Vr training system based on cognitive function deficiency characteristics of children with attention deficit hyperactivity disorder

Abstract

The application discloses a VR training system based on the cognitive function defect characteristics of children's attention deficit hyperactivity disorder, relates to the technical field of digital medical treatment, and comprises a data acquisition module, a data evaluation module, a VR training module, a dynamic adjustment module, a result output module and a collaborative work module; the application constructs an objective quantitative model by fusing multi-modal data, accurately identifies three ADHD subtypes and the degree of cognitive defect of each dimension, solves the problem that traditional evaluation is highly subjective, and divides cold / hot executive function double-path training based on the CAPS theory, dynamically customizes task matching according to the subtypes, realizes accurate matching of defects and training, additionally, through real-time monitoring of eye movement and physiological indexes, automatically adjusts task difficulty and feedback strategies, maintains the best challenge level, can effectively improve the participation and training continuity of children, and finally achieves the goal of high-efficiency, high-interest and side-effect-free personalized digital therapy.

Description

Technical Field

[0001] This invention relates to the field of digital medical technology, and in particular to a VR training system based on the cognitive function deficit characteristics of children with attention deficit hyperactivity disorder. Background Technology

[0002] Attention Deficit Hyperactivity Disorder (ADHD), also known as hyperactivity, is one of the most common neurodevelopmental disorders in childhood. Clinically, it is characterized by age-inappropriate attention deficits, hyperactivity, and impulsivity that persist for more than 6 months. These symptoms can have a significant impact on a patient's learning and daily life.

[0003] In traditional ADHD diagnosis and treatment techniques, drug therapy has significant side effects (insomnia, loss of appetite, etc.), while non-drug therapies (such as behavioral intervention) are limited by environment and teachers, resulting in low adoption rates. Furthermore, existing cognitive training content is monotonous and cannot be customized for different subtypes of ADHD (ADHD-I / ADHD-HI / ADHD-C) with cold / hot executive function deficits, leading to unstable efficacy. In addition, traditional training systems fix the difficulty of tasks and lack a real-time dynamic optimization mechanism based on physiological signals (such as eye movements and skin conductance), making it difficult to cope with the fluctuations in children's attention and emotional changes, and thus failing to achieve the ideal treatment goals. Therefore, this invention proposes a VR training system based on the cognitive function deficit characteristics of children with ADHD to solve the problems existing in the prior art. Summary of the Invention

[0004] To address the aforementioned problems, the present invention aims to propose a VR training system based on the cognitive function deficit characteristics of children with attention deficit hyperactivity disorder (ADHD). This system solves the problems in traditional ADHD diagnosis and treatment techniques, such as significant side effects of drug therapy, limitations of non-drug therapy in terms of environment and qualified teachers, inability to customize solutions for cold / hot executive function deficits in different subtypes of ADHD, and difficulty in coping with fluctuations in children's attention and emotional changes.

[0005] To achieve the objectives of this invention, the invention is implemented through the following technical solution: a VR training system based on the cognitive function deficit characteristics of children with attention deficit hyperactivity disorder (ADHD), comprising: The data acquisition module is used to collect multimodal data, including clinical behavioral data, neurocognitive gamification assessment data, and physiological signal data. The data assessment module is used to perform fusion analysis on multimodal data, generate individualized cognitive function assessment reports, and identify ADHD subtypes and the degree of cognitive deficits in each dimension. The VR training module is based on the CAPS theoretical model to construct cold and hot dual-executive function subtype training units, and matches targeted VR immersive training tasks for children with different ADHD subtypes according to the assessment report. The dynamic adjustment module dynamically optimizes the difficulty, stimulus intensity, and feedback strategy of the training task by collecting physiological signals and behavioral data in real time during the training process. The results output module is used to output training progress reports, cognitive function improvement analysis, and subsequent intervention suggestions.

[0006] A further improvement is that the data acquisition module includes a clinical data acquisition unit for collecting children's clinical behavioral data, a neurocognitive data acquisition unit for collecting children's neurocognitive gamification assessment data, and a physiological signal acquisition unit for collecting children's physiological signal data.

[0007] Further improvements are made in that the data evaluation module includes a multi-source data fusion unit for feature extraction and analysis of multimodal data, a subtype identification unit for identifying ADHD subtypes and quantifying the core defect dimensions of each subtype, and an evaluation report generation unit for generating individualized evaluation reports.

[0008] Further improvements include: the VR training module includes a cold executive function training unit for training cognitive processing abilities, a hot executive function training unit for training emotional and motivational regulation abilities, and a VR interaction unit for providing children with an immersive virtual scene experience.

[0009] Further improvements are made in that: the tasks of the cold executive function training unit include attention switching training tasks, working memory training tasks, and inhibitory control training tasks, while the tasks of the hot executive function training unit include risk decision simulation tasks, delayed gratification selection tasks, and emotional conflict resolution tasks.

[0010] Further improvements are made in that: the dynamic adjustment module includes a process data acquisition unit, a state recognition unit, and an adaptive adjustment unit. The process data acquisition unit collects eye movement data, skin conductance response data, and heart rate variability data of children in real time during the training process. The state recognition unit identifies the child's level of attention and emotional arousal. The adaptive adjustment unit dynamically adjusts the training task parameters.

[0011] A further improvement is that the result output module includes a training report generation unit for generating training visualization reports and a VR device display, tablet computer, and mobile phone terminal for synchronously outputting the training visualization reports to the user.

[0012] Further improvements include a subtype-customized configuration unit and a collaborative working module. The subtype-customized configuration unit is used to assign differentiated training weights and task ratios to the cold and hot dual-execution function training units based on the ADHD subtype identification results. The collaborative working module executes the following closed-loop process: Initial assessment phase: Baseline assessment of cognitive function is completed through the data acquisition module and the data assessment module; Intervention training phase: The VR training module initiates subtype training, dynamically adjusting the module's parameters in real time; Effect verification phase: Repeat the evaluation after the training cycle ends, compare the efficacy and iteratively adjust the plan.

[0013] The beneficial effects of this invention are as follows: By integrating clinical behavioral data, neurocognitive gamified assessment data, and physiological signal data, this invention constructs a multi-dimensional feature vector, realizing an objective quantitative analysis of cognitive function deficits in children. It can accurately identify the three subtypes of ADHD-I, ADHD-HI, and ADHD-C, as well as the degree of deficit in each dimension, improving the pertinence of digital therapy. It solves the problems of strong subjectivity and ambiguous subtype identification in traditional assessments, and provides a reliable individualized basis for subsequent subtyping training. Furthermore, ADHD intervention training is divided into two dimensions: cold executive function and hot executive function, and training tasks are customized for different ADHD subtypes: ADHD-I type focuses on strengthening cold functions, ADHD-HI type focuses on regulating hot functions, and ADHD-C type dynamically balances the two-dimensional training ratio. This breaks through the shortcomings of traditional cognitive training, achieves precise matching between training content and core defects, and significantly improves the pertinence and effectiveness of intervention. In addition, by collecting eye-tracking data, physiological signals, and behavioral data in real time during the training process, the system dynamically identifies the child's attentional state and emotional arousal level, and automatically adjusts the difficulty, stimulation intensity, or feedback strategy of the training task accordingly. This solves the problem of traditional training programs being fixed and unable to adapt to changes in the child's real-time state, ensuring that the training is always kept at the optimal level of challenge. This effectively maintains the child's participation and the continuity of the training effect, improves the participation of children with ADHD, and achieves the treatment goals of high efficiency, few restrictions, strong interest, and few side effects. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the connection framework of the various modules of the VR training system of the present invention; Figure 2 This is a schematic diagram of the training method flow of the VR training system of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] According to the Diagnostic and Statistical Manual of Mental Disorders (DSM-V), ADHD can be divided into three subtypes: Attention-Deficit Predominant (ADHD-I), Hyperactivity-Impulsivity Predominant (ADHD-HI), and Mixed Subtype (ADHD-C). Current research suggests that the core symptoms of ADHD are cognitive, with different clinical subtypes exhibiting impairments in different dimensions of executive function. ADHD-I often shows more pronounced impairments in intellectual capacity and attention switching ability, ADHD-C shows worse performance in working memory and response inhibition, while ADHD-HI exhibits milder impairments in executive function. Researchers have used cognitive tasks combined with imaging techniques to explore brain networks and activation in different subtypes of ADHD. ADHD-C patients primarily exhibit abnormalities in the frontostriatal circuit, while ADHD-I patients primarily exhibit abnormalities in the fronto-parietal system. These studies demonstrate that the cognitive deficits and their severity vary among different subtypes of ADHD, and these differences in cognitive impairment can serve as a basis for clinical classification and treatment.

[0017] Digital therapy, based on the modern medical model, leverages emerging technologies to combine digital intelligence with various novel sensing systems and AI algorithms, creating software-driven products that offer preventative, managerial, and therapeutic benefits tailored to each patient's condition. Digital therapy offers significant advantages in ADHD treatment. First, it is entirely non-invasive, using high-tech intelligent computation and program refinement to guide patients correctly, improving their cognitive structure, attention, and executive functions. Second, compared to traditional drug treatments, digital therapy has fewer side effects and higher patient compliance. With the widespread use of the internet, mobile phones, tablets, and other electronic devices, game-based digital therapy can be conducted on various mobile devices. Corresponding ADHD children's skill training systems can provide therapeutic training to children through different application platforms, increasing collaboration between therapists and families and enhancing treatment effectiveness.

[0018] Example 1 according to Figure 1 As shown, this embodiment provides a VR training system based on the cognitive function deficit characteristics of children with attention deficit hyperactivity disorder (ADHD). The VR training system consists of a data acquisition module, a data evaluation module, a VR training module, a dynamic adjustment module, a result output module, and a collaborative working module, wherein: The data acquisition module is used to collect multimodal data from children, including clinical behavioral data, neurocognitive gamification assessment data, and physiological signal data. The data assessment module communicates with the data acquisition module and is used to perform fusion analysis on the multimodal data collected by the data acquisition module, generate individualized cognitive function assessment reports, and identify ADHD subtypes (attention deficit predominant: ADHD-I, hyperactivity-impulsivity predominant: ADHD-H, mixed type: ADHD-C) and the degree of cognitive deficit in each dimension; The VR training module communicates and connects with the data assessment module. Based on the cognitive-emotional personality system (CAPS) theoretical model, it constructs cold and hot dual executive function subtype training units. According to the individualized cognitive function assessment report generated by the data assessment module, it matches targeted VR immersive training tasks for children with different ADHD subtypes. The dynamic adjustment module communicates and connects with the VR training module. By collecting physiological signals and behavioral data from the training process in real time, it dynamically optimizes the difficulty, stimulation intensity, and feedback strategy of the training task. The results output module communicates with the dynamic adjustment module and is used to output training progress reports, cognitive function improvement analysis and subsequent intervention suggestions. The collaborative work module is used to execute the following closed-loop process: Initial assessment phase: Baseline assessment of cognitive function is completed through the data acquisition module and the data assessment module; Intervention training phase: The VR training module initiates subtype training, dynamically adjusting the module's parameters in real time; Effect verification phase: Repeat the evaluation after the training cycle ends, compare the efficacy and iteratively adjust the plan.

[0019] In this embodiment, the data acquisition module includes: The clinical data collection unit obtains basic information, behavioral performance and medical history data of children through structured questionnaires and doctors' diagnostic records. The structured questionnaire is dynamically generated based on Bayesian algorithm and is used to collect typical symptoms of different ADHD subtypes. The neurocognitive data acquisition unit collects data on children's neurocognitive abilities, such as attention shifting, working memory, and inhibitory control, through gamified assessment tasks built into VR devices. The gamified assessment tasks include interactive operations such as target tracking, object classification, and rule switching in virtual scenes. The physiological signal acquisition unit synchronously collects children's facial micro-expressions, skin conductance response (GSR), and heart rate variability (HRV) data through wearable devices to identify emotional states and attention fluctuations during training (using local encrypted storage and anonymized transmission protocols to ensure the security of children's physiological data).

[0020] In this embodiment, the data evaluation module includes: The multi-source data fusion unit performs feature extraction and correlation analysis on clinical behavioral data, neurocognitive gamified assessment data, and physiological signal data to construct a feature vector of children's cognitive function. The subtype identification unit, based on cognitive function feature vectors, identifies ADHD subtypes (ADHD-I, ADHD-HI, ADHD-C) through a machine learning classification model, and quantifies the core deficit dimensions of each ADHD subtype (such as the degree of attentional switching impairment in ADHD-I, the impulsive decision-making tendency in ADHD-HI, and the mixed deficit level in ADHD-C). The assessment report generation unit generates an individualized assessment report based on the ADHD subtype identification results and core deficit dimensions, including the location of cognitive function deficiencies, severity level, and intervention priority.

[0021] In this embodiment, the VR training module includes: The Cold Executive Function Training Unit targets cognitive processing abilities such as working memory and inhibitory control. Through VR virtual scenarios, tasks such as number sorting, image matching, and symbol conversion are designed. The complexity of the tasks is dynamically matched to the child's current working memory span and ability to suppress interference. The following tasks are performed: Attention switching training tasks: In the virtual classroom scenario, children need to switch learning materials of different subjects according to instructions (such as switching from math question cards to Chinese question cards). The frequency of task switching increases as attention switching ability improves. Working memory training task: In a virtual supermarket shopping scenario, children need to remember the names, quantities, and locations of designated items and complete a shopping list within a specified time. The types and quantities of items increase as the breadth of working memory improves. Inhibition control training task: On a virtual road, children need to ignore distracting virtual objects (such as suddenly appearing animated characters) and only perform the target action (such as pressing a button of a specified color to cross the road). The intensity of the distraction increases as the ability of inhibition control improves. The Hot Executive Function Training Unit trains children's ability to regulate emotions and motivations. Through VR virtual scenarios, it designs tasks such as risk decision-making simulation, delayed gratification choices, and emotional conflict resolution. The task scenarios are adapted to children's emotion recognition abilities and impulse control levels, and the following tasks are performed: Risk decision simulation task: In a virtual arcade scenario, children need to choose between "receiving a small reward immediately" and "receiving a large reward later". The reward delay time and the intensity of temptation are adjusted as their impulse control ability improves. Delayed gratification selection task: In a virtual birthday party scenario, children need to wait for their virtual friends to finish the game before sharing the cake together. During the waiting period, they can watch animations but cannot get the cake in advance. The appeal of the animations decreases as their ability to delay gratification improves. Emotional conflict resolution task: In a virtual classroom setting, children need to choose to respond calmly or leave when a classmate provokes them (the virtual character says negative things). The intensity of the virtual character's emotions and the frequency of provocation change as their ability to regulate emotional motivation improves. The VR interaction unit integrates motion capture, spatial positioning, and voice interaction functions to provide children with an immersive virtual scene experience, supporting multimodal interaction methods such as gesture operation and head rotation.

[0022] In this embodiment, the dynamic adjustment module includes: The process data acquisition unit collects real-time data on children's fixation time, saccade count, and pupil diameter during training using an eye tracker, and simultaneously collects data on skin conductance response (GSR) and heart rate variability (HRV) using a wearable device (physiological signal data). The state recognition unit identifies the child's level of attention (e.g., high focus, moderate focus, distraction) based on eye-tracking data and the child's level of emotional arousal (e.g., calm, excitement, anxiety) based on physiological signal data. The adaptive adjustment unit dynamically adjusts the training task parameters based on the state recognition results, including: When attention is distracted, reduce the intensity of task interference or provide voice prompts; when feeling anxious, switch to a low-wake virtual scene; when the task completion rate is below the threshold or the response time exceeds the preset range, automatically reduce the task complexity or provide guidance prompts.

[0023] In this embodiment, the result output module includes: The training report generation unit generates a visual report based on multimodal data during the training process. This report includes curves showing changes in cognitive function indicators (such as the rate of improvement in working memory span and the rate of increase in impulse control scores), subtype feature matching analysis (such as the fit between training tasks and core deficiencies), and training suggestions (such as adjusting training frequency and supplementing auxiliary interventions). The multi-terminal output unit supports the synchronous output of the training report to VR device displays, tablets, and mobile terminals, allowing parents and doctors to view training progress and intervention effects through different terminals.

[0024] In this embodiment, the system also includes a subtype-customized configuration unit, which is communicatively connected to the data evaluation module and the VR training module. This unit is used to assign differentiated training weights and task ratios to the cold and hot dual-executive function training units based on the ADHD subtype identification results. For ADHD-I (Attention Deficit Type), increase the task proportion of cold executive function training units (e.g., 70%), and focus on strengthening working memory and inhibitory control training; For ADHD-HI (hyperactive and impulsive type), increase the proportion of tasks in the hot executive function training unit (e.g., 70%), and focus on strengthening emotional motivation regulation and impulse control training; For ADHD-C (hybrid type), the task ratio of cold and hot executive function training units is dynamically adjusted (e.g., the initial ratio is 1:1, and it is adjusted to 2:3 or 3:2 according to the training progress) to take into account the synergistic improvement of the two executive functions.

[0025] Example 2 See Figure 2 This embodiment provides a training method for a VR training system based on the cognitive function deficit characteristics of children with attention deficit hyperactivity disorder, including the following steps: 1. Multimodal evaluation and personalized solution generation 1.1 Startup of the data acquisition module Children's behavioral characteristics and medical history were collected through structured questionnaires (dynamically generated using a Bayesian algorithm); VR gamification assessment tasks quantify cognitive abilities such as attention shifting / working memory; Wearable devices simultaneously capture facial micro-expressions and skin conductance response (GSR) to establish a physiological baseline; Output: A raw dataset comprising clinical data, neurocognitive data, and physiological signals; 1.2 Evaluation Module Analysis and Decision Making The multi-source data fusion unit extracts feature vectors (such as attention error rate + GSR fluctuation correlation value); The subtype identification unit classifies ADHD subtypes (I / HI / C) through machine learning. Generate personalized reports: Mark core defect dimensions (such as severe working memory deficit in ADHD-C type). Output: Quantitative report on cognitive deficits and intervention priority plan; 2. VR-based classification training and dynamic optimization 2.1 Training Module Task Matching Automatically allocate the cold / hot execution function training ratio according to subtype: ADHD-I type: 70% of cold functional tasks (pay attention to switching training in virtual classrooms); ADHD-HI type: 80% hot functional tasks (arcade delayed gratification decision); ADHD-C type: Dynamically balance hot and cold tasks (initially 1:1, adjusted according to performance). 2.2 Implementation of Immersive Training Cold functional scenarios: Virtual supermarket product memory (working memory enhancement); Virtual road interference filtering animation (suppression control training); Hot functional scenarios: Waiting for the cake at the birthday party (impulse control); How to deal with verbal provocation in the classroom (resolving emotional conflicts); 2.3 Real-time intervention via dynamic adjustment module Eye tracker + physiological sensors monitor training status: Pupil dilation >15% detected → anxiety assessed → switch to low-stimulation forest scene; A sudden 50% increase in skin conductance triggers a voice prompt to "take a deep breath"; Behavioral data feedback: Three consecutive errors → Reduce task complexity (number of items -30%). Accuracy > 90% → Increase challenge level (interference animation speed +200%) 3. Efficacy verification and closed-loop optimization 3.1 Output module generates report Compare key metrics before and after training: Working memory span (5.1→8.2); Impulsive decision-making error rate (42%→18%) Visualize the improvement curves and subtype matching degree analysis; 3.2 Collaborative Module Iteration Startup When the therapeutic threshold is not reached (e.g., improvement rate <30%): Reassess the subtype classification (e.g., reclassify ADHD-C as ADHD-I); Adjust the ratio of cold to hot tasks (from 1:1 to 3:1); Supplemental training (such as adding a neurofeedback module); Parents can receive home training suggestions via a mobile app.

[0026] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A VR training system based on the cognitive deficit characteristics of children with attention deficit hyperactivity disorder (ADHD), characterized in that, include: The data acquisition module is used to collect multimodal data, including clinical behavioral data, neurocognitive gamification assessment data, and physiological signal data. The data assessment module is used to perform fusion analysis on multimodal data, generate individualized cognitive function assessment reports, and identify ADHD subtypes and the degree of cognitive deficits in each dimension. The VR training module is based on the CAPS theoretical model to construct cold and hot dual executive function subtype training units, and matches targeted VR immersive training tasks for children with different ADHD subtypes according to the assessment report. The dynamic adjustment module dynamically optimizes the difficulty, stimulus intensity, and feedback strategy of the training task by collecting physiological signals and behavioral data in real time during the training process. The results output module is used to output training progress reports, cognitive function improvement analysis, and subsequent intervention suggestions.

2. The VR training system based on the cognitive function deficit characteristics of children with attention deficit hyperactivity disorder according to claim 1, characterized in that: The data acquisition module includes a clinical data acquisition unit for collecting children's clinical behavioral data, a neurocognitive data acquisition unit for collecting children's neurocognitive gamification assessment data, and a physiological signal acquisition unit for collecting children's physiological signal data.

3. The VR training system based on the cognitive function deficit characteristics of children with attention deficit hyperactivity disorder according to claim 1, characterized in that: The data evaluation module includes a multi-source data fusion unit for feature extraction and analysis of multimodal data, a subtype identification unit for identifying ADHD subtypes and quantifying the core defect dimensions of each subtype, and an evaluation report generation unit for generating individualized evaluation reports.

4. The VR training system based on the cognitive function deficit characteristics of children with attention deficit hyperactivity disorder according to claim 1, characterized in that: The VR training module includes a cold executive function training unit for training cognitive processing abilities, a hot executive function training unit for training emotional and motivational regulation abilities, and a VR interaction unit for providing children with immersive virtual scene experiences.

5. The VR training system based on the cognitive function deficit characteristics of children with attention deficit hyperactivity disorder according to claim 4, characterized in that: The tasks of the cold executive function training unit include attention switching training tasks, working memory training tasks, and inhibitory control training tasks, while the tasks of the hot executive function training unit include risk decision-making simulation tasks, delayed gratification selection tasks, and emotional conflict resolution tasks.

6. The VR training system based on the cognitive function deficit characteristics of children with attention deficit hyperactivity disorder according to claim 1, characterized in that: The dynamic adjustment module includes a process data acquisition unit, a state recognition unit, and an adaptive adjustment unit. The process data acquisition unit collects eye movement data, skin conductance response data, and heart rate variability data in real time during the child's training process. The state recognition unit identifies the child's level of attention concentration and emotional arousal. The adaptive adjustment unit dynamically adjusts the training task parameters.

7. The VR training system based on the cognitive function deficit characteristics of children with attention deficit hyperactivity disorder according to claim 1, characterized in that: The result output module includes a training report generation unit for generating training visualization reports and a VR device display, tablet computer, and mobile phone terminal for synchronously outputting the training visualization reports to the user's terminal.

8. The VR training system based on the cognitive function deficit characteristics of children with attention deficit hyperactivity disorder according to claim 1, characterized in that: It also includes a subtype-customized configuration unit and a collaborative working module. The subtype-customized configuration unit is used to assign differentiated training weights and task ratios to the cold and hot dual-execution function training units according to the ADHD subtype identification results. The collaborative working module executes the following closed-loop process: Initial assessment phase: Baseline assessment of cognitive function is completed through the data acquisition module and the data assessment module; Intervention training phase: The VR training module initiates subtype training, dynamically adjusting the module's parameters in real time; Effect verification phase: Repeat the evaluation after the training cycle ends, compare the efficacy and iteratively adjust the plan.

Images