequency unit; Traverse all time-frequency units and determine the abnormality value of each time-frequency unit; Enhancement modules for: Comparing the abnormality level of each time-frequency unit with a preset abnormality level threshold, taking the time-frequency unit corresponding to the abnormality level greater than or equal to the preset abnormality level threshold as a noise time-frequency unit, and obtaining a plurality of noise time-frequency units; performing noise reduction and suppression on the plurality of noise time-frequency units based on a noise reduction algorithm; Traverse all speech frames to complete the enhancement of the speech signal to be enhanced.

5. The virtual platform-based mental health assessment and intervention system for lung cancer patients according to claim 2 is characterized in that: Analyze the patient's voice tone and speech speed changes, and record the patient's body movements and facial expressions, specifically: Intonation analysis: By comparing with the normal range of intonation, the abnormality of the patient's intonation changes can be determined; Speech rate change analysis: Analyze the changes in speech rate and judge the patient's emotional state based on the absolute value and change trend of speech rate; Body movement analysis: Analyze the patient's body movement characteristics based on the amplitude, frequency, and speed of the body movements. By analyzing the pattern of body movements, the patient's behavioral intention in the virtual environment can be determined. Expression analysis: Based on the expression recognition results, the duration and frequency of the patient's expressions in different expression categories are counted.

6. The virtual platform-based mental health assessment and intervention system for lung cancer patients according to claim 1 is characterized in that: The mental health assessment module includes: A data collection module is configured to collect a large amount of historical data as training samples and label the training samples with emotion types and severity of psychological problems; The model building module is configured to combine the random forest algorithm and the linear regression algorithm to construct a hybrid mental health assessment model, and the labeled training samples are used to train and optimize the constructed mental health assessment model; The report generation module is configured to input the comprehensive psychological state data into the trained mental health assessment model, output the patient's emotional type and the severity of psychological problems through the mental health assessment model, generate a corresponding mental health assessment report based on the output results and display it visually.

7. The virtual platform-based mental health assessment and intervention system for lung cancer patients according to claim 1 is characterized in that: The mental health intervention module includes: The report analysis module is configured to interpret the mental health assessment report, extract the patient's anxiety level, depression level and stress sources, and collect and organize the patient's information to understand the background of the patient's psychological problems. The psychological status information in the mental health assessment report is combined with the patient's basic information to comprehensively analyze the causes and development trends of the patient's psychological problems; A program development module is configured to clarify the main goal of the intervention based on the comprehensive analysis results, select an appropriate method combination from the three main methods of psychological counseling dialogue, relaxation training, and cognitive behavioral therapy in a virtual environment based on the main intervention goal, and then customize the specific content according to the selected psychological intervention program. For psychological counseling dialogue, the dialogue script is written according to the patient's problem situation; for relaxation training, relaxation techniques suitable for the patient are selected; and for cognitive behavioral therapy, the type of cognitive distortion that needs to be corrected and the behavior change plan are determined; A virtual interaction module is configured to create a suitable virtual character in a virtual environment based on the patient information collected and organized above, and interact with the patient through the virtual character; The effectiveness evaluation module is configured to collect various data during the intervention process, including conversation records, relaxation training process records, and cognitive behavioral therapy implementation records. Based on the collected data, the effectiveness of the psychological intervention program is evaluated. By comparing the patient's psychological scale scores before and after the intervention, it is determined whether anxiety and depression have been alleviated, and the patient's behavioral changes are evaluated. Based on the intervention effectiveness evaluation report, the psychological intervention program is adjusted; If the intervention is effective, continue with the original plan; If the intervention effect is not good, adjust the combination of intervention methods, modify the intervention content or reset the intervention goals.

8. The virtual platform-based mental health assessment and intervention system for lung cancer patients according to claim 7 is characterized in that: The effect evaluation module includes: The collection module is used to collect conversation records, relaxation training process records and cognitive behavioral therapy implementation records during the intervention process; A feature extraction module is used to extract features from the conversation records, relaxation training process records, and cognitive behavioral therapy implementation records; A first acquisition module is used to acquire a preset evaluation template set, wherein the evaluation template set includes a plurality of evaluation templates; A first evaluation module is configured to evaluate the conversation records, relaxation training process records, and cognitive behavioral therapy implementation records multiple times based on the multiple evaluation templates to obtain multiple evaluation values; A second acquisition module is used to obtain the template weight of the evaluation template; A second evaluation module is used to determine the evaluation value of the psychological intervention program based on the evaluation value of each evaluation template and the weight value of the corresponding template; in, It represents the evaluation value of the psychological intervention program; Indicates the total number of evaluation templates; Indicates the The first evaluation template was obtained after multiple evaluations of the conversation records, relaxation training process records and cognitive behavioral therapy implementation records. An assessment value; Indicates the The total number of evaluation values ​​obtained after multiple evaluations of the conversation records, relaxation training process records, and cognitive behavioral therapy implementation records using the evaluation templates; represents an intermediate variable; Indicates the The weight value of each evaluation template; Indicates the preset template weight threshold; when When , it means that the evaluation value obtained by evaluating the conversation record, relaxation training process record and cognitive behavioral therapy implementation record based on the evaluation template is not credible; A query module, configured to query a preset evaluation value-psychological scale score based on the evaluation value of the psychological intervention program, and determine a target psychological scale score corresponding to the evaluation value of the psychological intervention program; Judgment module, used to: Compare the target psychological scale scores with the patient's pre-intervention psychological scale scores; If the target psychological scale score is lower than the patient's psychological scale score before intervention, it means that the patient's anxiety and depression have been alleviated and the original plan should be continued; If the target psychological scale score is greater than or equal to the patient's psychological scale score before intervention, it means that the patient's anxiety and depression are facing worsening, and it is necessary to adjust the combination of intervention methods, modify the intervention content, or reset the intervention goals.

9. The virtual platform-based mental health assessment and intervention system for lung cancer patients according to claim 7, characterized in that: Interact with patients through virtual characters, specifically: In a virtual environment, psychological counseling dialogues are conducted with patients through virtual characters: Based on customized psychological counseling dialogue scripts, we guide patients to express their inner feelings, help them understand the causes of their emotions, and provide support and suggestions; In a virtual environment, patients are guided through relaxation training by virtual characters: Based on customized relaxation training content, the virtual character demonstrates relaxation movements and guides patients to perform breathing regulation or muscle relaxation exercises; In a virtual environment, virtual characters are used to help patients conduct cognitive behavioral therapy: Following a customized cognitive behavioral therapy plan, the avatar guides the patient in identifying and correcting irrational cognitions.

10. The virtual platform-based mental health assessment and intervention system for lung cancer patients according to claim 1, characterized in that: The virtual environment creation module includes: The scene management module is configured to store, classify and manage all created virtual environments, specifically: Establish a clear directory structure and add rich metadata tags for each virtual environment; Provide powerful search function and visual preview function; Control the access and operation permissions of different roles to the virtual environment; It also provides a standardized process for importing new virtual environments and passing the created virtual environments to the personalized configuration module for subsequent interactive configuration. The personalized configuration module is configured to select target scenarios from the scenario management module and perform personalized configuration and dynamic adjustment during runtime based on the patient's condition, treatment stage and evaluation purpose.

Images