A postoperative rehabilitation state monitoring method, system, terminal and storage medium

Abstract

The application relates to the technical field of medical rehabilitation monitoring, and discloses a postoperative rehabilitation state monitoring method, a postoperative rehabilitation state monitoring system, a terminal and a storage medium. The method comprises the following steps: through a wearable monitoring device attached to the skin on both sides of the neck of a patient, real-time collection of voice vibration signals, motion posture signals and electromyographic signals of the patient; pretreatment and feature extraction are performed on the voice vibration signals, the motion posture signals and the electromyographic signals, so as to obtain voice feature parameters, shoulder and neck activity degree parameters and muscle strength parameters; the voice feature parameters, the shoulder and neck activity degree parameters and the muscle strength parameters are compared and analyzed with a preset rehabilitation evaluation standard library, and real-time rehabilitation scores of the patient are calculated; according to the real-time rehabilitation scores and historical monitoring data, personalized rehabilitation tasks and a weekly rehabilitation report are generated, and the personalized rehabilitation tasks and the weekly rehabilitation report are respectively pushed to a patient end and a medical end. The application significantly improves the objectivity, accuracy and efficiency of postoperative rehabilitation management of thyroid glands.

Description

Technical Field

[0001] This invention relates to the field of medical rehabilitation monitoring technology, and in particular to a method, system, terminal, and computer-readable storage medium for monitoring postoperative rehabilitation status. Background Technology

[0002] In the field of post-thyroid surgery rehabilitation, existing assessment and management models have significant limitations. At the monitoring level, reliance on non-wearable specialized equipment such as laryngoscopes and electromyography (EMG) devices, while offering acceptable accuracy, is hampered by their bulky size, complex operation, and strict limitation to hospital settings. This prevents long-term, continuous monitoring at home, resulting in severe data discontinuity. Regarding assessment methods, traditional clinical assessments heavily depend on the subjective experience and judgment of medical staff and the patient's ambiguous complaints. Assessment results are mostly qualitative descriptions, lacking objective and standardized quantitative criteria, leading to poor consistency among assessors and difficulty in accurately tracking and comparing rehabilitation progress. In terms of management models, current rehabilitation interventions are passive and intermittent. Rehabilitation plans are often generic and lack personalization, requiring medical staff to manually process information, resulting in low efficiency and a relatively crude management approach. These shortcomings collectively lead to inaccurate and untimely post-operative rehabilitation status assessments, hindering the improvement of rehabilitation management quality and efficiency.

[0003] Therefore, existing technologies still need to be improved and developed. Summary of the Invention

[0004] The main objective of this invention is to provide a method, system, terminal, and computer-readable storage medium for monitoring postoperative rehabilitation status. This invention aims to address the shortcomings of existing postoperative rehabilitation monitoring and assessment methods for thyroid surgery, such as limited equipment and scenarios, subjective assessment lacking quantitative standards, and passive and inefficient management models. These shortcomings lead to inaccurate and untimely assessment of rehabilitation status and low efficiency in rehabilitation management.

[0005] To achieve the above objectives, the present invention provides a method for monitoring postoperative rehabilitation status, the method comprising the following steps: Wearable monitoring devices attached to the skin on both sides of the patient's neck can collect the patient's acoustic vibration signals, motion posture signals and electromyographic signals in real time. The acoustic vibration signal, the motion posture signal, and the electromyographic signal are preprocessed and feature extracted to obtain voice feature parameters, shoulder and neck range of motion parameters, and muscle strength parameters, respectively. The voice feature parameters, shoulder and neck range of motion parameters, and muscle strength parameters are compared and analyzed with a preset rehabilitation assessment standard library to calculate the patient's real-time rehabilitation score. Based on the real-time rehabilitation score and historical monitoring data, personalized rehabilitation tasks and weekly rehabilitation reports are generated and pushed to the patient's end and the medical staff's end, respectively.

[0006] Furthermore, the acoustic vibration signal includes the vocal cord vibration frequency, vibration amplitude, and phonation frequency when the patient speaks; the motion posture signal includes the neck movement angle, movement amplitude, and movement completion degree when the patient performs shoulder and neck rehabilitation exercises; and the electromyographic signal includes the electromyographic peak value when the patient swallows and the neck muscles contract.

[0007] Furthermore, the calculation yields the patient's real-time rehabilitation score, specifically including: Map the voice feature parameters to voice scoring sub-items; The shoulder and neck mobility parameters are mapped to shoulder and neck function scoring sub-items; The muscle strength parameters are mapped to swallowing function scoring items; Obtain the preset weight coefficients corresponding to each scoring sub-item, and perform a weighted summation of the voice scoring sub-item, the shoulder and neck function scoring sub-item, and the swallowing function scoring sub-item based on all the preset weight coefficients to obtain the real-time rehabilitation score.

[0008] Furthermore, the generation of personalized rehabilitation tasks specifically includes: The dimension with the lowest score in the real-time rehabilitation score is identified as the current weakness in rehabilitation. Match training programs targeting the current rehabilitation weakness from a pre-set rehabilitation program library. These training programs include specific shoulder and neck exercises, vocalization exercises, or swallowing exercises. Generate task instructions that include links to motion guidance videos for the training program, suggested training frequency, and expected target scores.

[0009] Furthermore, the weekly rehabilitation report includes: Overall trend statistics of patient recovery data; The daily change curve of an individual patient's comprehensive rehabilitation score; Automatically labeled list of abnormal indicators and statistical duration of abnormalities; Recommended intervention measures generated based on the changing trends of historical monitoring data include at least one of the following: recommending follow-up visits, adjusting training intensity, and conducting manual telephone follow-ups.

[0010] Furthermore, the labeling rules for abnormal indicators in the abnormal indicator list include: If the peak electromyography value of N consecutive swallowing actions is lower than the preset percentage of the preset baseline value, it is marked as "suspected decline in swallowing function"; If the frequency fluctuation of the sound vibration during N consecutive vocalizations exceeds the normal threshold range, it is marked as "abnormal voice recovery". If the shoulder and neck range of motion fails to reach the target angle set in the rehabilitation exercises for N consecutive times, it will be marked as "limited shoulder and neck range of motion". Where N is an integer greater than or equal to 3.

[0011] Furthermore, the step of pushing the personalized rehabilitation task and the weekly rehabilitation report to the patient's end and the medical staff's end respectively, further includes: When the patient receives the personalized rehabilitation task and starts execution, the wearable monitoring device is controlled to enter a high sampling rate task monitoring mode; Real-time acquisition of patients' movement posture signals and electromyographic signals during training programs; The collected signals are compared with standard action templates to calculate the action standard score; If the action standard score is greater than or equal to the preset qualified threshold, it is recorded as a valid check-in and the historical monitoring data is updated. If the action standard score is less than the preset qualified threshold, real-time correction feedback information is output through the patient terminal.

[0012] Furthermore, to achieve the above objectives, the present invention also provides a postoperative rehabilitation status monitoring system, which is used to implement the postoperative rehabilitation status monitoring method described above, wherein the postoperative rehabilitation status monitoring system includes: The data acquisition module is used to collect the patient's acoustic vibration signals, motion posture signals and electromyographic signals in real time through wearable monitoring devices attached to the skin on both sides of the patient's neck; The data processing module is used to preprocess and extract features from the acoustic vibration signal, the motion posture signal and the electromyographic signal to obtain voice feature parameters, shoulder and neck range of motion parameters and muscle strength parameters, respectively. The scoring calculation module is used to compare and analyze the voice feature parameters, the shoulder and neck range of motion parameters, and the muscle strength parameters with a preset rehabilitation assessment standard library to calculate the patient's real-time rehabilitation score. The interactive feedback module is used to generate personalized rehabilitation tasks and weekly rehabilitation reports based on the real-time rehabilitation score and historical monitoring data, and push the personalized rehabilitation tasks and weekly rehabilitation reports to the patient's end and the medical staff's end, respectively.

[0013] In addition, to achieve the above objectives, the present invention also provides a terminal, wherein the terminal includes: a memory, a processor, and a postoperative rehabilitation status monitoring program stored in the memory and executable on the processor, wherein when the postoperative rehabilitation status monitoring program is executed by the processor, it implements the steps of the postoperative rehabilitation status monitoring method as described above.

[0014] In addition, to achieve the above objectives, the present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a postoperative rehabilitation status monitoring program, which, when executed by a processor, implements the steps of the postoperative rehabilitation status monitoring method as described above.

[0015] The beneficial effects of this invention are as follows: This invention enables the objective, continuous, and multimodal synchronous acquisition of key postoperative rehabilitation indicators' physiological signals. By integrating multiple sensors into a single adhesive device, data can be collected non-invasively in the patient's natural living state. This overcomes the limitations of traditional large-scale professional medical instruments (such as laryngoscopes and electromyography machines) in terms of application scenarios and their inability to be used for long-term daily monitoring, providing a real and rich raw data foundation for subsequent quantitative analysis. This invention transforms raw, mixed physical signals into standardized quantitative characteristic parameters with clear clinical significance. Through signal processing and feature extraction, subjective descriptions are transformed into objective values, enabling the description of rehabilitation status to move from qualitative to quantitative. This lays the data foundation for establishing a standardized assessment system and makes the rehabilitation progress comparable between different time points and different patients.

[0016] This invention maps and integrates multi-dimensional feature parameters into a single score, enabling patients and medical staff to quickly and clearly grasp the overall level of rehabilitation and progress in each dimension. This significantly lowers the professional threshold for data interpretation, allowing patients to intuitively perceive their recovery progress and enabling medical staff to efficiently screen a large number of patients. This invention forms an intelligent management closed loop of "monitoring-assessment-intervention," realizing personalized customization of rehabilitation plans and precise allocation of medical resources. Generating personalized tasks ensures the targeting of rehabilitation training and patient compliance; generating and pushing weekly reports greatly improves the work efficiency of medical staff, realizing a shift from passive review to proactive management. Ultimately, by closely integrating technical monitoring with clinical practice, the quality and efficiency of overall rehabilitation management are improved. Attached Figure Description

[0017] Figure 1 This is a flowchart of a preferred embodiment of the postoperative rehabilitation status monitoring method of the present invention; Figure 2 This is a structural diagram of a preferred embodiment of the postoperative rehabilitation status monitoring system of the present invention; Figure 3 This is a structural diagram of a preferred embodiment of the terminal of the present invention. Detailed Implementation

[0018] This application provides a method, system, terminal, and storage medium for monitoring postoperative rehabilitation status. To make the purpose, technical solution, and effects of this application clearer and more explicit, the following detailed description is provided with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining this application and are not intended to limit this application.

[0019] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0020] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0021] The postoperative rehabilitation status monitoring method described in the preferred embodiment of the present invention, such as... Figure 1 As shown, the postoperative rehabilitation status monitoring method includes the following steps: S10. The patient's acoustic vibration signals, motion posture signals and electromyographic signals are collected in real time through wearable monitoring devices attached to the skin on both sides of the patient's neck.

[0022] In step S10, the acoustic vibration signal includes the vocal cord vibration frequency, vibration amplitude, and phonation frequency when the patient speaks; the motion posture signal includes the neck movement angle, movement amplitude, and movement completion degree when the patient performs shoulder and neck rehabilitation exercises; and the electromyographic signal includes the electromyographic peak value when the patient swallows and the neck muscles contract.

[0023] The purpose of this step is to establish an objective, continuous, and multimodal foundation for physiological signal acquisition, overcoming the limitations of traditional subjective assessment and intermittent monitoring. Wearable devices enable continuous sensing of key rehabilitation indicators in patients' daily lives, providing raw data support for subsequent quantitative analysis.

[0024] In this embodiment, the wearable monitoring device is specifically a flexible patch-type monitoring device designed for post-thyroidectomy patients. The device integrates three core sensing units onto the same flexible substrate: an acoustic vibration sensor, an inertial measurement unit (IMU), and a surface electromyography (sEMG) sensor. The device is lightweight and breathable, using highly biocompatible medical-grade silicone to encapsulate the sensors, which are then fixed to the skin with a hypoallergenic medical adhesive layer. The device is worn on both sides of the patient's neck, strictly avoiding the surgical incision area, typically on the skin surface of the mid-section of the sternocleidomastoid muscle. This location effectively collects muscle activity signals related to speech and swallowing, and also monitors neck flexion and rotation.

[0025] The specific operation of acoustic vibration signal acquisition: The device's built-in miniature acoustic vibration sensor (such as a contact microphone or accelerometer) is closely fitted to the skin on the side of the neck to capture the mechanical vibration signals transmitted through the neck tissue by vocal cord vibration. When the patient engages in daily conversation, reading, or performs specific vocal exercises, the sensor records the vibration waveform at a high sampling rate (e.g., 1024 Hz). The acquired raw signal contains rich information such as the fundamental frequency of vocal cord vibration (reflecting pitch), amplitude (reflecting volume or hoarseness), and harmonic-to-noise ratio. Simultaneously, by analyzing the number of vibration events per unit time, the system can count the frequency of the patient's active vocalizations, serving as an indirect indicator of their speech activity. This contact-based acquisition method effectively filters out environmental noise interference, focusing on physiological vibration signals.

[0026] The specific operation of motion posture signal acquisition: The device's built-in nine-axis inertial measurement unit, including a three-axis accelerometer, a three-axis gyroscope, and a three-axis magnetometer, is used to accurately sense the spatial movement and posture changes of the neck. When the patient performs standardized shoulder and neck rehabilitation exercises (such as neck flexion, extension, lateral flexion, and rotation) according to the rehabilitation plan, the IMU continuously outputs raw data. Through sensor fusion algorithms (such as complementary filtering or Kalman filtering), this data is calculated into absolute or relative angles, angular velocities, and accelerations of the neck. Key parameters include: the maximum angle of movement (range), the smoothness of the movement (characterized by the angular velocity curve), and the composite angle of multi-axis movements. For example, in the "rotate to the left" movement, the system can accurately calculate whether the actual angle of rotation of the patient's neck reaches the target value (such as 60 degrees), and whether the trajectory during the rotation is smooth and without shaking.

[0027] The specific operation of electromyography (EMG) signal acquisition: The device's built-in surface electromyography (sEMG) sensor electrode pairs are arranged at a specific interval to detect the electrical activity of subcutaneous muscle groups. The main monitoring scenarios include voluntary swallowing and specific isometric contraction training of neck muscles. During swallowing, the system records EMG signals from related muscles such as the suprahyoid and thyrohyoid muscles, extracting features such as peak EMG amplitude, duration of muscle activation, and integrated EMG value. These features are directly related to the force and efficiency of swallowing. During neck muscle strength training, the contraction intensity of muscles such as the sternocleidomastoid and upper trapezius is monitored. The acquired raw sEMG signals are amplified, filtered (typically bandpass filtered at 20-500 Hz to remove interference from ECG, motion artifacts, etc.), and rectified and smoothed to obtain a linear envelope signal suitable for analysis.

[0028] The device has a built-in low-power Bluetooth module for transmitting multiple raw signals to the patient's smartphone (patient device) in real time or at regular intervals for initial caching. The device has two operating modes: daily monitoring mode (low sampling rate, periodic acquisition, for all-weather background monitoring) and task monitoring mode (high sampling rate, for accurate assessment of the quality of rehabilitation task execution).

[0029] It should be noted that this step miniaturizes and integrates multiple physiological signal acquisition modules, adapting them to the specific and sensitive area of ​​the neck, to achieve non-invasive and long-term continuous monitoring. This differs from large devices such as laryngoscopes and standalone electromyography machines, which can only perform single, momentary examinations within a hospital setting, and also from traditional rehabilitation assessments that rely on the patient's subjective recollection or brief observations by medical staff. Therefore, it truly achieves digital and comprehensive recording of the postoperative rehabilitation process.

[0030] S20. The acoustic vibration signal, the motion posture signal and the electromyographic signal are preprocessed and feature extracted to obtain voice feature parameters, shoulder and neck range of motion parameters and muscle strength parameters, respectively.

[0031] The purpose of this step is to transform the raw, mixed physical signals obtained from sensors into standardized feature parameters that can characterize specific physiological functional states and can be used for quantitative assessment. Preprocessing aims to improve signal quality, while feature extraction aims to uncover the information dimensions most relevant to rehabilitation goals.

[0032] In this embodiment, signal processing is performed within the computing unit of the terminal (such as a smartphone or cloud server), and the specific operation is as follows: 1. Acoustic and vibration signal processing: Preprocessing: The acquired raw vibration signal is bandpass filtered (e.g., 80-1000 Hz) to remove extremely low frequency displacement noise and high frequency electronic noise, and the signal is normalized to eliminate baseline differences caused by different sensor attachment pressures.

[0033] Feature extraction: Time-domain and frequency-domain features are extracted from the preprocessed signal to form a set of voice feature parameters. Key parameters include: Fundamental frequency and its standard deviation: These reflect the basic pitch and stability of the voice. Postoperative vocal cord paralysis or edema often leads to abnormal fluctuations in the fundamental frequency.

[0034] Amplitude and dynamic range: reflect the intensity of the voice. Weak voice is a common symptom after surgery.

[0035] Harmonic-to-noise ratio (HNR): This reflects the ratio of the periodic component of vocal cord vibration to the noise component, and is the objective gold standard for quantifying the degree of hoarseness.

[0036] Jitter and Shimmer: These measure the perturbations in fundamental frequency and amplitude, respectively, and their increases are highly correlated with vocal cord pathology.

[0037] Maximum vocalization time: The longest duration of sustaining the vowel "ah" after a deep inhalation, used to assess lung function and glottal closure efficiency.

[0038] 2. Motion attitude signal processing: Preprocessing: The raw acceleration and angular velocity data of the IMU are calibrated, denoised, and gravitational acceleration compensated. Attitude is calculated by quaternion or direction cosine matrix algorithm to obtain stable and accurate three-dimensional Euler angles (pitch angle, roll angle, yaw angle) of the neck.

[0039] Feature extraction: During the patient's standardized rehabilitation exercise cycle, parameters reflecting shoulder and neck range of motion were extracted. Range of motion: The maximum angle actually reached for each prescribed movement (such as flexion, extension, lateral flexion, rotation), compared with the healthy side or the standard value.

[0040] Movement smoothness: Quantified by calculating the root mean square value of the derivative (acceleration) of the movement angular velocity curve. The smaller the value, the smoother the movement, and the lower the pain or stiffness may be.

[0041] Movement symmetry: Compare the ratio of range of motion for leftward and rightward rotation, and for leftward and rightward flexion.

[0042] Movement completion time: The time required to complete one standard movement cycle, reflecting movement speed and confidence.

[0043] 3. Electromyographic signal processing: Preprocessing: The raw sEMG signal is subjected to a 50Hz power frequency notch filter to remove power supply interference, then full-wave rectification is performed, and then a low-pass filter (e.g., 5Hz) is used to obtain a linear envelope, which represents the trend of muscle activation intensity changing over time.

[0044] Feature extraction: For swallowing movements and isometric muscle contraction tasks, muscle strength parameters are extracted. Peak amplitude: The maximum value of the electromyographic envelope during a single swallow or contraction, directly reflecting the instantaneous maximum force of muscle contraction.

[0045] Mean electromyography (EMG) value: The average value of the EMG envelope over the duration of a single movement, reflecting the average activation level of the muscle.

[0046] Median frequency or average power frequency: calculated from the power spectrum of electromyographic signals, its trend can reflect the state of muscle fatigue.

[0047] Activation delay time: The time from the issuance of the action command to the electromyographic signal exceeding the preset threshold, reflecting the efficiency of neuromuscular control.

[0048] It is important to note that feature extraction in this step is the cornerstone of subsequent quantitative assessment. The extracted parameters all clearly correspond to the physical meaning of clinical rehabilitation assessments, ensuring that the results obtained by the algorithm are intuitively understood and accepted by clinicians. For example, "shoulder and neck range of motion parameters" directly correspond to "joint range of motion measurement" in clinical practice, and the peak amplitude in "muscle strength parameters" corresponds to a quantitative version of "manual muscle strength testing." This ensures effective integration between the technical approach and medical practice.

[0049] S30. The voice feature parameters, the shoulder and neck range of motion parameters, and the muscle strength parameters are compared and analyzed with a preset rehabilitation assessment standard library to calculate the patient's real-time rehabilitation score.

[0050] In step S30, the calculation of the patient's real-time rehabilitation score specifically includes: mapping the voice feature parameters to voice scoring sub-items; mapping the shoulder and neck mobility parameters to shoulder and neck function scoring sub-items; mapping the muscle strength parameters to swallowing function scoring sub-items; obtaining the preset weight coefficients corresponding to each scoring sub-item; and performing a weighted summation of the voice scoring sub-items, the shoulder and neck function scoring sub-items, and the swallowing function scoring sub-items based on all the preset weight coefficients to obtain the real-time rehabilitation score.

[0051] The purpose of this step is to integrate multi-dimensional physiological characteristic parameters into one or more intuitive and traceable quantitative scores to achieve a comprehensive and objective evaluation of the patient's recovery status and to support horizontal and vertical comparisons across time and across patients.

[0052] In this embodiment, the "preset rehabilitation assessment standard library" is a database stored in the cloud or locally, containing standard values, thresholds, and mapping functions derived from statistical modeling of data from a large number of healthy individuals and patients at different stages of rehabilitation after thyroid surgery. The specific operation for calculating the real-time rehabilitation score is as follows: 1. Sub-item rating mapping: Voice Scoring Sub-item (Sv): The set of voice feature parameters extracted in step S20 (such as fundamental frequency, harmonic-to-noise ratio, amplitude, etc.) is input into a trained evaluation model (such as a multiple linear regression model or a machine learning model). This model maps these parameters to a sub-item score of 0-100. For example, the higher the harmonic-to-noise ratio and the lower the vibration, the higher the score. The model is constructed and calibrated based on a correlation study between the Clinical Voice Disorder Index score and objective acoustic parameters.

[0053] The shoulder and neck function scoring sub-item (Sm) compares parameters such as joint range of motion, smoothness of movement, and symmetry with the target values ​​required by standard rehabilitation exercises. Each parameter is scored based on the percentage of its actual value relative to the target value, and then combined into a shoulder and neck function score of 0-100. For example, achieving 60 degrees of left rotation (target value) is 100% and gets full marks, while achieving only 30 degrees gets 50 points.

[0054] The swallowing function scoring sub-item (Ss) is primarily based on the peak amplitude of electromyography (EMG) and the EMG pattern of swallowing movements. The current peak EMG value during swallowing is compared to the patient's personal baseline (e.g., the average value during the postoperative stable period) or a standard baseline for healthy individuals. Simultaneously, the timing pattern of EMG activation is analyzed to determine if it is normal. A swallowing function score of 0-100 is then derived.

[0055] 2. The weighted summation yields the comprehensive rehabilitation score (S): The system has preset weight coefficients for each dimension, for example: voice function weight (Wv) = 0.4, shoulder and neck function weight (Wm) = 0.35, and swallowing function weight (Ws) = 0.25. The weight settings can be personalized by the doctor within the system according to different postoperative stages (such as focusing more on swallowing in the early stage and more on voice in the later stage) or different surgical types (such as unilateral vs. central dissection).

[0056] The formula for calculating the real-time rehabilitation comprehensive score is: S = Sv * Wv + Sm * Wm + Ss * Ws. This score also falls within the range of 0-100 points, with a higher score indicating a better overall rehabilitation status.

[0057] In addition to the overall score, the system also displays three sub-scores, allowing patients and doctors to clearly understand the specific recovery status in each dimension.

[0058] It should be noted that the rehabilitation assessment standard library is dynamically updated. As more patient data is integrated and clinical research progresses, the standard values ​​and mapping models can be continuously optimized, making the scores more accurate. This step transforms complex multimodal signals into concise scores, greatly reducing the threshold for data interpretation. This allows patients to intuitively perceive their recovery progress (e.g., "Today's voice score is 85 points, an improvement of 5 points from yesterday"), and also enables medical staff to quickly screen the overall condition of a large number of patients.

[0059] S40. Based on the real-time rehabilitation score and historical monitoring data, generate personalized rehabilitation tasks and weekly rehabilitation reports, and push the personalized rehabilitation tasks and weekly rehabilitation reports to the patient's end and the medical staff's end, respectively.

[0060] The purpose of this step is to form a closed loop of monitoring-assessment-intervention, transform the assessment results into specific and actionable rehabilitation guidelines, provide decision support for the medical team, and achieve personalized and intelligent rehabilitation management.

[0061] Further, in step S40, generating the personalized rehabilitation task includes: The dimension with the lowest score in the real-time rehabilitation score is identified as the current weakness in rehabilitation. Match training programs targeting the current rehabilitation weakness from a pre-set rehabilitation program library. These training programs include specific shoulder and neck exercises, vocalization exercises, or swallowing exercises. Generate task instructions that include links to motion guidance videos for the training program, suggested training frequency, and expected target scores.

[0062] In this embodiment, the specific operation of generating personalized rehabilitation tasks is as follows: 1. Weakness Analysis: The system analyzes the patient's latest real-time rehabilitation score and identifies the lowest score among the three sub-items: voice (Sv), neck and shoulder (Sm), and swallowing (Ss), as the "current rehabilitation weakness." For example, if a patient's S=75, with Sv=80, Sm=60, and Ss=85, then "neck and shoulder function" is identified as the weakness.

[0063] 2. Task Matching: The system accesses a pre-set "Rehabilitation Plan Library." This library contains a large number of structured rehabilitation training programs, each targeting a specific functional dimension (voice, neck and shoulder, swallowing), different difficulty levels (beginner, intermediate, advanced), and different implementation formats (such as text and image instructions, video links). Based on the "current rehabilitation weakness," the system automatically matches a set of targeted training programs. For example, for "neck and shoulder functional weakness," it might match "Progressive Isometric Contraction Training of the Left Neck Flexor Muscles" or "Cervical Joint Range of Motion Stretching Exercise (Section 2)," etc.

[0064] 3. Task Instruction Generation: The system packages the matched training items into a clear "personalized rehabilitation task," which includes: Training program description: Specific actions or exercises to be performed (e.g., "practice the trilled tongue sound 20 times").

[0065] Link to instructional video: Click to watch a video demonstrating the standard movements.

[0066] Recommended training frequency: such as "3 sets per day, 10 repetitions per set".

[0067] Target score: The score range expected to be achieved in this weak area after completing this training phase.

[0068] Task validity period: For example, "to be completed within the next 3 days".

[0069] 4. Push notifications: The generated personalized rehabilitation tasks are sent instantly to the patient's smartphone app (patient side) via push notifications, in-app notifications, and other methods.

[0070] Further, in step S40, the weekly rehabilitation report includes: Overall trend statistics of patient recovery data; The daily change curve of an individual patient's comprehensive rehabilitation score; Automatically labeled list of abnormal indicators and statistical duration of abnormalities; Recommended intervention measures generated based on the changing trends of historical monitoring data include at least one of the following: recommending follow-up visits, adjusting training intensity, and conducting manual telephone follow-ups.

[0071] Furthermore, the labeling rules for the abnormal indicators include: If the peak electromyography value of N consecutive swallowing actions is lower than the preset percentage of the preset baseline value, it is marked as "suspected decline in swallowing function"; If the frequency fluctuation of the sound vibration during N consecutive vocalizations exceeds the normal threshold range, it is marked as "abnormal voice recovery". If the shoulder and neck range of motion fails to reach the target angle set in the rehabilitation exercises for N consecutive times, it will be marked as "limited shoulder and neck range of motion". Where N is an integer greater than or equal to 3.

[0072] In this embodiment, the specific steps for generating the weekly rehabilitation report are as follows: The system automatically summarizes all monitoring data, assessment scores, and task completion records for the patient over the past seven days each week, generating a visually appealing "Weekly Rehabilitation Report," which is then pushed to the management platform (medical staff terminal) used by doctors and nurses. The report includes: 1. Overall trend statistics of patient group rehabilitation data: The chart displays the trend of average comprehensive score of all monitored patients this week, the comparison of average scores in each dimension, and other statistical charts, organized by department or treatment group, to help medical staff grasp the general situation of group rehabilitation.

[0073] 2. Daily Comprehensive Rehabilitation Score Change Curve for Individual Patients: Clearly shows the fluctuation trend of the patient's daily comprehensive score (S) and three sub-scores (Sv, Sm, Ss) over the past week.

[0074] 3. Automatically labeled list of abnormal indicators and statistics on the duration of abnormalities: This is the core warning section of the report. The system automatically scans data and marks abnormal situations according to preset, quantified anomaly labeling rules. For example: Rule 1 (Swallowing function): If the peak electromyography value of swallowing actions in three consecutive swallowing actions (N=3) is lower than 70% of the individual's baseline value (usually the average value during the postoperative stable period) (X=70), it will be automatically marked as "suspected decline in swallowing function" and the duration of the abnormality will be recorded (e.g., "has lasted for 2 days").

[0075] Rule 2 (Voice Recovery): If the frequency fluctuation range (standard deviation) of spontaneous vocalizations exceeds the normal threshold range calculated based on the patient's historical data for 3 consecutive times (N=3), it is marked as "abnormal voice recovery".

[0076] Rule 3 (Shoulder and Neck Movement): If, during standard rehabilitation exercises, the angle of leftward rotation of the shoulder and neck does not reach 80% of the target angle (e.g., 60 degrees) for 5 consecutive times (N=5), it is marked as "Limited Leftward Rotation of Shoulder and Neck Movement".

[0077] 4. Recommended intervention measures based on historical monitoring data trends: The system combines scoring trends and anomaly markers, using pre-defined decision-making logic to generate text-based recommendations. For example: If the abnormal "limited neck and shoulder movement" persists for more than 3 days and the score shows a downward trend, it is recommended to "schedule a follow-up visit this week for a manual assessment by a therapist".

[0078] If a certain score plateaus without improvement for a week, it is recommended to "consider increasing the training intensity by 10% in the next training session".

[0079] If new abnormal markers appear, it is recommended to "conduct a follow-up telephone interview to understand the patient's subjective feelings".

[0080] Step S40 is followed by S50, a task execution verification step, which includes: When the patient receives the personalized rehabilitation task and starts execution, the wearable monitoring device is controlled to enter a high sampling rate task monitoring mode; Real-time acquisition of patients' movement posture signals and electromyographic signals during training programs; The collected signals are compared with standard action templates to calculate the action standard score; If the action standard score is greater than or equal to the preset passing threshold, it is recorded as a valid check-in and the historical monitoring data is updated; if it is less than the preset passing threshold, real-time correction feedback information is output through the patient terminal.

[0081] The specific steps are as follows: To ensure the effective execution of rehabilitation tasks, the system initiates a verification process after the task is pushed out: 1. When the patient's app receives a personalized rehabilitation task and clicks "Start Training", the app will send a command to the wearable monitoring device via Bluetooth to control it to switch to a high sampling rate task monitoring mode in order to obtain more accurate motion signals.

[0082] 2. The patient performs the training according to the video instructions (such as performing a specific neck lateral flexion movement). The device acquires high-precision motion posture signals and / or electromyographic signals in real time during this process.

[0083] 3. The acquired signals are transmitted to the processing unit in real-time or near real-time and compared with a standard action template. The standard action template is derived from a standard signal model collected during demonstrations by healthy individuals or therapists. Algorithms such as dynamic time warping are used to calculate the similarity between the currently executed action and the template, resulting in an action standardization score (0-100 points).

[0084] 4. Feedback and Records: If the score for standard of movement is greater than or equal to the preset passing threshold (e.g., 80 points), the system determines that this training session is a "valid check-in" and updates the historical monitoring data, while giving the patient positive encouragement (e.g., "Standard of movement, excellent performance!").

[0085] If the score is below the passing threshold, the system outputs real-time corrective feedback via voice or vibration through the patient's app, such as: "Please note that the angle of head rotation to the left is insufficient; please try increasing the amplitude." This guides the patient to adjust immediately, ensuring the quality of training. This attempt may not be recorded as a valid completion, and the patient is encouraged to try again.

[0086] The purpose of this step is to ensure the quality of rehabilitation training, prevent patients from experiencing poor results or even injury due to improper movements, and implement intelligent supervision in the final stage of rehabilitation implementation.

[0087] It should be noted that steps S40 and S50 achieve a closed loop from "data" to "insight" and then to "action." Personalized tasks solve the problem of "one-size-fits-all" rehabilitation plans, improving patient compliance and training relevance; weekly rehabilitation reports free medical staff from heavy data processing work, allowing them to focus on abnormal warnings and decision support; and task execution verification ensures the quality of home-based rehabilitation. The entire system constitutes a complete, data-driven post-thyroidectomy rehabilitation management solution.

[0088] The beneficial effects of this invention are as follows: This invention enables the simultaneous, objective, continuous, and multimodal acquisition of physiological signals for key postoperative rehabilitation indicators (voice, neck and shoulder movement, and swallowing function). It fundamentally solves the pain points of "high subjectivity" (replacing subjective judgment based on medical experience) and "intermittent monitoring" (achieving 24 / 7 or on-demand monitoring in a home environment). By integrating multiple sensors into a single adhesive device, data can be collected non-invasively in the patient's natural living state, overcoming the limitations of traditional large-scale professional medical instruments (such as laryngoscopes and electromyography machines) in terms of application scenarios and their inability to be used for long-term daily monitoring. This provides a real and rich foundation of raw data for subsequent quantitative analysis.

[0089] This invention transforms raw, mixed physical signals into standardized, quantitative characteristic parameters with clear clinical significance. Through signal processing and feature extraction, subjective descriptions such as "a feeling of tightness in the neck" and "a slightly hoarse voice" are converted into objective values ​​such as "the leftward rotation angle of the neck is 45 degrees" and "the harmonic-to-noise ratio is 25 dB." This shifts the description of rehabilitation status from qualitative to quantitative, laying a data foundation for establishing a standardized assessment system and making the rehabilitation progress comparable between different time points and different patients.

[0090] This invention automatically generates a unified, intuitive, and traceable comprehensive rehabilitation score through an algorithmic model and a standard database, achieving automation and standardization of assessment. It effectively addresses the challenge of inconsistent assessment results among different doctors. The system maps and integrates multi-dimensional feature parameters into a single score (and sub-scores), enabling patients and medical staff to quickly and clearly grasp the overall level of rehabilitation and progress in each dimension (e.g., "Today's voice score is 85 points"). This significantly lowers the professional threshold for data interpretation, allowing patients to intuitively perceive their recovery progress and enabling medical staff to efficiently screen a large number of patients.

[0091] This invention forms an intelligent management closed loop of "monitoring-assessment-intervention," enabling personalized customization of rehabilitation plans and precise allocation of medical resources. Addressing the shortcomings of traditional rehabilitation management, which "lacks systematic tracking and immediate feedback," it transforms assessment results into concrete actions. Generating personalized tasks ensures targeted rehabilitation training and patient compliance; generating and pushing weekly reports, especially automatically annotating abnormal indicators and generating intervention suggestions, greatly improves the work efficiency of medical staff, allowing them to focus on the timely handling of abnormal cases, thus realizing a shift from passive review to proactive management. Ultimately, this feature closely integrates technological monitoring with clinical practice, improving the overall quality and efficiency of rehabilitation management.

[0092] Furthermore, such as Figure 2 As shown, based on the above-described postoperative rehabilitation status monitoring method, the present invention also provides a postoperative rehabilitation status monitoring system, the postoperative rehabilitation status monitoring system comprising: The data acquisition module 51 is used to collect the patient's acoustic vibration signals, motion posture signals and electromyographic signals in real time through a wearable monitoring device attached to the skin on both sides of the patient's neck. Data processing module 52 is used to preprocess and extract features from the acoustic vibration signal, the motion posture signal and the electromyographic signal to obtain voice feature parameters, shoulder and neck range of motion parameters and muscle strength parameters, respectively. The scoring calculation module 53 is used to compare and analyze the voice feature parameters, the shoulder and neck range of motion parameters, and the muscle strength parameters with a preset rehabilitation assessment standard library to calculate the patient's real-time rehabilitation score. The interactive feedback module 54 is used to generate personalized rehabilitation tasks and weekly rehabilitation reports based on the real-time rehabilitation score and historical monitoring data, and push the personalized rehabilitation tasks and weekly rehabilitation reports to the patient end and the medical staff end respectively.

[0093] Furthermore, such as Figure 3 As shown, based on the above-mentioned postoperative rehabilitation status monitoring method and system, the present invention also provides a terminal, which includes a processor 10, a memory 20 and a display 30. Figure 3 Only some of the terminal components are shown; however, it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0094] In some embodiments, the memory 20 may be an internal storage unit of the terminal, such as a hard disk or memory. In other embodiments, the memory 20 may be an external storage device of the terminal, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc. Further, the memory 20 may include both internal and external storage devices. The memory 20 is used to store application software and various types of data installed on the terminal, such as program code installed on the terminal. The memory 20 can also be used to temporarily store data that has been output or will be output. In one embodiment, the memory 20 stores a postoperative rehabilitation status monitoring program 40, which can be executed by the processor 10 to implement the postoperative rehabilitation status monitoring method of this application.

[0095] In some embodiments, the processor 10 may be a central processing unit (CPU), a microprocessor, or other data processing chip, used to run program code stored in the memory 20 or process data, such as executing the postoperative rehabilitation status monitoring method.

[0096] In some embodiments, the display 30 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display 30 is used to display information on the terminal and to display a visual user interface. The components of the terminal communicate with each other via a system bus.

[0097] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a postoperative rehabilitation status monitoring program, which, when executed by a processor, implements the steps of the postoperative rehabilitation status monitoring method as described above.

[0098] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal that includes that element.

[0099] Of course, those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0100] It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A method for monitoring postoperative rehabilitation status, characterized in that, The postoperative rehabilitation status monitoring method includes the following steps: Wearable monitoring devices attached to the skin on both sides of the patient's neck can collect the patient's acoustic vibration signals, motion posture signals and electromyographic signals in real time. The acoustic vibration signal, the motion posture signal, and the electromyographic signal are preprocessed and feature extracted to obtain voice feature parameters, shoulder and neck range of motion parameters, and muscle strength parameters, respectively. The voice feature parameters, shoulder and neck range of motion parameters, and muscle strength parameters are compared and analyzed with a preset rehabilitation assessment standard library to calculate the patient's real-time rehabilitation score. Based on the real-time rehabilitation score and historical monitoring data, personalized rehabilitation tasks and weekly rehabilitation reports are generated and pushed to the patient's end and the medical staff's end, respectively.

2. The postoperative rehabilitation status monitoring method according to claim 1, characterized in that, The acoustic vibration signal includes the vocal cord vibration frequency, vibration amplitude, and phonation frequency when the patient speaks; the motion posture signal includes the neck movement angle, movement amplitude, and movement completion degree when the patient performs shoulder and neck rehabilitation exercises; and the electromyographic signal includes the electromyographic peak value when the patient swallows and the neck muscles contract.

3. The postoperative rehabilitation status monitoring method according to claim 1, characterized in that, The calculation yields the patient's real-time rehabilitation score, specifically including: Map the voice feature parameters to voice scoring sub-items; The shoulder and neck mobility parameters are mapped to shoulder and neck function scoring sub-items; The muscle strength parameters are mapped to swallowing function scoring items; Obtain the preset weight coefficients corresponding to each scoring sub-item, and perform a weighted summation of the voice scoring sub-item, the shoulder and neck function scoring sub-item, and the swallowing function scoring sub-item based on all the preset weight coefficients to obtain the real-time rehabilitation score.

4. The postoperative rehabilitation status monitoring method according to claim 1, characterized in that, The generation of personalized rehabilitation tasks specifically includes: The dimension with the lowest score in the real-time rehabilitation score is identified as the current weakness in rehabilitation. Match training programs targeting the current rehabilitation weakness from a pre-set rehabilitation program library. These training programs include specific shoulder and neck exercises, vocalization exercises, or swallowing exercises. Generate task instructions that include links to motion guidance videos for the training program, suggested training frequency, and expected target scores.

5. The postoperative rehabilitation status monitoring method according to claim 1, characterized in that, The weekly rehabilitation report includes: Overall trend statistics of patient recovery data; The daily change curve of an individual patient's comprehensive rehabilitation score; Automatically labeled list of abnormal indicators and statistical duration of abnormalities; Recommended intervention measures generated based on the changing trends of historical monitoring data include at least one of the following: recommending follow-up visits, adjusting training intensity, and conducting manual telephone follow-ups.

6. The postoperative rehabilitation status monitoring method according to claim 5, characterized in that, The labeling rules for abnormal indicators in the abnormal indicator list include: If the peak electromyography value of N consecutive swallowing actions is lower than the preset percentage of the preset baseline value, it is marked as "suspected decline in swallowing function"; If the frequency fluctuation of the sound vibration during N consecutive vocalizations exceeds the normal threshold range, it is marked as "abnormal voice recovery". If the shoulder and neck range of motion fails to reach the target angle set in the rehabilitation exercises for N consecutive times, it will be marked as "limited shoulder and neck range of motion". Where N is an integer greater than or equal to 3.

7. The postoperative rehabilitation status monitoring method according to claim 1, characterized in that, The process of pushing the personalized rehabilitation task and the weekly rehabilitation report to the patient's end and the medical staff's end respectively includes: When the patient receives the personalized rehabilitation task and starts execution, the wearable monitoring device is controlled to enter a high sampling rate task monitoring mode; Real-time acquisition of patients' movement posture signals and electromyographic signals during training programs; The collected signals are compared with standard action templates to calculate the action standard score; If the action standard score is greater than or equal to the preset qualified threshold, it is recorded as a valid check-in and the historical monitoring data is updated. If the action standard score is less than the preset qualified threshold, real-time correction feedback information is output through the patient terminal.

8. A postoperative rehabilitation status monitoring system, characterized in that, The postoperative rehabilitation status monitoring system is used to implement the postoperative rehabilitation status monitoring method as described in any one of claims 1-7, and the postoperative rehabilitation status monitoring system includes: The data acquisition module is used to collect the patient's acoustic vibration signals, motion posture signals and electromyographic signals in real time through wearable monitoring devices attached to the skin on both sides of the patient's neck; The data processing module is used to preprocess and extract features from the acoustic vibration signal, the motion posture signal and the electromyographic signal to obtain voice feature parameters, shoulder and neck range of motion parameters and muscle strength parameters, respectively. The scoring calculation module is used to compare and analyze the voice feature parameters, the shoulder and neck range of motion parameters, and the muscle strength parameters with a preset rehabilitation assessment standard library to calculate the patient's real-time rehabilitation score. The interactive feedback module is used to generate personalized rehabilitation tasks and weekly rehabilitation reports based on the real-time rehabilitation score and historical monitoring data, and push the personalized rehabilitation tasks and weekly rehabilitation reports to the patient's end and the medical staff's end, respectively.

9. A terminal, characterized in that, The terminal includes: a memory, a processor, and a postoperative rehabilitation status monitoring program stored in the memory and executable on the processor. When the postoperative rehabilitation status monitoring program is executed by the processor, it implements the steps of the postoperative rehabilitation status monitoring method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a postoperative rehabilitation status monitoring program, which, when executed by a processor, implements the steps of the postoperative rehabilitation status monitoring method as described in any one of claims 1-7.

Images