An intelligent medication assistance system based on cognitive impairment patients
By leveraging the multimodal interaction and adaptive reminder functions of the intelligent drug assistance system, the problem of insufficient reminders for visually and hearing impaired patients in existing devices has been solved. This enables the systematic management of drug identification and medication records, thereby improving medication adherence and safety for patients with cognitive impairment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- THE SIXTH AFFILIATED HOSPITAL OF SUN YAT SEN UNIV
- Filing Date
- 2026-04-07
- Publication Date
- 2026-07-10
AI Technical Summary
Existing smart medication reminder devices are not effective for reminding patients with visual impairments and hearing loss. They cannot adapt and adjust themselves, lack medication recognition functions, and their medication records are not systematic, making it impossible to effectively prevent incorrect or missed doses, thus affecting treatment outcomes.
It employs a terminal interaction layer, a perception and decision-making layer, a cloud service layer, and a user management layer to achieve multimodal interaction, drug identification, medication behavior verification, adaptive reminders, medication records, and strategy optimization. It also dynamically adjusts reminder strategies and interaction methods using a deep learning model.
It improves medication adherence in patients with cognitive impairment, reduces missed and incorrect doses, provides a systematic medication record, and enhances medication safety and treatment support.
Smart Images
Figure CN122369788A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of medical technology, and in particular to an intelligent drug-assisted system for patients with cognitive impairment. Background Technology
[0002] Patients with cognitive impairment (such as Alzheimer's disease) have poor medication adherence due to memory decline, poor concentration, and other problems. According to clinical data, the rates of missed, incorrect, and duplicate medications among patients with cognitive impairment are much higher than those among ordinary patients, which seriously affects the treatment effect and may even cause adverse drug reactions.
[0003] While smart pillboxes and medication reminder devices can provide timed reminders, they have the following shortcomings: 1. Existing medication reminder devices mostly use a single reminder method. For patients who also have visual impairment, hearing loss, or reduced ability to take medication, the reminders may be ineffective or undetectable, resulting in poor practicality.
[0004] 2. The reminder rules of existing devices are mostly preset with fixed parameters, which cannot be adaptively adjusted and optimized based on the patient's historical medication adherence data. It is difficult to form an intervention strategy for continuous improvement based on the patient's actual medication situation and response status, resulting in limited long-term effectiveness.
[0005] 3. Most reminder devices can only provide timed reminders and do not have the function of automatically identifying drug information and comparing prescriptions. They cannot effectively prevent patients from taking the wrong medicine or taking the wrong medicine, and it is difficult to effectively solve the risk of medication mis-dosing and accidental ingestion.
[0006] 4. While most existing systems record medication use, they lack a systematic, traceable, and searchable medication log, which fails to provide complete and accurate medication data support for medical staff and their families, hindering diagnosis, treatment assessment, and care decisions.
[0007] Therefore, designing an intelligent drug-assisted system for patients with cognitive impairment that can solve the above-mentioned technical problems is a technical issue that needs to be addressed. Summary of the Invention
[0008] To address the aforementioned problems, the present invention aims to provide an intelligent medication assistance system for patients with cognitive impairment, thereby enabling safe, standardized, intelligent, and humane medication management for these patients.
[0009] To achieve the above objectives, the present invention adopts the following technical solution: including a terminal interaction layer, a perception and decision-making layer, a cloud service layer, and a user management layer; The terminal interaction layer is used to enable multimodal interaction between the patient and the system; The perception and decision-making layer is used to perform drug identification, medication behavior verification, and reminder strategy generation; The cloud service layer is used to store medication history and patient records, and supports data retrospective analysis and model optimization; The user management layer is used to adjust the configuration of medication plans, monitor medication status, and adjust auxiliary strategies. The perception and decision-making layer includes an intelligent medication management module and a personalized adaptation module. The intelligent medication management module is used to perform multiple reminders, drug identification and verification, medication behavior recording, and medication record review according to a preset medication plan. The personalized adaptation module is used to dynamically adjust the reminder method and interaction strategy according to the patient's degree of cognitive impairment and visual impairment level.
[0010] Furthermore, the intelligent medication management module includes a multi-reminder submodule, a drug recognition submodule, and a medication record submodule; The multi-reminder submodule is used to perform multi-level reminders via voice, vibration, and light according to preset times; wherein, the reminder strategy of the multi-reminder submodule is dynamically optimized by a cloud-based deep learning model based on the patient's historical medication adherence data; The drug identification submodule is used to identify drug information through image acquisition and machine learning and compare it with the medication regimen. The medication record submodule is used to automatically record the medication time, medication, execution status, and supports historical backtracking.
[0011] Furthermore, the personalized adaptation module includes an obstacle level recognition submodule and a strategy adaptation submodule; The disability level identification submodule is used to classify the patient's disability level through initial cognitive and visual tests and analysis of daily behavior data; The strategy adaptation submodule is used to dynamically adjust the reminder method and interaction strategy according to the obstacle level.
[0012] Furthermore, the cloud service layer includes a data storage module, a model optimization module, and a communication module; The data storage module is used to encrypt and store medication records and patient files, and supports data backtracking by time and event dimensions; The model optimization module continuously trains and updates the deep learning model based on accumulated medication data to improve the system's adaptability and accuracy. The communication module is used for data synchronization and command transmission with the terminal interaction layer and the user management layer.
[0013] Furthermore, the user management layer includes a configuration module, a monitoring module, and a policy adjustment module; The configuration module is used to input patient information, medication plan, impairment level, and reminder rules; The monitoring module is used to push notifications of abnormal events and provide medication log query functions; The strategy adjustment module is used to dynamically adjust the auxiliary strategy based on patient medication data and feedback.
[0014] Furthermore, the terminal interaction layer includes one or more combinations of a smart pillbox, a voice prompt device, a vibration prompt device, and a PDA handheld terminal.
[0015] The present invention has the following beneficial effects: 1-This invention sets up an intelligent medication management module, which dynamically optimizes the reminder strategy through multiple reminder sub-modules and cloud models, and combines drug identification and medication record retrieval to effectively reduce missed or incorrect doses in patients with cognitive impairment.
[0016] 2. This invention features a personalized adaptation module that dynamically adjusts interaction strategies based on the patient's disability level, improving the patient's user experience and cooperation. Through the remote monitoring and strategy adjustment functions of the user management layer, the system can monitor the patient's medication status in real time, assisting in treatment strategies. Attached Figure Description
[0017] Figure 1 This is a schematic diagram of the present invention. Detailed Implementation
[0018] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments: See Figure 1 As shown, the solution includes a terminal interaction layer, a perception and decision-making layer, a cloud service layer, and a user management layer. The terminal interaction layer is used to enable multimodal interaction between the patient and the system; The perception and decision-making layer is used to perform drug identification, medication behavior verification, and reminder strategy generation; The cloud service layer is used to store medication history and patient records, and supports data retrospective analysis and model optimization; The user management layer is used to adjust the configuration of medication plans, monitor medication status, and adjust auxiliary strategies. The perception and decision-making layer includes an intelligent medication management module and a personalized adaptation module. The intelligent medication management module is used to perform multiple reminders, drug identification and verification, medication behavior recording, and medication record review according to a preset medication plan. The personalized adaptation module is used to dynamically adjust the reminder method and interaction strategy according to the patient's degree of cognitive impairment and visual impairment level.
[0019] Furthermore, the intelligent medication management module includes a multi-reminder submodule, a drug identification submodule, and a medication record submodule; The multi-reminder submodule is used to execute multi-level reminders via voice, vibration, and light according to preset times. The reminder strategy of this submodule is dynamically optimized by a cloud-based deep learning model based on the patient's historical medication adherence data. Specifically, the initial preset reminder rules are: a primary reminder 10 minutes before medication, an intermediate reminder at medication time, and an escalation reminder 5 minutes after medication is missed. The primary reminder triggers only a flashing light and slight vibration; the intermediate reminder triggers a flashing light, voice prompt, and vibration; and the escalation reminder triggers a flashing red LED light, a looping voice prompt, and continuous vibration, while simultaneously pushing warning information to the user management layer via the communication module. The reminder frequency and lead time can be dynamically adjusted based on the patient's medication adherence.
[0020] The drug identification submodule is used to identify drug information and compare it with the medication regimen through image acquisition and machine learning. Specifically, it uses an image acquisition camera to acquire drug images, and a lightweight machine learning model (based on CNN convolutional neural network) is used to pre-train a drug database (containing more than 50 commonly used drugs such as common cognitive impairment treatment drugs, antihypertensive drugs, and hypoglycemic drugs, covering features such as drug appearance, shape, color, and engraving). After image acquisition, the model extracts drug features and compares them with the patient's preset medication regimen synchronized in the cloud. If the comparison matches, the terminal interaction layer is triggered to perform an operation; if the comparison does not match, the terminal interaction layer is prohibited from performing an operation, and the abnormal information is recorded and synchronized to the cloud.
[0021] The medication record submodule is used to automatically record medication time, medication, and execution status, and supports historical backtracking. It automatically records key information for each medication administration, including medication time, medication name, medication dosage, and execution status (normal administration, missed dose, incorrect dose), and the recorded data is synchronized to the cloud service layer in real time; it supports historical backtracking queries by time and event dimensions through PDA handheld terminals or user management systems.
[0022] Furthermore, the personalized adaptation module includes an impairment level identification submodule and a strategy adaptation submodule; combining the two types of data, the patient's cognitive impairment level is classified into mild, moderate, and severe, and the visual impairment level is classified into normal, mild visual impairment, moderate visual impairment, and severe visual impairment.
[0023] The disability level identification submodule is used to classify the patient's disability level through initial cognitive and visual tests and analysis of daily behavior data; The strategy adaptation submodule is used to dynamically adjust the reminder method and interaction strategy according to the obstacle level.
[0024] Furthermore, the cloud service layer includes a data storage module, a model optimization module, and a communication module; The data storage module is used to encrypt and store medication records and patient files, and supports data backtracking by time and event dimensions; The model optimization module continuously trains and updates the deep learning model based on accumulated medication data to improve the system's adaptability and accuracy. The communication module is used for data synchronization and command transmission with the terminal interaction layer and the user management layer.
[0025] Furthermore, the user management layer includes a configuration module, a monitoring module, and a policy adjustment module; The configuration module is used to input patient information, medication regimens, impairment levels, and reminder rules. Medical staff and family members input patient information (name, age, impairment type, impairment level, etc.), medication regimens (drug name, dosage, medication time, medication frequency, etc.), impairment levels, and reminder rules (initial reminder lead time, reminder intensity, etc.) through a web page or mobile app. After input, the data is synchronized to the cloud service layer through the communication module, and then to the perception and decision layer to generate patient-specific assistance strategies, while also supporting real-time modification of the medication regimen.
[0026] The monitoring module is used to push notifications of abnormal events and provide medication log query functions; The strategy adjustment module is used to dynamically adjust the auxiliary strategy based on patient medication data and feedback.
[0027] Furthermore, the terminal interaction layer includes one or more combinations of a smart pillbox, a voice prompt device, a vibration prompt device, and a PDA handheld terminal.
[0028] Specifically, through the configuration module of the user management layer, patients' basic information, medication plans, impairment levels, and initial reminder rules are entered. After configuration, the data is synchronized to the perception and decision layer through the communication module of the cloud service layer. The personalized adaptation module of the perception and decision layer automatically adjusts the interaction strategy (enlarges the PDA font and increases the frequency of voice reminders) according to the patient's impairment level. The intelligent medication management module loads the medication plan and initial reminder strategy, and the system completes initialization.
[0029] Ten minutes before the preset medication time (e.g., 8:00 AM), the multi-level reminder submodule of the perception and decision layer triggers a multi-level reminder. After hearing the reminder, the patient takes out the medication and places it in the image acquisition area of the smart pillbox. The drug recognition submodule captures the image of the medication through the camera and uses a machine learning model to identify the medication information (identified as donepezil hydrochloride tablets). It compares the information with the preset medication plan. If the comparison is consistent, the terminal interaction layer is triggered to perform the operation; if the comparison is inconsistent, the terminal interaction layer is prohibited from performing the operation, and the abnormal information is recorded and synchronized to the cloud.
[0030] During system operation, the impairment level identification submodule continuously collects patients' daily operation data (such as medication confirmation response time, PDA operation accuracy, etc.) and dynamically updates the patient's impairment level based on the initial test data. If the patient's recent operation accuracy declines, the level is adjusted to moderate cognitive impairment, and the strategy adaptation submodule automatically increases the reminder frequency and simplifies the interaction steps. At the same time, the model optimization module in the cloud service layer optimizes the reminder strategy of the multi-reminder submodule based on the patient's recent medication data (such as no missed doses or incorrect doses), adjusting the lead time for the primary reminder to 5 minutes to reduce unnecessary reminder interference.
[0031] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0032] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0033] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0034] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1The steps of the function specified in one or more boxes.
[0035] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.
Claims
1. An intelligent drug-assisted system for patients with cognitive impairment, characterized in that: This includes the terminal interaction layer, the perception and decision-making layer, the cloud service layer, and the user management layer; The terminal interaction layer is used to enable multimodal interaction between the patient and the system; The perception and decision-making layer is used to perform drug identification, medication behavior verification, and reminder strategy generation; The cloud service layer is used to store medication history and patient records, and supports data retrospective analysis and model optimization; The user management layer is used to adjust the configuration of medication plans, monitor medication status, and adjust auxiliary strategies. The perception and decision-making layer includes an intelligent medication management module and a personalized adaptation module. The intelligent medication management module is used to perform multiple reminders, drug identification and verification, medication behavior recording, and medication record review according to the preset medication plan; the personalized adaptation module is used to dynamically adjust the reminder method and interaction strategy according to the patient's degree of cognitive impairment and visual impairment level.
2. The intelligent drug-assisted system for patients with cognitive impairment according to claim 1, characterized in that: The intelligent medication management module includes a multi-reminder submodule, a drug identification submodule, and a medication record submodule; The multi-reminder submodule is used to perform multi-level reminders via voice, vibration, and light according to preset times; wherein, the reminder strategy of the multi-reminder submodule is dynamically optimized by a cloud-based deep learning model based on the patient's historical medication adherence data; The drug identification submodule is used to identify drug information through image acquisition and machine learning and compare it with the medication regimen. The medication record submodule is used to automatically record the medication time, medication, execution status, and supports historical backtracking.
3. The intelligent drug-assisted system for patients with cognitive impairment according to claim 1, characterized in that: The personalized adaptation module includes an obstacle level recognition submodule and a strategy adaptation submodule. The disability level identification submodule is used to classify the patient's disability level through initial cognitive and visual tests and analysis of daily behavior data; The strategy adaptation submodule is used to dynamically adjust the reminder method and interaction strategy according to the obstacle level.
4. The intelligent drug-assisted system for patients with cognitive impairment according to claim 1, characterized in that: The cloud service layer includes a data storage module, a model optimization module, and a communication module; The data storage module is used to encrypt and store medication records and patient files, and supports data backtracking by time and event dimensions; The model optimization module continuously trains and updates the deep learning model based on accumulated medication data to improve the system's adaptability and accuracy. The communication module is used for data synchronization and command transmission with the terminal interaction layer and the user management layer.
5. The intelligent drug-assisted system for patients with cognitive impairment according to claim 1, characterized in that: The user management layer includes a configuration module, a monitoring module, and a policy adjustment module; The configuration module is used to input patient information, medication plan, impairment level, and reminder rules; The monitoring module is used to push notifications of abnormal events and provide medication log query functions; The strategy adjustment module is used to dynamically adjust the auxiliary strategy based on patient medication data and feedback.
6. The intelligent drug-assisted system for patients with cognitive impairment according to claim 1, characterized in that: The terminal interaction layer includes one or more combinations of a smart pillbox, a voice prompt device, a vibration prompt device, and a PDA handheld terminal.