Intelligent drainage method and system based on cerebrospinal fluid pressure

Abstract

The invention relates to the technical field of medical data processing, in particular to an intelligent drainage method and system based on cerebrospinal fluid pressure, and the method comprises the steps: obtaining a pressure signal curve and a pulse period of a cerebrospinal fluid intelligent drainage system at each detection position; obtaining a synchronous conduction error value according to the signal change trend in the pulse period of the to-be-analyzed signal curves of each detection position and other detection positions in combination with the signal change trend between the historical signal curves; according to the signal delay condition between pulse periods in the signal curve to be analyzed of each detection position and other detection positions, combining the signal delay condition of the historical signal curve to obtain a signal delay characteristic value; according to the synchronous conduction error value, the signal delay characteristic value and the signal change trend, analyzing the signal artifacts of each pulse period in the signal curve to be analyzed to obtain an artifact-removed signal; the de-artifact signal is used for a drainage operation. According to the invention, the occurrence of misrecognition is reduced, and the reliability of intelligent drainage decision is ensured.

Description

Technical Field

[0001] This invention relates to the field of medical data processing technology, specifically to an intelligent drainage method and system based on cerebrospinal fluid pressure. Background Technology

[0002] Cerebrospinal fluid (CSF) drainage is a crucial treatment for neurological disorders such as hydrocephalus and intracranial hypertension. Traditional drainage methods rely on the physician's clinical experience to manually adjust the drainage valve opening, monitoring CSF pressure. With the development of precision medicine and intelligent sensing technology, automated, closed-loop control based on real-time intracranial pressure (ICP) monitoring has become a research hotspot. This method dynamically adjusts CSF flow using drainage valves or pumps to maintain intracranial pressure within physiological fluctuations, thereby improving treatment safety and efficacy. Therefore, the accuracy of CSF intracranial pressure monitoring results is paramount for CSF drainage procedures.

[0003] However, in the process of real-time monitoring of intracranial pressure, the actual clinical environment is easily affected by a variety of non-pathological factors, including transient physiological fluctuations such as breathing and coughing, artifacts in the tubing system itself (such as air bubbles and blockages), and human operation (such as body position adjustment or valve opening and closing). This means that when using the preset normal range of intracranial pressure to identify abnormal conditions, there is a possibility that environmental interference abnormalities may be misidentified as pathological intracranial hypertension, thus failing to guarantee the effectiveness of cerebrospinal fluid drainage. Summary of the Invention

[0004] To address the technical problem of misidentification when using a preset normal range of intracranial pressure to identify abnormalities, the present invention aims to provide an intelligent drainage method and system based on cerebrospinal fluid pressure. The specific technical solution adopted is as follows: In a first aspect, the present invention provides an intelligent drainage method based on cerebrospinal fluid pressure, comprising: The pressure signal curves of the intelligent cerebrospinal fluid drainage system at each detection location and each pulse period of the pressure signal are obtained. The pressure signal curves include the signal curve to be analyzed and historical signal curves. Based on the difference in signal change trends between the signal curves to be analyzed at each detection position and other detection positions within the same pulse period, and combined with the difference in signal trend between the signal curves to be analyzed at each detection position and all historical signal curves in each pulse period, the synchronization propagation error value of each detection position in each pulse period of the signal curves to be analyzed between other detection positions is obtained. Based on the signal delay between corresponding pulse periods in the signal curves to be analyzed at each detection position and other detection positions, and combined with the signal delay between historical signal curves at each detection position and other detection positions, the signal delay characteristic value of each pulse period in the signal curves to be analyzed at each detection position and other detection positions is obtained. Based on the synchronous conduction error value and the signal delay characteristic value combined with the signal change trend, the signal artifacts in each pulse period of the signal curve to be analyzed are obtained to obtain the artifact-free signal; the artifact-free signal is used for the drainage operation.

[0005] Preferably, the step of obtaining the synchronization propagation error value of each detection position and the signal curves to be analyzed at other detection positions within the same pulse period based on the difference in signal change trends of each detection position and other detection positions within the same pulse period, combined with the difference in signal trend of each detection position's signal curve and all historical signal curves in each pulse period, specifically includes: Based on the difference between the pulse trend of the signal curve to be analyzed at each detection location in each pulse cycle and the pulse trend of all historical signal curves in each pulse cycle, the trend change factor of the signal curve to be analyzed at each detection location in each pulse cycle is obtained. Based on the differences in the trend change factors of the signal curves to be analyzed at each detection location and the other detection locations within the same pulse cycle, as well as the differences in the pressure signals, the degree of characteristic synchronization between the signal curves to be analyzed at each detection location and the other detection locations within each pulse cycle is obtained. Based on the characteristic synchronization degree between the signal curve to be analyzed at each detection position and other detection positions in each pulse cycle, and the balance of the characteristic synchronization degree of the same detection position in all pulse cycles of the historical signal curve, the synchronization propagation error value of each detection position in each pulse cycle of the signal curve to be analyzed with other detection positions is obtained.

[0006] Preferably, the step of obtaining the trend change factor of the signal curve to be analyzed at each detection location in each pulse cycle based on the difference between the pulse trend of the signal curve to be analyzed at each detection location in each pulse cycle and the pulse trend of all historical signal curves in each pulse cycle specifically includes: Any pulse period of the signal curve to be analyzed at any detection position is recorded as the target pulse period. The peak-to-peak value of the target pulse period is used as the first trend parameter of the target pulse period. Based on the fitting trend of the pressure signal within the target pulse period, the second trend parameter of the target pulse period is determined. The mean of the first trend parameter in each pulse period of each historical signal curve is taken as the first historical trend value, and the mean of the second trend parameter in each pulse period of each historical signal curve is taken as the second historical trend value. The trend change factor of the target pulse period is determined based on the difference between the first trend parameter and the first historical trend value of the target pulse period, and the difference between the second trend parameter and the second historical trend value of the target pulse period.

[0007] Preferably, the step of obtaining the characteristic synchronization degree between the signal curve to be analyzed at each detection location and other detection locations in each pulse cycle based on the difference in the trend change factor of the signal curve to be analyzed at each detection location and the difference in the pressure signal, specifically includes: The pulse period with the greatest time overlap between the signal curve to be analyzed at each other detection position and the target pulse period within the same time interval is determined as the control pulse period corresponding to the target pulse period at each other detection position. Based on the negative correlation coefficient of the difference in trend change factors between the target pulse cycle and each control pulse cycle, the trend consistency coefficient between the target pulse cycle and each control pulse cycle is determined. Based on the balance of the negative correlation coefficient between the pressure signals at corresponding times within the overlapping time interval, the waveform matching coefficient between the target pulse period and each control pulse period is determined. The characteristic synchronization degree between the target pulse period and each control pulse period is obtained by combining the trend consistency coefficient and the waveform matching coefficient.

[0008] Preferably, the step of obtaining the synchronization propagation error value of each detection position in each pulse cycle of the signal curve to be analyzed between each detection position and other detection positions based on the balance of the characteristic synchronization degree between each detection position and the characteristic synchronization degree of the same detection position in all pulse cycles of the historical signal curve, specifically includes: The balance of all feature synchronization degrees between the detection position corresponding to the target pulse period and the corresponding pulse period of other detection positions on the historical signal curve is obtained, and is denoted as the historical synchronization feature factor of the detection position corresponding to the target pulse period. Based on the difference between the characteristic synchronization degree between the target pulse cycle and each control pulse cycle and the historical synchronization characteristic factor, and combined with the negative correlation coefficient of the characteristic synchronization degree, the synchronization transmission error value of the target pulse cycle and each control pulse cycle is determined.

[0009] Preferably, the step of obtaining the signal delay characteristic value of each pulse period in the signal curve to be analyzed between each detection position and other detection positions based on the signal delay between corresponding pulse periods in the signal curve to be analyzed between each detection position and other detection positions, combined with the signal delay between historical signal curves of each detection position and other detection positions, specifically includes: Based on the differences in the start and end times of the corresponding pulse periods in the signal curves to be analyzed between each detection position and other detection positions, the time difference factor of the corresponding pulse periods in the signal curves to be analyzed between each detection position and other detection positions is obtained. Obtain the equalization coefficient of the time difference factor of the corresponding pulse period in the historical signal curves of each detection position and other detection positions, and denote it as the historical time coefficient; Based on the difference between the time difference factor of the corresponding pulse period in the signal curve to be analyzed at each detection position and the historical time coefficient, the signal delay characteristic value of each pulse period in the signal curve to be analyzed at each detection position and the signal curve to be analyzed at the other detection positions is determined.

[0010] Preferably, the step of obtaining the time difference factor of the corresponding pulse period in the signal curve to be analyzed between each detection position and other detection positions based on the difference in start time and end time of the corresponding pulse period in the signal curve to be analyzed between each detection position and other detection positions specifically includes: The time interval between the start time of the target pulse cycle and the start time of each control pulse cycle is used as the first time coefficient; the time interval between the end time of the target pulse cycle and the end time of each control pulse cycle is used as the second time coefficient; based on the balance between the first time coefficient and the second time coefficient, the time difference factor between the target pulse cycle and each control pulse cycle is determined.

[0011] Preferably, the step of analyzing the signal artifacts in each pulse period of the signal curve to be analyzed based on the synchronization transmission error value and the signal delay characteristic value combined with the signal change trend specifically includes: By combining the balanced distribution of the synchronous conduction error value and time difference factor between the target pulse cycle and each control pulse cycle, the abnormal conduction characteristic value of the target pulse cycle is determined. The degree of signal artifact in the target pulse period is determined by the product of the trend change factor of the target pulse period and the abnormal conduction characteristic value. Whether to mark the target pulse period is determined based on the degree of signal artifacts in the target pulse period.

[0012] Preferably, determining whether to mark the target pulse period based on the degree of signal artifacts in the target pulse period specifically includes: When the signal artifact level of the target pulse period is greater than the preset artifact threshold, the target pulse period is marked as an artifact signal; when the signal artifact level of the target pulse period is less than or equal to the preset artifact threshold, the target pulse period is not marked.

[0013] Secondly, the present invention provides an intelligent drainage system based on cerebrospinal fluid pressure, which is used to perform the steps of an intelligent drainage method based on cerebrospinal fluid pressure, the intelligent drainage system based on cerebrospinal fluid pressure comprising: The data acquisition module is used to acquire the pressure signal curve of the intelligent cerebrospinal fluid drainage system at each detection location and each pulse period of the pressure signal, wherein the pressure signal curve includes the signal curve to be analyzed and the historical signal curve; The synchronous transmission analysis module is used to obtain the synchronous transmission error value of each detection position and the signal curves of other detection positions in each pulse period based on the difference in signal change trends of each detection position and other detection positions within the same pulse period, combined with the signal trend differences of each detection position's signal curve and all historical signal curves in each pulse period. The time delay analysis module is used to obtain the signal delay characteristic value of each pulse period in the signal curves to be analyzed between each detection position and other detection positions based on the signal delay between corresponding pulse periods in the signal curves to be analyzed between each detection position and other detection positions, combined with the signal delay between historical signal curves of each detection position and other detection positions. The artifact detection and analysis module is used to analyze the artifact nature of the signal in each pulse period of the signal curve to be analyzed based on the synchronous transmission error value and the signal delay characteristic value combined with the signal change trend, and to obtain the artifact-free signal; the artifact-free signal is used for the drainage operation.

[0014] The embodiments of the present invention have at least the following beneficial effects: This invention first analyzes the pressure changes of cerebrospinal fluid at various locations during the drainage process by deploying pressure sensors at multiple points, providing a data foundation for subsequent intelligent identification of non-pathological intracranial pressure trends. Then, it analyzes the differences in signal change trends between the corresponding single-pulse cycles of the signal curve at different detection locations, and quantifies the synchronization transmission error by combining this with the signal trend performance of historical signal curves. Secondly, it analyzes the signal delay characteristics between the corresponding single-pulse cycles of the signal curve at different detection locations, and also combines this with the signal delay data of historical signal curves to quantify the degree of signal time delay. Furthermore, by analyzing the synchronization of pressure signal fluctuations at multiple locations and compensating for signal time delay characteristics, it effectively eliminates false synchronization fluctuations caused by sensor drift, tubing compliance, or air bubble artifacts, reducing misidentification and ensuring the reliability of intelligent drainage decisions. This invention achieves accurate identification of respiratory artifacts, mechanical noise, and human interference. It effectively isolates non-pathological signal components, ensuring that trend interpretation reflects only the true changes in intracranial physiological state, providing a basis for anti-interference control of the closed-loop system. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of the steps of an intelligent drainage method based on cerebrospinal fluid pressure provided by the present invention; Figure 2 This is a flowchart of the steps of the method for obtaining synchronous transmission error values ​​provided by the present invention; Figure 3 This is a partial signal schematic diagram of a signal curve to be analyzed provided by the present invention; Figure 4 This is a flowchart of the steps for obtaining the trend change factor provided by the present invention; Figure 5 This is a flowchart of the steps in the method for obtaining signal delay feature values ​​provided by the present invention; Figure 6 This is a schematic diagram of a module of an intelligent drainage system based on cerebrospinal fluid pressure provided by the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an intelligent drainage method and system based on cerebrospinal fluid pressure proposed in accordance with the present invention.

[0018] Before introducing the specific solutions provided in the embodiments of this application, the implementation background of this application is explained in order to facilitate understanding by those skilled in the art, and is not intended to limit the uses of this application.

[0019] External cerebrospinal fluid drainage (EVD) is used to drain bloody or contaminated cerebrospinal fluid outside the cranium while monitoring and controlling intracranial pressure.

[0020] 1. Patient position: The patient is placed in a supine position (for ventricular puncture), keeping the body flexed to enlarge the drilling position in the skull.

[0021] 2. Puncture and catheter placement: Ventricular puncture (suitable for intracranial pressure monitoring): A catheter is placed into the lateral ventricle through a burr hole in the skull to directly measure the intracranial pressure.

[0022] 3. Sensor connection: (1) Intracranial end sensor: MEMS (microelectromechanical system) pressure sensor is implanted in the lateral ventricle or subarachnoid space.

[0023] (2) Sensors in the middle of the drainage tube: Flexible strain sensors are deployed at the intracranial end and every 10–15 cm (avoiding bending parts) to fit the inner wall of the catheter and resist fluid impact.

[0024] (3) Sensor at the outlet of the drainage tube: A piezoresistive sensor is deployed at the end of the drainage tube (before connecting to the reservoir or external drainage bag).

[0025] 4. Pressure signal acquisition and processing: (1) Real-time monitoring: The sensor converts the pressure signal into an electrical signal, which is then filtered by an amplifier (to remove high-frequency noise such as breathing and heartbeat) and transmitted to the monitor.

[0026] (2) Waveform analysis: Display pressure curve. The normal intracranial pressure (ICP) range is 5–15 mmHg (adult supine position). Abnormalities may show plateau waves (indicating intracranial hypertension) or low-level waveforms (excessive drainage).

[0027] In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments may be combined in any suitable form.

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0029] The following description, in conjunction with the accompanying drawings, details the specific scheme of the intelligent drainage method and system based on cerebrospinal fluid pressure provided by this invention.

[0030] The specific implementation scenario addressed by this invention is an intelligent cerebrospinal fluid (CSF) drainage system. This system dynamically adjusts drainage based on changes in intracranial pressure (ICP). When an abnormal trend in ICP is detected, the closed-loop control system controls the CSF flow rate via an electric drainage pump. Therefore, the accuracy of ICP changes directly affects the safety of CSF drainage. Generally, EVD drainage is recommended for 7-10 days. During this period, patient activity and the use of drainage equipment may cause pressure signal artifacts. The intelligent drainage system is prone to misinterpreting these artifacts and automatically adjusting CSF drainage, potentially leading to serious safety incidents. Therefore, this invention analyzes CSF pressure fluctuations to identify the authenticity of abnormal trends.

[0031] Please see Figure 1 The diagram illustrates a flowchart of a smart drainage method based on cerebrospinal fluid pressure according to an embodiment of the present invention. The method includes the following steps: Step S100: Obtain the pressure signal curve of the intelligent cerebrospinal fluid drainage system at each detection location and each pulse period of the pressure signal. The pressure signal curve includes the signal curve to be analyzed and the historical signal curve.

[0032] In clinical practice, pressure changes caused by transient fluctuations (such as coughing or changes in body position) should be transmitted synchronously to all measurement points, while local interference (such as air bubbles or blockages) only affects signals at specific locations. Therefore, artifacts can be identified through multi-node trend comparison. Specifically, this embodiment collects pressure signal data at multiple detection locations using sensors. Detection locations include intracranial pressure detection locations and detection locations at different parts of the drainage tube, such as the starting end of the drainage tube and different contact surfaces within the drainage tube.

[0033] It should be noted that, under normal circumstances, intracranial pressure and pressure fluctuations at the initiation end (closest to the intracranial end) of the drainage tube should be strongly correlated. While the pressure at the distal end of the drainage tube may experience amplitude attenuation due to drainage resistance, pressure changes caused by transient fluctuations are synchronous. To avoid amplitude discrepancies affecting the analysis results, spatiotemporal alignment and amplitude calibration of multi-sensor signals are necessary during pressure signal data acquisition. Specifically, the data acquisition frequency of each sensor is kept consistent, and transmission delays are eliminated through timestamp alignment. Simultaneously, the pressure signal data acquired by each sensor is standardized to avoid differences in the amplitude of changes in detection results at different locations.

[0034] Furthermore, in this embodiment, pressure signal curves are acquired using days as the time unit. Specifically, the curve formed by the pressure signal data collected on the current day is the signal curve to be analyzed, and the curves formed by the pressure signal data from each day prior to the current day are each historical signal curve. Thus, at each detection location, one corresponding signal curve to be analyzed and multiple historical signal curves can be obtained.

[0035] Finally, considering that intracranial pressure signals exhibit typical cardiac-related pulse waveforms, the pressure change trend is analyzed using single-pulse cycles as units. Specifically, for each pressure signal curve, the MOCAIP framework can be used to detect peak candidate points at the rising and falling edges based on the second derivative. For each detected peak, its adjacent valleys are used as pulse boundaries to complete the single-pulse cycle segmentation. It should be noted that the method for obtaining a complete pulse cycle is a well-known technique and will not be elaborated upon here.

[0036] Step S200: Based on the difference in signal change trends of the signal curves to be analyzed at each detection position and other detection positions within the same pulse period, and combining the signal trend differences of the signal curves to be analyzed at each detection position and all historical signal curves in each pulse period, the synchronization propagation error value of each detection position in each pulse period of the signal curves to be analyzed between other detection positions is obtained.

[0037] During drainage, when intracranial pressure changes abnormally, the fluid pressure at all different contact surfaces within the drainage tube will also change abnormally synchronously. Firstly, analyzing the consistency of pressure change trends at multiple locations can effectively quantify the synchronicity of pressure fluctuations.

[0038] In some embodiments, such as Figure 2 As shown, the method for obtaining the synchronous transmission error value can be implemented by steps S201 to S203.

[0039] Step S201: Based on the difference between the pulse trend of the signal curve to be analyzed at each detection position in each pulse cycle and the pulse trend of all historical signal curves in each pulse cycle, obtain the trend change factor of the signal curve to be analyzed at each detection position in each pulse cycle.

[0040] Firstly, before quantizing the signal synchronization characteristics, it is necessary to measure the signal variation trend of each single pulse period, such as... Figure 3 As shown, it illustrates a partial signal diagram of a signal curve to be analyzed, allowing for comparison of differences and similarities in trends. A single pulse cycle is used as an example for illustration.

[0041] As a specific implementation method, such as Figure 4As shown, the method for obtaining the trend change factor can be implemented by steps S2011 to S2014.

[0042] Step S2011: Record any pulse period of the signal curve to be analyzed at any detection position as the target pulse period.

[0043] Step S2012: The peak-to-peak value of the target pulse period is used as the first trend parameter of the target pulse period. Based on the fitting trend of the pressure signal within the target pulse period, the second trend parameter of the target pulse period is determined.

[0044] Specifically, the first trend parameter characterizes the amplitude of the pressure signal change during the target pulse period, while the second trend parameter characterizes the trend of the target pulse period. As a concrete example, a straight line is fitted to the pressure signal from the pulse start point to the peak point of the target pulse period, and the slope of the fitted line is used as the second trend parameter of the target pulse period. In this case, the second trend parameter reflects the signal rise rate of the target pulse period.

[0045] In other embodiments, curve fitting can be performed based on the peak point to the end point of the target pulse period. The slope of the fitted line is the signal descent rate of the target pulse period, and the signal descent rate is used as the corresponding second trend parameter. Furthermore, the average of the signal rise rate and signal fall rate of the target pulse period can also be used as the corresponding second trend parameter. It should be understood that the second trend parameter aims to characterize the rate of change of the pressure signal within the corresponding single pulse period, that is, to reflect the trend of signal change within the corresponding single pulse period.

[0046] Step S2013: Take the average value of the first trend parameter of each pulse period in each historical signal curve as the first historical trend value, and take the average value of the second trend parameter of each pulse period in each historical signal curve as the second historical trend value.

[0047] It should be understood that the first and second trend parameters for each pulse cycle in each historical signal curve are shared in the same way as those for each single pulse cycle in the signal curve to be analyzed. Here, by comprehensively considering the average values ​​of various parameters in all pulse cycles of the historical signal curve, the data changes of the pressure signal in the historical data are comprehensively characterized, providing a data basis for subsequent data comparison.

[0048] Step S2014: Based on the difference between the first trend parameter and the first historical trend value of the target pulse period, and the difference between the second trend parameter and the second historical trend value of the target pulse period, determine the trend change factor of the target pulse period.

[0049] The trend change factor of the target pulse period characterizes the difference between the pressure signal change trend of the target pulse period and its historical change trend. As a specific example, firstly, the absolute value of the difference between the first trend parameter and the first historical trend value of the target pulse period is calculated, reflecting the difference between the signal change range of the target pulse period and historical data. Secondly, the absolute value of the difference between the second trend parameter and the second historical trend value of the target pulse period is calculated, reflecting the difference between the signal abruptness of the target pulse period and historical data. Finally, the trend change factor of the target pulse period is obtained by combining the results of the first and second aspect calculations. This combination can be achieved through summation or product; specifically, the trend change factor of the target pulse period is obtained by multiplying the absolute values ​​of the differences in the first and second aspects.

[0050] Furthermore, the process of measuring the consistency of pressure change trends at multiple locations mainly includes, firstly, the analysis of differences in waveform change trends, and secondly, the analysis of data matching. The specific analysis process is as follows: steps S202 and S203.

[0051] Step S202: Based on the differences in the trend change factors of the signal curves to be analyzed at each detection location and the other detection locations in the same pulse period, as well as the differences in the pressure signals, the characteristic synchronization degree of the signal curve to be analyzed at each detection location between the other detection locations in each pulse period is obtained.

[0052] First, the differences in waveform variation trends of the signal curves under analysis at different detection positions within a single pulse period are analyzed. This embodiment still uses a single pulse period from the signal curve under analysis at any detection position as an example, specifically the target pulse period. Then, the single pulse periods corresponding to the target pulse period in the detection positions and those at other detection positions are obtained. In other words, the single pulse period with the largest temporal intersection within the time period of the target pulse period indicates a correspondence between them.

[0053] The first step is to determine the pulse period with the highest degree of time overlap between the target pulse period and the signal curve to be analyzed at each of the other detection positions within the same time interval. This pulse period is then used as the control pulse period for the target pulse period at each of the other detection positions. It can be understood that the target pulse period and the control pulse period are single pulse periods that have a corresponding relationship on the corresponding signal curve to be analyzed at different detection positions.

[0054] The second step involves determining the trend consistency coefficient between the target pulse period and each control pulse period based on the negative correlation coefficient of the difference in trend change factors between the target pulse period and each control pulse period. The greater the difference in trend change factors between the target pulse period and each control pulse period, the smaller the consistency of the signal change trends between the corresponding single pulse periods, and consequently, the smaller the value of the corresponding trend consistency coefficient.

[0055] As a concrete example, the nth pulse period of the signal curve to be analyzed at the i-th detection position is taken as the target pulse period, where i represents the index of the detection position and takes the value of , This represents the total number of detection positions, where n represents the sequence number of the pulse period in the signal curve to be analyzed, and its value is... , This indicates the total number of pulse cycles.

[0056] Furthermore, the trend consistency coefficient between the target pulse period and any control pulse period can be expressed as: ,in, The trend consistency coefficient between the target pulse period and the kth control pulse period is represented by i, where i represents the i-th detection position, n represents the n-th pulse period of the signal curve to be analyzed at the i-th detection position, which is also the index of the target pulse period in the signal curve to be analyzed, and k represents the index of the control pulse period corresponding to the target pulse period. The trend change factor representing the target pulse period. The trend change factor for the k-th control pulse period corresponding to the target pulse period is represented by exp, which represents an exponential function with the natural constant e as the base.

[0057] The third step is to determine the waveform matching coefficient between the target pulse period and each control pulse period based on the balance of the negative correlation coefficient between the pressure signals at corresponding times within the overlapping time interval, according to the target pulse period and each control pulse period.

[0058] As a specific example, the local time intervals where the target pulse cycle and each control pulse cycle completely overlap are obtained. The absolute value of the difference between the pressure signals of the target pulse cycle and each control pulse cycle at the same moment is calculated. Then, the mean of the results of negative correlation processing of the absolute values ​​of the differences at all moments within the local time intervals where the time overlap exists is used as the waveform matching coefficient between the target pulse cycle and each control pulse cycle. The method of negative correlation processing is a well-known technique, for example, it can be processed in the form of exp(-x), which will not be described in detail here.

[0059] In other embodiments, the DTW distance of the pressure signal between the target pulse period and each control pulse period can be calculated and negative correlation processing can be performed to obtain the waveform matching coefficient between the target pulse period and each control pulse period.

[0060] The waveform matching coefficient represents the balance of signal differences between the target pulse period and each control pulse period across all corresponding moments. The smaller the signal difference at corresponding moments, the greater the signal similarity between the two, reflecting the greater the signal matching degree between different detection positions, which in turn corresponds to a larger value of the waveform matching coefficient.

[0061] Finally, the characteristic synchronization degree between the target pulse period and each control pulse period is obtained by combining the trend consistency coefficient and the waveform matching coefficient.

[0062] Firstly, the trend consistency coefficient reflects the consistency of signal change trends between different detection positions. Secondly, the waveform matching coefficient reflects the similarity of signal distribution between different detection positions. Furthermore, the feature analysis results of the two aspects can be fused by summation or product to comprehensively characterize the feature synchronization between two different detection positions.

[0063] As a concrete example, the product of the trend consistency coefficient and waveform matching coefficient of the target pulse period and each control pulse period is used as the characteristic synchronization degree between the target pulse period and each control pulse period.

[0064] Step S203: Based on the characteristic synchronization degree between each detection position and other detection positions in each pulse cycle of the signal curve to be analyzed, and the balance of the characteristic synchronization degree of the same detection position in all pulse cycles of the historical signal curve, the synchronization propagation error value of each detection position in each pulse cycle of the signal curve to be analyzed with other detection positions is obtained.

[0065] Non-pathological abnormal fluctuations may cause differences in the corresponding pressure waveforms at different locations. If the synchronous transmission characteristics of the pressure waveform at different locations differ from those in historical data, it indicates that some interference factor has caused pressure change errors. Based on the synchronous transmission characteristics of the corresponding pressure waveforms, the synchronous transmission error of pressure signals at different detection locations is analyzed.

[0066] The degree of feature synchronization reflects the synchronization of signal features between corresponding single-pulse periods at different detection positions in the signal curve under analysis. It is a result of synchronization analysis based on feature comparison in the spatial dimension. Furthermore, a feature comparison operation in the time dimension can be performed for each detection position to analyze the synchronization of feature performance corresponding to each detection position and the historical signal curve.

[0067] Specifically, the balance of all feature synchronization levels between the detection position corresponding to the target pulse period and the corresponding pulse periods of other detection positions on the historical signal curve for each pulse period is obtained, and denoted as the historical synchronization feature factor of the detection position corresponding to the target pulse period. The balance is characterized by averaging.

[0068] Based on the difference between the characteristic synchronization degree between the target pulse cycle and each control pulse cycle and the historical synchronization characteristic factor, and combined with the negative correlation coefficient of the characteristic synchronization degree, the synchronization transmission error value of the target pulse cycle and each control pulse cycle is determined.

[0069] As a concrete example, the formula for calculating the synchronization propagation error value of the target pulse period and any control pulse period can be expressed as: ,in This represents the synchronization propagation error value between the target pulse period and the kth control pulse period. i represents the i-th detection position, n represents the n-th pulse period of the signal curve to be analyzed at the i-th detection position, which is also the index of the target pulse period in the signal curve to be analyzed, and k represents the index of the control pulse period corresponding to the target pulse period. This indicates the degree of characteristic synchronization between the target pulse period and the k-th control pulse period. represents the historical synchronization characteristic factor of the detection position corresponding to the target pulse period, and exp represents an exponential function with the natural constant e as the base.

[0070] The purpose of adding one to the difference is to avoid the difference calculation result being 0, which would affect the calculation result. That is, when the synchronization characteristics between the data to be analyzed and the historical data are relatively consistent, we can directly focus on the signal synchronization within the single pulse period of the signal curve to be analyzed. The stronger the signal synchronization, the smaller the corresponding synchronization error, and the smaller the value of the synchronization transmission error.

[0071] The synchronization propagation error value characterizes the degree of error in the synchronization of signals between each detection position and other detection positions. The larger the value, the worse the synchronization of each detection position compared to other detection positions, and the larger the error. The smaller the value, the better the synchronization of each detection position compared to other detection positions, and the smaller the error.

[0072] Step S300: Based on the signal delay between corresponding pulse periods in the signal curves to be analyzed at each detection position and other detection positions, and combined with the signal delay between historical signal curves at each detection position and other detection positions, obtain the signal delay characteristic value for each pulse period in the signal curves to be analyzed at each detection position and other detection positions.

[0073] The highest point of the cerebrospinal fluid drainage device is higher than when the patient is lying flat, and the drainage tube is curved, resulting in time differences in pressure fluctuations transmitted to different locations. During cerebrospinal fluid drainage, which can last for days, patients may move and engage in uncontrolled physiological behaviors, causing pressure fluctuations that are simultaneously transmitted to different pressure collection points. However, sudden physiological behaviors such as coughing can cause instantaneous pressure fluctuations that may exceed the normal cerebrospinal fluid pressure threshold. Simultaneously, due to the rapid pressure, the time difference of transmission to different drainage locations will also change. This can be compared with the historical average transmission time difference to obtain the time delay variation of the pressure waveform at different locations.

[0074] In some embodiments, such as Figure 5 As shown, the method for obtaining the signal delay characteristic value can be implemented by steps S301 to S303.

[0075] Step S301: Based on the difference in start time and end time of the corresponding pulse period in the signal curve to be analyzed between each detection position and other detection positions, obtain the time difference factor of the corresponding pulse period in the signal curve to be analyzed between each detection position and other detection positions.

[0076] The time difference factor characterizes the time delay between the corresponding single-pulse cycles at different detection positions in the signal curve to be analyzed. In this embodiment, an example is taken of any single-pulse cycle at any detection position in the signal curve to be analyzed. Specifically, this embodiment uses the target pulse cycle and each control pulse cycle that has a corresponding relationship with the single-pulse cycle as examples.

[0077] Specifically, the time interval between the start time of the target pulse cycle and the start time of each control pulse cycle is used as the first time coefficient; the time interval between the end time of the target pulse cycle and the end time of each control pulse cycle is used as the second time coefficient; and based on the balance between the first and second time coefficients, the time difference factor between the target pulse cycle and each control pulse cycle is determined.

[0078] The average of the first and second time coefficients is calculated as the time difference factor between the target pulse cycle and each corresponding control pulse cycle. The average of the differences in the start and end times of single pulse cycles corresponding to different detection positions is used to characterize the degree of time delay between single pulse cycles corresponding to different detection positions.

[0079] Step S302: Obtain the equalization coefficient of the time difference factor of the corresponding pulse period in the historical signal curves of each detection position and other detection positions, and denot it as the historical time coefficient.

[0080] The historical time coefficient characterizes the performance of the historical average transmission time difference. That is, it is obtained by calculating the mean of the time difference factor between all detection positions and each other detection position in the historical signal curve where there is a corresponding relationship.

[0081] Step S303: Based on the difference between the time difference factor of the corresponding pulse period in the signal curve to be analyzed between each detection position and other detection positions and the historical time coefficient, determine the signal delay characteristic value of each pulse period in the signal curve to be analyzed between each detection position and other detection positions.

[0082] As a concrete example, the difference between data can be quantified by calculating the absolute value of the difference. Specifically, the absolute value of the difference between the time difference factor and the historical time coefficient between the target pulse period and each control pulse period is used as the signal delay characteristic value between the target pulse period and each control pulse period.

[0083] Signal delay eigenvalues ​​characterize the differences between the signal delay variations at two different detection locations on the signal curve to be analyzed and the historical average delay.

[0084] Step S400: Based on the synchronous conduction error value and the signal delay characteristic value combined with the signal change trend, analyze the signal artifacts of each pulse period in the signal curve to be analyzed, and obtain the artifact-removed signal; the artifact-removed signal is used for the drainage operation.

[0085] Asynchronous pressure conduction implies that the abnormal pressure trend may be a signal artifact caused by interference factors. If the synchronous conduction deviation is large, it may be an error originating at that location. If the deviation is small, combining it with time delay changes can further determine the authenticity of the abnormal pressure fluctuations. Based on the synchronous conduction deviation of pressure waveforms at different locations, combined with the time delay changes in cerebrospinal fluid pressure conduction, the abnormal conduction characteristics of each intracranial pressure waveform during drainage and the artifact characteristics exhibited by the signal under analysis are analyzed.

[0086] The first step is to determine the abnormal conduction characteristic value of the target pulse cycle by combining the synchronous conduction error value and the time difference factor of the target pulse cycle with the balanced distribution of the time difference factor of each control pulse cycle.

[0087] First, the synchronization transmission error value and time difference factor between the target pulse period and each control pulse period need to be fused. Then, the balance of the fused results corresponding to the target pulse period and all control pulse periods is calculated to obtain the abnormal transmission characteristic value. This value characterizes the difference in signal characteristics of a single pulse period in the signal curve to be analyzed at a selected detection position relative to the comprehensive aspects of other detection positions, reflecting the likelihood of abnormal characteristics in the pressure signal.

[0088] As a concrete example, the product of the synchronization transmission error value and the time difference factor of the target pulse period and any control pulse period is calculated. The mean of the products of the target pulse period and all control pulse periods is taken as the abnormal transmission characteristic value of the target pulse period. The greater the difference in signal change trend between the target pulse period and the pulse periods corresponding to other detection positions, the greater the signal time delay, and the greater the possibility that the current target pulse period has abnormal transmission characteristics.

[0089] The second step is to determine the degree of signal artifacts in the target pulse period based on the product of the trend change factor of the target pulse period and the abnormal conduction characteristic value.

[0090] Based on the abnormal conduction characteristics of the single pulse cycle, the trend change factor of the single pulse cycle is further combined to consider that when the abnormal change trend of cerebrospinal fluid pressure is asynchronously transmitted to other detection locations, it indicates that the possibility of pressure signal artifacts is greater at this location.

[0091] Based on this, the product of the trend change factor and the abnormal conduction characteristic value of the target pulse period is normalized to obtain the signal artifact degree of the target pulse period. A larger value for the trend change factor indicates a stronger signal trend change within the target pulse period. Simultaneously, a larger value for the abnormal conduction characteristic value indicates a greater likelihood of asynchronous conduction within the target pulse period, and a higher probability that the signal distribution within the corresponding single pulse period is an artifact.

[0092] The third step is to determine whether to mark the target pulse period based on the degree of signal artifacts in the target pulse period.

[0093] The signal artifact severity of the target pulse cycle characterizes the likelihood that the pressure signal within that single pulse cycle is an artifact. When physiological artifacts exist, such as coughing, changes in body position, tubing problems like air bubbles or blockages, or human error such as adjusting drainage tubes or valves, these abnormal physiological conditions cause pressure fluctuations and a certain degree of time delay, leading to a higher quantified signal artifact severity value. A higher value indicates a greater likelihood that the pressure signal within the corresponding single pulse cycle is an artifact, or more likely, a non-physiological abnormal signal. In such cases, these signals need to be marked for subsequent targeted data cleaning. Conversely, a lower value indicates a lower likelihood that the pressure signal within the corresponding single pulse cycle is an artifact, and in these cases, marking is unnecessary.

[0094] Specifically, when the signal artifact level of the target pulse cycle is greater than a preset artifact threshold, it indicates that the fluctuation of the pressure signal in the target pulse cycle is more likely to be a non-physiological abnormal fluctuation, and a marking operation is required, that is, marking the target pulse cycle as an artifact signal. When the signal artifact level of the target pulse cycle is less than or equal to the preset artifact threshold, it indicates that the fluctuation of the pressure signal in the target pulse cycle is less likely to be a non-physiological abnormal fluctuation, and no marking is performed on the target pulse cycle. In this embodiment, the artifact threshold is set to 0.7. It should be understood that the artifact analysis method for each single pulse cycle in the signal curve to be analyzed is the same for each detection location. At the same time, considering that the artifact signal of intracranial pressure is the main part affecting the operation, in some embodiments, each pulse cycle in the signal curve to be analyzed corresponding to intracranial pressure can also be used as a target for artifact judgment.

[0095] Furthermore, after identifying artifact signals, data cleaning operations can be used to process the intracranial pressure signal to obtain artifact-free signals. This also alerts relevant medical personnel to review the portions containing artifacts and promptly investigate any issues. The data cleaning method is a well-known technique and will not be described in detail here; for example, no abnormality warning is issued for signal portions within a period containing artifact markers.

[0096] It should be understood that the data cleaning operation only needs to effectively avoid misidentification caused by signal components containing artifacts. Implementers can choose the appropriate method based on the specific implementation scenario. The cleaned signal effectively avoids misidentification, and the intelligent cerebrospinal fluid drainage device uses the cleaned signal for drainage, thus ensuring the effectiveness of cerebrospinal fluid drainage to a certain extent. It should also be noted that the main purpose of this embodiment is to provide a method for identifying artifact signals. For real-time monitored pressure signals, the same method can be used within a complete single-pulse cycle. Historical data is stored in a cloud database, and relevant medical personnel can review the data by viewing the historical database.

[0097] In summary, by deploying miniature pressure sensors at multiple locations and comparing the pressure waveforms, phase differences, and time delays at different acquisition points, this invention can distinguish in real time non-pathological fluctuations caused by artifacts such as sensor drift, pipeline compliance, or air bubbles. This allows for the intelligent elimination of false signals and the accurate capture of genuine pathological intracranial hypertension trends, significantly reducing false alarm rates and improving the safety and response accuracy of the closed-loop drainage system.

[0098] By monitoring the morphology of the cerebrospinal fluid pressure curve in real time, abnormal trends in the pressure curve can be analyzed and identified, enabling more sensitive detection of latent hypertension or worsening compliance. When intracranial pressure changes abruptly, the actual changes should be synchronously mapped to pressure sensors at multiple detection points in the tubing. By analyzing and quantifying the correlation and phase lag between intracranial pressure and tubing pressure, non-real fluctuations can be captured, preventing the closed-loop algorithm from incorrectly opening or closing valves due to false fluctuations. The conduction of cerebrospinal fluid between the ventricles and tubing exhibits physical lag, which is affected by flow rate and tubing characteristics. By fitting the time delay of each tubing segment using historical data, it is possible to distinguish between real pathological pressure changes (such as intracranial hypertension continuously exceeding the time delay window) and normal delayed conduction phenomena (such as signal lag caused by slow flow rate). Physiological behaviors can cause instantaneous increases in intracranial pressure, but the propagation of its pressure waveform is characterized by short-term suddenness and multi-location asynchrony. By detecting the inconsistency of pressure changes within the time delay window, false alarms can be further filtered out, reducing over-response to transient physiological fluctuations and ensuring that the system only triggers intervention when cerebrospinal fluid pressure is truly abnormal.

[0099] In some embodiments, such as Figure 6 As shown, a smart drainage system based on cerebrospinal fluid pressure is also provided. This system is used to perform the steps of a smart drainage method based on cerebrospinal fluid pressure. The smart drainage system based on cerebrospinal fluid pressure includes: The data acquisition module is used to acquire the pressure signal curve of the intelligent cerebrospinal fluid drainage system at each detection location and each pulse period of the pressure signal, wherein the pressure signal curve includes the signal curve to be analyzed and the historical signal curve; The synchronous transmission analysis module is used to obtain the synchronous transmission error value of each detection position and the signal curves of other detection positions in each pulse period based on the difference in signal change trends of each detection position and other detection positions within the same pulse period, combined with the signal trend differences of each detection position's signal curve and all historical signal curves in each pulse period. The time delay analysis module is used to obtain the signal delay characteristic value of each pulse period in the signal curves to be analyzed between each detection position and other detection positions based on the signal delay between corresponding pulse periods in the signal curves to be analyzed between each detection position and other detection positions, combined with the signal delay between historical signal curves of each detection position and other detection positions. The artifact detection and analysis module is used to analyze the artifact nature of the signal in each pulse period of the signal curve to be analyzed based on the synchronous transmission error value and the signal delay characteristic value combined with the signal change trend, and to obtain the artifact-free signal; the artifact-free signal is used for the drainage operation.

[0100] The modules in the apparatus or system of this application can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0101] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A smart drainage method based on cerebrospinal fluid pressure, characterized in that, The method includes the following steps: The pressure signal curves of the intelligent cerebrospinal fluid drainage system at each detection location and each pulse period of the pressure signal are obtained. The pressure signal curves include the signal curve to be analyzed and historical signal curves. Based on the difference in signal change trends between the signal curves to be analyzed at each detection position and other detection positions within the same pulse period, and combined with the difference in signal trend between the signal curves to be analyzed at each detection position and all historical signal curves in each pulse period, the synchronization propagation error value of each detection position in each pulse period of the signal curves to be analyzed between other detection positions is obtained. Based on the signal delay between corresponding pulse periods in the signal curves to be analyzed at each detection position and other detection positions, and combined with the signal delay between historical signal curves at each detection position and other detection positions, the signal delay characteristic value of each pulse period in the signal curves to be analyzed at each detection position and other detection positions is obtained. Based on the synchronous conduction error value and the signal delay characteristic value combined with the signal change trend, the signal artifacts in each pulse period of the signal curve to be analyzed are obtained to obtain the artifact-free signal; the artifact-free signal is used for the drainage operation.

2. The intelligent drainage method based on cerebrospinal fluid pressure according to claim 1, characterized in that, The method involves determining the difference in signal change trends between the signal curves to be analyzed at each detection position and other detection positions within the same pulse period, and combining this with the signal trend differences between the signal curves to be analyzed at each detection position and all historical signal curves for each pulse period. Specifically, this includes: Based on the difference between the pulse trend of the signal curve to be analyzed at each detection location in each pulse cycle and the pulse trend of all historical signal curves in each pulse cycle, the trend change factor of the signal curve to be analyzed at each detection location in each pulse cycle is obtained. Based on the differences in the trend change factors of the signal curves to be analyzed at each detection location and the other detection locations within the same pulse cycle, as well as the differences in the pressure signals, the degree of characteristic synchronization between the signal curves to be analyzed at each detection location and the other detection locations within each pulse cycle is obtained. Based on the characteristic synchronization degree between the signal curve to be analyzed at each detection position and other detection positions in each pulse cycle, and the balance of the characteristic synchronization degree of the same detection position in all pulse cycles of the historical signal curve, the synchronization propagation error value of each detection position in each pulse cycle of the signal curve to be analyzed with other detection positions is obtained.

3. The intelligent drainage method based on cerebrospinal fluid pressure according to claim 2, characterized in that, The step of obtaining the trend change factor of the signal curve to be analyzed at each detection location in each pulse cycle based on the difference between the pulse trend of the signal curve to be analyzed at each detection location in each pulse cycle and the pulse trend of all historical signal curves in each pulse cycle specifically includes: Any pulse period of the signal curve to be analyzed at any detection position is recorded as the target pulse period. The peak-to-peak value of the target pulse period is used as the first trend parameter of the target pulse period. Based on the fitting trend of the pressure signal within the target pulse period, the second trend parameter of the target pulse period is determined. The mean of the first trend parameter in each pulse period of each historical signal curve is taken as the first historical trend value, and the mean of the second trend parameter in each pulse period of each historical signal curve is taken as the second historical trend value. The trend change factor of the target pulse period is determined based on the difference between the first trend parameter and the first historical trend value of the target pulse period, and the difference between the second trend parameter and the second historical trend value of the target pulse period.

4. The intelligent drainage method based on cerebrospinal fluid pressure according to claim 3, characterized in that, The method of obtaining the characteristic synchronization degree between the signal curve to be analyzed at each detection location and other detection locations in each pulse cycle based on the difference in the trend change factor of the signal curve to be analyzed at each detection location and the difference in pressure signal, specifically includes: The pulse period with the greatest time overlap between the signal curve to be analyzed at each other detection position and the target pulse period within the same time interval is determined as the control pulse period corresponding to the target pulse period at each other detection position. Based on the negative correlation coefficient of the difference in trend change factors between the target pulse cycle and each control pulse cycle, the trend consistency coefficient between the target pulse cycle and each control pulse cycle is determined. Based on the balance of the negative correlation coefficient between the pressure signals at corresponding times within the overlapping time interval, the waveform matching coefficient between the target pulse period and each control pulse period is determined. The characteristic synchronization degree between the target pulse period and each control pulse period is obtained by combining the trend consistency coefficient and the waveform matching coefficient.

5. The intelligent drainage method based on cerebrospinal fluid pressure according to claim 4, characterized in that, The step of obtaining the synchronization propagation error value for each pulse cycle in the signal curve to be analyzed between each detection position and other detection positions based on the characteristic synchronization degree between each detection position and other detection positions in each pulse cycle, and the balance of the characteristic synchronization degree of the same detection position in all pulse cycles of the historical signal curve, specifically includes: The balance of all feature synchronization degrees between the detection position corresponding to the target pulse period and the corresponding pulse period of other detection positions on the historical signal curve is obtained, and is denoted as the historical synchronization feature factor of the detection position corresponding to the target pulse period. Based on the difference between the characteristic synchronization degree between the target pulse cycle and each control pulse cycle and the historical synchronization characteristic factor, and combined with the negative correlation coefficient of the characteristic synchronization degree, the synchronization transmission error value of the target pulse cycle and each control pulse cycle is determined.

6. The intelligent drainage method based on cerebrospinal fluid pressure according to claim 4, characterized in that, The step of obtaining the signal delay characteristic value for each pulse period in the signal curves to be analyzed between each detection position and other detection positions, based on the signal delay between corresponding pulse periods in the signal curves to be analyzed between each detection position and other detection positions, and in conjunction with the signal delay between historical signal curves of each detection position and other detection positions, specifically includes: Based on the differences in the start and end times of the corresponding pulse periods in the signal curves to be analyzed between each detection position and other detection positions, the time difference factor of the corresponding pulse periods in the signal curves to be analyzed between each detection position and other detection positions is obtained. Obtain the equalization coefficient of the time difference factor of the corresponding pulse period in the historical signal curves of each detection position and other detection positions, and denote it as the historical time coefficient; Based on the difference between the time difference factor of the corresponding pulse period in the signal curve to be analyzed at each detection position and the historical time coefficient, the signal delay characteristic value of each pulse period in the signal curve to be analyzed at each detection position and the signal curve to be analyzed at the other detection positions is determined.

7. The intelligent drainage method based on cerebrospinal fluid pressure according to claim 6, characterized in that, The step of obtaining the time difference factor of the corresponding pulse period in the signal curves to be analyzed between each detection position and other detection positions based on the difference in start time and end time of the corresponding pulse period in the signal curves to be analyzed between each detection position and other detection positions specifically includes: The time interval between the start time of the target pulse cycle and the start time of each control pulse cycle is used as the first time coefficient; the time interval between the end time of the target pulse cycle and the end time of each control pulse cycle is used as the second time coefficient; based on the balance between the first time coefficient and the second time coefficient, the time difference factor between the target pulse cycle and each control pulse cycle is determined.

8. The intelligent drainage method based on cerebrospinal fluid pressure according to claim 7, characterized in that, The step of analyzing the signal artifacts in each pulse period of the signal curve under analysis based on the synchronization transmission error value, the signal delay characteristic value, and the signal change trend specifically includes: By combining the balanced distribution of the synchronous conduction error value and time difference factor between the target pulse cycle and each control pulse cycle, the abnormal conduction characteristic value of the target pulse cycle is determined. The degree of signal artifact in the target pulse period is determined by the product of the trend change factor of the target pulse period and the abnormal conduction characteristic value. Whether to mark the target pulse period is determined based on the degree of signal artifacts in the target pulse period.

9. The intelligent drainage method based on cerebrospinal fluid pressure according to claim 8, characterized in that, The step of determining whether to mark the target pulse period based on the degree of signal artifacts in the target pulse period specifically includes: When the signal artifact level of the target pulse period is greater than the preset artifact threshold, the target pulse period is marked as an artifact signal; when the signal artifact level of the target pulse period is less than or equal to the preset artifact threshold, the target pulse period is not marked.

10. An intelligent drainage system based on cerebrospinal fluid pressure, characterized in that, The system is used to perform the steps of a smart drainage method based on cerebrospinal fluid pressure as described in any one of claims 1-9, wherein the smart drainage system based on cerebrospinal fluid pressure comprises: The data acquisition module is used to acquire the pressure signal curve of the intelligent cerebrospinal fluid drainage system at each detection location and each pulse period of the pressure signal, wherein the pressure signal curve includes the signal curve to be analyzed and the historical signal curve; The synchronous transmission analysis module is used to obtain the synchronous transmission error value of each detection position and the signal curves of other detection positions in each pulse period based on the difference in signal change trends of each detection position and other detection positions within the same pulse period, combined with the signal trend differences of each detection position's signal curve and all historical signal curves in each pulse period. The time delay analysis module is used to obtain the signal delay characteristic value of each pulse period in the signal curves to be analyzed between each detection position and other detection positions based on the signal delay between corresponding pulse periods in the signal curves to be analyzed between each detection position and other detection positions, combined with the signal delay between historical signal curves of each detection position and other detection positions. The artifact detection and analysis module is used to analyze the artifact nature of the signal in each pulse period of the signal curve to be analyzed based on the synchronous transmission error value and the signal delay characteristic value combined with the signal change trend, and to obtain the artifact-free signal; the artifact-free signal is used for the drainage operation.

Images