Data evaluation and comparison system before and after postoperative rehabilitation of posterior fossa brain tumor

Abstract

The invention relates to the technical field of medical data evaluation, in particular to a data evaluation and comparison system before and after posterior fossa brain tumor operation rehabilitation, and provides an accurate comparison reference for subsequent recovery prediction and progress monitoring through standardized processing of preoperative physiological data and extraction of key indexes. Through continuous monitoring and comparison of real-time physiological data, abnormal fluctuation can be found in the rehabilitation process, a treatment scheme can be adjusted in time, secondary health risks caused by recovery lag or abnormity are avoided, a hidden Markov model is adopted, a scientific basis is provided for careful division of rehabilitation stages through time series data analysis, and the method is suitable for clinical application and popularization. According to the method and the system, all rehabilitation stages are accurately identified according to the actual state of the patient, and the principal component analysis method and comprehensive evaluation of multi-dimensional data are combined, so that the precision of recovery progress monitoring is improved, challenges possibly encountered by the patient in the rehabilitation process can be identified through analysis of the recovery bottleneck, and the pertinence and effectiveness of a treatment scheme are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data evaluation, and in particular to a system for evaluating and comparing data before and after rehabilitation of posterior fossa brain tumor surgery. Background Art

[0002] The field of data evaluation technology mainly focuses on collecting, processing and analyzing patients' medical data, evaluating and monitoring key medical indicators such as treatment effects and rehabilitation progress. By comparing and analyzing the patient's data before and after treatment, it can monitor the patient's recovery after surgery in real time, helping medical staff adjust treatment plans and optimize the rehabilitation process.

[0003] The posterior fossa brain tumor surgery pre- and post-operative data evaluation and comparison system is designed to evaluate the patient's postoperative recovery by analyzing the data before and after brain tumor surgery. The purpose is to help doctors and rehabilitation experts understand the patient's postoperative changes in real time through digital means, provide scientific basis, and promote the formulation of more accurate personalized rehabilitation plans.

[0004] Existing technologies lack dynamic monitoring and detailed phased analysis of physiological changes during the rehabilitation process, and are unable to provide detailed, quantitative division and evaluation of rehabilitation stages. They are also unable to capture individual differences that occur during the patient's rehabilitation process in real time. Lacking a real-time feedback mechanism, existing technologies are unable to deeply integrate and analyze multiple physiological indicators and motor skills information, resulting in incomplete and inaccurate assessments of rehabilitation effects, making it difficult to detect potential problems in a timely manner, and causing the evaluation results to lag and lack specificity. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a data evaluation and comparison system for posterior fossa brain tumor surgery before and after recovery.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A system for evaluating and comparing data before and after rehabilitation of posterior fossa brain tumor surgery includes:

[0007] Preoperative health benchmark analysis module: Based on the patient's preoperative heart rate, body temperature, blood oxygen data and imaging examination results, a health assessment is conducted to obtain the patient's preoperative physiological status data. The physiological parameters are sorted in combination with the medical history, key physiological indicators are extracted, and the difference with the baseline values is calculated to identify potential health problems and generate preoperative health benchmark data.

[0008] Postoperative physiological status monitoring module: Based on the real-time collected patient physiological status data, sensors continuously monitor and compare data change trends, identify abnormal fluctuations, record gait, motor skills and sleep status, and compare them with the preoperative health baseline data to obtain real-time postoperative physiological data;

[0009] Rehabilitation stage division module: Based on the real-time postoperative physiological data, the module uses a hidden Markov model to analyze time series data to identify the fluctuation range of key indicators such as heart rate and gait, judge the rehabilitation status, and divide the patient into acute phase, recovery phase, and consolidation phase. Each phase is divided by gait analysis and motor ability data to generate patient rehabilitation stage data;

[0010] Recovery prediction and progress monitoring module: Based on the preoperative health baseline data and postoperative real-time physiological data, principal component analysis is used to calculate the recovery progress, compare the current progress with historical data, monitor heart rate and gait changes, detect whether the rehabilitation progress meets expectations, and conduct deviation assessment to generate recovery deviation data;

[0011] Multi-dimensional recovery status analysis module: Based on the recovery deviation data and real-time postoperative physiological data, combined with various key indicators of the recovery period, heart rate, gait and blood oxygen data are calculated, integrated according to different weights, and bottlenecks in the recovery process are analyzed. Key features are extracted through data combination to generate multi-dimensional recovery status data;

[0012] Phased recovery effect evaluation module: Based on the multi-dimensional recovery status data and the patient's rehabilitation stage data, combined with heart rate, blood oxygen and gait indicators, it determines the recovery effect of each stage, compares the key indicator values with the predetermined threshold value, evaluates the rehabilitation effect, identifies potential recovery obstacles, and generates phased rehabilitation effect data;

[0013] Final evaluation result output module: Based on the staged rehabilitation effect data, multi-dimensional recovery status data and patient rehabilitation stage data, through data analysis, combined with physiological indicators and motion data, output the patient's overall recovery progress report, integrate the recovery status of different stages, and generate complete evaluation result data.

[0014] As a further solution of the present invention, the preoperative health benchmark analysis module includes a physiological data collation submodule, a key physiological indicator extraction submodule and a health benchmark data generation submodule, wherein:

[0015] Physiological data organization submodule: Based on the patient's preoperative heart rate, body temperature, blood oxygen data and imaging examination results, combined with the patient's medical history, organizes physiological status data in a standardized and structured manner, and pre-processes each data item, such as removing noise, filling missing values, and unifying data formats, to generate preoperative physiological status data;

[0016] Key physiological indicator extraction submodule: Based on the preoperative physiological status data, key physiological indicators such as heart rate, body temperature, and blood oxygen are selected, the indicator values are extracted item by item and standardized, and the stability and deviation of each indicator are further evaluated through data analysis to generate key physiological indicator data;

[0017] Health baseline data generation submodule: Based on the key physiological indicator data, the difference between each indicator and the baseline value is calculated, the change trend is analyzed, and the patient's preoperative health baseline data is generated according to specific rules.

[0018] As a further solution of the present invention, the postoperative physiological status monitoring module includes a real-time physiological data acquisition submodule, a data trend comparison submodule and a postoperative real-time data generation submodule, wherein:

[0019] Real-time physiological data acquisition submodule: Based on the real-time collected patient physiological status data, including heart rate, body temperature, blood oxygen, gait, exercise ability and sleep status, data cleaning and preprocessing are performed to remove abnormal data and correct data deviations during the processing process to generate real-time postoperative physiological data;

[0020] Data trend comparison submodule: Based on the real-time postoperative physiological data, continuously track and compare the change trend, identify the fluctuation points in the data, and use the difference comparison method to analyze, identify abnormal fluctuations and deviations, and generate data change trend analysis results;

[0021] Postoperative real-time data generation submodule: compares and analyzes postoperative real-time physiological data with preoperative health baseline data, identifies the differences between postoperative data and baseline data, and generates the patient's real-time physiological status data based on the differences, thereby generating postoperative real-time physiological data.

[0022] As a further solution of the present invention, the rehabilitation stage division module includes a data fluctuation identification submodule, a stage division submodule and a rehabilitation stage data generation submodule, wherein:

[0023] Data Fluctuation Identification Submodule: Based on the real-time postoperative physiological data, a hidden Markov model is used to analyze time series data to identify the fluctuation range of key physiological indicators such as heart rate and gait. A window sliding method is used to obtain fluctuation data for each time period, and data segmentation and comparison are performed to generate fluctuation range analysis results.

[0024] Stage division submodule: Based on the fluctuation range analysis results, statistical analysis is performed on key indicators such as heart rate and gait to calculate the fluctuation range of each indicator. Combined with the clinically defined acute phase, recovery phase, and consolidation phase standards, the physiological state of each phase is divided to generate the rehabilitation stage division results;

[0025] Rehabilitation stage data generation submodule: Based on the results of the rehabilitation stage division, combined with gait analysis and motor ability data, the key indicator values of each stage are counted through data aggregation methods, and each stage is further evaluated and data merged to generate patient rehabilitation stage data.

[0026] As a further solution of the present invention, the recovery prediction and progress monitoring module includes a recovery progress calculation submodule, a progress comparison and analysis submodule, and a recovery deviation data generation submodule, wherein:

[0027] Recovery progress calculation submodule: Based on the preoperative health baseline data and postoperative real-time physiological data, principal component analysis is used to compare the recovery status of each physiological indicator item by item. By calculating the change rate and difference of each indicator, a numerical representation of the recovery progress is obtained, and a recovery progress calculation result is generated;

[0028] Progress comparison and analysis submodule: Based on the recovery progress calculation results, historical data is compared with existing data to monitor changes in key indicators such as heart rate and gait, identify deviations in the current recovery progress, and generate progress comparison and analysis results;

[0029] Recovery deviation data generation submodule: Based on the progress comparison and analysis results, calculate the deviation between the current recovery progress and the expected progress, use the data difference calculation method to obtain recovery deviation data, and generate recovery deviation data.

[0030] As a further solution of the present invention, the multi-dimensional recovery status analysis module includes a data fusion submodule, a bottleneck analysis submodule and a recovery status data generation submodule, wherein:

[0031] Data fusion submodule: Based on the recovery deviation data and postoperative real-time physiological data, by selecting key indicators such as heart rate, gait, blood oxygen, etc., weighted merging of each data according to the set weights, and processing the fused data using the weighted summation method to generate a fused data set;

[0032] Bottleneck analysis submodule: Based on the fused data set, analyzes bottlenecks that occur during the recovery process, compares data based on the rate of change of recovery indicators, identifies stagnation points and abnormal fluctuations in the data, and generates bottleneck analysis results by identifying bottleneck areas.

[0033] Recovery status data generation submodule: Based on the bottleneck analysis results, combined with the key feature extraction method, various key indicators are aggregated, and multi-dimensional recovery status data is generated using data integration technology.

[0034] As a further solution of the present invention, the staged recovery effect evaluation module includes an effect judgment submodule, a threshold comparison submodule and a rehabilitation effect data generation submodule, wherein:

[0035] Effect judgment submodule: Based on the multi-dimensional recovery status data and the patient's rehabilitation stage data, combined with key indicators such as heart rate, blood oxygen, and gait, the recovery effect of each stage is evaluated, and a judgment result of the staged recovery effect is generated by comparing the indicator values with the predetermined standards;

[0036] Threshold comparison submodule: Based on the staged recovery effect judgment results, key indicator values such as heart rate, blood oxygen, gait, etc. are compared with the set thresholds one by one to calibrate potential recovery obstacles and generate threshold comparison results;

[0037] Rehabilitation effect data generation submodule: Based on the threshold comparison results, the staged recovery effect is comprehensively evaluated, and combined with the key indicator values, the patient's staged rehabilitation effect data is generated.

[0038] As a further solution of the present invention, the final evaluation result output module includes a data integration submodule, a recovery progress calculation submodule and an evaluation result generation submodule, wherein:

[0039] Data integration submodule: Based on the staged rehabilitation effect data, multi-dimensional recovery status data and patient rehabilitation stage data, the data of each stage are summarized, and the physiological data of each stage are integrated one by one through the data matching method to generate an integrated data set;

[0040] Recovery progress calculation submodule: Based on the integrated data set, the physiological indicators and exercise data are analyzed item by item, the recovery progress of the patient at each stage is calculated, and the recovery progress of each indicator is compared with the predetermined standard to generate the recovery progress calculation result;

[0041] Evaluation result generation submodule: Based on the recovery progress calculation results, by comprehensively analyzing the recovery status of each stage, the patient's overall recovery progress is integrated, and the recovery data is merged to generate complete evaluation result data.

[0042] Compared with the prior art, the advantages and positive effects of the present invention are:

[0043] 1. In this invention, by standardizing preoperative physiological data and extracting key indicators, an accurate comparison benchmark is provided for subsequent recovery prediction and progress monitoring. Continuous monitoring and comparison of real-time physiological data can detect abnormal fluctuations during the recovery process, allowing timely adjustments to treatment plans and avoiding secondary health risks caused by delayed or abnormal recovery.

[0044] 2. This invention uses a hidden Markov model and time series data analysis to provide a scientific basis for the detailed division of rehabilitation stages. It also accurately identifies each rehabilitation stage based on the patient's actual condition, providing support for the development of personalized treatment plans.

[0045] 3. In this invention, the combination of principal component analysis and comprehensive evaluation of multi-dimensional data not only improves the accuracy of recovery progress monitoring, but also can identify the challenges that patients may encounter during the rehabilitation process through analysis of recovery bottlenecks, ensuring the targetedness and effectiveness of treatment plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a system flow chart of the present invention;

[0047] Figure 2 is a flow chart of the present invention;

[0048] Figure 3 Schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0050] See also Figure 1 The present invention provides a technical solution: a system for evaluating and comparing data before and after rehabilitation of posterior fossa brain tumor surgery, comprising:

[0051] Preoperative health benchmark analysis module: Based on the patient's preoperative heart rate, body temperature, blood oxygen data and imaging examination results, a health assessment is conducted to obtain the patient's preoperative physiological status data. The physiological parameters are sorted in combination with the medical history, key physiological indicators are extracted, and the difference with the baseline values is calculated to identify potential health problems and generate preoperative health benchmark data.

[0052] Postoperative physiological status monitoring module: Based on the real-time collected patient physiological status data, sensors continuously monitor and compare data change trends, identify abnormal fluctuations, record gait, exercise capacity and sleep status, and compare with preoperative health baseline data to obtain real-time postoperative physiological data;

[0053] Rehabilitation stage division module: Based on real-time postoperative physiological data, the module uses a hidden Markov model to analyze time series data to identify the fluctuation range of key indicators such as heart rate and gait, judge the rehabilitation status, and divide the patient into acute, recovery, and consolidation phases. Each phase is divided through gait analysis and motor ability data to generate patient rehabilitation stage data.

[0054] Recovery prediction and progress monitoring module: Based on preoperative health baseline data and postoperative real-time physiological data, principal component analysis is used to calculate recovery progress, compare current progress with historical data, monitor heart rate and gait changes, detect whether the recovery progress meets expectations, and conduct deviation assessments to generate recovery deviation data;

[0055] Multi-dimensional recovery status analysis module: Based on recovery deviation data and real-time postoperative physiological data, combined with key recovery indicators, heart rate, gait, and blood oxygen data are calculated and integrated according to different weights to analyze bottlenecks in the recovery process. Key features are extracted through data combination to generate multi-dimensional recovery status data;

[0056] Phased recovery effect evaluation module: Based on multi-dimensional recovery status data and patient rehabilitation stage data, combined with heart rate, blood oxygen and gait indicators, it determines the recovery effect of each stage. By comparing key indicator values with predetermined thresholds, it evaluates the recovery effect, identifies potential recovery obstacles, and generates phased recovery effect data;

[0057] Final evaluation result output module: Based on the phased rehabilitation effect data, multi-dimensional recovery status data and patient rehabilitation stage data, through data analysis, combined with physiological indicators and motion data, outputs the patient's overall recovery progress report, integrates the recovery status of different stages, and generates complete evaluation result data.

[0058] See also Figure 3 The preoperative health benchmark analysis module includes a physiological data collation submodule, a key physiological indicator extraction submodule, and a health benchmark data generation submodule, among which:

[0059] Physiological data organization submodule: Based on the patient's preoperative heart rate, body temperature, blood oxygen data and imaging examination results, combined with the patient's medical history, organizes physiological status data in a standardized and structured manner, and pre-processes each data item, such as removing noise, filling missing values, and unifying data formats, to generate preoperative physiological status data;

[0060] Key physiological indicator extraction submodule: Based on preoperative physiological status data, key physiological indicators such as heart rate, body temperature, and blood oxygen are selected, and the indicator values are extracted and standardized item by item. The stability and deviation of each indicator are further evaluated through data analysis to generate key physiological indicator data;

[0061] Health baseline data generation submodule: Based on key physiological indicator data, the module calculates the difference between each indicator and the baseline value, analyzes the change trend, and generates the patient's preoperative health baseline data according to specific rules;

[0062] Physiological data collation submodule: Based on preoperative heart rate, body temperature, blood oxygen data and imaging examination results, a standardized algorithm is used to organize various data. First, a data denoising algorithm is used, specifically a median filter algorithm, to smooth each data item. The parameter is set to a filter window size of 3, and a sliding window operation is performed. Then, the KNN algorithm is used to fill missing values. The specific parameter is that the k value is set to 5, and the missing data is filled with the average value of the nearest neighbor. The Z-Score normalization algorithm is used with the parameters set to 0 and the standard deviation to 1. All data are uniformly formatted to finally generate preoperative physiological status data;

[0063] Key physiological indicator extraction submodule: Based on preoperative physiological status data, key physiological indicators such as heart rate, body temperature, and blood oxygen are selected, and the data is subjected to dimensionality reduction processing. The specific operation is to convert the data matrix into a covariance matrix, calculate its eigenvalues and eigenvectors, select the eigenvectors with a cumulative variance of 95%, extract the value of each physiological indicator and perform normalization processing. The normalization method is Min-Max normalization, which scales the data to between 0 and 1. The stability and deviation of each indicator are then evaluated through standard deviation analysis. During the specific execution, the standard deviation of each indicator is calculated and compared with the set threshold to generate key physiological indicator data;

[0064] Health baseline data generation submodule: Based on key physiological indicator data, a baseline difference calculation algorithm is used. Specifically, the difference between each indicator and the baseline value is calculated. The algorithm calculates the difference between each indicator to obtain the difference value, and uses a sliding average algorithm to smooth the difference value. The sliding window size is set to 10, and each difference value is smoothed to reduce short-term fluctuations. Then, a trend analysis algorithm is used to calculate the changing trend of the difference of each indicator. Specifically, a linear regression model is used to fit the difference value. The parameters of the model include the training data set and time series as input to generate the patient's preoperative health baseline data.

[0065] See also Figure 3 The postoperative physiological status monitoring module includes a real-time physiological data acquisition submodule, a data trend comparison submodule, and a postoperative real-time data generation submodule, wherein:

[0066] Real-time physiological data acquisition submodule: Based on the real-time collected patient physiological status data, including heart rate, body temperature, blood oxygen, gait, exercise ability and sleep status, data cleaning and preprocessing are performed to remove abnormal data and correct data deviations during the processing process to generate real-time postoperative physiological data;

[0067] Data trend comparison submodule: Based on real-time postoperative physiological data, it continuously tracks and compares change trends, identifies fluctuation points in the data, and uses difference comparison methods to analyze, identify abnormal fluctuations and deviations, and generate data change trend analysis results;

[0068] Postoperative real-time data generation submodule: compares and analyzes postoperative real-time physiological data with preoperative health baseline data, identifies the differences between postoperative data and baseline data, and generates the patient's real-time physiological status data based on the differences, generating postoperative real-time physiological data;

[0069] Real-time physiological data acquisition submodule: Based on the real-time collected patient physiological status data, including heart rate, body temperature, blood oxygen, gait, exercise ability and sleep status, the interquartile range algorithm is used to detect outliers in the data. The upper and lower quartiles Q1 and Q3 are set, and the interquartile range is calculated. After removing abnormal data, the least squares method is used to correct data deviations. The fit is set to 0.95. The optimal fit curve of the data is calculated by the least squares method to correct the deviations in the data and generate real-time postoperative physiological data;

[0070] Data trend comparison submodule: Based on real-time postoperative physiological data, the module continuously tracks and compares the changing trends. The module uses an autoregressive integral sliding average model to model the data. First, the data is differentiated with the differential order set to 1. After generating the differential data, the model parameters are set to autoregressive order 2, differential order 1, and sliding average order 2. The model is used to predict the data trend. Then, the difference between the actual value and the predicted value at each time point is compared to identify abnormal fluctuations and deviations, and generate data trend analysis results.

[0071] Postoperative real-time data generation submodule: The postoperative real-time physiological data is compared and analyzed with the preoperative health baseline data. The Euclidean distance algorithm is used to calculate the difference between each postoperative data and the preoperative baseline data to generate a difference value; then, based on the weighted difference calculation method, different weights are assigned to each indicator. The weight of heart rate is 0.3, body temperature is 0.2, blood oxygen is 0.2, gait is 0.1, exercise ability is 0.1, and sleep status is 0.1. The patient's real-time physiological status data is generated through weighted calculation, and finally the postoperative real-time physiological data is generated.

[0072] See also Figure 3 The rehabilitation stage division module includes a data fluctuation identification submodule, a stage division submodule and a rehabilitation stage data generation submodule, wherein:

[0073] Data Fluctuation Identification Submodule: Based on real-time postoperative physiological data, the Hidden Markov Model is used to analyze time series data to identify the fluctuation range of key physiological indicators such as heart rate and gait. The window sliding method is used to obtain fluctuation data for each time period, and data segmentation and comparison are performed to generate fluctuation range analysis results.

[0074] Stage division submodule: Based on the fluctuation range analysis results, through statistical analysis of key indicators such as heart rate and gait, the fluctuation range of each indicator is calculated. Combined with the clinically defined acute phase, recovery phase, and consolidation phase standards, the physiological state of each phase is divided to generate the rehabilitation stage division results;

[0075] Rehabilitation stage data generation submodule: Based on the results of rehabilitation stage division, combined with gait analysis and motor ability data, the key indicator values of each stage are counted through data aggregation methods. Each stage is further evaluated and data is merged to generate patient rehabilitation stage data;

[0076] Data Fluctuation Identification Submodule: Based on real-time postoperative physiological data, a hidden Markov model is used to analyze time series data to identify the fluctuation range of key physiological indicators such as heart rate and gait. First, the state space of the hidden Markov model is set to 3, and the state transition probability matrix is observed. A forward algorithm is calculated based on the observed data at each moment. A window sliding method is used, and the window size is set to 30 seconds. Fluctuation data within each time period is obtained. Then, data segmentation and comparison are performed, and the K-means clustering algorithm is used to classify the fluctuation data, ultimately generating fluctuation range analysis results.

[0077] Stage division submodule: Based on the results of the fluctuation range analysis, statistical analysis is performed on key indicators such as heart rate and gait, and the fluctuation range of each indicator is calculated. Specifically, the variance analysis method is used to calculate the variance values of heart rate and gait, and then compared with the set thresholds. Combined with the clinically defined acute phase, recovery phase, and consolidation phase standards, the fluctuation range thresholds for the acute phase are set at heart rate ±20%, the fluctuation range thresholds for the recovery phase are set at heart rate ±10%, and the fluctuation range thresholds for the consolidation phase are set at heart rate ±5%. Based on these standards, the physiological state of each stage is divided, and the final result of the rehabilitation stage division is generated;

[0078] Rehabilitation stage data generation submodule: Based on the results of rehabilitation stage division, combined with gait analysis and motor ability data, the weighted average method is used to assign different weights to the key indicators of each stage. The weights are set to 0.4 for gait, 0.3 for motor ability, 0.2 for heart rate, and 0.1 for body temperature. The indicators of each stage are statistically analyzed using the weighted average method, and the data merging algorithm is used to merge the data of each stage. The data interpolation method is used during the merging, and linear interpolation is used to process missing data to generate patient rehabilitation stage data.

[0079] Hidden Markov model, according to the formula:

[0080]

[0081] Where: P(H t |θ,w1,w2,δ) represents the heart rate H observed at time point t tThe probability of i represents the probability of the previous moment of the current state i, b i (H t ) represents the state i, N represents the number of hidden states, θ represents the parameter set of the hidden Markov model, w1 represents the weight coefficient, w2 represents the weight coefficient, and δ represents the bias weight coefficient;

[0082] Execution process: First, the collected heart rate data H t It is divided into multiple time periods. For each time period, the model calculates the observation probability b of the heart rate at that time i (H t ), and combined with the probability α of each state i at the previous moment i , and then two weight coefficients w1 and w2 are introduced, where w1 is used to control the influence of each state transition in the hidden Markov model and enhance the model's response to different state changes, and w2 is used to adjust the deviation between the heart rate fluctuation and the historical mean. The deviation weight coefficient δ is further adjusted by adjusting the heart rate and the historical mean. The contribution of the difference to the overall calculation is enhanced, the response to data with large fluctuations is enhanced, the probability of heart rate fluctuation at each moment is calculated, the fluctuation range of heart rate in each time period is obtained, and the heart rate fluctuation evaluation results are generated to provide an accurate reference for the patient's rehabilitation data evaluation.

[0083] See also Figure 3 The recovery prediction and progress monitoring module includes a recovery progress calculation submodule, a progress comparison and analysis submodule, and a recovery deviation data generation submodule, among which:

[0084] Recovery progress calculation submodule: Based on preoperative health baseline data and postoperative real-time physiological data, principal component analysis is used to compare the recovery status of each physiological indicator item by item. By calculating the change rate and difference of each indicator, a numerical representation of the recovery progress is obtained, and the recovery progress calculation result is generated;

[0085] Progress comparison and analysis submodule: Based on the recovery progress calculation results, historical data is compared with existing data to monitor changes in key indicators such as heart rate and gait, identify deviations in the current recovery progress, and generate progress comparison and analysis results;

[0086] Recovery deviation data generation submodule: Based on the progress comparison analysis results, calculate the deviation between the current recovery progress and the expected progress, use the data difference calculation method to obtain the recovery deviation data, and generate the recovery deviation data;

[0087] Recovery progress calculation submodule: Based on preoperative health baseline data and postoperative real-time physiological data, principal component analysis is used. First, the data of each physiological indicator is constructed into a matrix, and the covariance matrix is calculated. The eigenvalue decomposition method is used to obtain eigenvalues and eigenvectors. The eigenvectors with a cumulative variance of 95% are selected. The PCA algorithm is used to reduce the dimension of each physiological indicator. The rate of change of each indicator is calculated, and then the indicator difference is calculated to obtain the difference of each indicator, and finally the recovery progress calculation result is generated;

[0088] Progress Comparison Analysis Submodule: Based on the recovery progress calculation results, the historical data comparison analysis method is used. First, historical data is selected as a comparison sample. Through the regression analysis algorithm, specifically using linear regression, setting time as the independent variable and recovery progress as the dependent variable, the historical data is fitted. After obtaining the regression coefficient, the existing data is predicted. Then, the deviation calculation method is used to calculate the difference between historical data and existing data, identify the deviation of the current recovery progress, and finally generate the progress comparison analysis results.

[0089] Recovery deviation data generation submodule: Based on the progress comparison analysis results, the deviation between the current recovery progress and the expected progress is calculated. The data difference calculation method is adopted, specifically the weighted average difference method. The weights of various physiological indicators are set, such as heart rate set to 0.4, gait set to 0.3, blood oxygen set to 0.2, and exercise capacity set to 0.1. The difference of each physiological indicator is calculated, and the difference values are weighted averaged to finally generate the recovery deviation data.

[0090] Principal component analysis, according to the formula:

[0091]

[0092] Where: P is the principal component matrix after dimensionality reduction, X is the original data matrix, Q is the new eigenvector matrix, γ1 is the weight coefficient, ΔX is the difference between the postoperative data and the preoperative data, γ2 is the standard deviation weight coefficient, σ X is the standard deviation of the original data, γ3 is the balance coefficient, N is the number of samples in the data set, X i is the data of the i-th sample;

[0093] Implementation process: First, collect preoperative and postoperative physiological data and organize them into the original data matrix X. Then, reduce the dimension of the data through the eigenvector matrix Q, convert the high-dimensional data into the main change direction, and obtain the principal component matrix P after dimensionality reduction. Then, introduce three new weight coefficients, calculate the difference ΔX between the postoperative data and the preoperative data, and weight it by the weight coefficient γ1 to enhance the influence of postoperative changes. Then calculate the standard deviation σ of the original data. XAnd adjust the influence of fluctuation amplitude on recovery progress calculation through weight coefficient γ2, and finally calculate the mean square sum of original data By adjusting the influence of the balance coefficient γ3 on the data mean, a numerical representation of the recovery progress of each physiological indicator is generated to help evaluate the patient's postoperative recovery and ensure the accuracy and rationality of the recovery progress calculation.

[0094] See also Figure 3 The multi-dimensional recovery status analysis module includes a data fusion submodule, a bottleneck analysis submodule, and a recovery status data generation submodule, wherein:

[0095] Data fusion submodule: Based on the recovery deviation data and postoperative real-time physiological data, by selecting key indicators such as heart rate, gait, and blood oxygen, each data is weighted and merged according to the set weights, and the fused data is processed using the weighted summation method to generate a fused data set;

[0096] Bottleneck analysis submodule: Based on the fused data set, it analyzes bottlenecks that occur during the recovery process, compares data based on the rate of change of recovery indicators, identifies stagnation points and abnormal fluctuations in the data, and generates bottleneck analysis results by identifying bottleneck areas.

[0097] Recovery status data generation submodule: Based on the bottleneck analysis results and combined with the key feature extraction method, various key indicators are aggregated and multi-dimensional recovery status data is generated using data integration technology;

[0098] Data fusion submodule: Based on the recovery deviation data and postoperative real-time physiological data, by selecting key indicators such as heart rate, gait, and blood oxygen, a weighted summation method is used. First, the weight of each physiological indicator is set to 0.4 for heart rate, 0.3 for gait, and 0.3 for blood oxygen. The weighted summation method is used to process each data item. The specific operation is to multiply the data of each physiological indicator by the corresponding weight and sum them to obtain the fused data, and finally generate a fused data set;

[0099] Bottleneck Analysis Submodule: Based on the fused data set, it analyzes bottlenecks that arise during the recovery process. Using a rate-of-change calculation method, it first calculates the rate of change of each key indicator by calculating the difference between the current data and the previous data. Then, it applies a stagnation point detection algorithm and sets a threshold for the rate of change. When the rate of change falls below the threshold twice consecutively, a bottleneck is considered to exist. Furthermore, an anomaly detection algorithm, using a distance-based outlier detection method, calculates the mean and standard deviation of each indicator, identifies abnormal fluctuations in the data, and ultimately generates bottleneck analysis results.

[0100] Recovery status data generation submodule: Based on the bottleneck analysis results and combined with the key feature extraction method, the key indicators are aggregated, and the feature selection algorithm is adopted. The chi-square test method is used to select features with a high correlation with the recovery progress. The significance level is set to 0.05, and the features with the strongest correlation with the recovery progress are selected. Further application of data integration technology and multidimensional data fusion method are used to integrate the selected key indicators according to dimensions such as time and stage to generate multidimensional recovery status data.

[0101] See also Figure 3 The phased recovery effect evaluation module includes an effect judgment submodule, a threshold comparison submodule, and a rehabilitation effect data generation submodule, among which:

[0102] Effect judgment submodule: Based on multi-dimensional recovery status data and patient rehabilitation stage data, combined with key indicators such as heart rate, blood oxygen, and gait, it evaluates the recovery effect of each stage, and generates a staged recovery effect judgment result by comparing the indicator values with the predetermined standards;

[0103] Threshold comparison submodule: Based on the staged recovery effect judgment results, key indicator values such as heart rate, blood oxygen, and gait are compared with the set thresholds item by item to calibrate potential recovery obstacles and generate threshold comparison results;

[0104] Rehabilitation effect data generation submodule: Based on the threshold comparison results, the phased recovery effect is comprehensively evaluated, and combined with the key indicator values, the patient's phased rehabilitation effect data is generated;

[0105] Effect judgment submodule: Based on multi-dimensional recovery status data and patient rehabilitation stage data, combined with key indicators such as heart rate, blood oxygen, and gait, a comparative analysis method is used. First, predetermined standards for each stage are set, with the heart rate standard set at 60-80bpm, the blood oxygen standard set at above 95%, and the gait standard set at a stride length of more than 50cm. Each indicator is compared item by item, and the difference calculation method is used to calculate the difference between the actual value and the predetermined standard. The threshold is set at a difference of more than 10%, which is considered non-compliant. Finally, the staged recovery effect judgment result is generated;

[0106] Threshold comparison submodule: Based on the staged recovery effect judgment results, the values of key indicators such as heart rate, blood oxygen, and gait are compared with the set thresholds one by one. First, the threshold of each indicator is set: the heart rate threshold is set to 65bpm, the blood oxygen threshold is set to 93%, and the gait threshold is set to 55cm. The actual value of each indicator is compared with the set threshold through the threshold comparison method. The anomaly detection algorithm is used to calculate the deviation between each indicator and the threshold to determine whether it exceeds the set threshold, calibrate potential recovery obstacles, and finally generate the threshold comparison results;

[0107] Rehabilitation effect data generation submodule: Based on the threshold comparison results, the stage-by-stage recovery effect is comprehensively evaluated. The weighted scoring method is used to set different weights for each indicator. The heart rate weight is set to 0.4, the blood oxygen weight is set to 0.3, and the gait weight is set to 0.3. The weighted summation method is used to combine the actual value of each indicator with the standard deviation to calculate the comprehensive score and finally generate the stage-by-stage rehabilitation effect data.

[0108] See also Figure 3 The final evaluation result output module includes a data integration submodule, a recovery progress calculation submodule, and an evaluation result generation submodule, among which:

[0109] Data integration submodule: Based on the staged rehabilitation effect data, multi-dimensional recovery status data and patient rehabilitation stage data, the data of each stage are summarized and the physiological data of each stage are integrated one by one through the data matching method to generate an integrated data set;

[0110] Recovery progress calculation submodule: Based on the integrated data set, the physiological indicators and exercise data are analyzed item by item to calculate the patient's recovery progress at each stage. The recovery progress of each indicator is compared with the predetermined standard to generate the recovery progress calculation results;

[0111] Evaluation result generation submodule: Based on the recovery progress calculation results, the recovery status of each stage is comprehensively analyzed to integrate the patient's overall recovery progress and merge the recovery data to generate complete evaluation result data;

[0112] Data integration submodule: Based on the phased rehabilitation effect data, multi-dimensional recovery status data, and patient rehabilitation stage data, a data matching algorithm is used. First, data matching is performed using the KNN algorithm with the K value set to 5. By calculating the similarity of the data points in each stage, the K nearest data points are selected and weighted averaged to generate the integrated data for each stage. Finally, the data for each stage are aligned and merged one by one to generate an integrated data set.

[0113] Recovery progress calculation submodule: Based on the integrated data set, physiological indicators and exercise data are analyzed item by item. The recovery progress of each indicator is calculated using the item-by-item difference calculation method. The physiological data of each stage is first compared with the predetermined standard. The standard values are set as heart rate 60-80bpm, blood oxygen value above 95%, and gait stride greater than 50cm. The progress of each stage is calculated using the difference analysis method. The weighted average method is used to set different weights for each indicator. The weight of heart rate is set to 0.4, blood oxygen is set to 0.3, and gait is set to 0.3. Finally, the recovery progress calculation result is generated;

[0114] Evaluation result generation submodule: Based on the recovery progress calculation results, through comprehensive analysis of the recovery status of each stage, the stage weighted synthesis method is adopted to assign different weights to the recovery progress of each stage. The acute option is reset to 0.5, the recovery option is reset to 0.3, and the consolidation option is reset to 0.2. The weighted summation method is used to combine the recovery data of each stage to generate the final comprehensive evaluation data. Finally, the recovery data of each stage are merged to generate complete evaluation result data.

[0115] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A system for evaluating and comparing data before and after rehabilitation of posterior fossa brain tumor surgery, characterized in that: The system comprises: Preoperative health benchmark analysis module: Based on the patient's preoperative heart rate, body temperature, blood oxygen data and imaging examination results, a health assessment is conducted to obtain the patient's preoperative physiological status data. The physiological parameters are sorted in combination with the medical history, key physiological indicators are extracted, and the difference with the baseline values is calculated to identify potential health problems and generate preoperative health benchmark data. Postoperative physiological status monitoring module: Based on the real-time collected patient physiological status data, sensors continuously monitor and compare data change trends, identify abnormal fluctuations, record gait, motor skills and sleep status, and compare them with the preoperative health baseline data to obtain real-time postoperative physiological data; Rehabilitation stage division module: Based on the real-time postoperative physiological data, the module uses a hidden Markov model to analyze time series data to identify the fluctuation range of key indicators such as heart rate and gait, judge the rehabilitation status, and divide the patient into acute phase, recovery phase, and consolidation phase. Each phase is divided by gait analysis and motor ability data to generate patient rehabilitation stage data; Recovery prediction and progress monitoring module: Based on the preoperative health baseline data and postoperative real-time physiological data, principal component analysis is used to calculate the recovery progress, compare the current progress with historical data, monitor heart rate and gait changes, detect whether the rehabilitation progress meets expectations, and conduct deviation assessment to generate recovery deviation data; Multi-dimensional recovery status analysis module: Based on the recovery deviation data and real-time postoperative physiological data, combined with various key indicators of the recovery period, heart rate, gait and blood oxygen data are calculated, integrated according to different weights, and bottlenecks in the recovery process are analyzed. Key features are extracted through data combination to generate multi-dimensional recovery status data; Phased recovery effect evaluation module: Based on the multi-dimensional recovery status data and the patient's rehabilitation stage data, combined with heart rate, blood oxygen and gait indicators, it determines the recovery effect of each stage, compares the key indicator values with the predetermined threshold value, evaluates the rehabilitation effect, identifies potential recovery obstacles, and generates phased rehabilitation effect data; Final evaluation result output module: Based on the staged rehabilitation effect data, multi-dimensional recovery status data and patient rehabilitation stage data, through data analysis, combined with physiological indicators and motion data, output the patient's overall recovery progress report, integrate the recovery status of different stages, and generate complete evaluation result data.

2. The posterior fossa brain tumor surgery recovery data evaluation and comparison system according to claim 1 is characterized in that: The preoperative health benchmark analysis module includes a physiological data collation submodule, a key physiological index extraction submodule, and a health benchmark data generation submodule, wherein: Physiological data organization submodule: Based on the patient's preoperative heart rate, body temperature, blood oxygen data and imaging examination results, combined with the patient's medical history, organizes physiological status data in a standardized and structured manner, and pre-processes each data item, such as removing noise, filling missing values, and unifying data formats, to generate preoperative physiological status data; Key physiological indicator extraction submodule: Based on the preoperative physiological status data, key physiological indicators such as heart rate, body temperature, and blood oxygen are selected, the indicator values are extracted item by item and standardized, and the stability and deviation of each indicator are further evaluated through data analysis to generate key physiological indicator data; Health baseline data generation submodule: Based on the key physiological indicator data, the difference between each indicator and the baseline value is calculated, the change trend is analyzed, and the patient's preoperative health baseline data is generated according to specific rules.

3. The system for evaluating and comparing pre- and post-operative data of posterior fossa brain tumor surgery according to claim 1, characterized in that: The postoperative physiological status monitoring module includes a real-time physiological data acquisition submodule, a data trend comparison submodule and a postoperative real-time data generation submodule, wherein: Real-time physiological data acquisition submodule: Based on the real-time collected patient physiological status data, including heart rate, body temperature, blood oxygen, gait, exercise ability and sleep status, data cleaning and preprocessing are performed to remove abnormal data and correct data deviations during the processing process to generate real-time postoperative physiological data; Data trend comparison submodule: Based on the real-time postoperative physiological data, continuously track and compare the change trend, identify the fluctuation points in the data, and use the difference comparison method to analyze, identify abnormal fluctuations and deviations, and generate data change trend analysis results; Postoperative real-time data generation submodule: compares and analyzes postoperative real-time physiological data with preoperative health baseline data, identifies the differences between postoperative data and baseline data, and generates the patient's real-time physiological status data based on the differences, thereby generating postoperative real-time physiological data.

4. The posterior fossa brain tumor surgery recovery data evaluation and comparison system according to claim 1, characterized in that: The rehabilitation stage division module includes a data fluctuation identification submodule, a stage division submodule, and a rehabilitation stage data generation submodule, wherein: Data fluctuation identification submodule: Based on the real-time postoperative physiological data, the hidden Markov model is used to identify the fluctuation range of key physiological indicators such as heart rate and gait through time series data analysis. The window sliding method is used to obtain the fluctuation data of each time period, and the data is cut and compared to generate the fluctuation range analysis results.

5. The system for evaluating and comparing pre- and post-operative data of posterior fossa brain tumor surgery according to claim 1, characterized in that: The rehabilitation stage division module includes a data fluctuation identification submodule, a stage division submodule, and a rehabilitation stage data generation submodule, wherein: Stage division submodule: Based on the fluctuation range analysis results, statistical analysis is performed on key indicators such as heart rate and gait to calculate the fluctuation range of each indicator. Combined with the clinically defined acute phase, recovery phase, and consolidation phase standards, the physiological state of each phase is divided to generate the rehabilitation stage division results; Rehabilitation stage data generation submodule: Based on the results of the rehabilitation stage division, combined with gait analysis and motor ability data, the key indicator values of each stage are counted through data aggregation methods, and each stage is further evaluated and data merged to generate patient rehabilitation stage data.

6. The system for evaluating and comparing pre- and post-operative data of posterior fossa brain tumor surgery according to claim 1, characterized in that: The recovery prediction and progress monitoring module includes a recovery progress calculation submodule, a progress comparison and analysis submodule, and a recovery deviation data generation submodule, wherein: Recovery progress calculation submodule: Based on the preoperative health baseline data and postoperative real-time physiological data, principal component analysis is used to compare the recovery status of each physiological indicator item by item. By calculating the change rate and difference of each indicator, a numerical representation of the recovery progress is obtained, and a recovery progress calculation result is generated; Progress comparison and analysis submodule: Based on the recovery progress calculation results, historical data is compared with existing data to monitor changes in key indicators such as heart rate and gait, identify deviations in the current recovery progress, and generate progress comparison and analysis results; Recovery deviation data generation submodule: Based on the progress comparison and analysis results, calculate the deviation between the current recovery progress and the expected progress, use the data difference calculation method to obtain recovery deviation data, and generate recovery deviation data.

7. The system for evaluating and comparing pre- and post-operative data of posterior fossa brain tumor surgery according to claim 1, characterized in that: The multi-dimensional recovery status analysis module includes a data fusion submodule, a bottleneck analysis submodule and a recovery status data generation submodule, wherein: Data fusion submodule: Based on the recovery deviation data and postoperative real-time physiological data, by selecting key indicators such as heart rate, gait, blood oxygen, etc., weighted merging of each data according to the set weights, and processing the fused data using the weighted summation method to generate a fused data set; Bottleneck analysis submodule: Based on the fused data set, analyzes bottlenecks that occur during the recovery process, compares data based on the rate of change of recovery indicators, identifies stagnation points and abnormal fluctuations in the data, and generates bottleneck analysis results by identifying bottleneck areas. Recovery status data generation submodule: Based on the bottleneck analysis results, combined with the key feature extraction method, various key indicators are aggregated, and multi-dimensional recovery status data is generated using data integration technology.

8. The system for evaluating and comparing pre- and post-operative data of posterior fossa brain tumor surgery according to claim 1, characterized in that: The staged recovery effect evaluation module includes an effect judgment submodule, a threshold comparison submodule, and a rehabilitation effect data generation submodule, wherein: Effect judgment submodule: Based on the multi-dimensional recovery status data and the patient's rehabilitation stage data, combined with key indicators such as heart rate, blood oxygen, and gait, the recovery effect of each stage is evaluated, and a judgment result of the staged recovery effect is generated by comparing the indicator values with the predetermined standards; Threshold comparison submodule: Based on the staged recovery effect judgment results, key indicator values such as heart rate, blood oxygen, gait, etc. are compared with the set thresholds one by one to calibrate potential recovery obstacles and generate threshold comparison results; Rehabilitation effect data generation submodule: Based on the threshold comparison results, the staged recovery effect is comprehensively evaluated, and combined with the key indicator values, the patient's staged rehabilitation effect data is generated.

9. The system for evaluating and comparing pre- and post-operative data of posterior fossa brain tumor surgery according to claim 1, characterized in that: The final evaluation result output module includes a data integration submodule, a recovery progress calculation submodule and an evaluation result generation submodule, wherein: Data integration submodule: Based on the staged rehabilitation effect data, multi-dimensional recovery status data and patient rehabilitation stage data, the data of each stage are summarized, and the physiological data of each stage are integrated one by one through the data matching method to generate an integrated data set; Recovery progress calculation submodule: Based on the integrated data set, the physiological indicators and exercise data are analyzed item by item, the recovery progress of the patient at each stage is calculated, and the recovery progress of each indicator is compared with the predetermined standard to generate the recovery progress calculation result; Evaluation result generation submodule: Based on the recovery progress calculation results, by comprehensively analyzing the recovery status of each stage, the patient's overall recovery progress is integrated, and the recovery data is merged to generate complete evaluation result data.

Images