Cognitive disorder intervention method and system, storage medium and program product

Abstract

The invention discloses a cognitive disorder intervention method and system, a storage medium and a program product, and relates to the technical field of digital medical treatment, and the method comprises the steps: obtaining a life trajectory material of a to-be-intervened object with driving experience, and generating a personalized virtual driving scene based on the life trajectory material; in the virtual driving scene, a driving training task is issued to the to-be-intervened object, and multi-modal monitoring data of the to-be-intervened object during execution of the driving training task is collected in real time; based on the multi-modal monitoring data, a cognitive load index of the to-be-intervened object is obtained through analysis; according to the cognitive load index, task parameters of the driving training task are adjusted, so that the cognitive load index of the to-be-intervened object is maintained in a preset cognitive load index interval until the to-be-intervened object completes the driving training task. Simulated driving is introduced for intervention, and the problems that in the prior art, due to the lack of situational training, the cognitive decline delay effect is limited, and the intervention efficiency is poor are solved.

Description

Technical Field

[0001] This application relates to the field of digital health technology, and in particular to cognitive impairment intervention methods, cognitive impairment intervention systems, storage media, and computer program products. Background Technology

[0002] Globally, with the accelerating aging of the population, the number of people with mild to moderate cognitive impairment (MCI) and early-stage Alzheimer's disease awaiting intervention is rising sharply. People with cognitive impairment often experience symptoms such as memory loss, slowed thinking, and disorientation, severely impacting their ability to perform daily activities and participate in social activities. This not only seriously erodes the quality of life of those awaiting intervention, gradually causing them to lose their ability to live independently, but also places a heavy economic and psychological burden on family caregivers.

[0003] Studies have shown that drivers' cognitive function declines relatively slowly. Driving behavior prompts the brain to continuously optimize neural network connections when processing complex traffic situations, thus providing a certain degree of protection for cognitive function. Current technologies for cognitive training that incorporate driving scenarios exist, but their scenario construction often uses generalized, standardized road models, making it difficult to effectively activate the relevant memory and cognitive functions of the subjects. Furthermore, the task design in existing cognitive training methods largely relies on fixed scripts or a limited number of pre-set scenarios, failing to reflect individual differences in cognitive ability and load tolerance, and failing to achieve dynamic difficulty adjustment based on real-time physiological and behavioral feedback during training. Therefore, it is difficult to consistently maintain the cognitive training load within the ideal range that induces neuroplasticity.

[0004] In summary, existing methods that incorporate driving scenarios remain limited in their effectiveness in delaying cognitive decline, and the interventions for cognitive impairment are ineffective. Summary of the Invention

[0005] The main objective of this application is to provide a cognitive impairment intervention method, a cognitive impairment intervention system, a storage medium, and a computer program product, aiming to solve the technical problem that existing methods of introducing driving scenarios are still limited in their effectiveness in delaying cognitive decline and have poor cognitive impairment intervention effects.

[0006] To achieve the above objectives, this application proposes a method for intervening in cognitive impairment, the method comprising: Acquire the life trajectory data of the subject to be intervened who has driving experience, and generate a personalized virtual driving scenario based on the life trajectory data; In the virtual driving scenario, a driving training task is issued to the object to be intervened in, and multimodal monitoring data of the object to be intervened in during the execution of the driving training task is collected in real time; Based on the multimodal monitoring data, the cognitive load index of the subject to intervention was obtained through analysis. Based on the cognitive load index, the task parameters of the driving training task are adjusted to maintain the cognitive load index of the subject to be intervened within a preset cognitive load index range until the subject to be intervened completes the driving training task.

[0007] In one embodiment, the step of generating a personalized virtual driving scene based on the life trajectory material includes: The life trajectory materials are reconstructed in three dimensions to generate an initial three-dimensional scene model; Based on the high-frequency trajectory information in the life trajectory material, multiple points of interest of the object to be intervened are marked in the initial three-dimensional scene model; The initial 3D scene model is fine-tuned based on the marked points of interest to generate a personalized virtual driving scene.

[0008] In one embodiment, the step of generating a personalized virtual driving scene based on the life trajectory material further includes: The life trajectory materials are reconstructed in three dimensions to generate an initial three-dimensional scene model; If there are missing scene areas in the life trajectory material, then a similar scene fragment that matches the missing scene area is called from the preset scene completion library; The similar scene fragments are fused into the initial 3D scene model to obtain a fused 3D scene model; Based on the high-frequency trajectory information in the life trajectory material, multiple points of interest of the object to be intervened are marked in the fused 3D scene model; The fused 3D scene model is fine-tuned based on the marked points of interest to generate a personalized virtual driving scene.

[0009] In one embodiment, the step of analyzing and obtaining the cognitive load index of the subject to intervention based on the multimodal monitoring data includes: The multimodal monitoring data is timestamped and fused to obtain a unified time-series data stream. Extract multi-dimensional features related to cognitive load from the time-series data stream, wherein the multi-dimensional features include eye movement dynamics features, driving operation response features and physiological index features; The multi-dimensional features are input into a pre-trained cognitive state assessment model to obtain the cognitive load index of the subject to be intervened.

[0010] In one embodiment, the step of issuing a driving training task to the object to be intervened in the virtual driving scenario includes: In response to the training mode selected by the object to be intervened in the training mode selection interface, a driving training task corresponding to the training mode is issued to the object to be intervened. The training mode includes a lifestyle mode based on a personalized virtual driving scenario and an enjoyment mode based on a preset route. Alternatively, based on the profile information of the target object, the current training mode can be automatically selected and the driving training task corresponding to the current training mode can be issued to the target object. The profile information of the target object includes cognitive level data, training history data, and physiological indicator data.

[0011] In one embodiment, the method further includes: During the execution of the driving training task corresponding to the training mode, the cognitive load index of the subject to be intervened is analyzed at preset time intervals. When the cognitive load index exceeds the upper limit threshold of the preset cognitive load index range, the training mode is switched to rest mode.

[0012] In one embodiment, after the step of real-time acquisition of multimodal monitoring data of the object to be intervened in while performing the driving training task, the method further includes: When the multimodal monitoring data exceeds a predefined safety threshold, an alarm is triggered and an emergency alarm message containing the location information of the object to be intervened is pushed to a preset monitoring terminal. A remote two-way communication channel is established between the monitoring terminal and the training site to view real-time images of the training site and provide medical guidance through the monitoring terminal.

[0013] In addition, to achieve the above objectives, this application also proposes a cognitive impairment intervention system, which includes a client device and a cloud server device. When the cognitive impairment intervention system is executed, it implements the steps of the cognitive impairment intervention method described above.

[0014] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the cognitive impairment intervention method described above.

[0015] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the cognitive impairment intervention method described above.

[0016] One or more technical solutions proposed in this application have at least the following technical effects: Existing methods for introducing driving scenarios remain limited in their effectiveness in delaying cognitive decline, resulting in poor intervention outcomes for cognitive impairment. This application overcomes the disconnect between generalized standardized road models and the life experiences of the subjects by acquiring real-life trajectory materials and generating personalized virtual driving scenarios. It utilizes familiar visual-spatial cues to activate autobiographical memories and related neural circuits. Furthermore, by publishing driving training tasks within personalized virtual driving scenarios and collecting multimodal monitoring data in real time, continuous and objective monitoring of the subjects' cognitive states is achieved, providing data support for accurate assessment. Subsequently, the cognitive load index is obtained from the multimodal monitoring data analysis, transforming the previously vague cognitive state into a quantifiable regulatory indicator, overcoming the limitations of relying on fixed scripts or lagging statistics for difficulty adjustment. Finally, the task parameters of the driving training tasks are dynamically adjusted based on this cognitive load index and maintained within a preset range, ensuring that the cognitive load remains within the ideal range for inducing neural plasticity. This avoids ineffective training under low load and prevents training interruption due to high load. Through the above steps, this application achieves personalized scenario construction and refined control of cognitive load, effectively solving the problem of poor intervention effect caused by the weak relevance of existing technologies to life scenarios and insufficient adaptive ability. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A first flowchart illustrating an embodiment of the cognitive impairment intervention method of this application; Figure 2 This is a schematic diagram of the second process provided in an embodiment of the cognitive impairment intervention method of this application; Figure 3 This is a schematic diagram of the third process provided in an embodiment of the cognitive impairment intervention method of this application; Figure 4 This is a schematic diagram of the fourth process provided in an embodiment of the cognitive impairment intervention method of this application; Figure 5 This is a simplified schematic diagram of the cognitive impairment intervention system of this application.

[0020] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of this application and are not intended to limit this application.

[0022] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0023] Globally, with the accelerating aging of the population, the number of people with mild to moderate cognitive impairment (MCI) and early-stage Alzheimer's disease awaiting intervention is rising sharply. According to statistics from authoritative international research institutions, approximately 10 million new cases of cognitive impairment requiring intervention are added globally each year, with MCI cases accounting for as much as 35%. People with cognitive impairment often experience symptoms such as memory loss, slowed thinking, and disorientation, severely impacting their daily living abilities and participation in social activities. This not only severely erodes the quality of life of those awaiting intervention, gradually causing them to lose their ability to live independently, but also places a heavy economic and psychological burden on family caregivers. From a family perspective, caregivers often need to sacrifice work time or even resign from their jobs to accompany their children to medical appointments and manage their daily lives, leading to a sharp decrease in family income. At the same time, the constant mental stress can easily trigger psychological problems such as anxiety and depression, and family relationships may also become strained due to the caregiving pressure. From a societal perspective, the emergence of a large-scale cognitive impairment population is subtly altering community structures. Neighborhood support networks are facing a double strain on human resources and resources due to surging care demands, and public spaces and transportation facilities urgently need age-friendly renovations. Furthermore, the frequent occurrence of missing persons and accidental injuries among those requiring intervention for cognitive impairment poses new challenges to grassroots governance and public safety. From an economic perspective, those requiring intervention for cognitive impairment require long-term professional care and medical support. It is estimated that by 2030, medical expenses related to cognitive impairment alone will account for more than 15% of total social medical expenditures, undoubtedly placing enormous pressure on the healthcare system. Against this backdrop, developing home-based, low-cost, highly adherent, and easily scalable cognitive intervention methods has become a pressing public health issue globally.

[0024] Numerous studies have shown that, compared to other populations, drivers experience a relatively slower decline in cognitive function. Driving is essentially a complex, multi-domain cognitive task involving the coordinated operation of multiple brain networks, including spatial navigation, working memory, executive control, attention allocation, and emotion regulation. During driving, continuous lane keeping and judging the distance to the vehicle ahead place a high demand on the parietal-occipital visuospatial network, activating the prefrontal cortex's executive control area. Facing sudden stimuli such as traffic lights and pedestrian crossings requires constant switching of attention, strengthening the alertness network. The immediate feedback from speed control and error monitoring activates the dopamine reward pathway, inducing neuroplasticity and thus, to some extent, slowing down the cognitive decline process. Long-term driving behavior prompts the brain to continuously optimize neural network connections when processing complex traffic situations, thereby providing a certain degree of protection for cognitive function.

[0025] Although there are existing technical solutions for cognitive training that incorporate driving scenarios, the scenario construction often uses general standardized road models, which makes it difficult to effectively activate the relevant memory and cognitive functions of the subjects to be intervened. At the same time, the task design in existing cognitive training methods mostly relies on fixed scripts or a limited number of preset scenarios, which cannot reflect the differences in individual cognitive abilities and load tolerance, nor can it achieve dynamic difficulty adjustment based on real-time physiological and behavioral feedback during training. Therefore, it is difficult to continuously maintain the cognitive training load within the ideal range that can induce neuroplasticity.

[0026] To address the limitations and insufficient personalization and adaptability of existing cognitive intervention methods that introduce driving scenarios, this application fundamentally solves the problem of the disconnect between generalized standardized road models and individual life experiences by acquiring real-life trajectory data of the subjects to be intervened in and generating highly personalized virtual driving scenarios. Specifically, the constructed virtual driving scenarios are derived from the subjects' familiar daily environments, eliminating the need for them to undergo cognitive mapping and environmental adaptation processes to new spaces, thereby significantly reducing additional cognitive load and demonstrating significant cognitive-friendly characteristics.

[0027] Building upon this foundation, this application further collects multimodal monitoring data in real time within a personalized virtual driving scenario and dynamically quantifies the cognitive load index accordingly. Based on this cognitive load index, the task parameters of the driving training task are then dynamically adjusted to achieve closed-loop adaptive control of training difficulty, ensuring that the cognitive training process remains within a preset cognitive load range—the optimal load window for inducing neuroplasticity. This application achieves synergistic optimization across three levels: scenario construction, state assessment, and task adjustment. This not only improves the accuracy and individual suitability of the intervention but also provides a feasible path for home-based, low-burden intervention for mild to moderate cognitive impairment.

[0028] It is particularly important to note that one of the core intervention targets of the virtual driving training task designed in this application is the significant decline in spatial orientation and navigation functions in the early stages of mild cognitive impairment. Driving itself is a complex activity that highly relies on spatial navigation ability, involving the integration of egocentric localization and environmental central reference frames, path planning and execution, as well as landmark recognition and route memorization. By completing a series of tasks ranging from simple lane keeping to complex route replanning in a high-fidelity, personalized virtual scenario, the visuospatial perception, mental rotation, path integration, and environmental learning sub-functions of the subject can be systematically and progressively challenged and trained, thereby achieving precise intervention in the spatial cognitive dimension.

[0029] It should be noted that the implementing entity of this embodiment can be a cognitive impairment intervention system. The following uses a cognitive impairment intervention device as an example to describe this embodiment and the following embodiments.

[0030] Additionally, it should be noted that the cognitive impairment intervention system includes client devices (hereinafter referred to as the client) and cloud server devices (hereinafter referred to as the cloud). The aforementioned client devices or cloud server devices can be computing service devices with data processing, network communication, and program execution functions, such as tablet computers, personal computers, and mobile phones.

[0031] Based on this, embodiments of this application provide a method for intervening in cognitive impairment, referring to... Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the cognitive impairment intervention method of this application.

[0032] In this embodiment, the cognitive impairment intervention method includes steps S10 to S40: Step S10: Obtain the life trajectory data of the subject to be intervened who has driving experience, and generate a personalized virtual driving scenario based on the life trajectory data; It should be noted that the cognitive impairment intervention system obtains life trajectory materials from the intervention subjects who have driving experience. These life trajectory materials may include short video clips of the intervention subjects' daily travels, historical map trajectory data, and optional panoramic photos.

[0033] Understandably, the method in this application involves cognitive impairment intervention training in a driving scenario, making it suitable for individuals with prior driving experience. By combining high-fidelity virtual driving scenarios with multi-dimensional cognitive training tasks, it achieves targeted training of cognitive dimensions such as memory, executive function, visuospatial awareness, attention, and emotional regulation, thereby slowing down or improving the cognitive decline process. Applicable scenarios include home settings or other settings such as communities, clinics, and professional rehabilitation institutions. It is a low-cost and easily scalable cognitive impairment intervention method applied to cognitive impairment intervention systems.

[0034] For example, the life trajectory data can cover real landmarks such as the entrance to the community, the supermarket or park that the person to be intervened in, which they frequently visit. In a specific implementation, the cognitive impairment intervention system uses a cloud server device to perform a lightweight neural radiation field algorithm (NeRF-Light) to reconstruct the life trajectory data in three dimensions, generating a 1:1 personalized virtual driving scene based on the real life trajectory of the person to be intervened in.

[0035] Understandably, using real-world scenarios instead of abstract ones is not only because familiarity significantly reduces cognitive load, allowing limited training resources to be focused on the target cognitive dimension, but also because familiar visual-spatial cues can reactivate the "cognitive map" in the subject's long-term memory, thereby promoting the reconsolidation of hippocampal-entorhinal cortex neural circuits and intervening in the core mechanisms of declining spatial navigation ability. Simultaneously, real-world scenarios contain rich multi-sensory details; whether it's a prominent shop sign, a clear zebra crossing, or trees swaying in the wind along the roadside, they all provide multimodal stimulation, further enhancing the transfer effect of training to daily life. More importantly, virtual scenarios constructed based on personal life trajectories have spatial layouts and navigation logic that are completely consistent with the subject's real-world experience, enabling the most effective cognitive training of spatial navigation and orientation abilities. Furthermore, individualized real-world scenarios are often intertwined with the subject's autobiographical memories, triggering deep emotional resonance and significantly improving training motivation and compliance.

[0036] In one feasible implementation, step S10 includes steps S11 to S13: Step S11: Perform 3D reconstruction on the life trajectory material to generate an initial 3D scene model; It should be noted that the life trajectory materials may include, for example, geographic trajectory data generated by map applications, scene video data captured by mobile devices, and scene image data that records the real environment.

[0037] In specific implementations, the life trajectory data typically covers areas where the subject of intervention frequently engages in daily life, such as streets they often travel, supermarket entrances they frequently visit, or community parks.

[0038] Understandably, the collection of life trajectory materials can provide rich and complementary spatial and visual information for subsequent scene reconstruction, thus laying a data foundation for building a highly realistic virtual environment.

[0039] The cognitive impairment intervention system uses collected life trajectory materials to perform 3D reconstruction, generating an initial 3D scene model. In a specific implementation, this process can be achieved using algorithms such as cloud-based lightweight neural radiation field (NeRF-Light) to transform 2D images and video sequences into 3D scenes with geometric structures and surface materials.

[0040] For example, the reconstruction process includes steps such as sparse point cloud generation, Poisson surface reconstruction, and physically rendered material mapping.

[0041] Understandably, through 3D reconstruction technology, cognitive impairment intervention systems can transform the real-world environment familiar to the person being intervened in into an interactive virtual space.

[0042] Step S12: Based on the high-frequency trajectory information in the life trajectory material, mark multiple points of interest of the object to be intervened in the initial three-dimensional scene model; It should be noted that the cognitive impairment intervention system automatically marks multiple points of interest of the subject to intervention in the generated initial 3D scene model based on high-frequency trajectory information extracted from life trajectory materials.

[0043] In a specific implementation, high-frequency trajectory information can be identified by analyzing the recurring stops and path nodes in the historical travel data of the subject to intervention, such as intersections that are passed every day, farmers' markets visited every week, or bus stops that are frequently stopped at.

[0044] For example, a cognitive impairment intervention system can automatically identify and label five to seven key points of interest using a trajectory hotspot analysis algorithm. Understandably, labeling these points of interest can closely link the virtual scenario with the individual's personal life experiences, further enhancing the relevance and immersion of the training context.

[0045] Step S13: Fine-tune the initial 3D scene model based on the marked points of interest to generate a personalized virtual driving scene.

[0046] It should be noted that the cognitive impairment intervention system performs personalized fine-tuning of the initial 3D scene model based on the marked points of interest, thereby generating the final usable virtual driving scene.

[0047] In a specific implementation, the fine-tuning process may include operations such as enhancing visual details in the area surrounding the point of interest, optimizing path connectivity, or unifying the style of scene elements.

[0048] Understandably, through refined adjustments based on points of interest, the cognitive impairment intervention system can output a personalized driving scenario that not only closely matches the individual's life trajectory but also possesses good visual realism and functional integrity, thus providing an effective training scenario for subsequent cognitive training.

[0049] For example, such as Figure 2 As shown, a schematic diagram of scene reconstruction is presented. After obtaining life trajectory materials from the client, they are uploaded to the cloud. The cloud generates an initial 3D scene model based on this. Then, based on the high-frequency trajectory information in the life trajectory materials, multiple points of interest of the object to be intervened are marked in the initial 3D scene model. Next, the initial 3D scene model is fine-tuned according to the marked points of interest, and finally, a personalized virtual driving scene is generated.

[0050] In this embodiment, a cognitive training environment deeply coupled with the individual's autobiographical memory is constructed by transforming the subject's real-life trajectory into an interactive 3D virtual scene. For example, when the subject sees a crossroads they pass every day or the entrance to a supermarket they frequent in the virtual scene, these familiar visual-spatial cues can effectively activate the hippocampal-entorhinal cortex neural circuit, thereby promoting the consolidation and retrieval of spatial memory.

[0051] Understandably, this implementation method, based on points of interest marked with high-frequency trajectory information, not only provides the subjects with navigation paths that match their actual travel patterns, but also triggers deep emotional resonance through scenario recreation, significantly enhancing training motivation and long-term adherence. This approach of constructing a cognitive training environment based on real-life scenarios allows cognitive resources to be more focused on training the target dimension rather than the environmental adaptation process. This maintains an appropriate cognitive load while continuously inducing neuroplasticity, ultimately improving and enhancing multidimensional cognitive functions such as memory, executive function, and visuospatial abilities.

[0052] In one possible implementation, step S10 further includes steps D14 to D18; Step D14: Perform 3D reconstruction on the life trajectory material to generate an initial 3D scene model; It should be noted that step D14 is similar to step S11, and will not be described again here.

[0053] Step D15: If there are missing scene areas in the life trajectory material, then call similar scene segments that match the missing scene areas from the preset scene completion library. It should be noted that after receiving life trajectory materials, the cognitive impairment intervention system will detect whether there are any missing areas in the scene, such as local blanks caused by limited shooting range or incomplete data. For example, a section of a street may lack image materials.

[0054] In a specific implementation, the cognitive impairment intervention system automatically identifies areas with missing scenes by comparing the coverage of the materials with geographic trajectory data, and then retrieves similar scene fragments with matching visual styles and structures from a pre-built scene completion library. For example, it uses architectural models or road elements from other areas of the same city for filling in the missing scenes.

[0055] Understandably, this step can effectively compensate for the deficiencies of the original materials, ensure the continuity and integrity of the initial 3D scene model, and lay the foundation for subsequent personalized adjustments.

[0056] Step D16: Merge similar scene fragments into the initial 3D scene model to obtain a merged 3D scene model; It should be noted that the cognitive impairment intervention system integrates similar scene fragments into the initial 3D scene model reconstructed from life trajectory materials, thereby obtaining a fused 3D scene model.

[0057] In specific implementations, the fusion process involves operations such as geometric alignment, texture blending, and lighting consistency adjustment. For example, a style transfer algorithm is used to make the added fragments visually seamless with the original scene, avoiding abrupt transitions.

[0058] Understandably, this integration not only solves the problem of missing scenes, but also maintains the realism and coherence of the overall environment, making the virtual driving scene closer to the real-life experience of the object to be intervened in terms of spatial layout and visual details.

[0059] Step D17: Based on the high-frequency trajectory information in the life trajectory material, mark multiple points of interest of the object to be intervened in the fused 3D scene model; Step D18: Fine-tune the fused 3D scene model based on the marked points of interest to generate a personalized virtual driving scene.

[0060] It should be noted that steps D17-D18 are similar to steps S12-S13, and will not be repeated here.

[0061] This embodiment addresses the issue of partially missing or incomplete life trajectory footage. By employing scene completion and fusion mechanisms, it overcomes the limitations of raw data quality, constructing a virtual driving environment that maintains both high personalization and complete spatial continuity. For example, when road segments are missing from streets frequently visited by the subject due to limited shooting range, a scene completion library with a consistent style is retrieved to fill in the gaps, ensuring the continuity of the driving path. This completion not only maintains visual consistency but also preserves key spatial relationships and navigation logic within the subject's personal trajectory, enabling subsequent task design based on high-frequency points of interest to realistically reflect their daily life experiences.

[0062] Understandably, this implementation method, by constructing a complete and highly fitting virtual scene, can effectively activate the hippocampal-entorhinal cortex circuit associated with autobiographical memory to strengthen spatial memory, and ensure that the cognitive load is always focused on training in the target dimension rather than adapting to an unfamiliar environment. This can maximize the promotion of neuroplasticity while improving training motivation and compliance, and achieve multi-dimensional and precise intervention for mild to moderate cognitive impairment.

[0063] Step S20: In the virtual driving scenario, a driving training task is issued to the subject to be intervened, and multimodal monitoring data of the subject to be intervened is collected in real time when performing the driving training task. It should be noted that after generating a personalized virtual driving scenario, the cognitive impairment intervention system issues driving training tasks to the individuals to be intervened with.

[0064] Optionally, driving training tasks may include lane keeping, traffic signal recognition, or obstacle avoidance.

[0065] Multimodal monitoring data is collected in real time during driving training tasks. For example, the collected data includes eye-tracking data, driving behavior data, and physiological indicator data. In a specific implementation, a multi-degree-of-freedom control unit, a high-fidelity visual output unit, and a multimodal physiological behavior acquisition array can work collaboratively to ensure the comprehensiveness and synchronization of data collection.

[0066] Understandably, this real-time, multi-dimensional data collection can accurately reflect the cognitive and physiological state of the subjects being intervened in during the training process.

[0067] Step S30: Based on multimodal monitoring data, analyze and obtain the cognitive load index of the subject to be intervened; It should be noted that the cognitive load index of the target group is obtained by analyzing the real-time collected multimodal monitoring data through the AI ​​engine on the cloud server.

[0068] In a specific implementation, the analysis process includes denoising, feature extraction, and cognitive state vector generation of the original multimodal monitoring data, which can integrate various data such as eye-tracking data, driving behavior data, and physiological indicator data.

[0069] For example, the cognitive load index can be dynamically calculated within a sliding time window using a reinforcement learning model to quantify the current cognitive stress level of the subject to intervention.

[0070] Understandably, this cognitive load index serves as a basis for assessing the cognitive state of the individuals to be intervened in and adjusting the difficulty of training, ensuring that the intervention process always revolves around the real-time capabilities of the individuals to be intervened in.

[0071] Step S40: Adjust the task parameters of the driving training task according to the cognitive load index to maintain the cognitive load index of the subject to be intervened within the preset cognitive load index range until the subject to be intervened completes the driving training task.

[0072] It should be noted that the cognitive impairment intervention system dynamically adjusts the task parameters of the driving training task based on the cognitive load index to adjust the difficulty of the driving training task, such as traffic density, speed limit range, or probability of sudden obstacle occurrence. The aim is to maintain the cognitive load index of the subject to intervention within the preset cognitive load index range during the execution of the driving training task.

[0073] Optionally, when the subject of intervention completes the driving training task, a cognitive training report will also be generated for the subject of intervention.

[0074] In a specific implementation, the operation of adjusting the task parameters of the driving training task is completed by the policy network of the cloud server device within seconds and sent to the client device in real time. For example, if the cognitive load index remains high, the cognitive impairment intervention system can automatically adjust the task parameters of the driving training task to reduce the task complexity or insert rest intervals; conversely, it can increase challenging content to stimulate neuroplasticity.

[0075] Understandably, this adaptive adjustment mechanism ensures that the training load always falls within a manageable yet challenging window, thereby continuously promoting the improvement and enhancement of cognitive function.

[0076] This embodiment provides a cognitive impairment intervention method. By acquiring real-life trajectory materials of the subject to intervention and generating personalized virtual driving scenarios, it overcomes the problem of the disconnect between general standardized road models and the subject's life experience. It utilizes familiar visual-spatial cues to activate autobiographical memories and related neural circuits. Furthermore, by issuing driving training tasks in the personalized virtual driving scenarios and collecting multimodal monitoring data in real time, it achieves continuous and objective monitoring of the subject's cognitive state, providing data support for accurate assessment. Subsequently, the multimodal monitoring data is analyzed to obtain a cognitive load index, transforming the originally vague cognitive state into a quantifiable control indicator, overcoming the limitations of relying on fixed scripts or lagging statistics for difficulty adjustment. Finally, based on the cognitive load index, the task parameters of the driving training task are dynamically adjusted and maintained within a preset range, ensuring that the cognitive load is always within the ideal range that can induce neural plasticity. This avoids ineffective training under low load and prevents training interruption caused by high load. Through the above steps, this application achieves personalized scenario construction and refined control of cognitive load, effectively solving the problem of poor intervention effects caused by weak relevance to life scenarios and insufficient adaptive capabilities in existing technologies.

[0077] In one feasible implementation, the cognitive impairment intervention method further includes steps S50-S60: Step S50: During the execution of the driving training task corresponding to the training mode, the cognitive load index of the subject to be intervened is analyzed at preset time intervals. It should be noted that during the execution of driving training tasks corresponding to the training mode, the cognitive impairment intervention system will continuously analyze and obtain the cognitive load index of the subject to intervention at preset time intervals.

[0078] In a specific implementation, the analysis process is based on real-time collected multimodal monitoring data, and the cognitive load index of the subject to be intervened is obtained through the method described in step S30. The preset time interval can be 2 seconds, 3 seconds or other values ​​set based on experience or actual conditions, and is not limited here.

[0079] Understandably, periodic cognitive state assessments can dynamically capture the psychological load level of the subject under the current training task, providing a quantitative basis for subsequent intervention decisions.

[0080] Step S60: When the cognitive load index exceeds the upper limit threshold of the preset cognitive load index range, switch the training mode to the rest mode.

[0081] It should be noted that when the cognitive impairment intervention system detects that the cognitive load index exceeds the upper limit of the preset cognitive load index range, it will immediately switch the training mode to the rest mode.

[0082] In a specific implementation, the switching operation can be triggered within a short time delay. For example, the cognitive impairment intervention system will automatically pause the current driving training task and instead present a soothing natural scene or guide deep breathing training, while continuously monitoring the recovery of physiological indicators such as heart rate.

[0083] In this embodiment, the instantaneous mode switching based on critical load can not only effectively prevent training resistance caused by cognitive over-fatigue, but also maintain the plasticity induction window of the nervous system by providing appropriate recovery intervals, thereby optimizing the long-term intervention effect while ensuring training safety.

[0084] Based on the above embodiments of this application, in another embodiment of this application, the same or similar content as the above embodiments can be referred to the above description, and will not be repeated hereafter. In this regard, step S30 includes steps S31 to S33: Step S31: Perform timestamp alignment and data fusion processing on the multimodal monitoring data to obtain a unified time-series data stream; It should be noted that the cognitive impairment intervention system performs timestamp alignment and data fusion processing on multimodal monitoring data from multiple sources of sensors to form a unified time-series data stream.

[0085] In specific implementations, multimodal monitoring data include, for example, eye-tracking data such as visual tracking sequences collected by an eye tracker, driving behavior data captured by the steering wheel and pedals, and physiological indicators such as heart rate variability recorded by a wrist sensor. These data have different sampling frequencies and time series references. Through hardware-level time synchronization and end-to-end encryption processing, the time deviation between different modal data can be eliminated, and a unified time series data stream that is precisely corresponding in the time domain can be constructed.

[0086] Step S32: Extract multi-dimensional features related to cognitive load from the time-series data stream. The multi-dimensional features include eye-tracking dynamics features, driving operation response features, and physiological index features. It should be noted that the cognitive impairment intervention system extracts multi-dimensional features closely related to cognitive load from the aligned temporal data stream. In specific implementations, multi-dimensional features include, for example, eye-tracking dynamics features (such as fixation duration and saccade complexity), driving operation response features (such as steering angular velocity and braking response time), and physiological indicators (such as the frequency domain component of heart rate variability).

[0087] Understandably, by characterizing the cognitive state of the subjects to be intervened from both behavioral and physiological perspectives, a more comprehensive assessment of cognitive load can be obtained, avoiding misjudgments that may result from a single indicator.

[0088] Step S33: Input the multi-dimensional features into the pre-trained cognitive state assessment model to obtain the cognitive load index of the subject to be intervened.

[0089] It should be noted that the cognitive impairment intervention system inputs the extracted multi-dimensional features into a pre-trained cognitive state assessment model to obtain the cognitive load index of the subject to intervention.

[0090] In a specific implementation, the cognitive state assessment model may optionally be a regression or classification model trained using deep learning or reinforcement learning algorithms based on a large amount of historical training data, which can map high-dimensional features into a quantified cognitive load score.

[0091] For example, the cognitive load index of the subject to intervention can also be obtained based on the mapping table between multidimensional features and cognitive load index.

[0092] In this embodiment, time-stamp alignment and fusion processing of multimodal monitoring data from multiple sources effectively eliminates temporal deviations caused by different sampling frequencies, forming a unified temporal data stream. Multidimensional features, such as eye-tracking dynamics, driving response characteristics, and physiological indicators, are then extracted from the temporal data stream. By inputting the fused multidimensional features into a pre-trained cognitive state assessment model, complex multimodal signals are transformed into an intuitive cognitive load index. This provides a basis for real-time adjustment of training difficulty, ensuring that the training load remains within the optimal range for inducing neural plasticity. Furthermore, it establishes quantifiable assessment indicators, laying a data foundation for long-term intervention effect tracking and improving the accuracy of cognitive impairment interventions.

[0093] Based on the above embodiments of this application, in another embodiment of this application, the same or similar content as the above embodiments can be referred to the above description, and will not be repeated hereafter. Based on this, before step S20, the cognitive impairment intervention method includes steps S21-S22: Step S21: In response to the training mode selected by the subject to intervention in the training mode selection interface, a driving training task corresponding to the training mode is issued to the subject to intervention. The training modes include a lifestyle mode based on a personalized virtual driving scenario and an enjoyment mode based on a preset route. It should be noted that the cognitive impairment intervention system responds to the active selection made by the subject in the training mode selection interface and issues driving training tasks in the corresponding training mode to the subject.

[0094] Additionally, it should be noted that this also includes Mixed Reality (MR) training mode, which is a mode that overlays virtual task elements of driving training tasks onto the real-life environment of the subject to be intervened in.

[0095] In a specific implementation, for the MR mode, the system uses an MR head-mounted display device to integrate and interact in real time with virtual driving task elements (such as navigation arrows, traffic signals, and virtual obstacles) and the real physical environment (such as a living room or rehabilitation room) in which the subject is located. This mode utilizes simultaneous localization and mapping (SLAM) technology for environmental perception and spatial anchoring, ensuring that virtual elements are stably superimposed on the real world. In this mode, the subject can operate a physical steering wheel and pedals in a safe real space to complete virtual driving tasks. This not only enhances the sense of presence and fun of the training but also promotes the generalization and application of cognitive functions through complex scenarios combining virtual and real elements. It is particularly suitable for improving spatial orientation and executive functions in a familiar home environment.

[0096] In specific implementations, the training modes include, for example, a lifestyle mode based on personalized virtual driving scenarios and an enjoyment mode based on preset routes. The lifestyle mode focuses on daily life routes familiar to the subject of intervention, while the enjoyment mode provides a variety of carefully selected scenic routes.

[0097] For example, when the subject of intervention selects the lifestyle mode based on personalized scenarios, the cognitive impairment intervention system effectively activates autobiographical memory and spatial cognitive networks by restoring familiar real landmarks such as supermarkets and neighborhoods; while when the subject selects the enjoyment mode, the preset scenic route expands the breadth of visual scanning through novel stimuli.

[0098] Understandably, by giving the subjects the right to choose, we can respect their personal preferences and train diverse cognitive functions in a targeted manner based on the characteristics of different scenarios, thereby effectively improving the participation and adaptability of the training process.

[0099] Step S22, or, based on the profile information of the subject to be intervened, automatically select the current training mode and issue the driving training task corresponding to the current training mode to the subject to be intervened, wherein the profile information of the subject to be intervened includes cognitive level data, training history data and physiological indicator data.

[0100] It should be noted that the cognitive impairment intervention system also provides an automated training mode selection method, which automatically determines the current training mode and issues the corresponding driving training task based on the profile information of the person to be intervened.

[0101] In a specific implementation, the profile information of the subject to intervention includes multi-dimensional information such as cognitive level data, training history data, and real-time physiological indicator data. By analyzing the profile information of the subject to intervention, the most suitable training mode for the subject to intervention can be determined. For example, a less difficult life version route can be automatically selected for the subject to intervention with a high cognitive load.

[0102] Understandably, this intelligent decision-making mechanism can dynamically adapt to changes in the cognitive state of the subject to intervention, avoiding inappropriate training load caused by subjective selection bias or state fluctuations of the subject to intervention. Thus, while ensuring training safety, it ensures that each intervention can accurately match the immediate needs of the subject to intervention, thereby optimizing the effect of personalized cognitive training.

[0103] In this embodiment, by constructing a training system that includes both a lifestyle version and an enjoyment version, and supplementing it with an intelligent decision-making mechanism, a high degree of adaptation between the training content and the needs of the subjects to be intervened is achieved. It is understandable that this configuration strategy, which balances autonomous selection and intelligent recommendation, respects the subjective preferences of the subjects to be intervened to enhance their motivation to participate, while also achieving precise matching of training intensity through data-driven approaches. This ensures that cognitive intervention remains within a safe and effective workload range while maintaining the novelty and challenge of training, ultimately promoting the improvement of multi-dimensional cognitive functions.

[0104] In one feasible implementation, after step S20, the cognitive impairment intervention method includes steps D21-D22: Step D21: When the multimodal monitoring data exceeds the predefined safety threshold, an alarm is triggered and an emergency alarm message containing the location information of the object to be intervened is pushed to the preset monitoring terminal. It should be noted that when the cognitive impairment intervention system detects that multimodal monitoring data (such as heart rate, movement posture, or eye movement trajectory) exceeds their respective predefined safety thresholds, it will automatically trigger an alarm mechanism and push an emergency alarm message containing the location information of the person to be intervened to a preset monitoring terminal. Optionally, the alarm message may also include a timestamp, real-time video clips, and physiological data.

[0105] For example, the safety threshold may be dynamically set based on the baseline of the person to be intervened. For instance, if the heart rate is detected to be higher than 180 bpm or the fall posture lasts for more than 3 seconds, a local audible and visual alarm will be activated immediately and an SOS message will be sent to the family member's mobile phone or cloud monitoring platform via a safety protocol.

[0106] Understandably, this kind of safety monitoring can effectively prevent accidents during training, promptly notify caregivers to intervene, and thus ensure the physical and mental safety of the person to be intervened in.

[0107] Step D22: Establish a remote two-way communication channel between the monitoring terminal and the training site to view real-time images of the training site and provide medical guidance through the monitoring terminal.

[0108] It should be noted that after the alarm is triggered, a remote two-way communication channel will be automatically established between the monitoring terminal and the training site, enabling the monitoring party to view the real-time video of the training site and provide medical guidance.

[0109] In a specific implementation, this communication channel can be achieved through a highly reliable lightweight messaging protocol and audio / video streaming technology. For example, after receiving an alarm in the WeChat mini-program, family members can initiate a two-way audio / video call with one click to remotely observe the status of the person to be intervened and interact with the scene. At the same time, the cognitive impairment intervention system can automatically connect to the nearest on-duty doctor to join the conversation.

[0110] Optionally, the communication channel can integrate data sharing functionality, allowing healthcare professionals to access the historical training records and physiological trends of the subjects to be intervened in real time to optimize guidance strategies.

[0111] Understandably, this real-time remote intervention mechanism not only breaks down spatial limitations and provides professional support in emergencies, but also alleviates caregiving anxiety through visual communication methods and enhances the credibility of the overall training system.

[0112] In this implementation, by constructing an intelligent safety monitoring and emergency response mechanism, comprehensive protection is provided for cognitive training in the home environment. This not only effectively reduces safety risks during training alone, but also significantly enhances the psychological safety of both the individuals to be intervened and their caregivers through the immediate accessibility of professional support. Thus, while ensuring the safety of training, it enhances the trust and acceptance of the cognitive impairment intervention system, laying the foundation for the home-based promotion of cognitive impairment intervention.

[0113] In a preferred embodiment, such as Figure 3 As shown, after automatically or manually selecting a training mode (Lifestyle Mode or Enjoyment Mode), the corresponding driving training task is issued to the subject to be intervened, and the corresponding training scenario is loaded; while the subject to be intervened performs the driving training task, multimodal monitoring data is collected in real time; then, based on the analysis of the multimodal monitoring data, the cognitive load index of the subject to be intervened is obtained, and the cognitive load of the subject to be intervened is evaluated based on the cognitive load index; when the driving training task is completed, a cognitive training report is automatically generated.

[0114] In such Figure 4 In one preferred embodiment shown, the cognitive impairment intervention method includes: Step 101: The subject to intervention registers and logs in; if it is the first time the subject to intervention logs in, the registration information of the subject to intervention is collected and stored in the cloud; after successful login, the baseline assessment is completed on the client and the assessment results are sent to the cloud. The baseline assessment refers to establishing an initial profile of the cognitive ability of each subject to intervention and quantifying the baseline level of the subject to intervention in key cognitive dimensions (such as memory, executive function, visuospatial ability, attention and emotion regulation).

[0115] Step 102: The person to be intervened or their family members can upload life trajectory materials to the cloud through the client. Life trajectory materials can include short videos of 15 to 30 seconds taken casually (covering the entrance and exit of the community, frequently visited supermarkets, frequently visited parks, etc.); or historical trajectory data retained by frequently used map applications; or panoramic photos of city streets that have been visited, etc.

[0116] The cloud-based NeRF-Light service container performs sparse reconstruction, dense point cloud processing, surface reconstruction, and material mapping. It also uses semantic segmentation to identify buildings, lane lines, traffic lights, and speed limit signs, and saves them as vectorized, editable layers. Next, it marks points of interest based on high-frequency trajectory information in the life trajectory materials. When the life trajectory materials are incomplete, it calls similar city segments from a preset scene completion library for style transfer to ensure visual consistency. Finally, it obtains a scene package, which is a personalized virtual driving scene.

[0117] The virtual driving scenario is pushed to the client. Optionally, the person to be intervened or their family members can be asked to rescan at regular intervals to incrementally update the life trajectory data and keep it synchronized with reality.

[0118] Step 103: Before performing the driving training task, the interactive terminal devices connected to the client via wires or wirelessly, such as VR (Virtual Reality) headsets or TV / projection screens, force feedback steering wheels, accelerator pedals, brake pedals, clutch pedals, six-degree-of-freedom controllers, and wrist-type heart rate / blood oxygen sensors, perform self-tests. These devices sequentially test the steering wheel ±900° travel, pedal linearity, headset / TV display synchronization, and sensor sampling rate. Once all tests pass, the training interface is entered.

[0119] Step 104: Before conducting the driving training task, the person to be intervened or their family member can choose between the Lifestyle version, the Enjoyment version, or the Mixed Reality (MR) mode, and can freely switch between these modes according to the training plan or real-time status. Specifically, the Lifestyle version is the default training mode, which includes a 3-5km daily route. The core task load focuses on training spatial orientation and navigation capabilities, while also taking into account executive functions and contextual memory. It is a 1:1 three-dimensional map constructed based on the neighborhood, streets, supermarkets, parking lots, etc., which are familiar to the person to be intervened. The virtual driving distance is 3-5km, including 5-7 navigation decision points. The driving training task includes lane keeping, traffic signal recognition, route replanning, and sudden obstacle avoidance.

[0120] The Enjoy Edition includes 30+ carefully selected self-driving routes, each 30-50km long, which can be divided into training segments; driving training tasks include the cognitive impairment intervention system randomly inserting speed limit signs ranging from 40-120km / h, requiring the subject to adjust the vehicle speed in real time; sudden sheep flocks, rockfalls, and temporary construction, testing the subject's executive function and reaction inhibition; completing reverse parking within a limited time and taking photos to check in, strengthening spatial memory.

[0121] Meanwhile, the cloud records preferences, allowing the system to select the appropriate training mode for the day based on the individual's profile during subsequent cognitive impairment interventions. The cognitive impairment intervention system then loads the corresponding driving training task according to the selected training mode.

[0122] During the driving training task, the client collects data from the steering wheel, pedals, and handgrip in parallel, along with multimodal monitoring data such as eye movement, heart rate, and skeletal flow. After timestamp alignment and encryption, the data is uploaded to the cloud. The cloud analyzes the multimodal monitoring data to obtain the cognitive load index of the subject to be intervened. The cloud-based reinforcement learning model evaluates the cognitive load index every 2 seconds. When the cognitive load index falls outside the preset cognitive load index range, the task parameters of the driving training task are adjusted in real time, such as increasing or decreasing the density of speed limit signs, inserting sudden obstacles, or adjusting the route length.

[0123] After the driving training task is completed, the client can immediately display the score and heart rate curve, and automatically enter the rest countdown after a preset time interval (such as 5 seconds); it can also automatically generate a multi-dimensional report (five-dimensional scores of memory, executive, visuospatial, attention and emotion); and it can also push it to the family members of the person to be intervened or the family doctor of the person to be intervened.

[0124] Step 105: During cognitive training, the system will also continuously monitor whether the multimodal monitoring data exceeds their respective safety thresholds. If so, an alarm will be triggered, and an emergency alarm message containing the location information of the person to be intervened will be pushed to a preset monitoring terminal. For example, a fall detection camera and a heart rate wristband are also provided. The fall detection camera collects key skeletal points at 30fps. When a "fall posture" is detected for ≥3 seconds or the heart rate is >180 bpm, an audible and visual alarm will be triggered, and the location information and a real-time screenshot will be pushed to the family members or family doctor of the person to be intervened. The family members or family doctor can open a two-way audio and video channel with one click to remotely view the real-time video and provide medical guidance.

[0125] In addition, the cloud can store all data in three levels: hot (3 days), warm (30 days), and cold (≥5 years), and record every access through blockchain hash; the client can download or delete the data related to it at any time.

[0126] Additionally, it should be noted that the steps of the above-mentioned cognitive impairment intervention method can also be implemented when the client device is executed independently.

[0127] In such Figure 5 In one preferred embodiment, a schematic diagram of a cognitive impairment intervention system is shown, including a client device and a cloud server device. The two devices can communicate bidirectionally and transmit data to each other to jointly implement the cognitive impairment intervention method.

[0128] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the cognitive impairment intervention method of this application. Any simple variations based on this technical concept, such as the interaction and combination of various embodiments, are all within the protection scope of this application.

[0129] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0130] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to perform the cognitive impairment intervention method in the above embodiments.

[0131] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0132] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the computer to be intervened on, partially on the computer to be intervened on, as a standalone software package, partially on the computer to be intervened on and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the computer to be intervened on any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0133] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0134] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described cognitive impairment intervention method. This addresses the technical problem that existing methods involving the introduction of driving scenarios are still limited in their effectiveness in delaying cognitive decline, resulting in poor intervention effects for cognitive impairment. Compared to the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the cognitive impairment intervention method provided in the above embodiments, and will not be elaborated upon here.

[0135] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the cognitive impairment intervention method described above.

[0136] The computer program product provided in this application can solve the technical problem that existing methods for introducing driving scenarios are still limited in their effectiveness in delaying cognitive decline, resulting in poor intervention effects for cognitive impairment. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the cognitive impairment intervention methods provided in the above embodiments, and will not be repeated here.

[0137] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for intervening in cognitive impairment, characterized in that, The intervention methods for cognitive impairment include: Acquire the life trajectory data of the subject to be intervened who has driving experience, and generate a personalized virtual driving scenario based on the life trajectory data; In the virtual driving scenario, a driving training task is issued to the object to be intervened in, and multimodal monitoring data of the object to be intervened in during the execution of the driving training task is collected in real time; Based on the multimodal monitoring data, the cognitive load index of the subject to intervention was obtained through analysis; Based on the cognitive load index, the task parameters of the driving training task are adjusted to maintain the cognitive load index of the subject to be intervened within a preset cognitive load index range until the subject to be intervened completes the driving training task.

2. The cognitive impairment intervention method as described in claim 1, characterized in that, The step of generating a personalized virtual driving scene based on the life trajectory data includes: The life trajectory materials are reconstructed in three dimensions to generate an initial three-dimensional scene model; Based on the high-frequency trajectory information in the life trajectory material, multiple points of interest of the object to be intervened are marked in the initial three-dimensional scene model; The initial 3D scene model is fine-tuned based on the marked points of interest to generate a personalized virtual driving scene.

3. The cognitive impairment intervention method as described in claim 1, characterized in that, The step of generating a personalized virtual driving scene based on the life trajectory data further includes: The life trajectory materials are reconstructed in three dimensions to generate an initial three-dimensional scene model; If there are missing scene areas in the life trajectory material, then a similar scene fragment that matches the missing scene area is called from the preset scene completion library; The similar scene fragments are fused into the initial 3D scene model to obtain a fused 3D scene model; Based on the high-frequency trajectory information in the life trajectory material, multiple points of interest of the object to be intervened are marked in the fused 3D scene model; The fused 3D scene model is fine-tuned based on the marked points of interest to generate a personalized virtual driving scene.

4. The cognitive impairment intervention method as described in claim 1, characterized in that, The step of analyzing and obtaining the cognitive load index of the subject to intervention based on the multimodal monitoring data includes: The multimodal monitoring data is timestamped and fused to obtain a unified time-series data stream. Extract multi-dimensional features related to cognitive load from the time-series data stream, wherein the multi-dimensional features include eye movement dynamics features, driving operation response features and physiological index features; The multi-dimensional features are input into a pre-trained cognitive state assessment model to obtain the cognitive load index of the subject to be intervened.

5. The cognitive impairment intervention method as described in claim 1, characterized in that, The step of issuing a driving training task to the object to be intervened in the virtual driving scenario includes the following prior steps: In response to the training mode selected by the object to be intervened in the training mode selection interface, a driving training task corresponding to the training mode is issued to the object to be intervened. The training mode includes a lifestyle mode based on a personalized virtual driving scenario and an enjoyment mode based on a preset route. Alternatively, based on the profile information of the target object, the current training mode can be automatically selected and the driving training task corresponding to the current training mode can be issued to the target object. The profile information of the target object includes cognitive level data, training history data, and physiological indicator data.

6. The cognitive impairment intervention method as described in claim 1, characterized in that, The method further includes: During the execution of the driving training task corresponding to the training mode, the cognitive load index of the subject to intervention is analyzed at preset time intervals. When the cognitive load index exceeds the upper limit threshold of the preset cognitive load index range, the training mode is switched to rest mode.

7. The cognitive impairment intervention method as described in claim 1, characterized in that, Following the step of real-time acquisition of multimodal monitoring data of the subject to intervention during the driving training task, the method further includes: When the multimodal monitoring data exceeds a predefined safety threshold, an alarm is triggered and an emergency alarm message containing the location information of the object to be intervened is pushed to a preset monitoring terminal. A remote two-way communication channel is established between the monitoring terminal and the training site to view real-time images of the training site and provide medical guidance through the monitoring terminal.

8. A cognitive impairment intervention system, characterized in that, The cognitive impairment intervention system includes a client device and a cloud server device, and when the cognitive impairment intervention system is executed, it implements the steps of the cognitive impairment intervention method as described in any one of claims 1 to 7.

9. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the cognitive impairment intervention method as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the cognitive impairment intervention method as described in any one of claims 1 to 7.

Images