Neurosurgery postoperative rehabilitation monitoring method and system based on wearable device

Abstract

The invention discloses a neurosurgery postoperative rehabilitation monitoring method and system based on wearable equipment, relates to the technical field of medical intelligent monitoring, and provides the following scheme that the neurosurgery postoperative rehabilitation monitoring method comprises the steps that a body position angle signal, a cerebral blood flow speed signal and a neck vibration energy signal of a patient are obtained, the body position angle change rate is calculated in real time based on the body position angle signal, and the body position angle change rate is calculated; if the change rate exceeds the preset change rate threshold value, the current moment is recorded as the triggering moment, the body position impact amount in the fixed time period before the triggering moment is calculated, the body position impact amount is the absolute integral value of the body position angle change acceleration, and a set of delay values are generated according to the body position impact amount and the response delay reference value calibrated before the operation. According to the method, the cerebrospinal fluid-blood vessel coupling regulation index is constructed by fusing the maximum vibration energy value and the cerebral blood flow speed change rate at the child window termination moment of the maximum vibration energy value, the defect that abnormal intracranial pressure cannot be judged by a static single parameter is overcome, and quantitative recognition of physiological and pathological intracranial pressure fluctuation is achieved.

Description

Technical Field

[0001] This invention relates to the field of medical intelligent monitoring technology, and more specifically, this application relates to a method and system for postoperative rehabilitation monitoring in neurosurgery based on wearable devices. Background Technology

[0002] Postoperative rehabilitation in neurosurgery is a crucial stage affecting the recovery of patients' neurological function, especially after major surgeries such as resection of brain tumors and repair of vascular malformations. Maintaining brain tissue stability and intracranial environment balance directly determines the quality of rehabilitation. During this period, dynamic monitoring of intracranial pressure (ICP) becomes a core indicator for assessing postoperative complications (such as cerebral edema and cerebral hemorrhage). Clinical practice shows that approximately 15-30% of neurosurgical patients experience delayed rehabilitation or require secondary surgery due to abnormal fluctuations in postoperative intracranial pressure.

[0003] Traditional monitoring methods usually rely on invasive sensor implantation or intermittent imaging examinations. The former carries the risk of infection and restricts patient activity, while the latter is difficult to capture instantaneous intracranial pressure changes caused by changes in body position. During the recovery period, patients often experience involuntary postural changes (such as suddenly sitting up or turning over) due to factors such as pain and confusion. Such actions can induce a sudden increase in intracranial pressure by altering cerebral hemodynamics and cerebrospinal fluid distribution, thus masking pathological abnormal signals.

[0004] In intracranial pressure monitoring, individual differences in patients' cerebrospinal fluid buffering capacity and vascular regulation function make it difficult to quantify and predict the trajectory of intracranial pressure fluctuations under the same positional changes. This causes existing monitoring systems to rely on static numerical capture, making it difficult to distinguish between physiological intracranial pressure fluctuations caused by positional changes and pathological intracranial pressure abnormalities, resulting in an inherent risk of confusion in signal interpretation. Furthermore, patients are prone to non-standard positional changes during rehabilitation due to pain, further increasing this risk. Therefore, a neurosurgical postoperative rehabilitation monitoring method and system based on wearable devices is proposed to solve this problem. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method and system for postoperative rehabilitation monitoring in neurosurgery based on wearable devices. This technical solution resolves the issues raised in the background section.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows: In a first aspect, this application provides a method for postoperative rehabilitation monitoring in neurosurgery based on a wearable device, the method comprising: Acquire patient position angle signals, cerebral blood flow velocity signals, and neck vibration energy signals; The rate of change of body angle is calculated in real time based on the body angle signal. If it exceeds the preset rate of change threshold, the current time is recorded as the trigger time, and the body impact amount within a fixed period before the trigger time is calculated. The body impact amount is the absolute integral value of the acceleration of the change of body angle. A set of delay values ​​is generated based on the postural impact volume and the preoperatively calibrated response delay baseline value. These values ​​are added to the trigger time to determine two endpoints, and a closed interval is constructed using the two endpoints as the delay time window. Based on the real-time rate of change of body position angle, an exponential decay function is constructed to perform nonlinear compensation on the cerebral blood flow velocity signal to obtain the compensated cerebral blood flow velocity. Within the delay time window, sub-windows are divided in real time based on the phase correlation between the real-time rate of change of body position angle and the compensating cerebral blood flow velocity. Energy integration is performed on the neck vibration energy signal in each sub-window and the maximum energy value is selected. Based on the termination time of the sub-window corresponding to the maximum energy value, the rate of change of compensating cerebral blood flow velocity is calculated, and the rate of change is weighted and fused with the maximum energy value to generate the cerebrospinal fluid-vascular coupling regulation index.

[0007] Secondly, this application provides a neurosurgical postoperative rehabilitation monitoring system based on wearable devices, used to implement the neurosurgical postoperative rehabilitation monitoring method based on wearable devices described in any of the above claims, including: The multi-source signal acquisition module is used to acquire the patient's body position angle signal, cerebral blood flow velocity signal, and neck vibration energy signal; The body position analysis and trigger detection module is used to calculate the rate of change of body position angle in real time based on the body position angle signal. If it exceeds the preset rate of change threshold, the current time is recorded as the trigger time, and the body position impact amount within a fixed period before the trigger time is calculated. The body position impact amount is the absolute integral value of the acceleration of the change of body position angle. The delay time window dynamic adjustment module is used to dynamically adjust the patient's preoperatively calibrated delay parameters according to the impact volume of the body position, and determine the delay time window at the trigger time. The cerebral blood flow signal compensation module is used to construct an exponential decay function based on the real-time rate of change of body position angle to perform nonlinear compensation on the cerebral blood flow velocity signal, thereby obtaining the compensated cerebral blood flow velocity. The sub-window segmentation and energy filtering module is used to divide the body angle change rate and the phase correlation of the compensating cerebral blood flow velocity in real time within the delay time window, perform energy integration on the neck vibration energy signal in each sub-window, and filter the maximum energy value. The cerebrospinal fluid-vascular coupling analysis module is used to calculate the rate of change of compensating cerebral blood flow velocity based on the termination time of the sub-window corresponding to the maximum energy value, and to generate the cerebrospinal fluid-vascular coupling regulation index by weighted fusion of the rate of change and the maximum energy value.

[0008] Thirdly, this application provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described feedback adjustment method based on haptic interaction.

[0009] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described feedback adjustment method based on haptic interaction.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: This application achieves individualized coverage of pathological response duration by preoperatively calibrating brainstem nerve response benchmark values ​​and vascular smooth muscle response benchmark values, and dynamically calculating the delay time window by combining real-time calculated postural impact volume, thus avoiding the missed detection of critical events caused by a fixed window. This application constructs an exponential decay function based on the rate of change of body position angle to perform nonlinear compensation on cerebral blood flow velocity signals, thereby eliminating the compensatory blood flow inhibition of the autonomic nervous system caused by sudden changes in body position, realizing the true restoration of cerebral blood flow signals, and providing a pure data basis for distinguishing between physiological and pathological fluctuations. This application solves the risk of interpretation confusion caused by asynchronous mixing of multiple source signals by dividing the sub-window based on the phase correlation between the rate of change of body position angle and the compensating cerebral blood flow velocity within the delay window, and achieves precise spatiotemporal alignment of vibration energy signals and hemodynamic events, thus separating the physiological noise associated with sudden changes in body position from the root cause. This application constructs a cerebrospinal fluid-vascular coupling regulation index by integrating the maximum vibration energy value with the rate of change of cerebral blood flow velocity at the end of its sub-window. This overcomes the inability of static single parameters to determine intracranial pressure abnormalities and achieves quantitative identification of physiological and pathological intracranial pressure fluctuations. Attached Figure Description

[0011] The disclosure of this invention is illustrated with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. Wherein: Figure 1 This is a flowchart of the neurosurgical postoperative rehabilitation monitoring method based on wearable devices proposed in this invention; Figure 2 This is a flowchart of the process for obtaining the maximum energy value in this invention; Figure 3 This is a structural block diagram of the neurosurgical postoperative rehabilitation monitoring system based on wearable devices proposed in this invention. Detailed Implementation

[0012] It is readily understood that, based on the technical solution of this invention, those skilled in the art can propose various interchangeable structural methods and implementations without altering the essential spirit of the invention. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this invention and should not be considered as the entirety of the invention or as limitations or restrictions on the technical solution of this invention.

[0013] In current technologies, postoperative rehabilitation monitoring in neurosurgery mainly relies on invasive sensor implantation or intermittent imaging examinations. Traditional invasive monitoring methods pose risks of infection and limit patient mobility, while imaging methods cannot capture transient intracranial pressure changes caused by postural shifts. Because patients often experience involuntary postoperative postoperative postoperative post-operative post-operative post-operative changes, these movements may mask pathological intracranial pressure abnormalities by altering cerebral hemodynamics and cerebrospinal fluid distribution. Current technologies struggle to quantify and predict intracranial pressure fluctuations caused by individual differences, leading to a risk of confusion between physiological and pathological signals.

[0014] To address the aforementioned issues, research analysis revealed that the phase correlation between the rate of change in body position angle and cerebral blood flow velocity can reflect the dynamic process of cerebrospinal fluid-vascular regulation. Further research showed that neck vibration energy signals can serve as an indirect characterization of cerebrospinal fluid pressure fluctuations. Based on this, a breakthrough direction is proposed: multimodal signal fusion analysis. This involves constructing a nonlinear compensation mechanism to eliminate the interference of sudden changes in body position on cerebral blood flow signals, and designing a dynamic delay window to match individualized differences in neural responses.

[0015] Reference Figure 1-2 As shown, this application proposes a method for postoperative rehabilitation monitoring in neurosurgery based on wearable devices, including: Acquire patient position angle signals, cerebral blood flow velocity signals, and neck vibration energy signals; Exemplarily, the wearable device of the present invention adopts a dual-module heterogeneous architecture, which achieves multimodal physiological signal acquisition through the collaboration of the head-mounted module and the neck ring module: the body position angle signal is directly measured by the 9-axis IMU sensor in the head-mounted module to measure the spatial pose change of the skull, with the temporal-occipital junction as the monitoring point, eliminating measurement distortion caused by neck compensatory movements or external braces; the cerebral blood flow velocity signal is captured by a miniature transcranial Doppler probe embedded in the head-mounted module, and the temporal window is adaptively located based on the preoperative CT three-dimensional reconstruction data, and a stepper motor is used to dynamically compensate for head displacement to ensure continuous and stable monitoring of the middle cerebral artery blood flow signal after surgery; the neck vibration energy signal is collected by a flexible piezoelectric sensor array in the lining of the neck ring module, and the basic cardiopulmonary vibration interference is suppressed by bandpass filtering, so as to capture sudden muscle tremors and vascular pulsation energy in a non-contact manner, avoiding compression of the surgical wound; It should be noted that this wearable device, through its wireless integrated design, can use Bluetooth for transmission, solving the problem of traditional wired monitors restricting patient movement. Furthermore, by utilizing spatial near-field coupling, it controls the signal synchronization error within an effective time range, such as 5-20ms, providing a hardware foundation for the real-time calculation of the cerebrospinal fluid-vascular coupling regulation index. Compared to separate sensor solutions, its integrated structure reduces the verification error of the spatiotemporal correlation between body position angle, cerebral blood flow, and neck vibration. The rate of change of body position angle is calculated in real time based on the body position angle signal. If it exceeds a preset rate of change threshold, the current moment is recorded as the trigger moment, and the postural impact amount within a fixed period before the trigger moment is calculated. The postural impact amount is the absolute integral value of the acceleration of the change of body position angle. For example, based on the monitoring data of natural body position changes within 48 hours after surgery for various types of neurosurgical patients (including mainstream surgical procedures such as meningioma resection, glioma resection, and craniocerebral trauma surgery), the 95th percentile of the mean rate of change of body position angle is statistically analyzed. To cover a safe range of 85% of postoperative patients, the preset rate of change threshold is set to 22. This value is a clinically validated conservative safety threshold; A set of delay values ​​is generated based on the postural impact volume and the preoperatively calibrated response delay baseline value. These values ​​are added to the trigger time to determine two endpoints, and a closed interval is constructed using the two endpoints as the delay time window. Based on the real-time rate of change of body position angle, an exponential decay function is constructed to perform nonlinear compensation on the cerebral blood flow velocity signal to obtain the compensated cerebral blood flow velocity. Within the delay time window, sub-windows are divided in real time based on the phase correlation between the real-time rate of change of body position angle and the compensating cerebral blood flow velocity. Energy integration is performed on the neck vibration energy signal in each sub-window and the maximum energy value is selected. Based on the termination time of the sub-window corresponding to the maximum energy value, the rate of change of compensating cerebral blood flow velocity is calculated, and the rate of change is weighted and fused with the maximum energy value to generate the cerebrospinal fluid-vascular coupling regulation index.

[0016] In an optional embodiment, a set of delay values ​​is generated based on the postural impact volume and the preoperatively calibrated response delay baseline value. These values ​​are then added to the trigger time to determine two endpoints. A closed interval is constructed using these two endpoints as the delay time window. Specifically, this includes: Obtain the patient's preoperative test dataset, which includes the impact volume of the test position and its corresponding brainstem nerve response delay value and vascular smooth muscle response delay value. It should be noted that the test dataset was generated using an inclined plane to implement three levels of postural impact. (tilt change), each stage lasting 45-60 seconds, with a fixed rate of 22°. The system simulates the impact of body position on the environment; it induces brainstem neural electrical activity through acoustic stimulation (such as a clicking sound), and records the time from the start of stimulation to the appearance of wave III (representing neural conduction from the superior olivary nucleus to the lateral lemniscus) as the brainstem neural response delay value, which is used to quantify the conduction efficiency of the brainstem neural pathway; it also detects the frequency shift of erythrocyte motion-scattered light by emitting low-intensity laser light to the capillary bed of the fingertip, reflecting changes in microcirculation blood perfusion in real time, and uses the time difference from body position change to the peak of blood flow velocity decrease as the vascular smooth muscle response delay value to assess the contraction and relaxation response speed of vascular smooth muscle; For example, after meningioma resection, brainstem edema increases the brainstem nerve response delay to 1.8-2.5s (normal 1.0-1.4s), leading to decreased autonomic nerve compensation efficiency during positional changes; glioma patients have impaired vasomotor function postoperatively, with the delay value rising to 3.2-4.0s (normal 2.2-2.8s), and the duration of intracranial pressure fluctuations is prolonged by 120%. The formula for calculating postural impact is: ; In the formula, This refers to the postural impact force. The acceleration is the change in body position angle. The starting point of a fixed time period before the trigger time. The trigger time; for example, based on the autonomic nervous system response delay period, a fixed time period of 1.2s is taken; The brainstem response delay values ​​and vascular smooth muscle response delay values ​​were denoised and their mean values ​​were calculated to generate the corresponding brainstem neural response baseline values. and vascular smooth muscle response benchmark value Denoising can be performed using wavelet thresholding, with the db4 wavelet basis selected and a decomposition level of 5. Only valid data with a coefficient of variation <15% from 3 repeated experiments are retained. The scaling factor is the ratio of the brainstem nerve response baseline value minus the brainstem nerve response baseline value to the impact amount of the test position. It should be noted that the scaling factor , To test the impact force of body position. This represents the neural response efficiency per unit impact quantity; when the neural response efficiency is high (i.e., When the scaling factor is small, the monitoring time window is increased to accommodate efficient compensation; conversely, when the scaling factor is large (such as postoperative nerve injury), the monitoring time window is shortened. The delay value at the end of the delay time window is calculated based on the baseline constraint formula, and the delay time window is determined as follows: ; It should be noted that the starting point of the delay time window is the moment of brainstem nerve response activation, capturing the initial compensatory signal of the autonomic nervous system, and its ending point is the sum of the moment of body position change and the preset delay value, defining the maximum duration of intracranial pressure fluctuation and covering the complete response cycle of vascular smooth muscle; the brainstem nerve response benchmark value quantifies the nerve conduction efficiency, and its scaling factor dynamically adjusts the length of the delay time window; the vascular smooth muscle response benchmark value characterizes vascular vasomotor inertia, determines the basic length of the delay time window and reflects individual regulatory ability; Compared with existing technologies, traditional methods use a fixed delay time window to analyze intracranial pressure signals, which cannot distinguish individual physiological response differences caused by postural impact, leading to missed pathological signals. This approach establishes a dual-benchmark parameter system through preoperative testing, and dynamically adjusts the delay time window in combination with real-time postural impact, while simultaneously preserving individualized physiological characteristics and introducing a dynamic compensation mechanism. The benchmark constraint formula is: ; In the formula, The delay value at the termination time. This is the scaling factor. This represents the postural impact magnitude at the trigger moment. This represents the vascular smooth muscle response delay value. Through the above technical solution, this application can dynamically adjust the monitoring time window according to the patient's individual physiological characteristics and real-time postural impact intensity, effectively solving the problem of mismatch between the fixed delay time window and the time-varying characteristics of physiological response.

[0017] In an optional embodiment, an exponential decay function is constructed based on the real-time rate of change of body position angle to perform nonlinear compensation on the cerebral blood flow velocity signal, thereby obtaining the compensated cerebral blood flow velocity, specifically including: It should be noted that the necessity of nonlinear compensation lies in the physiological abnormalities of patients after neurosurgery: sudden changes in body position trigger excessive autonomic inhibition through brainstem pressure reflex, resulting in a false decrease in the original cerebral blood flow velocity signal, which masks the pathological blood flow abnormality; nonlinear compensation through an exponential decay function can dynamically eliminate the neural inhibition effect using the preoperatively calibrated brainstem response benchmark value, restore the vascular regulation function signal, and provide distortion-free input for subsequent phase correlation analysis and cerebrospinal fluid-vascular coupling regulation index; Based on the baseline value of brainstem neural response, the product of its reciprocal and a preset proportional coefficient is used as the blood flow inhibition coefficient; whereby the blood flow inhibition coefficient reflects the intensity of vasomotor inhibition caused by brainstem compression, and is negatively correlated with the rate of change of body position, and the preset proportional coefficient can be taken as 0.4-0.8. Substituting the blood flow inhibition coefficient and the real-time rate of change of body position angle into the exponential function, an exponential decay function is generated; It should be noted that when the angular velocity of body position increases, the brainstem vasomotor center triggers a compensatory vasoconstriction response via pressure reflex, resulting in a decay response characteristic of cerebral blood flow velocity that is first rapidly inhibited and then gradually approaches saturation. This characteristic is highly consistent with the mathematical characteristics of exponential functions. The cerebral blood flow velocity signal is multiplied point by point with the exponential decay function at the same time to obtain the compensated cerebral blood flow velocity. The point-by-point multiplication at the same time ensures that the cerebral blood flow velocity signal is time-aligned with the body position angle signal, avoiding compensation distortion caused by phase shift. The exponential decay function is: ; In the formula, It is an exponentially decaying function. The blood flow inhibition coefficient and , This represents the real-time rate of change of body position angle. This is the baseline value for brainstem neural response. This is a preset proportional coefficient; Through the above technical solution, this application can effectively eliminate the interference of body position changes on cerebral blood flow velocity signals and accurately restore the real hemodynamic characteristics. This compensation mechanism solves the problem of compensation inaccuracy caused by individual differences in neural regulation in traditional methods by dynamically coupling the neural response benchmark value with body position changes, and provides a reliable data basis for the accurate identification of subsequent intracranial pressure abnormal signals.

[0018] In an optional embodiment, sub-windows are divided in real time based on the phase correlation between the real-time rate of change of body position angle and the compensating cerebral blood flow velocity, specifically including: Hilbert transforms were performed on the real-time rate of change of body position angle and the cerebral blood flow velocity after compensation, and the instantaneous phase of the two was extracted and the phase difference was calculated. Within the delay time window, several sub-windows are divided, starting from the local minimum of the phase difference and with the neural response cycle as the window length. The neural response cycle is the brainstem neural response baseline value multiplied by a preset window coefficient. It should be noted that during the change of body position, when the brainstem nerve command and the vascular smooth muscle relaxation / contraction response achieve temporal synchronization, the instantaneous phase difference reaches a minimum value. This minimum value corresponds to the time phase-lock point between the rising edge of the nerve impulse signal and the peak of the vascular response wave, marking the peak effect of neurovascular coupling. Using this minimum value as the starting point of the sub-window can optimally capture the blood flow event dominated by coupling. The preset window coefficient is obtained by linearly regressing the brainstem nerve response delay value measured by the preoperative tilt table test with the brainstem nerve response benchmark value. The goodness of fit is required to be greater than 0.85 and the sample size is greater than 30 cases. Specifically, 1.2 can be used. The fluctuation range of the phase difference is calculated for each sub-window. If it exceeds the preset fluctuation threshold, the preset window coefficient is dynamically adjusted through the smoothing limit function until the fluctuation range is lower than the preset fluctuation threshold. It should be noted that the fluctuation range is the difference between the maximum and minimum phase difference within the sub-window, quantifying the synchronization stability of the neurovascular signal. When the fluctuation range exceeds the preset fluctuation threshold, the preset window coefficient is reduced by a smoothing constraint function (e.g., from 1.2 to 1.04). Starting from the original minimum phase difference value, the sub-windows are re-divided according to the updated window length. This process is iterated until the fluctuation range of all sub-windows meets the preset fluctuation threshold requirement, ensuring that the signal synchronization meets the standard. The preset fluctuation threshold was obtained based on tilt table tests of healthy volunteers (more than 100 cases). Body position angular velocity and cerebral blood flow velocity signals were recorded simultaneously. The instantaneous phase difference sequence was extracted using Hilbert transform, and the 95th percentile was used as the base threshold (typical value 0.15-0.25 rad). Postoperative application involved dynamic fine-tuning based on preoperative test data: for brainstem compression patients due to nerve conduction inhibition, the preset fluctuation threshold was reduced to 0.9 times the typical value; for vascular lesion patients due to endothelial dysfunction, the preset fluctuation threshold was expanded to 1.25 times the typical value. Verification after fine-tuning: if fluctuations exceeded the limit for three consecutive sub-windows, a second calibration was performed in ±0.03 rad increments until the exceedance rate was <5%. For example, the smoothing constraint function is: In the formula, The preset window size after scaling down. For the preset window coefficient, This is the smoothing intensity coefficient, with a default value of 5. For phase difference, The preset fluctuation threshold is used; Compared with existing technologies, traditional methods typically use fixed time windows for signal segmentation, which is difficult to adapt to the dynamic fluctuations in phase difference caused by sudden changes in body position, resulting in a mismatch between sub-window division and physiological response cycle. However, this application effectively solves the mismatch between sub-window length and neural response cycle under sudden changes in body position by using phase synchronization start point detection and dynamic window adjustment mechanism. Through the above technical solution, this application achieves precise division of sub-windows for brainstem nerve response feature adaptation, ensures the effective time range for neck vibration energy integral calculation, reduces the risk of incorrect energy value selection caused by phase difference fluctuations, and lays a reliable data foundation for the accurate generation of the cerebrospinal fluid-vascular coupling regulation index.

[0019] In an optional embodiment, the nonlinear compensation of the cerebral blood flow velocity signal further includes a dynamic feedback mechanism, specifically comprising: Calculate the standard deviation of the phase difference fluctuation range of each sub-window. If it exceeds the preset standard deviation threshold, update the preset scaling factor. It should be noted that the instantaneous phase difference sequence at the time of obtaining the preset fluctuation threshold is extracted, the standard deviation of the phase difference sequence fluctuation for each volunteer is calculated, and the 95th percentile of the standard deviation distribution of all volunteers is taken as the basic threshold. The threshold is then dynamically fine-tuned and verified in combination with preoperative test data. The preset standard deviation threshold can be specifically set to 0.15±0.03 rad. The adjustment direction is determined based on the positive or negative trend of the standard deviation changing with the preset proportional coefficient value; The deviation rate between the standard deviation and the preset standard deviation threshold is multiplied by the preset learning rate coefficient to generate the adjustment range; The preset proportional coefficient is iteratively updated based on the adjustment direction and adjustment range; The specific adjustment process of the preset proportional coefficient is as follows: Calculate the slope of the change of the standard deviation with respect to the preset proportional coefficient. Based on the three most recent valid historical records and with an absolute value of the slope greater than 0.01, decrease the coefficient if the slope is positive and increase it if it is negative to determine the adjustment direction; calculate the ratio of the absolute value of the difference between the standard deviation and the preset threshold to the threshold as the deviation rate; multiply the preset learning rate coefficient (which can be 0.02) by the deviation rate to obtain the adjustment range; the product of the adjustment direction and the adjustment range is the adjustment amount of the preset proportional coefficient. The compensation cerebral blood flow velocity is recalculated using the updated preset proportional coefficient, and the sub-windows are re-divided. If the standard deviation of the phase difference fluctuation of each sub-window does not exceed the preset threshold after re-division, the iteration is terminated. Otherwise, the iteration is terminated when the absolute value of the adjustment amount is less than 0.005 for three consecutive times or the total number of iterations is greater than 50. If the standard deviation still exceeds the threshold at this time, an alarm is triggered and manual intervention is initiated. Through the above technical solution, this application solves the signal distortion problem caused by fixed parameters in the traditional cerebral blood flow velocity compensation process. By dynamically adjusting the preset proportional coefficient, the compensated cerebral blood flow velocity maintains a stable phase relationship with the change in body position angle, thereby improving the accuracy of subsequent energy integral and coupling regulation index calculation.

[0020] In an optional embodiment, the neck vibration energy signal within each sub-window is integrated and the maximum energy value is selected, specifically including: The neck vibration energy signal within each sub-window is integrated to generate a sequence of integral values. Calculate the mean and standard deviation of the integral value sequence; If there exists an integral value that exceeds the sum of the mean and a fixed multiple of the standard deviation, then that integral value is excluded from the sequence of integral values; for example, the fixed multiple can be 3. The sequence of excluded integral values ​​is sorted in ascending order and its mean is updated. If the difference between the maximum and the second largest value exceeds twice the effective data range, the second largest value is determined to be the maximum energy value. Otherwise, the maximum value is taken as the maximum energy value. The effective data range is the updated mean minus the integral value of the first term of the sequence. Compared with existing technologies, traditional methods directly select the maximum value of the integral value sequence as the energy feature, which is easily affected by transient noise or abnormal signal interference caused by sudden changes in body position. This scheme introduces a statistical screening mechanism to coordinate dynamic threshold adjustment, and reliably identifies the maximum energy value that reflects the true physiological state.

[0021] In an optional embodiment, the cerebrospinal fluid-vascular coupling modulation index is: ; In the formula, The cerebrospinal fluid-vascular coupling regulatory index. To compensate for the rate of change in cerebral blood flow velocity, The maximum energy value, The total duration of the delay time window. It is the attenuation constant and its value range is [0.05, 0.2]. , The weighting coefficients and , The time at which the sub-window terminates corresponds to the maximum energy value. For example, a patient with vascular disorders. cerebrospinal fluid disorder patients ; Specifically, the cerebrospinal fluid-vascular coupling regulation index (VEI) accurately distinguishes between physiological and pathological intracranial pressure fluctuations through the following mechanisms: S1. Quantitative item on vascular regulation capacity ( ): This item is positive in physiological fluctuations (such as compensatory vasoconstriction in the brain). >0), offsetting blood pressure fluctuations caused by postural changes; in pathological fluctuations, this item is negative or low (e.g., vasoparesis, ≤0), the blood vessels lose their compensatory ability; S2. Quantitative item of cerebrospinal fluid buffering strength ( ),in For the Sigmoid function: Physiological fluctuations Low, and energy is released promptly ( The Sigmoid value is approximately 0.5. Pathological fluctuations Abnormally high, and release delayed ( The Sigmoid value approaches 1. S3. Temporal Coordination Determination (Sigmoid function characterizes the delayed properties of cerebrospinal fluid pressure transmission): During physiological fluctuations, vasoconstriction and energy release are synchronized, and the Sigmoid value remains stable (around 0.5), reflecting cerebrospinal fluid-vascular coupling. Temporal decoupling between the two in pathological fluctuations: Vascular disorders are characterized by hysteresis of neck vibration energy, with signal characteristics as follows: The amplitude is normal, but After a delay of 3 seconds or more, the Sigmoid value increases slightly (typical value). At that time, the Sigmoid value rose to around 0.75. Cerebrospinal fluid disorders manifest as cervical vibration energy retention, with signal characteristics as follows: The amplitude increased abnormally and the high-energy state persisted, and the Sigmoid value increased significantly (typical value). At that time, the Sigmoid value rose to around 0.95. S4. Weighted fusion of quantitative terms on vascular regulation capacity and cerebrospinal fluid buffering strength: Physiological fluctuations The mean is 0.53 ± 0.07, and the distribution is concentrated. Pathological fluctuations: Vascular disorders The mean value was 0.76 ± 0.12, in the cerebrospinal fluid disorder group. The mean was 0.82 ± 0.09; Compared with existing technologies, traditional methods usually use a single blood flow parameter or static threshold to judge intracranial pressure abnormalities, which cannot effectively separate physiological fluctuations from pathological factors that interfere with the monitoring signal. This scheme integrates vascular regulation capacity and cerebrospinal fluid buffering strength, and constructs a cerebrospinal fluid-vascular coupling regulation index based on the delayed characteristics of cerebrospinal fluid pressure transmission to quantify the dynamic synergy of the vascular-cerebrospinal fluid system. Combined with the numerical range of the cerebrospinal fluid-vascular coupling regulation index, it can more accurately distinguish between physiological intracranial pressure fluctuations and pathological intracranial pressure abnormalities caused by postural changes.

[0022] See Figure 3 As shown, this solution proposes a neurosurgical postoperative rehabilitation monitoring system based on wearable devices to implement the aforementioned neurosurgical postoperative rehabilitation monitoring method based on wearable devices, including: The multi-source signal acquisition module is used to acquire the patient's body position angle signal, cerebral blood flow velocity signal, and neck vibration energy signal; The body position analysis and trigger detection module is used to calculate the rate of change of body position angle in real time based on the body position angle signal. If it exceeds the preset rate of change threshold, the current time is recorded as the trigger time, and the body position impact amount within a fixed period before the trigger time is calculated. The body position impact amount is the absolute integral value of the acceleration of the change of body position angle. The delay time window dynamic adjustment module is used to dynamically adjust the patient's preoperatively calibrated delay parameters according to the impact volume of the body position, and determine the delay time window at the trigger time. The cerebral blood flow signal compensation module is used to construct an exponential decay function based on the real-time rate of change of body position angle to perform nonlinear compensation on the cerebral blood flow velocity signal, thereby obtaining the compensated cerebral blood flow velocity. The sub-window segmentation and energy filtering module is used to divide the body angle change rate and the phase correlation of the compensating cerebral blood flow velocity in real time within the delay time window, perform energy integration on the neck vibration energy signal in each sub-window, and filter the maximum energy value. The cerebrospinal fluid-vascular coupling analysis module is used to calculate the rate of change of compensating cerebral blood flow velocity based on the termination time of the sub-window corresponding to the maximum energy value, and to generate the cerebrospinal fluid-vascular coupling regulation index by weighted fusion of the rate of change and the maximum energy value.

[0023] In another embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above embodiments.

[0024] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps described above.

[0025] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and executes the computer instructions, causing the computer device to perform the steps described above.

[0026] The technical scope of this invention is not limited to the content described above. Those skilled in the art can make various modifications and variations to the above embodiments without departing from the technical concept of this invention, and all such modifications and variations should fall within the protection scope of this invention.

Claims

1. A method for postoperative rehabilitation monitoring in neurosurgery based on wearable devices, characterized in that, The method includes: Acquire patient position angle signals, cerebral blood flow velocity signals, and neck vibration energy signals; The rate of change of body angle is calculated in real time based on the body angle signal. If it exceeds the preset rate of change threshold, the current time is recorded as the trigger time, and the body impact amount within a fixed period before the trigger time is calculated. The body impact amount is the absolute integral value of the acceleration of the change of body angle. A set of delay values ​​is generated based on the postural impact volume and the preoperatively calibrated response delay baseline value. These values ​​are added to the trigger time to determine two endpoints, and a closed interval is constructed using the two endpoints as the delay time window. Based on the real-time rate of change of body position angle, an exponential decay function is constructed to perform nonlinear compensation on the cerebral blood flow velocity signal to obtain the compensated cerebral blood flow velocity. Within the delay time window, sub-windows are divided in real time based on the phase correlation between the real-time rate of change of body position angle and the compensating cerebral blood flow velocity. Energy integration is performed on the neck vibration energy signal in each sub-window and the maximum energy value is selected. Based on the termination time of the sub-window corresponding to the maximum energy value, the rate of change of compensating cerebral blood flow velocity is calculated, and the rate of change is weighted and fused with the maximum energy value to generate the cerebrospinal fluid-vascular coupling regulation index.

2. The method according to claim 1, characterized in that, A set of delay values ​​is generated based on the postural impact volume and the preoperatively calibrated response delay baseline value. These values ​​are added to the trigger time to determine two endpoints. The interval between the two endpoints is recorded as the delay time window, which specifically includes: Obtain the patient's preoperative test dataset, which includes the impact volume of the test position and its corresponding brainstem nerve response delay value and vascular smooth muscle response delay value. The brainstem response delay values ​​and vascular smooth muscle response delay values ​​were denoised and their mean values ​​were calculated to generate the corresponding brainstem neural response baseline values. and vascular smooth muscle response benchmark value ; The scaling factor is the ratio of the brainstem nerve response baseline value minus the brainstem nerve response baseline value to the impact amount of the test position. The delay value at the end of the delay time window is calculated based on the baseline constraint formula, and the delay time window is determined as follows: ; The benchmark constraint formula is: ； In the formula, The delay value at the termination time. This is the scaling factor. This represents the postural impact magnitude at the trigger moment. This represents the delay value for the vascular smooth muscle response.

3. The method according to claim 2, characterized in that, Based on the real-time rate of change of body position angle, an exponential decay function is constructed to perform nonlinear compensation on the cerebral blood flow velocity signal, thereby obtaining the compensated cerebral blood flow velocity, specifically including: Based on the brainstem neural response baseline, the product of its reciprocal and a preset proportional coefficient is used as the blood flow inhibition coefficient; Substituting the blood flow inhibition coefficient and the real-time rate of change of body position angle into the exponential function, an exponential decay function is generated; The cerebral blood flow velocity signal is multiplied point by point at the same time to obtain the compensated cerebral blood flow velocity; The exponential decay function is: ； In the formula, It is an exponentially decaying function. The blood flow inhibition coefficient and , This represents the real-time rate of change of body position angle. This is the baseline value for brainstem neural response. This is the preset scaling factor.

4. The method according to claim 3, characterized in that, The system divides the brain into sub-windows in real time based on the phase correlation between the rate of change of body position angle and the compensating cerebral blood flow velocity, specifically including: Hilbert transforms were performed on the real-time rate of change of body position angle and the cerebral blood flow velocity after compensation, and the instantaneous phase of the two was extracted and the phase difference was calculated. Within the delay time window, several sub-windows are divided, starting from the local minimum of the phase difference and with the neural response cycle as the window length. The neural response cycle is the brainstem neural response baseline value multiplied by a preset window coefficient. The fluctuation range of the phase difference is calculated for each sub-window. If it exceeds the preset fluctuation threshold, the preset window coefficient is dynamically adjusted through a smoothing limit function until the fluctuation range is lower than the preset fluctuation threshold.

5. The method according to claim 4, characterized in that, Nonlinear compensation for cerebral blood flow velocity signals also includes a dynamic feedback mechanism, specifically including: Calculate the standard deviation of the phase difference fluctuation range of each sub-window. If it exceeds the preset standard deviation threshold, update the preset scaling factor. The adjustment direction is determined based on the positive or negative trend of the standard deviation changing with the preset proportional coefficient value; The deviation rate between the standard deviation and the preset standard deviation threshold is multiplied by the preset learning rate coefficient to generate the adjustment range; The preset proportional coefficient is iteratively updated based on the adjustment direction and adjustment range; The compensated cerebral blood flow velocity is recalculated using the updated preset scaling factor, and the sub-windows are redefined.

6. The method according to claim 1, characterized in that, The energy of the neck vibration energy signal within each sub-window is integrated, and the maximum energy value is selected. Specifically, this includes: The neck vibration energy signal within each sub-window is integrated to generate a sequence of integral values. Calculate the mean and standard deviation of the integral value sequence; If there exists an integral value that exceeds the sum of the mean and a fixed multiple of the standard deviation, then that integral value is excluded from the sequence of integral values. The sequence of excluded integral values ​​is sorted in ascending order and its mean is updated. If the difference between the maximum and the second largest value exceeds twice the effective data range, the second largest value is determined to be the maximum energy value; otherwise, the maximum value is taken as the maximum energy value. The effective data range is the updated mean minus the integral value of the first term of the sequence.

7. The method according to claim 1, characterized in that, The cerebrospinal fluid-vascular coupling regulatory index is: ； In the formula, The cerebrospinal fluid-vascular coupling regulatory index. To compensate for the rate of change in cerebral blood flow velocity, The maximum energy value, The total duration of the delay time window. The attenuation constant is , These are the weighting coefficients. The time at which the sub-window terminates corresponds to the maximum energy value. The trigger time.

8. A neurosurgical postoperative rehabilitation monitoring system based on wearable devices, characterized in that, The method for implementing postoperative neurosurgical rehabilitation monitoring based on a wearable device as described in any one of claims 1-7 includes: The multi-source signal acquisition module is used to acquire the patient's body position angle signal, cerebral blood flow velocity signal, and neck vibration energy signal; The body position analysis and trigger detection module is used to calculate the rate of change of body position angle in real time based on the body position angle signal. If it exceeds the preset rate of change threshold, the current time is recorded as the trigger time, and the body position impact amount within a fixed period before the trigger time is calculated. The body position impact amount is the absolute integral value of the acceleration of the change of body position angle. The delay time window dynamic adjustment module is used to dynamically adjust the patient's preoperatively calibrated delay parameters according to the impact volume of the body position, and determine the delay time window at the trigger time. The cerebral blood flow signal compensation module is used to construct an exponential decay function based on the real-time rate of change of body position angle to perform nonlinear compensation on the cerebral blood flow velocity signal, thereby obtaining the compensated cerebral blood flow velocity. The sub-window segmentation and energy filtering module is used to divide the body angle change rate and the phase correlation of the compensating cerebral blood flow velocity in real time within the delay time window, perform energy integration on the neck vibration energy signal in each sub-window, and filter the maximum energy value. The cerebrospinal fluid-vascular coupling analysis module is used to calculate the rate of change of compensating cerebral blood flow velocity based on the termination time of the sub-window corresponding to the maximum energy value, and to generate the cerebrospinal fluid-vascular coupling regulation index by weighted fusion of the rate of change and the maximum energy value.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the method described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-7.

Images