An emergency postoperative colon cancer quantitative evaluation method, system and storage medium

Abstract

This invention relates to the field of postoperative assessment technology, and more particularly to a quantitative assessment method, system, and storage medium for emergency colon cancer surgery. It includes: collecting postoperative recovery data from patients, including preliminary indicators such as the time point of bowel function recovery, the range of bowel movement frequency fluctuations, and the intensity of overall physical activity; combining time series analysis and activity intensity information to identify potential high-risk time points; based on the comparison results of the high-risk time point set with preset thresholds, data fusion analysis of the interaction of preliminary indicators to construct a comprehensive recovery trend vector, and determining recovery pattern classification through pattern matching; based on the recovery pattern classification, constructing a dynamic adjustment matrix, adjusting the weights of each indicator, and calculating the overall recovery score through weighted average; and generating a personalized intervention adjustment sequence based on the relationship between the overall recovery score and the preset threshold. This invention solves the problems of strong subjectivity and poor real-time performance in postoperative recovery assessment, and achieves accurate and personalized recovery assessment and intervention adjustment.

Description

Technical Field

[0001] This invention relates to the field of postoperative assessment technology, and in particular to a quantitative assessment method, system and storage medium for emergency colon cancer surgery. Background Technology

[0002] Colorectal cancer is one of the most common malignant tumors worldwide, and postoperative recovery has a significant impact on patient prognosis. However, postoperative recovery is a complex physiological process influenced by various factors, such as the patient's physical condition, type of surgery, postoperative complications, and treatment. Traditional postoperative recovery assessment relies primarily on the physician's experience and clinical observation, which has inherent subjectivity and limitations. Therefore, achieving more accurate and objective postoperative recovery assessment has become an important issue in the current management of colorectal cancer surgery.

[0003] Currently, with the development of sensor technology and big data analytics, an increasing number of studies are attempting to monitor patients' postoperative recovery in real time using multidimensional data. In particular, by collecting physiological data such as bowel function, defecation frequency, and activity intensity, and combining this with advanced data analysis methods, more accurate assessments of patient recovery can be provided. However, existing technologies still have shortcomings in the accuracy and timeliness of data analysis, especially when processing large-scale patient data. Ensuring the accuracy and personalization of recovery assessments remains a pressing issue.

[0004] This invention provides a data-driven quantitative assessment method and system for emergency colon cancer surgery, effectively addressing the problems of insufficient assessment accuracy and inability to dynamically adjust in real time in existing technologies. By using sensor devices to collect postoperative recovery-related data in real time, combined with time series analysis, data fusion, and dynamic adjustment matrix technology, this invention can accurately identify potential high-risk time points, quantitatively assess the patient's recovery trend, and generate personalized intervention and adjustment plans, demonstrating significant clinical application value. Summary of the Invention

[0005] This invention provides a method, system, and storage medium for quantitative assessment after emergency colon cancer surgery, which aims to provide quantitative assessment and personalized intervention decisions for patients' postoperative recovery through accurate and objective data analysis.

[0006] In a first aspect, the present invention provides a method for quantitative assessment after emergency colon cancer surgery, the method comprising: Step S1: Collect postoperative recovery-related data of colon cancer patients through sensor devices to obtain a preliminary indicator set; perform time series analysis on the preliminary indicator set, and combine it with activity intensity information to determine a set of potential high-risk time nodes; Step S2: Based on the comparison between the number of nodes in the high-risk time node set and the preset number of nodes, the interaction of indicators in the preliminary indicator set is analyzed through data fusion to construct a comprehensive recovery trend vector; the comprehensive recovery trend vector is matched with a preset benchmark to determine the recovery mode classification. Step S3: Based on the interaction between the assessment indicators of the recovery mode classification, construct a dynamic adjustment matrix and dynamically adjust the weights of each indicator; use the dynamic adjustment matrix to perform weighted calculation on the preliminary indicator set to obtain the overall recovery score; Step S4: Determine the intervention adjustment sequence based on the relationship between the overall recovery score and the preset threshold.

[0007] As a preferred embodiment of the present invention, in step S1, a preliminary set of indicators is obtained, including: The system collects real-time data on the patient's intestinal function recovery time point and defecation frequency fluctuation range using sensor devices, while also recording information on the intensity level of overall physical activity. The collected raw data is processed using data cleaning techniques to remove noise and standardize the format, generating a preliminary indicator set that includes at least the intestinal function recovery time point, defecation frequency fluctuation range, and physical activity intensity level. Outliers in the preliminary indicator set are then removed.

[0008] As a preferred embodiment of the present invention, step S1, determining the set of potential high-risk time nodes, includes: For the data on intestinal function recovery time points and defecation frequency fluctuation range in the preliminary indicator set, time series dependency analysis is performed to extract fluctuation trend features; combined with the whole-body activity intensity level in the preliminary indicator set, risk node identification logic is applied to analyze the abnormal performance of each indicator in the time dimension; based on the analysis results, potential high-risk time nodes are screened to form the high-risk time node set; and the high-risk time node set is verified.

[0009] As a preferred embodiment of the present invention, step S2, constructing a comprehensive recovery trend vector, includes: If the number of nodes in the high-risk time node set exceeds the preset number of nodes, the correlation coefficient between each indicator in the preliminary indicator set is calculated through data fusion; based on the correlation coefficient and the preset data fusion influence factor, the interaction strength between each indicator is analyzed; based on the analysis results, the comprehensive recovery trend vector containing indicator priority ranking rules is constructed; the comprehensive recovery trend vector is standardized to obtain the standardized comprehensive recovery trend vector.

[0010] As a preferred embodiment of the present invention, step S2, determining the recovery mode classification, includes: Obtain a pre-established historical data comparison benchmark library and extract similarity vector results based on recovery pattern matching criteria; compare the comprehensive recovery trend vector with the similarity vector results and calculate the matching degree between the two; based on the matching degree results, determine the recovery pattern classification that is closest to the current patient.

[0011] As a preferred embodiment of the present invention, step S3, constructing the dynamic adjustment matrix, includes: Based on the recovery mode classification, the interaction between different indicators in the preliminary indicator set is quantitatively evaluated using the interactive influence dynamic matrix technology; combined with the indicator priority ranking rules and the preset weight update trigger conditions, the dynamic weight allocation scheme of each indicator is determined; based on the dynamic weight allocation scheme, the dynamic adjustment matrix containing the weight allocation of recovery indicators is constructed.

[0012] As a preferred embodiment of the present invention, step S3, obtaining the overall recovery score, includes: The preliminary indicator set is weighted using the dynamic adjustment matrix, and the weighted value of each indicator is calculated by combining the weight allocation of recovery indicators and the interaction quantification standard. The overall recovery score is generated by summing the weighted values. The overall recovery score is calibrated to ensure the reliability of the calculation results. The overall recovery score is then stored.

[0013] As a preferred embodiment of the present invention, step S4, determining the intervention adjustment sequence, includes: If the overall recovery score is lower than a preset warning threshold, optimization criteria related to personalized intervention parameter adjustments are extracted from a preset intervention rule base; the recovery characteristics of the current patient are analyzed in conjunction with a pre-established historical data comparison benchmark; based on the analysis results, a postoperative intervention adjustment sequence for the current patient is generated; the intervention adjustment sequence is verified to ensure its applicability and relevance.

[0014] Secondly, the present invention also provides a quantitative assessment system for emergency colon cancer surgery, used to implement the above-mentioned method, the system comprising: The data acquisition unit is used to collect postoperative recovery-related data of colorectal cancer patients through sensor devices to obtain a preliminary indicator set; and to perform time series analysis on the preliminary indicator set, combined with activity intensity information, to determine a set of potential high-risk time nodes. The data fusion unit is used to construct a comprehensive recovery trend vector by analyzing the interaction of indicators in the preliminary indicator set through data fusion analysis based on the comparison results between the number of nodes in the high-risk time node set and the preset number of nodes. The pattern acquisition unit is used to perform pattern matching between the comprehensive recovery trend vector and a preset benchmark to determine the recovery pattern classification. The matrix construction and analysis unit is used to construct a dynamic adjustment matrix based on the interaction between the assessment indicators according to the recovery mode classification, and dynamically adjust the weights of each indicator; the overall recovery score is obtained by weighting the preliminary indicator set through the dynamic adjustment matrix. An intervention adjustment unit is used to determine an intervention adjustment sequence based on the relationship between the overall recovery score and a preset threshold.

[0015] Thirdly, the present invention also provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the above-described method.

[0016] The beneficial effects of this invention are as follows: This invention collects various physiological data of patients after surgery using sensor devices, including the time point of bowel function recovery, the range of bowel frequency fluctuations, and the level of overall activity intensity. This data is processed through time series analysis and combined with activity intensity information to capture changes in the patient's recovery process in real time. By identifying potential high-risk time point sets, it can detect possible postoperative risks early. Based on the comparison between the number of high-risk time point sets and preset thresholds, data fusion analysis is performed to quantify the interactions between various indicators in the preliminary indicator set. By calculating correlation coefficients and combining them with preset influencing factors, a comprehensive recovery trend vector is constructed. Through pattern matching with preset benchmarks, the patient's recovery pattern classification is ultimately determined, ensuring accurate positioning of the patient's recovery trend. Furthermore, based on the determined recovery pattern classification, further... By assessing the interactions between various indicators, constructing a dynamic adjustment matrix, and dynamically adjusting the weights of each indicator, a weighted calculation is performed on the initial indicator set to obtain the patient's overall recovery score. This score accurately reflects the impact of each recovery indicator on the patient's overall recovery under different recovery modes, ensuring the reliability and accuracy of the assessment results. Based on the relationship between the calculated overall recovery score and a preset warning threshold, a personalized intervention adjustment sequence is generated to trigger corresponding intervention measures, ensuring timely and effective adjustment of the treatment plan to promote patient recovery. Through the synergy of the above technical solutions, not only can the postoperative recovery progress of patients be monitored in real time, but potential risks can also be identified in a timely manner, and precise intervention plans can be provided based on individual differences. This effectively improves the accuracy and efficiency of postoperative recovery management for colorectal cancer and reduces the occurrence of complications. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a quantitative assessment method for emergency colon cancer surgery, as illustrated in this embodiment. Figure 2 This is a schematic diagram illustrating the identification of abnormal behavior in the embodiment; Figure 3 This is a flowchart of the method for constructing the comprehensive recovery trend vector in the embodiment; Figure 4 This is a structural diagram of an emergency postoperative quantitative assessment system for colon cancer, as shown in the embodiment. Detailed Implementation

[0019] This invention provides a method, system, and storage medium for quantitative assessment after emergency colon cancer surgery. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0020] For ease of understanding, the specific process of the embodiments of the present invention will be described below, such as... Figure 1 As shown in this embodiment of the invention, a quantitative assessment method for emergency colon cancer surgery includes: Step S1: Collect postoperative recovery-related data of colon cancer patients through sensor devices to obtain a preliminary indicator set; perform time series analysis on the preliminary indicator set, and combine it with activity intensity information to determine a set of potential high-risk time nodes; In step S1, a preliminary set of indicators is obtained, including: The system collects real-time data on the patient's intestinal function recovery time point and defecation frequency fluctuation range using sensor devices, while also recording information on the intensity level of overall physical activity. The collected raw data is processed using data cleaning techniques to remove noise and standardize the format, generating a preliminary indicator set that includes at least the intestinal function recovery time point, defecation frequency fluctuation range, and physical activity intensity level. Outliers in the preliminary indicator set are then removed.

[0021] Specifically, the system collects postoperative recovery data from patients in real time using sensor devices. This includes data on the time points of bowel function recovery, the range of bowel movement frequency fluctuations, and the level of overall activity intensity. Wearable sensor devices deployed on the patient's abdomen, combined with accelerometers and bioelectrical impedance sensors, capture intestinal peristalsis signals to identify the time points of bowel function recovery, particularly when the peristalsis frequency returns to a preset normal threshold. Simultaneously, a smart bracelet records the patient's overall activity intensity in real time, categorizing activity into low, medium, and high levels based on step count and heart rate data, thus providing a quantitative basis for activity intensity analysis. Furthermore, data on the range of bowel movement frequency fluctuations is obtained through patient input or automatic sensor logs recording the daily number of bowel movements. The fluctuation range value is calculated statistically as the standard deviation, serving as the basis for further analysis.

[0022] Next, data cleaning techniques are used to denoise and standardize the format of the collected raw data, generating a preliminary set of organized indicators. Specifically, median filtering is used to remove noise, ensuring data consistency and accuracy, and the data format is standardized as a timestamp sequence. To further ensure data quality, the Z-score method is applied to identify and remove outliers. If the absolute value of the Z-score of a data point exceeds a corresponding set value, it is considered outlier and removed, thereby eliminating data deviations caused by sensor noise or other factors. For example, if extreme values ​​such as 20 times per day appear in the defecation frequency data, they can be removed using Z-score analysis, thus improving the subsequent accuracy. The reliability of the analysis is beneficial for accurately identifying recovery risks. After outlier removal, all data are normalized and scaled to the range of 0 to 1 for easier subsequent processing. Simultaneously, the correlation coefficient method is used to verify the relationships between different data points, ensuring data consistency and eliminating inconsistencies, thereby improving the reliability of the analysis. This technical solution, through precise multidimensional data acquisition and processing, combined with advanced data cleaning and outlier removal, lays the foundation for accurately identifying high-risk time points in the postoperative recovery process, ensuring the timeliness and accuracy of postoperative recovery assessment, and providing a reliable basis for subsequent personalized intervention adjustments.

[0023] Further, in step S1, the set of potential high-risk time nodes is determined, including: For the data on intestinal function recovery time points and defecation frequency fluctuation range in the preliminary indicator set, time series dependency analysis is performed to extract fluctuation trend features; combined with the whole-body activity intensity level in the preliminary indicator set, risk node identification logic is applied to analyze the abnormal performance of each indicator in the time dimension; based on the analysis results, potential high-risk time nodes are screened to form the high-risk time node set; and the high-risk time node set is verified.

[0024] Specifically, through precise time series analysis, risk node identification, and verification, potential high-risk time nodes were successfully identified, providing crucial information for intervention decisions during postoperative recovery. Specifically, time series dependency analysis was performed on the intestinal function recovery time points and defecation frequency fluctuation range data in the preliminary indicator set to extract fluctuation trend features. The continuity and data density of the intestinal function recovery time point sequence help reveal the regularity of postoperative intestinal recovery. For example, intestinal recovery may exhibit different fluctuation patterns at different postoperative time points, and these patterns can be revealed through time series analysis, providing a basis for subsequent intervention decisions. Therefore, time series dependency analysis was first performed on the intestinal function recovery time point sequence in the preliminary indicator set. This involved calculating the autocorrelation function to identify periodic patterns in the data sequence, and then quantifying the dependency between time points using the autocorrelation coefficient under lag order, thereby extracting fluctuation trend features. For the defecation frequency fluctuation range data, a moving average model was used to smooth short-term fluctuations, extracting long-term trend features, and calculating the mean and standard deviation of the above fluctuation trend features to form a fluctuation trend feature vector. This ensures a smooth transition from the original data state to feature extraction, contributing to the accuracy and stability of subsequent analysis.

[0025] Furthermore, combining the overall activity intensity level information from the preliminary indicator set, and applying risk node identification logic, the abnormal performance of each indicator over time was analyzed. First, the fluctuation trend features extracted from the time series analysis were standardized, normalizing the feature values ​​of the intestinal function recovery time point and the fluctuation range of defecation frequency to the same dimension, eliminating dimensional differences between data. The overall activity intensity level was introduced as an auxiliary variable, specifically calculating the linear relationship between activity intensity and fluctuation trend features using the Pearson correlation coefficient formula. If the absolute value of the correlation coefficient exceeds a preset threshold, the two are considered strongly correlated. Then, a threshold-based risk node identification logic was used to scan data points over time. An abnormal performance threshold is set at 1.5 times the standard deviation of the fluctuation trend. When the indicator value at a certain time point exceeds this threshold and the activity intensity is at a low level, that time point is marked as an abnormal performance. Specifically, a time series matrix is ​​constructed, where rows represent time points and columns represent different indicators. The deviation value (actual value minus expected trend value) is calculated row by row. This is achieved by analyzing historical data of patients during similar postoperative recovery processes, calculating the historical average trend value of each indicator at the corresponding time points, which is then used as the expected trend value. If the deviation value positively exceeds the set threshold and the activity intensity decreases, it is considered an abnormal performance. This ensures timely identification of potential risks during postoperative recovery, especially in the early stages of recovery. Figure 2 As shown, for patient A after emergency colon cancer surgery, the time point data of bowel function recovery showed a delay on the 3rd day after surgery, with the frequency of defecation fluctuating from 1 to 3 times per day, and the fluctuation trend showing a downward trend. Combined with patient A's general activity intensity being at a moderate level, the correlation coefficient between activity intensity level and fluctuation trend was 0.6, indicating a strong correlation. Scanning the time series revealed that the deviation value on the 4th day exceeded the abnormal threshold and was marked as an abnormal manifestation. The consideration of the recovery scenario of moderate activity intensity is helpful in identifying progressive abnormalities.

[0026] Based on the analysis results, the most risky time points are selected, forming a potential high-risk time point set. Specifically, the abnormal manifestations identified in the above steps are sorted according to the magnitude of the deviation value, and the top 10% of the time points are selected as high-risk time points, thus ensuring that the selected time points represent the most critical risk period in the recovery process. The high-risk time points are verified using a cross-validation method to ensure the accuracy of the time point identification. The cross-validation process divides the high-risk time point set into a training subset and a test subset. The training subset is fitted using a logistic regression model, and the accuracy of the test subset is calculated. If the accuracy is higher than a preset standard, such as 0.85, the time point identification process is considered to have passed the verification. If the verification fails, the threshold in the risk point identification logic is adjusted, for example, by reducing the threshold from 1.5 times to 1.2 times, and the screening and verification are repeated until a high-risk time point set that meets the accuracy requirements is obtained. Through the above technical solution, a verified high-risk time point set is finally obtained, which serves as the basis for subsequent recovery trend analysis and personalized intervention decisions, ensuring that high-risk time points can be identified and intervened in a timely and accurate manner during the postoperative recovery process, thereby significantly improving the patient's recovery effect and avoiding the occurrence of complications.

[0027] Step S2: Based on the comparison between the number of nodes in the high-risk time node set and the preset number of nodes, the interaction of indicators in the preliminary indicator set is analyzed through data fusion to construct a comprehensive recovery trend vector; the comprehensive recovery trend vector is matched with a preset benchmark to determine the recovery mode classification. In step S2, a comprehensive recovery trend vector is constructed, such as... Figure 3 As shown, it includes: The number of nodes in the set of high-risk time nodes is compared with the preset number of nodes. If the number of nodes exceeds the preset threshold, the correlation coefficient between each indicator in the preliminary indicator set is calculated through data fusion. Based on the correlation coefficient and the preset data fusion influence factor, the interaction strength between each indicator is analyzed. Based on the analysis results, a comprehensive recovery trend vector containing indicator priority ranking rules is constructed; the comprehensive recovery trend vector is then standardized to obtain a standardized comprehensive recovery trend vector.

[0028] Specifically, by comparing the number of nodes in the high-risk time node set with a preset number of nodes, when the number of nodes exceeds the preset number, the data fusion module further analyzes the data in the preliminary indicator set. Specifically, the number of nodes in the high-risk time node set is checked first. If the number of nodes exceeds a set threshold, such as 5, the correlation coefficient between each indicator is calculated. This mainly includes key indicators such as the time point of intestinal function recovery and the range of bowel frequency fluctuations. The Pearson correlation coefficient formula is used to calculate the correlation between the indicators. For example, in the recovery scenario after emergency colon cancer surgery, by calculating the correlation between the time point of intestinal function recovery and the range of bowel frequency fluctuations, a high correlation coefficient is obtained, indicating that there is a strong positive correlation between the two.

[0029] Based on the aforementioned correlation coefficients, a pre-defined data fusion influencing factor is further introduced, such as a weight of 0.5. By weighting the correlation coefficients between various indicators, the interaction strength between indicators is quantified. For example, if the correlation coefficient between whole-body activity intensity and defecation frequency is 0.7, then multiplying it by the pre-defined influencing factor of 0.6 yields an interaction strength value of 0.42, representing the interaction strength between indicators, which is helpful in identifying recovery risks. After performing similar analysis on all indicators, an interaction strength table is generated, recording the interaction strength of each pair of indicators. The degree of mutual influence between indicators is quantified by a weighted summation method, allowing for personalized adjustments for different patient groups. For example, the influencing factor can be adjusted separately for middle-aged and elderly patients, thereby improving the accuracy of the analysis and ensuring the accuracy of the recovery trend assessment.

[0030] Based on the analysis results, a comprehensive recovery trend vector was constructed, incorporating priority ranking rules for indicators. This involved sorting the data in the interaction intensity table according to the strength of their interactions, creating a set of priority-ranked indicators to further enhance the weight of key indicators. For example, in postoperative recovery monitoring, key factors influencing recovery were prioritized. If the interaction between recovery time and activity intensity was strong, recovery time was ranked as priority 1, followed by activity intensity and defecation frequency as priorities 2 and 3, respectively. Furthermore, by combining correlation coefficients, the above indicators were prioritized into a multi-dimensional vector, such as [Recovery Time: Priority 1, Defecation Frequency: Priority 2, Activity Intensity: Priority 3], further integrating the priority of each indicator and its corresponding correlation data to ensure the model accurately reflects the impact of each indicator on the recovery trend.

[0031] Finally, to ensure the applicability and accuracy of the comprehensive recovery trend vector, standardization was used to normalize the constructed vector. For example, the min-max standardization method was used to scale the values ​​of all vector elements to between 0 and 1, eliminating dimensional differences between different indicators and ensuring that the vector is compatible with the pattern matching of the benchmark library. The standardized comprehensive recovery trend vector can be used for comparison with historical data in subsequent recovery assessments and pattern matching, thereby improving the accuracy of the overall recovery score assessment and ensuring the timeliness and effectiveness of postoperative interventions. Through data processing, analysis, and standardization steps, combined with personalized patient information, this method can provide accurate and quantifiable assessments of the recovery process after emergency colon cancer surgery, providing data support for subsequent intervention decisions.

[0032] Further, in step S2, the recovery mode classification is determined, including: Obtain a pre-established historical data comparison benchmark library and extract similarity vector results based on recovery pattern matching criteria; compare the comprehensive recovery trend vector with the similarity vector results and calculate the matching degree between the two; based on the matching degree results, determine the recovery pattern classification that is closest to the current patient.

[0033] Specifically, in this embodiment, by matching historical data with a benchmark database, the patient's recovery pattern classification is accurately determined, providing data support for personalized recovery intervention. First, a pre-established historical data benchmark database is queried. This database stores multiple historical patients' recovery trend vectors, each containing quantitative data such as intestinal function recovery time points, bowel movement frequency fluctuation ranges, and overall activity intensity levels. This historical data provides important reference for subsequent analysis. Based on the current patient's information and preliminary indicator set (including patient gender, age, and surgical type), a subset of historical vectors similar to the current patient's age and surgical type is selected. From this subset, similar vectors based on recovery pattern matching are extracted. As a result, the aforementioned similar vectors were selected based on the similarity threshold of time series dependency analysis to ensure a high degree of similarity to the current patient's recovery process. After the similar vectors were extracted, the comprehensive recovery trend vector was compared with the historical similar vectors using the Euclidean distance formula to calculate the distance between them. The Euclidean distance was used as a standard to measure the similarity between vectors and to quantify the differences in recovery patterns among different patients. The smaller the distance, the closer the current patient's recovery trend is to a certain recovery pattern in the historical data. Then, the calculated distance values ​​were normalized and converted into a matching score. The higher the matching score, the stronger the similarity, which helps to identify the historical pattern that best matches the current patient's recovery characteristics.

[0034] In certain specific scenarios, such as those dominated by the time point of bowel function recovery or those dominated by the range of bowel frequency fluctuations, the parameters for calculating the matching degree are adjusted. Specifically, when the recovery time point has a significant impact on the recovery process, the weight of the recovery time point indicator is increased; while when the range of bowel frequency fluctuations has a significant impact on recovery, the weight of bowel frequency fluctuations is increased accordingly. Weighted Euclidean distance is used when calculating the matching degree, thus ensuring the accuracy of recovery pattern matching in different scenarios and adapting to the personalized recovery process of different patients. Based on the matching degree results calculated above, the recovery pattern classification that is closest to the current patient (i.e., the one with the shortest weighted Euclidean distance) is further determined. The recovery pattern classification corresponding to the most similar vector result with the highest matching degree is selected, such as the rapid recovery model. The system identifies the current recovery pattern of each patient, either in a fixed or delayed recovery mode. This classification provides a basis for subsequent recovery assessment and intervention. Finally, the determined recovery pattern classification results are recorded in the patient's recovery file, providing data support for subsequent recovery trend analysis and dynamic adjustment matrix weight updates. This data storage and update process ensures that the recovery pattern classification can be dynamically adjusted as the patient's recovery progresses, ensuring the real-time nature and targeted nature of the intervention plan, further improving the effectiveness of postoperative recovery management. This technical solution, combined with historical data comparison and personalized adjustments, can provide accurate recovery pattern classification for patients after emergency colon cancer surgery, thereby supporting personalized recovery assessment and intervention decisions, improving recovery outcomes, and reducing the risk of complications.

[0035] Step S3: Based on the interaction between the assessment indicators of the recovery mode classification, construct a dynamic adjustment matrix; use the dynamic adjustment matrix to perform weighted calculation on the preliminary indicator set to obtain the overall recovery score; In step S3, constructing the dynamic adjustment matrix includes: Based on the recovery mode classification, the interaction between different indicators in the preliminary indicator set is quantitatively evaluated using the interactive influence dynamic matrix technology; combined with the indicator priority ranking rules and the preset weight update trigger conditions, the dynamic weight allocation scheme of each indicator is determined; based on the dynamic weight allocation scheme, the dynamic adjustment matrix containing the weight allocation of recovery indicators is constructed.

[0036] Specifically, by constructing a dynamic adjustment matrix, the patient's recovery process is accurately assessed and personalized for intervention. This involves classifying recovery patterns and using interactive dynamic matrix technology to quantify the interactions between different indicators in the initial indicator set, ensuring that the recovery assessment can adapt to the individualized needs of different patients. First, the current patient's recovery type is extracted from the recovery pattern classification, such as a rapid recovery pattern or a delayed recovery pattern, and the priority of relevant data is determined based on this pattern. For example, in a rapid recovery pattern, the recovery time point may be the most critical indicator, while in a delayed recovery pattern, fluctuations in bowel movement frequency may have a greater impact on recovery. Then, the bowel function recovery time point, the range of bowel movement frequency fluctuations, and the level of overall activity intensity in the initial indicator set are used as row and column elements of the matrix. The influence coefficient between each indicator is calculated, typically using the Pearson correlation coefficient to quantify the linear correlation between indicators. The Pearson correlation coefficient is calculated by dividing the covariance by the product of the standard deviations of the two variables; the resulting value ranges between -1 and 1, with positive values ​​indicating positive correlation and negative values ​​indicating negative correlation. This technique reveals the dependencies between various indicators. For example, the correlation coefficient between the time point of intestinal function recovery and the range of bowel movement frequency fluctuations might be 0.7, indicating a moderate positive correlation. A dynamic matrix is ​​constructed based on these correlation coefficients, and the update frequency is adjusted by combining recovery mode classifications. For instance, in a rapid recovery mode, the matrix is ​​updated hourly to capture short-term fluctuations during the recovery process, thus accurately reflecting the interactions between indicators.

[0037] To ensure the accuracy of the recovery trend assessment, a dynamic weight allocation scheme for each indicator is determined by combining indicator priority ranking rules and preset weight update trigger conditions. This scheme is adjusted according to the importance of indicators in different scenarios. For example, during the recovery process after emergency colon cancer surgery, the time point for bowel function recovery may be assigned a higher initial weight, while the range of bowel frequency fluctuations and activity intensity receive corresponding weights based on the actual situation. The weight update trigger conditions include a recovery score falling below a set score or an increase in the number of high-risk time points. When these conditions are met, the weight allocation is dynamically adjusted; for example, if the trigger condition occurs, the weight of the recovery time point may be adjusted from 0.5 to 0.6, and the weight of the bowel frequency fluctuation range may be adjusted accordingly. The weight of the range of motion may be adjusted from 0.3 to 0.2, and the weight of the activity intensity may be adjusted from 0.2 to 0.1, thereby enhancing the focus on key recovery indicators and ensuring that the recovery assessment can reflect the patient's recovery progress in a timely manner. Based on the above dynamic weight allocation scheme, a dynamic adjustment matrix containing the weight allocation of recovery indicators is constructed. By embedding the above weight values ​​into the diagonal elements of the matrix, a multidimensional matrix is ​​formed, where each element in the matrix represents the strength of the interaction between different indicators. For example, the diagonal elements of the matrix may be 0.6, 0.3, and 0.1, corresponding to the weights of the recovery time point, the range of bowel movement frequency fluctuations, and the activity intensity level, respectively, ensuring that the matrix can effectively reflect the importance of different indicators and their interrelationships.

[0038] Further, in step S3, the overall recovery score is obtained, including: The preliminary indicator set is weighted using the dynamic adjustment matrix, and the weighted value of each indicator is calculated by combining the weight allocation of recovery indicators and the interaction quantification standard. The overall recovery score is generated by summing the weighted values. The overall recovery score is calibrated to ensure the reliability of the calculation results. The overall recovery score is stored as the basis for subsequent intervention decisions.

[0039] Specifically, in this embodiment, a dynamic adjustment matrix is ​​used to accurately assess the patient's recovery process to ensure the accuracy and timeliness of personalized intervention. Based on recovery mode classification, the dynamic adjustment matrix applies weighted calculation logic to each indicator in the preliminary indicator set. The weighted value of each indicator is calculated by combining the recovery indicator weight allocation and interaction quantification standards. In the preliminary indicator set, the intestinal function recovery time point, the range of bowel movement frequency fluctuations, and the level of overall activity intensity are assigned initial weight values. The recovery time point may be assigned a higher weight value as a core indicator to highlight its dominant role in postoperative recovery; for example, the recovery time point may be assigned a weight of 0.4 to ensure its important role in the weighted calculation.

[0040] Subsequently, a quantitative standard for interaction was introduced to assess the mutual influence between different indicators. For example, by calculating the correlation coefficient between the range of defecation frequency fluctuations and the level of whole-body activity intensity, if the correlation coefficient exceeds 0.5, it indicates a moderate positive correlation. The weight allocation was then adjusted to increase the weight value of the range of defecation frequency fluctuations, reflecting the amplifying effect of activity intensity on the recovery of intestinal function. Through the above weighted adjustment, the interaction between different recovery indicators and their comprehensive impact on postoperative recovery can be more accurately reflected.

[0041] Next, a weighted calculation logic is applied to multiply the raw data values ​​of all indicators to obtain the weighted value of each indicator, and a quantified indicator sequence is generated based on the weights. For example, the weighted value at the recovery time point is its original value multiplied by the adjusted weight, forming the final weighted data. This method quantifies the contribution of each indicator to the overall recovery process, ensuring that the combined impact of each key indicator is fully considered.

[0042] For different patient groups, especially elderly patients, the implementation plan can be extended to adjust the weighted values ​​for specific age groups. For example, in the recovery assessment of elderly patients, if the patient is over 65 years old, the initial weight of the activity intensity level may be reduced to avoid overestimating the recovery potential. In addition, for patients with large fluctuations in bowel movement frequency, interaction analysis may incorporate time-series dependency analysis. If the fluctuation range exceeds the normal threshold, the interaction coefficient is increased, and the weighted values ​​are further adjusted to ensure a more conservative and personalized weighted calculation. Based on the above weighted values, an overall recovery score is generated through summation. Specifically, the various weighted values ​​are summarized, for example, recovery time... The weighted values ​​of the points, the weighted values ​​of the range of bowel movement frequency fluctuations, and the weighted values ​​of activity intensity are added together to obtain a total score. The overall recovery score reflects the comprehensive performance of various indicators during the patient's postoperative recovery process. To ensure the reliability of the calculation results, the overall recovery score is calibrated. The calibration process first compares the score with the similarity vector results in the historical data benchmark library. If the deviation between the two exceeds a predetermined threshold, such as 5%, a calibration factor is applied to adjust the score. The calibration factor may be dynamically adjusted according to changes in the recovery pattern. For example, when the recovery pattern changes, the calibration factor may be different to ensure that it adapts to the patient's personalized recovery process.

[0043] The calibrated overall recovery score is stored as the basis for subsequent intervention decisions. The specific calibrated score is stored in a preset intervention rule base. During the patient's recovery process, if the overall recovery score is lower than a preset warning threshold, such as 0.7 (where the maximum overall recovery score is 1, indicating complete recovery), an intervention adjustment sequence is triggered. The correlation between the stored overall recovery score and the comprehensive recovery trend vector allows subsequent interventions to be dynamically adjusted in real time according to the patient's recovery status, thereby improving the efficiency and accuracy of postoperative recovery monitoring and ensuring that the patient's recovery path can be continuously optimized. This technical solution, through a comprehensive dynamic adjustment matrix method, enables precise assessment and personalized intervention of the recovery process of patients after emergency colon cancer surgery, greatly improving the effectiveness of postoperative recovery management and reducing the risk of complications.

[0044] Step S4: Determine the intervention adjustment sequence based on the relationship between the overall recovery score and the preset threshold; specifically, this includes: if the overall recovery score is lower than the preset warning threshold, extract optimization criteria related to personalized intervention parameter adjustment from the preset intervention rule base; analyze the current patient's recovery characteristics by combining the pre-established historical data comparison benchmark; generate a postoperative intervention adjustment sequence for the current patient based on the analysis results, and verify the intervention adjustment sequence.

[0045] Specifically, in this embodiment, a personalized intervention adjustment sequence is determined by comprehensively analyzing the relationship between the patient's overall recovery score and a preset warning threshold, in order to achieve more precise postoperative recovery management. Specifically, if the overall recovery score is lower than the preset warning threshold, such as 70 points, optimization criteria related to personalized intervention parameter adjustments are extracted from a preset intervention rule base. This intervention rule base includes various intervention parameters, such as rules for dietary adjustments and activity guidance. Based on the patient's current recovery status, corresponding intervention optimization criteria are extracted to ensure the targeted and personalized nature of the intervention plan. Combining this with a pre-established historical data comparison benchmark, the current patient's recovery characteristics are analyzed. This benchmark base stores a large amount of historical postoperative recovery data from colorectal cancer patients, including multiple indicators such as intestinal function recovery time points and bowel movement frequency fluctuation ranges. By loading the historical data comparison benchmark, a matching analysis is performed with the current patient's preliminary recovery indicator set. Specifically, similarity is calculated to assess the difference between the current patient's recovery process and historical patient data. For example, Euclidean distance is used to measure the deviation between the current patient and the historical benchmark, thereby determining the relative position of the current patient's recovery status.

[0046] Based on the analysis results, a postoperative intervention adjustment sequence for the current patient is generated. This involves first constructing a preliminary framework for the intervention sequence based on recovery characteristics extracted from historical benchmarks and optimization criteria, and then applying weighted updates to trigger conditions. For example, if the patient's recovery pattern shows delayed recovery, dietary intervention parameters are adjusted first. If the patient's bowel movement frequency fluctuates significantly, the monitoring frequency of intestinal function is increased, and the weight allocation of recovery indicators in the dynamic adjustment matrix is ​​taken into account. For instance, in cases of significant fluctuations in bowel movement frequency, dietary adjustments or increased drug treatment frequency are prioritized to reduce the pressure on intestinal function.

[0047] Furthermore, a complete intervention adjustment sequence is generated. This sequence may include adjustments to the daily activity intensity level or adjustments to the frequency of bowel function monitoring based on the patient's recovery. For example, for patients with low activity levels, the intervention sequence might include increasing mild activity on the first day and adjusting bowel frequency with medication on the second day, ensuring the intervention is gradual and safe to accelerate bowel function recovery. To ensure the applicability and effectiveness of the generated intervention adjustment sequence, it is validated. Specifically, the application effect of the sequence on the current patient's recovery trend vector is simulated, and the expected overall recovery score is calculated using a data fusion module. The sequence is then checked to ensure it meets personalized parameter requirements, such as verifying whether it can avoid high-risk situations. The overlapping of nodes ensures that interventions can be initiated promptly at critical moments. If the verification results meet the requirements, the intervention sequence is considered applicable; if not, the intervention sequence is iteratively adjusted to further optimize its content and parameters. For example, in some scenarios, if the patient's recovery pattern shows a slow recovery type, the generated intervention sequence may emphasize long-term intervention, such as weekly dynamic adjustment matrix assessment, to ensure that the intervention measures can be continuously optimized and reduce the risk of postoperative complications. Through the above iterative and optimization process, the final generated intervention adjustment sequence can not only improve the patient's overall recovery efficiency, but also be personalized according to the actual situation, thereby providing more accurate intervention decisions for postoperative recovery.

[0048] This invention also provides a quantitative assessment system for emergency colon cancer surgery, used to implement the above-mentioned method, such as... Figure 4 As shown, the system includes: The data acquisition unit is used to collect postoperative recovery-related data of colorectal cancer patients through sensor devices to obtain a preliminary indicator set; and to perform time series analysis on the preliminary indicator set, combined with activity intensity information, to determine a set of potential high-risk time nodes. The data fusion unit is used to construct a comprehensive recovery trend vector by analyzing the interaction of indicators in the preliminary indicator set through data fusion analysis based on the comparison results between the number of nodes in the high-risk time node set and the preset number of nodes. The pattern acquisition unit is used to perform pattern matching between the comprehensive recovery trend vector and a preset benchmark to determine the recovery pattern classification. The matrix construction and analysis unit is used to construct a dynamic adjustment matrix based on the interaction between the assessment indicators according to the recovery mode classification, and dynamically adjust the weights of each indicator; the overall recovery score is obtained by weighting the preliminary indicator set through the dynamic adjustment matrix. An intervention adjustment unit is used to determine an intervention adjustment sequence based on the relationship between the overall recovery score and a preset threshold.

[0049] The present invention also provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the above-described method.

[0050] In summary, this invention collects various physiological data of patients after surgery using sensor devices, including the time point of bowel function recovery, the range of bowel movement frequency fluctuations, and the level of overall activity intensity. This data is processed through time series analysis and combined with activity intensity information to capture changes in the patient's recovery process in real time. By identifying potential high-risk time point sets, it can detect possible postoperative risks early. Based on the comparison between the number of high-risk time point sets and preset thresholds, data fusion analysis is performed to quantify the interactions between various indicators in the preliminary indicator set. By calculating correlation coefficients and combining them with preset influencing factors, a comprehensive recovery trend vector is constructed. Through pattern matching with preset benchmarks, the patient's recovery pattern classification is ultimately determined, ensuring accurate positioning of the patient's recovery trend. Furthermore, based on the determined recovery pattern classification, further... This approach assesses the interactions between various indicators, constructs a dynamic adjustment matrix, and dynamically adjusts the weights of each indicator. By weighting the initial set of indicators, a patient's overall recovery score is obtained, accurately reflecting the impact of each recovery indicator on the patient's overall recovery under different recovery modes, ensuring the reliability and accuracy of the assessment results. Based on the relationship between the calculated overall recovery score and a preset warning threshold, a personalized intervention adjustment sequence is generated to trigger corresponding intervention measures, ensuring timely and effective adjustment of the treatment plan to promote patient recovery. Through the synergy of these technical solutions, not only can the postoperative recovery progress of patients be monitored in real time, but potential risks can also be identified in a timely manner, and precise intervention plans can be provided based on individual differences. This effectively improves the accuracy and efficiency of postoperative recovery management for colorectal cancer and reduces the occurrence of complications.

[0051] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0052] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0053] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for quantitative assessment of post-emergency colon cancer surgery, characterized by, The method includes: Step S1: Collect postoperative recovery-related data of colorectal cancer patients through sensor devices to obtain a preliminary indicator set; perform time series analysis on the preliminary indicator set, and combine it with activity intensity information to determine a set of potential high-risk time nodes; Step S2: Based on the comparison between the number of nodes in the high-risk time node set and the preset number of nodes, the interaction of indicators in the preliminary indicator set is analyzed through data fusion to construct a comprehensive recovery trend vector; the comprehensive recovery trend vector is matched with a preset benchmark to determine the recovery mode classification. Step S3: Based on the interaction between the assessment indicators of the recovery mode classification, construct a dynamic adjustment matrix and dynamically adjust the weights of each indicator; use the dynamic adjustment matrix to perform weighted calculation on the preliminary indicator set to obtain the overall recovery score; Step S4: Determine the intervention adjustment sequence based on the relationship between the overall recovery score and the preset threshold.

2. The method of claim 1, wherein, In step S1, a preliminary set of indicators is obtained, including: The system collects real-time data on the patient's intestinal function recovery time point and defecation frequency fluctuation range using sensor devices, while also recording information on the intensity level of overall physical activity. The collected raw data is processed using data cleaning techniques to remove noise and standardize the format, generating a preliminary indicator set that includes at least the intestinal function recovery time point, defecation frequency fluctuation range, and physical activity intensity level. Outliers in the preliminary indicator set are then removed.

3. The method of claim 2, wherein, In step S1, the set of potential high-risk time nodes is determined, including: For the intestinal function recovery time point and defecation frequency fluctuation range data in the preliminary indicator set, time series dependency analysis is performed to extract fluctuation trend features; combined with the whole body activity intensity level in the preliminary indicator set, abnormal performance of each indicator in the time dimension is analyzed through risk node identification; based on the analysis results, potential high-risk time nodes are screened to form the high-risk time node set; and the high-risk time node set is verified.

4. The method of claim 1, wherein, In step S2, a comprehensive recovery trend vector is constructed, including: If the number of nodes in the high-risk time node set exceeds the preset number of nodes, the correlation coefficient between each indicator in the preliminary indicator set is calculated through data fusion; based on the correlation coefficient and the preset data fusion influence factor, the interaction strength between each indicator is analyzed; based on the analysis results, the comprehensive recovery trend vector containing indicator priority ranking rules is constructed; the comprehensive recovery trend vector is standardized to obtain the standardized comprehensive recovery trend vector.

5. The method of claim 4, wherein, In step S2, the recovery mode classification is determined, including: Obtain a pre-established historical data comparison benchmark library and extract similarity vector results based on recovery pattern matching criteria; compare the comprehensive recovery trend vector with the similarity vector results and calculate the matching degree between the two; based on the matching degree results, determine the recovery pattern classification that is closest to the current patient.

6. The method of claim 1, wherein, In step S3, the dynamic adjustment matrix is ​​constructed, including: Based on the recovery mode classification, the interaction between different indicators in the preliminary indicator set is quantitatively evaluated using the interactive influence dynamic matrix technology; combined with the indicator priority ranking rules and the preset weight update trigger conditions, the dynamic weight allocation scheme of each indicator is determined; based on the dynamic weight allocation scheme, the dynamic adjustment matrix containing the weight allocation of recovery indicators is constructed.

7. The method as described in claim 1, characterized in that, In step S3, the overall recovery score is obtained, including: The preliminary indicator set is weighted using the dynamic adjustment matrix, and the weighted value of each indicator is calculated by combining the weight allocation of recovery indicators and the interaction quantification standard. The overall recovery score is generated by summing the weighted values. The overall recovery score is then calibrated and stored.

8. The method as described in claim 1, characterized in that, In step S4, the intervention adjustment sequence is determined, including: If the overall recovery score is lower than a preset warning threshold, optimization criteria related to personalized intervention parameter adjustments are extracted from a preset intervention rule base; the recovery characteristics of the current patient are analyzed in conjunction with a pre-established historical data comparison benchmark; based on the analysis results, a postoperative intervention adjustment sequence for the current patient is generated, and the intervention adjustment sequence is verified.

9. A quantitative assessment system for emergency colon cancer surgery, used to implement the method as described in any one of claims 1-8, characterized in that, The system includes: The data acquisition unit is used to collect postoperative recovery-related data of colorectal cancer patients through sensor devices to obtain a preliminary indicator set; and to perform time series analysis on the preliminary indicator set, combined with activity intensity information, to determine a set of potential high-risk time nodes. The data fusion unit is used to construct a comprehensive recovery trend vector by analyzing the interaction of indicators in the preliminary indicator set through data fusion analysis based on the comparison results between the number of nodes in the high-risk time node set and the preset number of nodes. The pattern determination unit is used to perform pattern matching between the comprehensive recovery trend vector and a preset benchmark to determine the recovery pattern classification. The matrix construction and analysis unit is used to construct a dynamic adjustment matrix based on the interaction between the assessment indicators according to the recovery mode classification, and dynamically adjust the weights of each indicator; the overall recovery score is obtained by weighting the preliminary indicator set through the dynamic adjustment matrix. An intervention adjustment unit is used to determine an intervention adjustment sequence based on the relationship between the overall recovery score and a preset threshold.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the method as described in any one of claims 1-8.

Images