Neurosurgery craniocerebral postoperative drainage monitoring method and system

Abstract

The invention provides a neurosurgery craniocerebral postoperative drainage monitoring method and system, and relates to the technical field of medical instruments, and the method comprises the steps: collecting intracranial pressure data, body position change data, drainage bag height data and liquid flow rate data; the intracranial pressure data, the drainage bag height data and the liquid flow rate data are preprocessed based on a local outlier factor and a Z-score standardization algorithm; performing coupling analysis on the preprocessed intracranial pressure data and the preprocessed liquid flow rate data based on a machine learning algorithm to obtain a dynamic drainage dynamic index; performing dynamic correction analysis on the dynamic drainage dynamic index and the preprocessed drainage bag height data based on a logistic regression algorithm to obtain a height correction drainage efficiency index, and further adjusting the height of the drainage bag; based on the neural network model, constructing a drainage bag automatic adjustment model, and then obtaining a height automatic adjustment command; automatically adjusting the height of the drainage bag according to the automatic height adjusting command; the nursing workload is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and specifically to a method and system for monitoring the drainage after neurosurgical craniocerebral surgery. Background Art

[0002] In the field of neurosurgery, for patients who have undergone craniocerebral surgery, effective drainage management after surgery is crucial for preventing complications and promoting the recovery of patients. During the drainage process, factors such as intracranial pressure, the height of the drainage bag, the liquid flow rate, and changes in the patient's body position are interrelated and jointly affect the drainage effect.

[0003] Currently, the clinical monitoring and adjustment of drainage mainly rely on the manual operation and empirical judgment of nurses. In terms of data collection, nurses need to manually measure data such as intracranial pressure, the height of the drainage bag, and the liquid flow rate at regular intervals, which not only consumes a large amount of time and energy, but also the frequency of manual measurement is limited, making it difficult to comprehensively and real-time grasp the patient's drainage status. At the same time, there are certain errors in manual measurement, which may lead to inaccurate data and affect subsequent analysis and decision-making.

[0004] In the data processing and analysis section, the traditional method lacks effective algorithms and model support. For the collected data, it is impossible to accurately identify outliers and noise, making the drainage adjustment decisions based on these data less scientific and reasonable. For example, when there are abnormal fluctuations in the patient's intracranial pressure or liquid flow rate, nurses may not be able to promptly and accurately determine the cause, making it difficult to take effective measures for adjustment.

[0005] In terms of adjusting the height of the drainage bag, nurses usually adjust the height of the drainage bag based on experience and limited monitoring data. However, this method cannot fully consider the individual differences of patients and the impact of factors such as body position changes on drainage. The conditions and physical conditions of different patients are different, and the optimal drainage height required by them also varies. At the same time, patients may frequently change their body positions after surgery, and the traditional drainage management method is difficult to adjust the height of the drainage bag in a timely manner according to the body position changes, easily resulting in insufficient or excessive drainage, increasing the risk of complications for patients, such as intracranial hemorrhage, brain hernia, etc., seriously affecting the treatment effect and recovery process of patients.

[0006] In addition, with the continuous development of medical technology and the increase in the number of patients, the nursing workload is becoming increasingly heavy. Nurses need to take care of multiple patients at the same time, and the energy invested in drainage management is limited, which further affects the quality and effect of drainage management.

[0007] Therefore, a method and system for monitoring the drainage after neurosurgical craniocerebral surgery are provided. Summary of the Invention

[0008] To solve the above technical problems, the object of the present invention is to provide a method and system for monitoring drainage after neurosurgical craniocerebral surgery.

[0009] To achieve the above object, the present invention provides the following technical solution: A method for monitoring drainage after neurosurgical craniocerebral surgery, the method comprising: Collecting intracranial pressure data, body position change data, drainage bag height data, and liquid flow rate data of the drainage tube of the patient; Preprocessing the intracranial pressure data, drainage bag height data, and liquid flow rate data based on the local outlier factor and the Z-score normalization algorithm to obtain preprocessed intracranial pressure data, preprocessed drainage bag height data, and preprocessed liquid flow rate data; Performing coupled analysis on the preprocessed intracranial pressure data and the preprocessed liquid flow rate data based on a machine learning algorithm to obtain a corresponding dynamic drainage power index; performing dynamic correction analysis on the dynamic drainage power index and the preprocessed drainage bag height data based on a logistic regression algorithm to obtain a height-corrected drainage efficiency index, and then adjusting the height of the drainage bag; Based on a neural network model, constructing an automatic drainage bag adjustment model, obtaining a height automatic adjustment command according to the automatic drainage bag adjustment model; automatically adjusting the height of the drainage bag according to the height automatic adjustment command.

[0010] According to one preferred embodiment of the present invention, the process of collecting intracranial pressure data, body position change data, drainage bag height data, and liquid flow rate data of the drainage tube of the patient includes: Setting a data acquisition device, the data acquisition device including an intracranial pressure acquisition unit, a body position change acquisition unit, a drainage bag height acquisition unit, and a drainage tube liquid flow rate acquisition unit; setting a corresponding acquisition period, the acquisition period consisting of a plurality of acquisition times, and synchronizing timestamps based on the PTP protocol; The intracranial pressure acquisition unit collects the intracranial pressure data of the patient according to the acquisition period; the body position change acquisition unit collects the body position change data of the patient according to the acquisition period; the drainage bag height acquisition unit collects the drainage bag height data according to the acquisition period; the drainage tube liquid flow rate acquisition unit collects the liquid flow rate data of the drainage tube according to the acquisition period.

[0011] According to one preferred embodiment of the present invention, the process of preprocessing the intracranial pressure data, drainage bag height data, and liquid flow rate data includes: Obtaining the intracranial pressure data, drainage bag height data, and liquid flow rate data; Performing outlier detection on the intracranial pressure data, drainage bag height data, and liquid flow rate data based on the local outlier factor, including: Record the intracranial pressure data, the drainage bag height data, and the liquid flow rate data as , and ; Preset the nearest neighborhood ; According to the nearest neighborhood , obtain the reachable distance , and the reachable distance is: ; where is the Euclidean distance, is the distance of the th nearest neighbor of the data point; According to the reachable distance , obtain the local reachability density , and the local reachability density is: ; where represents the data point corresponding to the same acquisition cycle and acquisition time; According to the local reachability density , obtain the corresponding local outlier factor ; the local outlier factor is: ; Preset the local outlier factor threshold ; when the local outlier factor , it is marked as an outlier point; otherwise, it is a normal point; Take the nearest neighborhood of the intracranial pressure data ; according to the local outlier factor , obtain the intracranial pressure local outlier factor ; preset the intracranial pressure local outlier factor threshold ; when the intracranial pressure local outlier factor is , it is marked as an intracranial pressure data outlier point; Take the nearest neighborhood of the drainage bag height data ; according to the local outlier factor , obtain the intracranial pressure local outlier factor ; preset the drainage bag height local outlier factor threshold ; when the drainage bag height local outlier factor is , it is marked as a drainage bag height data outlier point; Take the nearest neighborhood of the liquid flow rate data ; according to the local outlier factor ; according to the local outlier factor , obtain the local outlier factor of the liquid flow rate ; preset the threshold of the local outlier factor of the liquid flow rate ; when the local outlier factor of the liquid flow rate , it is marked as an outlier point of the liquid flow rate data; Standardize the intracranial pressure data, drainage bag height data, and liquid flow rate data after outlier detection based on the Z-score standardization algorithm, including: Remove the outlier points of the intracranial pressure data, drainage bag height data, and liquid flow rate data; Based on the linear interpolation method, repair and fill the outlier points of the intracranial pressure data; based on the mean value of the historical drainage bag height data, repair and fill the outlier points of the drainage bag height data; based on the hydrodynamic pressure difference formula, repair and fill the outlier points of the liquid flow rate data; Calculate the intracranial pressure data , drainage bag height data and liquid flow rate data corresponding mean and standard deviation; Based on the Z-score standardization formula, obtain the standardized intracranial pressure data , drainage bag height data and liquid flow rate data , denoted as the preprocessed intracranial pressure data, preprocessed drainage bag height data, and preprocessed liquid flow rate data.

[0012] According to one preferred embodiment of the present invention, the process of coupling and analyzing the preprocessed intracranial pressure data and the preprocessed liquid flow rate data to obtain the corresponding dynamic drainage power index includes: Denote the preprocessed intracranial pressure data and the preprocessed liquid flow rate data in the same acquisition cycle as and ; According to the preprocessed intracranial pressure data and the preprocessed liquid flow rate data in the same acquisition cycle, obtain the mean value of the preprocessed intracranial pressure data and the mean value of the preprocessed liquid flow rate data in the current acquisition cycle; According to the mean value of the preprocessed intracranial pressure data and the mean value of the preprocessed liquid flow rate data, obtain the normalized power term ; Calculate the dynamic change rate of the preprocessed intracranial pressure data and the dynamic change rate of the preprocessed liquid flow rate data; where is the differential at the acquisition moment; Based on a machine learning algorithm and according to the dynamic change rate of the preprocessed intracranial pressure data and the dynamic change rate of the preprocessed liquid flow rate data , a fluctuation correction term is obtained ; where and are the weight coefficients of the intracranial pressure data and the liquid flow rate data respectively; According to the normalized dynamic term and the fluctuation correction term , a dynamic drainage power index is obtained ; the dynamic drainage power index is: .

[0013] According to one preferred embodiment of the present invention, the process of constructing the dynamic optimal height model includes: Based on the basic formula of the physiological mechanism, a basic height model is constructed ; the basic height model is: ; where is the mean value of the preprocessed intracranial pressure data; is the ideal drainage power threshold; is the cerebrospinal fluid gravity constant; Based on the logistic regression algorithm and according to the basic height model , an individualized correction index is obtained , the individualized correction index is: ; where is the age correction; is the surgical type correction; is the postoperative days correction.

[0014] According to one preferred embodiment of the present invention, the process of performing dynamic correction analysis on the dynamic drainage power index and the preprocessed drainage bag height data to obtain the height-corrected drainage efficiency index includes: Denote the preprocessed drainage bag height data in the same acquisition cycle as ; obtain the mean value of the preprocessed drainage bag height data in the same acquisition cycle ; According to the individualized correction index and the mean value of the preprocessed drainage bag height data , obtain the height deviation index decay function ; where is the clinical fitting index, and the value is ; Set boundary constraint conditions , when , force the height to deviate from the exponential decay function ; is the lowest efficiency of the height deviation from the exponential decay; According to the height deviation exponential decay function and the dynamic drainage power index , obtain the height corrected drainage efficiency index ; The height corrected drainage efficiency index is: ; According to the height corrected drainage efficiency index , adjust the height of the drainage bag.

[0015] According to one preferred embodiment of the present invention, the process of constructing the automatic adjustment model of the drainage bag includes: Obtain several groups of height corrected drainage efficiency indexes and body position change data; Group and label several groups of height corrected drainage efficiency indexes and body position change data, denoted as is a natural number; Take groups of height corrected drainage efficiency indexes and body position change data as sample data, and is a natural number less than , and use the sample data to obtain the sample data mean, denoted as the sample set; Take the remaining several groups of height corrected drainage efficiency indexes and body position change data as the test set; According to the sample set and the test set, form the training sample set; Based on the neural network model, construct the standard height automatic adjustment model; And input the training sample set into the standard height automatic adjustment model to train the standard height automatic adjustment model, and denote the trained standard height automatic adjustment model as the drainage bag automatic adjustment model , the drainage bag automatic adjustment model is: ; Wherein, is the target height corrected drainage efficiency index; is the dynamic optimal height; is the height corrected drainage efficiency index of the current acquisition period; is the preprocessed drainage bag height data of the current acquisition period; is the body position change influence index; is the height correction drainage efficiency index is the proportionality coefficient of is the deviation coefficient is the body position compensation coefficient; The body position change influence index is: ; where is the angle-velocity coupling coefficient; is the body position change angle; is the body position change speed.

[0016] According to one preferred embodiment of the present invention, the process of obtaining the height automatic adjustment command according to the drainage bag automatic adjustment model includes: According to the drainage bag automatic adjustment model to obtain the height automatic adjustment command The height automatic adjustment command is: ; where is the safety clip function; are the minimum safety height and the maximum safety height respectively; The height correction drainage efficiency index of the current acquisition cycle When increasing, according to the drainage bag automatic adjustment model the height automatic adjustment command increases, and based on the human physiological effect, the height correction drainage efficiency index decreases, so as to realize that when the height correction drainage efficiency index is too high, the height of the drainage bag is lifted to slow down the drainage power; The height correction drainage efficiency index of the current acquisition cycle When decreasing, according to the drainage bag automatic adjustment model the height automatic adjustment command decreases, and based on the human physiological effect, the height correction drainage efficiency index increases, so as to realize that when the height correction drainage efficiency index is too low, the height of the drainage bag is lowered to increase the drainage power.

[0017] The present invention further provides a drainage monitoring system for neurosurgical craniocerebral surgery after operation, including: A data acquisition module for acquiring the intracranial pressure data of the patient, as well as the patient's body position change data, the drainage bag height data, and the liquid flow rate data of the drainage tube; A data preprocessing module preprocesses intracranial pressure data, drainage bag height data, and liquid flow rate data based on the local outlier factor and the Z-score normalization algorithm to obtain preprocessed intracranial pressure data, preprocessed drainage bag height data, and preprocessed liquid flow rate data; A data analysis module performs coupled analysis on the preprocessed intracranial pressure data and the preprocessed liquid flow rate data based on a machine learning algorithm to obtain corresponding dynamic drainage power indices; performs dynamic correction analysis on the dynamic drainage power indices and the preprocessed drainage bag height data based on a logistic regression algorithm to obtain a height-corrected drainage efficiency index, and then adjusts the height of the drainage bag; An intelligent monitoring module constructs an automatic drainage bag adjustment model based on a neural network model, obtains a height automatic adjustment command according to the automatic drainage bag adjustment model; automatically adjusts the height of the drainage bag according to the height automatic adjustment command.

[0018] The present invention further provides a computer-readable storage medium storing a computer program that can be executed by a processor to implement the described method for monitoring drainage after neurosurgical craniotomy.

[0019] Compared with the prior art, the beneficial effects of the present invention are: 1. Intracranial pressure data, body position change data, drainage bag height data, and liquid flow rate data are collected, and relevant data are preprocessed using the local outlier factor and the Z-score normalization algorithm to effectively filter abnormal data, providing a reliable data basis for subsequent analysis. On this basis, a machine learning algorithm performs coupled analysis on the preprocessed intracranial pressure data and liquid flow rate data to accurately generate dynamic drainage power indices, and a logistic regression algorithm further analyzes the dynamic drainage power indices and the preprocessed drainage bag height data to obtain a height-corrected drainage efficiency index, thereby accurately adjusting the height of the drainage bag, avoiding problems such as intracranial pressure imbalance caused by abnormal drainage, effectively ensuring the treatment safety of patients, and reducing the occurrence probability of complications.

[0020] 2. The automatic drainage bag adjustment model constructed based on the neural network model can quickly generate a height automatic adjustment command according to real-time data, and the response speed and adjustment accuracy far exceed manual operation. When the height-corrected drainage efficiency index is abnormal, the system can quickly make adjustments and solve problems in a timely manner, enabling nurses to devote more energy to the personalized care and comprehensive condition assessment of patients, promoting the transformation of nursing work towards intelligence and high efficiency, and improving the overall nursing work efficiency and quality. Description of the Drawings

[0021] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the accompanying drawings required in the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other accompanying drawings can also be obtained based on these drawings.

[0022] Figure 1 It is a schematic diagram of the steps of a method for monitoring the drainage after neurosurgical craniocerebral surgery.

[0023] Figure 2 It is a flowchart for judging a method for monitoring the drainage after neurosurgical craniocerebral surgery.

[0024] Figure 3 It is a schematic diagram of the modules of a system for monitoring the drainage after neurosurgical craniocerebral surgery. Detailed implementation manners

[0025] To make the objectives, technical solutions and advantages of the present invention clearer, the following will describe the technical solutions of the present invention in detail. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other implementation manners obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by the present invention.

[0026] As Figure 1 shown, a method for monitoring the drainage after neurosurgical craniocerebral surgery includes the following steps: Collect the intracranial pressure data, body position change data, drainage bag height data, and liquid flow rate data of the drainage tube of the patient; Preprocess the intracranial pressure data, drainage bag height data, and liquid flow rate data based on the local outlier factor and the Z-score normalization algorithm to obtain preprocessed intracranial pressure data, preprocessed drainage bag height data, and preprocessed liquid flow rate data; Perform coupled analysis on the preprocessed intracranial pressure data and the preprocessed liquid flow rate data based on a machine learning algorithm to obtain the corresponding dynamic drainage power index; perform dynamic correction analysis on the dynamic drainage power index and the preprocessed drainage bag height data based on a logistic regression algorithm to obtain a height-corrected drainage efficiency index, and then adjust the height of the drainage bag; Based on a neural network model, construct an automatic adjustment model for the drainage bag. According to the automatic adjustment model for the drainage bag, obtain a height automatic adjustment command; according to the height automatic adjustment command, automatically adjust the height of the drainage bag.

[0027] It should be further noted that in the specific implementation process, the specific process of collecting the intracranial pressure data, body position change data, drainage bag height data, and liquid flow rate data of the drainage tube of the patient includes: A data acquisition device is set up. The data acquisition device includes an intracranial pressure acquisition unit, a body position change acquisition unit, a drainage bag height acquisition unit, and a drainage tube liquid flow rate acquisition unit; a corresponding acquisition period is set. The acquisition period consists of several acquisition moments and synchronizes timestamps based on the PTP protocol; The intracranial pressure acquisition unit acquires the intracranial pressure data of the patient according to the acquisition period; the body position change acquisition unit acquires the body position change data of the patient according to the acquisition period; the drainage bag height acquisition unit acquires the drainage bag height data according to the acquisition period; the drainage tube liquid flow rate acquisition unit acquires the liquid flow rate data of the drainage tube according to the acquisition period.

[0028] It should be further noted that in the specific implementation process, the specific process of preprocessing the intracranial pressure data, the drainage bag height data, and the liquid flow rate data includes: Obtain the intracranial pressure data, the drainage bag height data, and the liquid flow rate data; Perform outlier detection on the intracranial pressure data, the drainage bag height data, and the liquid flow rate data based on the local outlier factor. The specific process includes: Denote the intracranial pressure data, the drainage bag height data, and the liquid flow rate data as , and ; Preset the nearest neighbor ; According to the nearest neighbor , obtain the reachable distance . The reachable distance is: ; where is the Euclidean distance, is the th nearest neighbor distance of the data point According to the reachable distance , obtain the local reachability density . The local reachability density is: ; where represents the data point corresponding to the same acquisition period and acquisition moment; According to the local reachability density , obtain the corresponding local outlier factor ; The local outlier factor is: ; Preset the local outlier factor threshold ; It should be further noted that when the local outlier factor is such and such, it is marked as an outlier; otherwise, it is a normal point; Take the nearest neighborhood of the intracranial pressure data ; ; According to the local outlier factor , obtain the local outlier factor of intracranial pressure ; Preset the threshold of the local outlier factor of intracranial pressure ; When the local outlier factor of intracranial pressure is such and such, it is marked as an outlier of intracranial pressure data; Take the nearest neighborhood of the drainage bag height data ; ; According to the local outlier factor , obtain the local outlier factor of intracranial pressure ; Preset the threshold of the local outlier factor of drainage bag height ; When the local outlier factor of drainage bag height is such and such, it is marked as an outlier of drainage bag height data; Take the nearest neighborhood of the liquid flow rate data ; ; According to the local outlier factor , obtain the local outlier factor of liquid flow rate ; Preset the threshold of the local outlier factor of liquid flow rate ; When the local outlier factor of liquid flow rate is such and such, it is marked as an outlier of liquid flow rate data; Standardize the intracranial pressure data, drainage bag height data, and liquid flow rate data after outlier detection based on the Z-score standardization algorithm. The specific process includes: Eliminate the outliers of intracranial pressure data, drainage bag height data, and liquid flow rate data; Based on the linear interpolation method, repair and fill the outliers of intracranial pressure data; based on the mean value of historical drainage bag height data, repair and fill the outliers of drainage bag height data; based on the hydrodynamic pressure difference formula, repair and fill the outliers of liquid flow rate data; Calculate the intracranial pressure data , drainage bag height data and liquid flow rate data The corresponding mean value and standard deviation; Based on the Z-score normalization formula, the standardized intracranial pressure data is obtained , the drainage bag height data and the liquid flow rate data , which are denoted as the preprocessed intracranial pressure data, the preprocessed drainage bag height data, and the preprocessed liquid flow rate data.

[0029] It should be further noted that in the specific implementation process, the specific process of coupling and analyzing the preprocessed intracranial pressure data and the preprocessed liquid flow rate data to obtain the corresponding dynamic drainage power index includes: Obtain the preprocessed intracranial pressure data and the preprocessed liquid flow rate data of the same acquisition cycle, which are respectively denoted as and ; According to the preprocessed intracranial pressure data and the preprocessed liquid flow rate data of the same acquisition cycle, obtain the mean value of the preprocessed intracranial pressure data and the mean value of the preprocessed liquid flow rate data ; ; ; where represents the corresponding acquisition moment; According to the mean value of the preprocessed intracranial pressure data and the mean value of the preprocessed liquid flow rate data , obtain the normalized power term ; Calculate the dynamic change rate of the preprocessed intracranial pressure data and the dynamic change rate of the preprocessed liquid flow rate data ; where is the difference of the acquisition moments; Based on the machine learning algorithm, and according to the dynamic change rate of the preprocessed intracranial pressure data and the dynamic change rate of the preprocessed liquid flow rate data , obtain the fluctuation correction term ; where , are respectively the weight coefficients of the intracranial pressure data and the liquid flow rate data; According to the normalized power term and the fluctuation correction term , obtain the dynamic drainage power index ; the dynamic drainage power index is: .

[0030] It should be further noted that in the specific implementation process, the specific process of dynamically correcting and analyzing the dynamic drainage power index and the preprocessed drainage bag height data to obtain the height-corrected drainage efficiency index includes: Construct a dynamic optimal height model; Based on the basic formula of the physiological mechanism, construct a basic height model ; The basic height model is: ; where is the mean value of the preprocessed intracranial pressure data; is the ideal drainage power threshold; is the cerebrospinal fluid gravity constant; Based on the logistic regression algorithm and according to the basic height model , obtain an individualized correction index , The individualized correction index is: ; where is the age correction; is the surgical type correction; is the postoperative days correction; Record the preprocessed drainage bag height data of the same acquisition period as ; Obtain the mean value of the preprocessed drainage bag height data of the same acquisition period ; According to the individualized correction index and the mean value of the preprocessed drainage bag height data , obtain a height deviation index decay function ; where is the clinical fitting index, and the value is , that is, the deviation efficiency decay ; Set boundary constraint conditions , that is When, force the height deviation index decay function ; It should be further noted that is the lowest efficiency of the height deviation index decay.

[0031] According to the height deviation index decay function and the dynamic drainage power index , obtain a height-corrected drainage efficiency index ; The height-corrected drainage efficiency index is: ; According to the height-corrected drainage efficiency index Adjust the height of the drainage bag.

[0032] It should be further noted that in the specific implementation process, the specific process of constructing the automatic adjustment model of the drainage bag includes: Obtain several groups of height-corrected drainage efficiency indexes and body position change data; For several groups of height-corrected drainage efficiency indexes and body position change data, conduct grouped labeling, denoted as where \(n\) is a natural number; Take the \(n\) groups of height-corrected drainage efficiency indexes and body position change data as sample data, and where \(n\) is a natural number less than and use the sample data to obtain the mean value of the sample data, denoted as the sample set; take the remaining several groups of height-corrected drainage efficiency indexes and body position change data as the test set; according to the sample set and the test set, form the training sample set; Based on the neural network model, construct the standard height automatic adjustment model; And input the training sample set into the standard height automatic adjustment model to train the standard height automatic adjustment model, and denote the trained standard height automatic adjustment model as the drainage bag automatic adjustment model The drainage bag automatic adjustment model is: ; where, \(y\) is the target height-corrected drainage efficiency index; \(h_{opt}\) is the dynamic optimal height; \(y_i\) is the height-corrected drainage efficiency index of the current acquisition cycle; \(h_i\) is the preprocessed drainage bag height data of the current acquisition cycle; \(k\) is the body position change influence index; \(y_j\) is the height-corrected drainage efficiency index \(\alpha\) is the proportionality coefficient of the height-corrected drainage efficiency index \(\beta\) is the deviation coefficient, \(\gamma\) is the body position compensation coefficient, set by relevant technical personnel; It should be further noted that the body position change influence index \(k\) is: ; where, \(k_{av}\) is the angle-velocity coupling coefficient; \(\theta\) is the body position change angle; \(\omega\) is the body position change speed.

[0033] It should be further noted that in the specific implementation process, the specific process of obtaining the height automatic adjustment command according to the drainage bag automatic adjustment model includes: According to the drainage bag automatic adjustment model , obtain the height automatic adjustment command , the height automatic adjustment command is: ; where is the safety clip function; are the minimum safety height and the maximum safety height respectively.

[0034] When the height correction drainage efficiency index of the current acquisition cycle increases, according to the drainage bag automatic adjustment model , the height automatic adjustment command increases, and based on the human physiological effect, the height correction drainage efficiency index decreases, realizing that when the height correction drainage efficiency index is too high, the height of the drainage bag is increased to slow down the drainage power; When the height correction drainage efficiency index of the current acquisition cycle decreases, according to the drainage bag automatic adjustment model , the height automatic adjustment command decreases, and based on the human physiological effect, the height correction drainage efficiency index increases, realizing that when the height correction drainage efficiency index is too low, the height of the drainage bag is decreased to increase the drainage power.

[0035] According to the height automatic adjustment command , the height of the drainage bag is automatically adjusted, providing effective management for the postoperative drainage care of craniocerebral patients and reducing the nursing workload, providing a basis for realizing rapid nursing; providing important guidance and reference for the rehabilitation of patients after treatment by the attending doctor; providing effective support for accelerating the patient's rehabilitation.

[0036] As Figure 3 shown, a postoperative drainage monitoring system for neurosurgical craniocerebral patients includes: A data acquisition module for acquiring the intracranial pressure data of the patient, as well as the patient's body position change data, the drainage bag height data, and the liquid flow rate data of the drainage tube; A data preprocessing module preprocesses the intracranial pressure data, the drainage bag height data, and the liquid flow rate data based on the local outlier factor and the Z-score normalization algorithm to obtain the preprocessed intracranial pressure data, the preprocessed drainage bag height data, and the preprocessed liquid flow rate data; The data analysis module performs coupled analysis on the preprocessed intracranial pressure data and the preprocessed liquid flow rate data based on machine learning algorithms to obtain corresponding dynamic drainage power indices; performs dynamic correction analysis on the dynamic drainage power indices and the preprocessed drainage bag height data based on logistic regression algorithms to obtain height-corrected drainage efficiency indices, and then adjusts the height of the drainage bag. The intelligent monitoring module constructs an automatic drainage bag adjustment model based on a neural network model, obtains a height automatic adjustment command according to the automatic drainage bag adjustment model; and automatically adjusts the height of the drainage bag according to the height automatic adjustment command.

[0037] The above embodiments are only used to illustrate the technical methods of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical methods of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A method for monitoring the drainage after neurosurgical craniocerebral surgery, characterized in that, The method includes: Collecting intracranial pressure data, body position change data, drainage bag height data, and liquid flow rate data of the drainage tube of the patient; Preprocessing the intracranial pressure data, drainage bag height data, and liquid flow rate data based on the local outlier factor and the Z-score normalization algorithm to obtain preprocessed intracranial pressure data, preprocessed drainage bag height data, and preprocessed liquid flow rate data; Performing coupled analysis on the preprocessed intracranial pressure data and the preprocessed liquid flow rate data based on a machine learning algorithm to obtain corresponding dynamic drainage power indexes; performing dynamic correction analysis on the dynamic drainage power indexes and the preprocessed drainage bag height data based on a logistic regression algorithm to obtain a height-corrected drainage efficiency index, and then adjusting the height of the drainage bag; Based on a neural network model, constructing an automatic drainage bag adjustment model, obtaining a height automatic adjustment command according to the automatic drainage bag adjustment model; automatically adjusting the height of the drainage bag according to the height automatic adjustment command.

2. The method for monitoring the drainage after neurosurgical craniocerebral surgery according to claim 1, wherein The process of collecting intracranial pressure data, body position change data, drainage bag height data, and liquid flow rate data of the drainage tube of the patient includes: Setting up a data collection device, the data collection device including an intracranial pressure collection unit, a body position change collection unit, a drainage bag height collection unit, and a drainage tube liquid flow rate collection unit; setting a corresponding collection period, the collection period consisting of a number of collection times, and synchronizing time stamps based on the PTP protocol; The intracranial pressure collection unit collects the intracranial pressure data of the patient according to the collection period; the body position change collection unit collects the body position change data of the patient according to the collection period; the drainage bag height collection unit collects the drainage bag height data according to the collection period; the drainage tube liquid flow rate collection unit collects the liquid flow rate data of the drainage tube according to the collection period.

3. The method for monitoring drainage after neurosurgical craniocerebral surgery according to claim 2, wherein, The process of preprocessing the intracranial pressure data, drainage bag height data, and liquid flow rate data includes: Obtaining the intracranial pressure data, drainage bag height data, and liquid flow rate data; Performing outlier detection on the intracranial pressure data, drainage bag height data, and liquid flow rate data based on the local outlier factor, including: Let the intracranial pressure data, the drainage bag height data, and the liquid flow rate data be denoted as , and ; Preset nearest field ; According to the nearest field , the reachable distance is obtained, and the reachable distance is: ; where is the Euclidean distance, and is the distance to the -th nearest neighbor of the data point According to the reachable distance , the local reachability density is obtained , and the local reachability density is: ; where represents the data point at the acquisition moment corresponding to the same acquisition period ; According to the local reachability density , the corresponding local outlier factor is obtained ; the local outlier factor is: ; Preset local outlier factor threshold ; When the local outlier factor , it is marked as an outlier; otherwise, it is a normal point; Take the nearest neighborhood of the intracranial pressure data ; Obtain the local outlier factor of intracranial pressure according to the local outlier factor ; Preset the local outlier factor threshold of intracranial pressure ; When the local outlier factor of intracranial pressure ; If the local outlier factor of intracranial pressure is met, it is marked as an outlier point of intracranial pressure data; The drainage bag height data The nearest neighbor of ; According to the local outlier factor , obtain the local outlier factor of intracranial pressure ; Preset drainage bag height local outlier factor threshold ; When the drainage bag height local outlier factor When , it is marked as an outlier point of drainage bag height data; Take the nearest neighborhood of the liquid flow rate data ; According to the local outlier factor , obtain the local outlier factor of the liquid flow rate ; Preset the local outlier factor threshold of the liquid flow rate ; When the local outlier factor of the liquid flow rate ; then mark it as an outlier point of the liquid flow rate data; ​ Normalizing the intracranial pressure data, drainage bag height data, and liquid flow rate data after outlier detection based on the Z-score normalization algorithm, including: Removing outlier points in the intracranial pressure data, outlier points in the drainage bag height data, and outlier points in the liquid flow rate data; Based on the linear interpolation method, repairing and filling the outlier points in the intracranial pressure data; based on the mean value of historical drainage bag height data, repairing and filling the outlier points in the drainage bag height data; based on the hydrodynamic pressure difference formula, repairing and filling the outlier points in the liquid flow rate data; Calculate the intracranial pressure data after repair and filling , the drainage bag height data and the liquid flow rate data and their corresponding means and standard deviations; Based on the Z-score normalization formula, the standardized intracranial pressure data is obtained , the drainage bag height data and the liquid flow rate data , which are denoted as the preprocessed intracranial pressure data, the preprocessed drainage bag height data, and the preprocessed liquid flow rate data.

4. A method for monitoring the drainage after neurosurgical craniocerebral surgery according to claim 3, characterized in that, The process of performing coupled analysis on the preprocessed intracranial pressure data and the preprocessed liquid flow rate data to obtain corresponding dynamic drainage power indexes includes: Denote the preprocessed intracranial pressure data and the preprocessed fluid flow rate data in the same acquisition period as and ; Based on the preprocessed intracranial pressure data of the same acquisition period and the preprocessed liquid flow rate data , obtain the mean value of the preprocessed intracranial pressure data for the current acquisition period and the mean value of the preprocessed liquid flow rate data ; Based on the mean value of the preprocessed intracranial pressure data and the mean value of the preprocessed liquid flow rate data , a normalized dynamic term is obtained ; Calculate the dynamic change rate of preprocessed intracranial pressure data and the dynamic change rate of preprocessed liquid flow rate data ; wherein is the differential at the acquisition time Based on a machine learning algorithm and according to the dynamic change rate of the preprocessed intracranial pressure data and the dynamic change rate of the preprocessed liquid flow rate data , a fluctuation correction term is obtained ; where and are the weight coefficients of the intracranial pressure data and the liquid flow rate data respectively; According to the normalized dynamic term and the fluctuation correction term , the dynamic drainage power index is obtained; the dynamic drainage power index is as follows: .

5. A method for monitoring the drainage after neurosurgical craniocerebral surgery according to claim 4, characterized in that, The process of constructing a dynamic optimal height model includes: Construct a basic height model based on the basic formula of physiological mechanism ; The basic height model is as follows: ; wherein, is the mean value of the preprocessed intracranial pressure data; is the ideal drainage power threshold; is the cerebrospinal fluid gravity constant; Based on the logistic regression algorithm and according to the said basic height model , an individualized correction index is obtained , and the said individualized correction index is as follows: ; among them, is the age correction; is the surgical type correction; is the postoperative day correction.

6. The method for monitoring the drainage after neurosurgical craniocerebral surgery according to claim 5, wherein The process of performing dynamic correction analysis on the dynamic drainage power indexes and the preprocessed drainage bag height data to obtain a height-corrected drainage efficiency index includes: Record the height data of the preprocessed drainage bag in the same collection cycle as ; Obtain the average value of the height data of the preprocessed drainage bag in the same collection cycle ; According to the individualized correction index and the mean value of the preprocessed drainage bag height data , obtain the height deviation index decay function ; where is the clinical fitting index, and its value is ; Set boundary constraint conditions When , force the height to deviate from the exponential decay function ; is the lowest efficiency of the height deviation from the exponential decay; According to the height deviation exponential decay function and the dynamic drainage power index , the height-corrected drainage efficiency index is obtained; the height-corrected drainage efficiency index is: ; According to the height-corrected drainage efficiency index , the height of the drainage bag is adjusted.

7. A method for monitoring the drainage after neurosurgical craniocerebral surgery according to claim 6, characterized in that, The process of constructing an automatic drainage bag adjustment model includes: Obtain several groups of height-corrected drainage efficiency indices and body position change data; Group several sets of height-corrected drainage efficiency indices and group numbers are assigned to the body position change data, denoted as where \(n\) is a natural number; Take the group height correction drainage efficiency index and the body position change data as sample data, and is less than a natural number, and use the sample data to obtain the sample data mean, denoted as the sample set; take the remaining several groups of height correction drainage efficiency indexes and the body position change data as the test set; according to the sample set and the test set, form the training sample set; Based on a neural network model, constructing a standard height automatic adjustment model; And input the training sample set into the standard height automatic adjustment model to train the standard height automatic adjustment model, and denote the trained standard height automatic adjustment model as the drainage bag automatic adjustment model , the drainage bag automatic adjustment model is as follows: ； Among them, is the target height correction drainage efficiency index; is the dynamic optimal height; is the height correction drainage efficiency index of the current acquisition cycle; is the preprocessed drainage bag height data of the current acquisition cycle; is the body position change influence index; is the height correction drainage efficiency index 's proportionality coefficient, is the deviation coefficient, is the body position compensation coefficient; The body position change influence index is defined as: wherein, is the angle-velocity coupling coefficient; is the body position change angle; is the body position change velocity.

8. A method for monitoring the drainage after neurosurgical craniocerebral surgery according to claim 7, characterized in that, According to the automatic adjustment model of the drainage bag, the process of obtaining the height automatic adjustment command includes: According to the automatic adjustment model of the drainage bag , obtain the height automatic adjustment command , the height automatic adjustment command is as follows: ; wherein, is a secure clipping function; are the minimum safe height and the maximum safe height, respectively; Height correction drainage efficiency index of the current collection cycle When increasing, according to the drainage bag automatic adjustment model the height automatic adjustment command increases, based on the human physiological effect, the height correction drainage efficiency index decreases, to achieve the height correction drainage efficiency index When it is too high, raise the height of the drainage bag to slow down the drainage power; Height correction drainage efficiency index of the current collection cycle When it decreases, according to the drainage bag automatic adjustment model Height automatic adjustment command Decrease, based on the physiological effect of the human body, the height correction drainage efficiency index Increase, to achieve the height correction drainage efficiency index When it is too low, lower the height of the drainage bag to enhance the drainage power.

9. A drainage monitoring system for neurosurgical craniocerebral surgery after operation, which executes a drainage monitoring method for neurosurgical craniocerebral surgery after operation according to any one of claims 1 to 8 above, characterized in that, The system includes: A data acquisition module, configured to acquire the intracranial pressure data of the patient, as well as the patient's body position change data, the drainage bag height data, and the liquid flow rate data of the drainage tube; A data preprocessing module, which preprocesses the intracranial pressure data, the drainage bag height data, and the liquid flow rate data based on the local outlier factor and the Z-score normalization algorithm to obtain preprocessed intracranial pressure data, preprocessed drainage bag height data, and preprocessed liquid flow rate data; A data analysis module, which performs coupled analysis on the preprocessed intracranial pressure data and the preprocessed liquid flow rate data based on a machine learning algorithm to obtain the corresponding dynamic drainage power index; performs dynamic correction analysis on the dynamic drainage power index and the preprocessed drainage bag height data based on a logistic regression algorithm to obtain a height-corrected drainage efficiency index, and then adjusts the height of the drainage bag; An intelligent monitoring module, which constructs an automatic adjustment model of the drainage bag based on a neural network model, and obtains a height automatic adjustment command according to the automatic adjustment model of the drainage bag; automatically adjusts the height of the drainage bag according to the height automatic adjustment command.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and the computer program can be executed by a processor to implement a drainage monitoring method for neurosurgical craniocerebral surgery after operation as described in any one of claims 1 to 8 above.

Images