Family participation type vascular intervention diagnosis and treatment postoperative cerebral infarction early warning method and device

Abstract

The invention discloses a family member participation type vascular intervention diagnosis and treatment postoperative cerebral infarction early warning method. The method comprises the following steps that patient data are acquired through a client module based on wrist strap data of a patient, and a cerebral infarction monitoring module is started; based on the patient data, binding the patient data with a client module through living body detection and information authentication of a cerebral infarction monitoring module, and starting a monitoring process of the cerebral infarction monitoring module; the behavior data and cerebral infarction risk level data of the patient are obtained through the monitoring process, and the monitoring process is a periodic task and is started and executed at fixed time in 1 hour, 3 hours, 6 hours, 12 hours and 24 hours after the operation; and acquiring early warning data through a medical care response module based on the patient behavior data and the patient cerebral infarction risk level data, wherein the early warning data is used for providing a postoperative cerebral infarction intervention voucher for the patient by medical staff.

Description

Technical Field

[0001] The present invention belongs to the technical field of early warning of cerebral infarction after vascular interventional diagnosis and treatment, and in particular relates to a method and device for early warning of cerebral infarction after vascular interventional diagnosis and treatment with family participation. Background Art

[0002] Vascular intervention, as a key technique in modern medicine, plays an increasingly important role in treating cerebrovascular and cardiovascular diseases. With the continuous advancement of interventional device technology and the increasing sophistication of surgical techniques, vascular intervention has become one of the preferred treatments for cerebrovascular diseases such as acute stroke, aneurysms, and arteriovenous malformations. According to statistics, over one million vascular interventions are performed annually in my country, with an annual increase of 15%. However, despite the advantages of minimal invasiveness and rapid recovery, vascular intervention carries a high risk of complications within the first 24 hours after surgery. Cerebral infarction is one of the most serious complications, occurring in approximately 2-8% of cases, severely impacting patient prognosis and quality of life. Existing studies have shown that approximately 60-70% of post-interventional cerebral infarctions occur within 6 hours of surgery, and 80-85% occur within 24 hours. Effective monitoring and timely intervention within this time window are crucial for improving patient outcomes. Current monitoring methods often rely on regular medical staff rounds, failing to provide real-time, continuous monitoring of patients. This can lead to subtle early symptoms being overlooked, potentially missing the optimal opportunity for treatment. Moreover, traditional early warning systems rarely consider the participation of patients' families. However, family members play an important role in the patient's recovery process. They can observe subtle changes in patients' daily behaviors more closely. If they can effectively participate in early warning, the timeliness and accuracy of the early warning will be greatly improved. Summary of the Invention

[0003] In order to solve the above problems, the present invention provides a method and device for warning of cerebral infarction after vascular diagnosis and treatment with family participation, in which family members assist patients in binding patient information, collect patient behavior data by setting a scheduled task, and obtain patient cerebral infarction risk level data through real-time processing, and obtain warning data through a medical response module based on the patient behavior data and the patient cerebral infarction risk level data. By family members participating in the process of warning of cerebral infarction after surgery, the diagnosis confirmation time and overall response time are shortened, and by family members binding patient information, the program automatically triggers the monitoring function, reducing the learning time and operation complexity of family members. A 24-hour continuous monitoring system is constructed by combining multimodal algorithms and the advantages of family members accompanying care to achieve monitoring coverage without blind spots. The timeliness of real-time warning is improved by real-time collection of patient behavior data, and the timeliness of abnormal monitoring delay is improved by obtaining authentication results by comparison based on the wristband data and user data and initiating abnormal processes. By using AI to assist in identifying the patient's cerebral infarction risk level data, the timeliness and accuracy of the prediction of the patient's cerebral infarction risk after vascular diagnosis and treatment are improved, and the investment of unnecessary medical resources is reduced.

[0004] A first aspect of the present invention provides a family-participated method for early warning of cerebral infarction after vascular interventional diagnosis and treatment, comprising the following steps: Based on the patient's wristband data, the client module obtains the patient's data and starts the cerebral infarction monitoring module; Based on the patient data, the patient data is bound to the client module through the liveness detection and information authentication of the cerebral infarction monitoring module and the monitoring process of the cerebral infarction monitoring module is started; The patient's behavior data and the patient's cerebral infarction risk level data are obtained through the monitoring process. The monitoring process is a periodic task that is scheduled to start and execute 1 hour, 3 hours, 6 hours, 12 hours, and 24 hours after the operation; Based on the patient behavior data and the patient cerebral infarction risk level data, early warning data is obtained through a medical response module, and the early warning data is used by medical staff to provide patients with credentials for postoperative cerebral infarction intervention.

[0005] Preferably, after obtaining the patient data through the client module based on the patient's wristband data, the method includes: Collecting the patient's facial data in real time through the collection end of the client module; obtaining user data of the patient through comparison based on the facial data; Based on the wristband data and user data, the authentication result is obtained by comparison and the abnormal process is started. Specifically, if the authentication result is successful, this scheduled task is ended and the experience scheduled task is started to execute the above steps; if the authentication result is failed, the scheduled task is started to warn the patient to move to the collection end. If the scheduled task is completed and the authentication result is failed, the client module is untied from the patient data and exits the monitoring process.

[0006] Preferably, the step of obtaining the patient's cerebral infarction risk level data through the monitoring process further includes: Acquire patient behavior data based on the guidance of the monitoring process, and acquire indicator score data based on the patient behavior data; Based on the index score data, the patient's cerebral infarction risk level data is obtained through a data processing module.

[0007] Preferably, the step of obtaining patient behavior data based on the guidance of the monitoring process and obtaining indicator score data based on the patient behavior data further includes: Instructing family members to collect the patient's facial data through voice, and obtaining facial index score data through the data processing module based on the patient's facial data; Instructing the family member by voice to collect limb data of the patient after raising both arms horizontally, and obtaining limb index score data through the data processing module based on the patient's limb data; The family member guides the patient to repeat preset content through voice guidance and collects patient voice data, and obtains voice scoring data through the data processing module based on the patient voice data.

[0008] Preferably, the steps further include: Based on the patient's facial data, facial symmetry, eye gaze deviation angle, and abnormal expression probability are obtained using MTCNN and Dlib models, ResNet model, and ResNet model respectively; Obtaining a facial symmetry score and a facial expression score by calculation based on the facial symmetry, the eye gaze deviation angle, and the abnormal expression probability; Based on the patient's limb data, limb symmetry, movement coordination, and abnormal jitter frequency are obtained through Kalman filtering, limb symmetry calculation, and motion trajectory reconstruction; Obtaining a limb symmetry score, a movement coordination score, and an abnormal shaking frequency score by calculation based on the limb symmetry, the movement coordination score, and the abnormal shaking frequency; Based on the patient's voice data, a voice matching score, a dysarthria score, and an abnormal pause score are obtained through a voice repetition evaluation model, MFCC cepstral coefficients, and a language fluency evaluation model.

[0009] Preferably, the step of obtaining the patient's cerebral infarction risk level data through a data processing module based on the indicator score data further includes: Based on the facial index scoring data, the limb index scoring data and the voice scoring data, respectively, a facial index, a limb index, a voice index, a facial weight, a limb weight and a voice weight are obtained by normalization processing and mapping weights; Postoperative cerebral infarction index is obtained by weighting based on facial index, limb index, voice index, facial weight, limb weight, and voice weight; Based on the postoperative cerebral infarction index, the patient's cerebral infarction risk level data is obtained through XGBoost.

[0010] Preferably, the step of obtaining early warning data through a medical response module based on the patient behavior data and the patient cerebral infarction risk level data further includes: The cerebral infarction warning is triggered by comparing the patient's cerebral infarction risk level data with the preset risk level. The specific logic is: if the patient's cerebral infarction risk level data is greater than or equal to the preset risk level, the warning data is triggered and the month-on-month change rate of each indicator score data for each patient and at different times is calculated.

[0011] The second aspect of the present invention provides a family-participated cerebral infarction early warning device for vascular interventional diagnosis and treatment, comprising: The client module is used to guide family members to collect patient behavior data through voice and to collect patient wristband data through family members; The cerebral infarction monitoring module is used to initiate the monitoring process based on patient data through liveness detection and information authentication, and to obtain the patient's cerebral infarction risk level data based on patient behavior data; The medical response module is used to obtain early warning data based on the patient behavior data and the patient's cerebral infarction risk level data.

[0012] The third aspect of the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the steps of any one of the above-mentioned methods for warning of cerebral infarction after vascular interventional diagnosis and treatment with family participation are implemented.

[0013] A fourth aspect of the present invention provides a computer-readable storage medium having instructions stored thereon, which, when executed by a processor, implement any of the above-mentioned family-participated methods for warning of cerebral infarction after vascular interventional diagnosis and treatment.

[0014] Due to the adoption of the above technical solutions, the present invention has the following advantages and positive effects compared with the prior art: family members assist patients in binding patient information, patient behavior data is collected by setting a scheduled task and patient cerebral infarction risk level data is obtained through real-time processing, and early warning data is obtained through the medical response module based on the patient behavior data and the patient cerebral infarction risk level data. By having family members participate in the process of postoperative cerebral infarction early warning, the diagnosis confirmation time and overall response time are shortened, and by having family members bind patient information, the program automatically triggers the monitoring function, reducing the learning time and operation complexity of family members, and combining the multimodal algorithm and the advantages of family members to build a 24-hour continuous monitoring system to achieve no-dead-angle monitoring coverage. The real-time collection of patient behavior data improves the timeliness of real-time early warnings, and the timeliness of abnormal monitoring delays is improved by obtaining authentication results by comparison based on the wristband data and user data and initiating abnormal processes. By using AI to assist in identifying patient cerebral infarction risk level data, the timeliness and accuracy of postoperative cerebral infarction risk prediction after vascular diagnosis and treatment of patients are improved, and unnecessary investment in medical resources is reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The specific embodiments of the present invention are further described in detail below with reference to the accompanying drawings, wherein: Figure 1 This is a schematic diagram of the main process of a family-participated method for early warning of cerebral infarction after vascular interventional diagnosis and treatment in the present invention. DETAILED DESCRIPTION

[0016] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. The advantages and features of the present invention will become more apparent from the following description and claims. It should be noted that the drawings are greatly simplified and not to exact ratios, and are intended solely to facilitate and clearly illustrate the embodiments of the present invention.

[0017] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the various components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0018] First embodiment See also Figure 1 The first aspect of the present invention provides a family-participated method for early warning of cerebral infarction after vascular interventional diagnosis and treatment, comprising the following steps: Based on the patient's wristband data, the client module obtains the patient's data and starts the cerebral infarction monitoring module; Based on the patient data, the cerebral infarction monitoring module uses liveness detection and information authentication to bind the patient data to the client module and start the monitoring process of the cerebral infarction monitoring module; The monitoring process is used to obtain patient behavior data and cerebral infarction risk level data. The monitoring process is a periodic task that is scheduled to start and execute 1 hour, 3 hours, 6 hours, 12 hours, and 24 hours after surgery. Based on patient behavior data and patient cerebral infarction risk level data, early warning data is obtained through the medical response module. The early warning data is used by medical staff to provide patients with credentials for postoperative cerebral infarction intervention.

[0019] By leveraging technology to facilitate family collaboration, a closed-loop system of "data collection-risk analysis-early warning response" was constructed, which significantly improved the accuracy, timeliness and comprehensiveness of early warning of cerebral infarction after vascular diagnosis and treatment. At the same time, it optimized the allocation of medical resources, and ultimately achieved the clinical goal of reducing the incidence of cerebral infarction and improving patient prognosis.

[0020] Optionally, early warning data can be displayed on the client module's display based on patient behavior data and the patient's cerebral infarction risk level. Displaying patient early warning data on the client display enables a three-way collaboration among data, family members, and medical staff: family members move from passive acceptance to active participation in postoperative cerebral infarction warning and monitoring, improving monitoring effectiveness; medical staff shift from relying on experience-based judgment to data-driven decision-making, improving intervention accuracy; and patients move from isolated treatment to transparent participation in rehabilitation, improving their overall experience. Ultimately, a closed-loop management system of early detection, rapid response, and accurate intervention is formed, significantly reducing the risk of postoperative cerebral infarction.

[0021] Preferably, after obtaining the patient data through the client module based on the patient's wristband data, the method includes: Collect the patient's facial data in real time through the acquisition end of the client module; Obtaining user data of the patient through comparison based on facial data; Based on the wristband data and user data, the authentication result is obtained by comparison and the abnormal process is started. Specifically: if the authentication result is successful, end this scheduled task and start the experience scheduled task to execute the above steps; if the authentication result is failed, start the scheduled task to warn the patient to move to the collection end. If the scheduled task is completed and the authentication result is failed, the client module is untied from the patient data and exits the monitoring process.

[0022] Patient identity authentication is achieved by combining wristband data with facial data. Furthermore, the patient's first user data is queried from the patient information database based on the wristband data. The second user data is then obtained by comparing the facial data with the patient information database. The user elements of the first user data are compared with the user elements of the second data. If the comparisons are consistent, authentication is successful; otherwise, authentication fails. By establishing a closed-loop system of identification, anomaly warning, and resource release, the response speed and resource utilization of postoperative monitoring are optimized while ensuring data reliability.

[0023] Preferably, the step of obtaining the patient's cerebral infarction risk level data through the monitoring process further includes: Obtain patient behavior data based on the guidance of the monitoring process, and obtain indicator score data based on the patient behavior data; Based on the index scoring data, the patient's cerebral infarction risk level data is obtained through the data processing module.

[0024] By collecting patient behavior data, memorizing patient behavior data to obtain indicator score data, and then obtaining patient cerebral infarction risk level data, data standardization and decision-making efficiency can be achieved, thereby improving the reliability and clinical practical value of postoperative cerebral infarction warning after vascular diagnosis and treatment.

[0025] Preferably, the step of obtaining patient behavior data based on the guidance of the monitoring process and obtaining indicator score data based on the patient behavior data further includes: The family members are guided by voice to collect the patient's facial data, and the facial index score data is obtained through the data processing module based on the patient's facial data; The family members are guided by voice to collect the limb data of the patient after raising both arms horizontally, and the limb index score data is obtained through the data processing module based on the patient's limb data; Through voice guidance, family members instruct patients to repeat preset content and collect patient voice data, and obtain voice scoring data based on the patient voice data through the data processing module.

[0026] By collecting behavioral data and standardized scoring, the risk of postoperative cerebral infarction is quantified, providing reliable data support for risk warning.

[0027] Preferably, the steps further include: Based on the patient's facial data, facial symmetry, eye gaze deviation angle, and abnormal expression probability were obtained using the MTCNN and Dlib models, the ResNet model, and the ResNet model respectively; The facial symmetry score and facial expression score were obtained by calculation based on facial symmetry, eye gaze deviation angle, and abnormal expression probability; Based on the patient's limb data, limb symmetry, movement coordination, and abnormal jitter frequency are obtained through Kalman filtering, limb symmetry calculation, and motion trajectory reconstruction; Limb symmetry score, motor coordination score, and abnormal jitter frequency score were calculated based on the limb symmetry score, motor coordination score, and abnormal jitter frequency score; Based on the patient's speech data, the speech matching score, articulation disorder score, and abnormal pause score are obtained through the speech repetition evaluation model, MFCC cepstral coefficient, and language fluency evaluation model.

[0028] Facial symmetry is used to detect signs of neurological damage, such as facial paralysis and ptosis, and to localize the responsible lesion in stroke. Eye gaze deviation is used to identify ophthalmoplegia or optic nerve pathway abnormalities, suggesting involvement of the brainstem or cortical gaze area. Abnormal expression probability is used to classify the degree of facial expression distortion using ResNet to quantify facial nerve dysfunction. Limb symmetry is used to identify differences in movement amplitude between left and right limbs. Movement coordination is used to analyze movement fluency to further identify cerebellar or basal ganglia lesions. Abnormal jitter frequency is used to identify extrapyramidal damage, such as high-frequency jitter, after Kalman filtering to eliminate physiological tremor. Speech matching is assessed by comparing MFCC cepstral coefficients to standard pronunciation, identifying unclear or missing parts of speech. Dysarthria is scored by assessing abnormal vocal cord vibration frequency. Abnormal pauses are scored by detecting excessively long pauses between sentences.

[0029] Preferably, the step of obtaining the patient's cerebral infarction risk level data through the data processing module based on the indicator score data further includes: Based on the facial index scoring data, the limb index scoring data and the voice scoring data, facial index, limb index, voice index, facial weight, limb weight and voice weight are obtained respectively through normalization processing and mapping weights; Postoperative cerebral infarction index is obtained by weighting based on facial index, limb index, voice index, facial weight, limb weight, and voice weight; Based on postoperative cerebral infarction indicators, XGBoost was used to obtain the patient's cerebral infarction risk level data.

[0030] Multi-dimensional data fusion is achieved through multi-dimensional weighting, reducing missed detections and misjudgments. The XGBoost model captures complex risk patterns and improves early warning sensitivity. Structured output guides intervention and optimizes resource allocation. Automated processes lower operational barriers, and transparent feedback enhances doctor-patient collaboration. The resulting full-cycle cerebral infarction prevention and control system significantly improves postoperative quality of life, the efficiency of medical resource utilization, and the timeliness of risk intervention after vascular diagnosis and treatment.

[0031] Preferably, the step of obtaining early warning data through the medical response module based on the patient behavior data and the patient's cerebral infarction risk level data further includes: The cerebral infarction warning is triggered by comparing the patient's cerebral infarction risk level data with the preset risk level. The specific logic is: if the patient's cerebral infarction risk level data is greater than or equal to the preset risk level, the warning data is triggered and the month-on-month change rate of each indicator score data for each patient and at different times is calculated.

[0032] By combining single-point risks with changes in risk trends, we can reduce missed diagnoses / misdiagnoses; locate core abnormal indicators, and optimize resources through graded responses; and improve the early warning and recognition rate of postoperative cerebral infarction after vascular diagnosis and treatment, thereby reducing the recurrence rate and disability rate of cerebral infarction.

[0033] Second embodiment The second aspect of the present invention provides a family-participated cerebral infarction early warning device for vascular interventional diagnosis and treatment, comprising: The client module is used to guide family members to collect patient behavior data through voice and to collect patient wristband data through family members; The cerebral infarction monitoring module is used to initiate the monitoring process based on patient data through liveness detection and information authentication, and to obtain the patient's cerebral infarction risk level data based on patient behavior data; The medical response module is used to obtain early warning data based on patient behavior data and patient cerebral infarction risk level data.

[0034] Optionally, the client module also includes a display terminal of the client module, which is used to display early warning data based on patient behavior data and patient cerebral infarction risk level data. Through the closed-loop process design of family-assisted collection-multimodal data fusion-dynamic risk warning-precise medical response, it is achieved: Quantified facial scoring data, limb scoring data, and voice scoring data, quantified postoperative cerebral infarction risk based on quantified facial scoring data, limb scoring data, and voice scoring data, capturing hidden cerebral infarction risk and reducing missed diagnosis rate; family participation improves compliance, and medical resources focus on high-risk patients; trend prediction supports recurrence prevention and improves postoperative quality of life. The final linked cerebral infarction prevention and control system significantly improves the postoperative management efficiency and patient prognosis after vascular diagnosis and treatment.

[0035] Optionally, the medical response module includes a medical monitoring platform, an early warning processing system, a remote guidance module, and a data processing module, wherein the data processing module is used to generate statistical reports based on all patient behavior data and patient cerebral infarction risk level data, analyze cerebral infarction risk indicator data, generate patient cerebral infarction risk level data and early warning data based on the prediction model, and generate treatment plans and suggestions based on the patient cerebral infarction risk level data and early warning data. The medical monitoring platform is used to display the monitoring devices of all patients in real time based on all patient behavior data and patient cerebral infarction risk level data, update the patient's cerebral infarction risk level data in real time, display different risk ranges in different colors and fonts, and query the patient-dimensional score data trend chart and historical score details data; the early warning processing system is used to push graded early warnings to the client module based on all patient cerebral infarction risk level data, trigger the medical staff response process, update treatment data and archive treatment effects, and the remote guidance module guides the associated medical staff and family members to provide effective and necessary treatment plans through the video call function, guides the treatment plan operation in real time and provides corrective guidance through screen sharing, and provides basic treatment guidance through operation demonstration playback.

[0036] Third embodiment The third aspect of the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the steps of any one of the above-mentioned methods for warning of cerebral infarction after vascular interventional diagnosis and treatment with family participation are implemented.

[0037] The program of the electronic device provided in this embodiment: by capturing cerebral infarction risk signals in real time, the treatment window period is seized; family members participate in the cerebral infarction early warning business process, and improve family care capabilities.

[0038] Fourth embodiment A fourth aspect of the present invention provides a computer-readable storage medium having instructions stored thereon, which, when executed by a processor, implement any of the above-mentioned family-participated methods for warning of cerebral infarction after vascular interventional diagnosis and treatment.

[0039] Computer-readable storage media seamlessly connects to the medical system, reducing manual intervention and communication costs to improve process efficiency. Through family participation, equipment monitoring, and medical staff intervention, the core carrier of the family-hospital-equipment collaborative network is ultimately formed, significantly improving the efficiency and accessibility of postoperative cerebral infarction risk management.

[0040] In the description of this application, it should be noted that the terms "inner" and "outer" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, or the orientations or positional relationships in which the product of this application is typically placed when in use. These terms are intended solely to facilitate the description of this application and simplify the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limitations on this application. Furthermore, the terms "first" and "second" and the like are used solely for distinction and should not be construed as indicating or implying relative importance.

[0041] It should also be noted that, unless otherwise expressly specified or limited, the terms "disposed" and "connected" should be understood broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to direct connections, indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art will understand the specific meanings of these terms in this application based on the specific circumstances.

[0042] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the identification content specifically executed by the above-described system and device can refer to the corresponding process in the aforementioned method embodiment.

[0043] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments. Even if various changes are made to the present invention, if these changes fall within the scope of the claims of the present invention and their equivalents, they still fall within the scope of protection of the present invention.

Claims

1. A family-participated method for early warning of cerebral infarction after vascular interventional diagnosis and treatment, characterized in that: The steps include: Based on the patient's wristband data, the client module obtains the patient's data and starts the cerebral infarction monitoring module; Based on the patient data, the patient data is bound to the client module through the liveness detection and information authentication of the cerebral infarction monitoring module and the monitoring process of the cerebral infarction monitoring module is started; The patient's behavior data and the patient's cerebral infarction risk level data are obtained through the monitoring process. The monitoring process is a periodic task that is scheduled to start and execute 1 hour, 3 hours, 6 hours, 12 hours, and 24 hours after the operation; Based on the patient behavior data and the patient cerebral infarction risk level data, early warning data is obtained through a medical response module, and the early warning data is used by medical staff to provide patients with credentials for postoperative cerebral infarction intervention.

2. The family-participated cerebral infarction early warning method after vascular interventional diagnosis and treatment according to claim 1, characterized in that: After obtaining patient data through the client module based on the patient's wristband data, it includes: Collecting the patient's facial data in real time through the collection end of the client module; obtaining user data of the patient through comparison based on the facial data; Based on the wristband data and user data, the authentication result is obtained by comparison and the abnormal process is started. Specifically, if the authentication result is successful, this scheduled task is ended and the experience scheduled task is started to execute the above steps; if the authentication result is failed, the scheduled task is started to warn the patient to move to the collection end. If the scheduled task is completed and the authentication result is failed, the client module is untied from the patient data and exits the monitoring process.

3. The family-participated cerebral infarction early warning method after vascular interventional diagnosis and treatment according to claim 1, characterized in that: The step of obtaining the patient's cerebral infarction risk level data through the monitoring process further includes: Acquire patient behavior data based on the guidance of the monitoring process, and acquire indicator score data based on the patient behavior data; Based on the index score data, the patient's cerebral infarction risk level data is obtained through a data processing module.

4. The family-participated cerebral infarction early warning method after vascular interventional diagnosis and treatment according to claim 3, characterized in that: Acquiring patient behavior data based on the guidance of the monitoring process, and acquiring indicator score data based on the patient behavior data further includes: Instructing family members to collect the patient's facial data through voice, and obtaining facial index score data through the data processing module based on the patient's facial data; Instructing the family member by voice to collect limb data of the patient after raising both arms horizontally, and obtaining limb index score data through the data processing module based on the patient's limb data; The family member guides the patient to repeat preset content through voice guidance and collects patient voice data, and obtains voice scoring data through the data processing module based on the patient voice data.

5. The family-participated cerebral infarction early warning method after vascular interventional diagnosis and treatment according to claim 4, characterized in that: The steps further include: Based on the patient's facial data, facial symmetry, eye gaze deviation angle, and abnormal expression probability are obtained using MTCNN and Dlib models, ResNet model, and ResNet model respectively; Obtaining a facial symmetry score and a facial expression score by calculation based on the facial symmetry, the eye gaze deviation angle, and the abnormal expression probability; Based on the patient's limb data, limb symmetry, movement coordination, and abnormal jitter frequency are obtained through Kalman filtering, limb symmetry calculation, and motion trajectory reconstruction; Obtaining a limb symmetry score, a movement coordination score, and an abnormal shaking frequency score by calculation based on the limb symmetry, the movement coordination score, and the abnormal shaking frequency; Based on the patient's voice data, a voice matching score, a dysarthria score, and an abnormal pause score are obtained through a voice repetition evaluation model, MFCC cepstral coefficients, and a language fluency evaluation model.

6. The family-participated cerebral infarction early warning method after vascular interventional diagnosis and treatment according to claim 4, characterized in that: The step of obtaining the patient's cerebral infarction risk level data through a data processing module based on the index score data further includes: Based on the facial index scoring data, the limb index scoring data and the voice scoring data, respectively, a facial index, a limb index, a voice index, a facial weight, a limb weight and a voice weight are obtained by normalization processing and mapping weights; Postoperative cerebral infarction index is obtained by weighting based on facial index, limb index, voice index, facial weight, limb weight, and voice weight; Based on the postoperative cerebral infarction index, the patient's cerebral infarction risk level data is obtained through XGBoost.

7. The family-participated cerebral infarction early warning method after vascular interventional diagnosis and treatment according to claim 1, characterized in that: The step of obtaining early warning data through a medical response module based on the patient behavior data and the patient cerebral infarction risk level data further includes: The cerebral infarction warning is triggered by comparing the patient's cerebral infarction risk level data with the preset risk level. The specific logic is: if the patient's cerebral infarction risk level data is greater than or equal to the preset risk level, the warning data is triggered and the month-on-month change rate of each indicator score data for each patient and at different times is calculated.

8. A family-participated cerebral infarction warning device for patients undergoing vascular interventional diagnosis and treatment, characterized in that: include: The client module is used to guide family members to collect patient behavior data through voice and to collect patient wristband data through family members; The cerebral infarction monitoring module is used to initiate the monitoring process based on patient data through liveness detection and information authentication, and to obtain the patient's cerebral infarction risk level data based on patient behavior data; The medical response module is used to obtain early warning data based on the patient behavior data and the patient's cerebral infarction risk level data.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the computer program is executed by a processor, the steps of the family-participated cerebral infarction early warning method after vascular interventional diagnosis and treatment are implemented as described in any one of claims 1 to 7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the family-participated vascular intervention diagnosis and treatment postoperative cerebral infarction warning method as described in any one of claims 1-7 is implemented.

Images