Long-term intelligent follow-up system for pediatric hypospadias patients

Abstract

This invention relates to the field of follow-up management technology, specifically to a long-term intelligent follow-up system for pediatric hypospadias patients. The system includes: a data collection module for acquiring the patient's surgical history, follow-up records, and follow-up addresses; a follow-up necessity analysis module for analyzing the necessity of follow-up based on the patient's historical recovery and surgical history; a follow-up rejection analysis module for clustering by follow-up addresses and planning optimal paths, analyzing the patient's follow-up necessity and their influence on the optimal path, obtaining potential rejection indicators, and combining these with the patient's distribution among other clusters to obtain follow-up rejection indices; and a follow-up planning module for incorporating the impact of the rejection indices on clustering, obtaining updated clusters for path planning and follow-up. This invention, when planning follow-up through clustering, analyzes the patient's follow-up necessity and the impact of their participation in cluster planning, adjusting the division of follow-up areas to improve follow-up efficiency.

Description

Technical Field

[0001] This invention relates to the field of management and follow-up technology, specifically to a long-term intelligent follow-up system for pediatric hypospadias patients. Background Technology

[0002] Hypospadias is a common congenital urogenital malformation that mainly occurs in male infants. Surgery is the primary treatment for hypospadias in children, but recovery after surgery is a long process. Children are in a stage of growth and development, and changes in their physical development may affect the recovery of hypospadias. At the same time, various complications may occur after hypospadias surgery. Long-term follow-up can not only help doctors detect early signs of these complications in time, but also help them adjust the treatment plan in a timely manner to ensure that patients receive the best medical care.

[0003] The distribution of pediatric hypospadias patients is random. Usually, before each follow-up visit, hospitals will group patients who are geographically close together into the same area, and the same group of medical staff will follow up with patients in the same area. When planning the clustering and follow-up of pediatric hypospadias patients, there is a lack of consideration for the recovery status of different patients and the necessity of follow-up for different patients. When the follow-up time is tight, the timeliness of follow-up for important patients cannot be guaranteed, which makes the follow-up planning inefficient. Summary of the Invention

[0004] To address the problem that existing technologies for cluster-based follow-up planning lack consideration for the recovery status and follow-up necessity of different patients, resulting in low efficiency in follow-up planning under tight time constraints, the present invention aims to provide a long-term intelligent follow-up system for pediatric hypospadias patients. The specific technical solution adopted is as follows: This invention provides a long-term intelligent follow-up system for pediatric hypospadias patients, the system comprising: The data collection module is used to obtain the follow-up address, surgery time, historical follow-up time and recovery score of patients to be followed up; The follow-up necessity analysis module is used to obtain recovery assessment indicators for each patient based on the time interval from surgery, the level of surgical difficulty, and the recovery score; and to determine the necessity of follow-up for each patient based on the recovery assessment indicators and the interval of historical follow-up time. The follow-up dropout analysis module is used to cluster patients based on their follow-up addresses to obtain address clusters. Within each address cluster, the optimal path for each address cluster is obtained based on the follow-up address distance and follow-up necessity between every two patients. The impact of each patient's dropout on the optimal path length of the address cluster is analyzed sequentially to obtain the possible dropout index for each patient. Based on each patient's follow-up necessity and dropout possible index, as well as the distance distribution between the patient and other address clusters and the recovery assessment index of patients in other address clusters, the follow-up dropout index for each patient is determined. The follow-up planning module is used to cluster patients based on follow-up dropout indicators and follow-up addresses to obtain updated clusters; and to perform follow-up planning based on the optimal path in each updated cluster.

[0005] Furthermore, the method for obtaining the recovery assessment indicators includes: For any given patient, the interval between the current follow-up time and the patient's surgery time is taken as the postoperative duration for that patient. The postoperative recovery index of the patient is obtained by multiplying the negative correlation mapping value of the surgical difficulty level of the patient with the postoperative duration; By combining the patient's postoperative duration, postoperative recovery indicators, and recovery score, recovery assessment indicators for the patient were obtained.

[0006] Furthermore, the method for obtaining the necessity of follow-up includes: For any given patient, the minimum of the patient's historical follow-up time and current follow-up time is taken as the patient's follow-up time necessity. The initial necessity of the patient is obtained by multiplying the negatively correlated value of the patient's recovery assessment index by the necessity of follow-up time. All patients to be followed up are sorted in ascending order of initial necessity to obtain the important follow-up sequence; The follow-up necessity of each patient is obtained by normalizing the product of the patient's sequence number in the important follow-up sequence and the initial necessity.

[0007] Furthermore, the method for obtaining the optimal path includes: For any address cluster, calculate the distance between the follow-up addresses of every two patients in the address cluster, and use it as the distance weight between every two patients; calculate the mean of the follow-up necessity between every two patients in the address cluster, and then perform negative correlation mapping to obtain the necessity weight between every two patients. By combining the distance weight and the required weight of each pair of patients in the address cluster, the edge weight of each pair of patients is obtained; Obtain a weighted graph of patients in the address cluster, where the weight of the edges connecting nodes is the edge weight. Use Dixtra's algorithm on the weighted graph to obtain the shortest path as the optimal path for the address cluster.

[0008] Furthermore, the method for obtaining the possible discard indicators includes: For any address cluster, each patient in the address cluster is taken as the patient to be analyzed; the number of connections between patients in the weighted graph of the address cluster is statistically analyzed, and this number is taken as the association importance of the patient to be analyzed. After excluding the patients in the analysis, the length of the shortest path for the remaining patients in the address cluster is obtained as the path discard length for the analyzed patients; the difference between the path discard length for the analyzed patients and the optimal path length for the address cluster is taken as the discard impact degree for the analyzed patients. The product of the negative correlation mapping of the patient's association importance and the discard impact is used as a possible indicator for discarding patients.

[0009] Furthermore, the method for obtaining the follow-up dropout index includes: The total length of the patient follow-up addresses corresponding to the optimal path in each address cluster is used as the processing time indicator for each address cluster. For any patient in an address cluster, if the processing time index of other address clusters outside the patient's address cluster is less than that of the patient's address cluster, the corresponding other address clusters will be recorded as the patient's reference cluster. Based on the distance distribution between the patient and each reference cluster, as well as the treatment time index and recovery assessment of the reference clusters, the non-treatment index of the patient and each reference cluster was obtained. By combining the non-treatment indicators of this patient with all reference clusters, and considering the necessity of follow-up for this patient, the discardable indicators for this patient were obtained. The product of the patient's discardable indicators and the possible discard indicators is used as the patient's follow-up discard indicator.

[0010] Furthermore, the method for obtaining the non-processing indicators includes: For any reference cluster of the patient, the distance between the center point of the follow-up addresses of all patients in the reference cluster and the follow-up address of the patient is taken as the divisibility distance between the patient and the reference cluster. Calculate the sum of recovery assessment indicators for all patients in the reference cluster and perform negative correlation mapping to obtain the time required for the reference cluster; The non-treatment index of the patient and the reference cluster is obtained by multiplying the divisibility distance between the patient and the reference cluster, the time required for the reference cluster, and the processing time index.

[0011] Furthermore, the method for obtaining the discardable indicators includes: The sum of the non-treatment metrics of the patient and all reference clusters is used as the criteria for discarding the fit for that patient. The product of the patient's discard fit index and the necessity of follow-up was negatively correlated and normalized to obtain the patient's discardable index.

[0012] Furthermore, the method for obtaining the updated clusters includes: The follow-up dropout index for each patient is used as a weight to weight the follow-up address. Clustering is then performed based on the weighted follow-up address of the patient to obtain the updated cluster.

[0013] Furthermore, the method for obtaining the address clusters includes: The number of clusters was obtained by using the elbow method for the follow-up addresses of all patients; the distance between the follow-up addresses of patients was used as a metric to perform K-means clustering algorithm on the patients to obtain the number of address clusters.

[0014] The present invention has the following beneficial effects: This invention assesses the necessity of patient follow-up by analyzing surgical history and historical recovery data, allowing subsequent planning analyses to consider the importance of follow-up and improving the rationality of current follow-up planning. When planning follow-up paths based on geographical location clustering, it combines the patient's follow-up necessity with their participation in the optimal path to analyze and obtain a potential rejection index for each patient. This index characterizes the degree to which each patient can be rejected in the current follow-up planning within a single cluster region, improving the optimal path planning for local areas. Furthermore, by considering the patient's distribution with other clusters, it reflects the degree to which the patient can participate in the planning of other cluster regions. Finally, it obtains the degree to which each patient can be rejected by regional planning based solely on address clusters—the follow-up rejection index—which is incorporated into cluster planning analysis, providing more reliable data support for cluster-based follow-up planning. This invention analyzes the patient's follow-up necessity and the impact of their participation in cluster planning during cluster-based follow-up planning, adjusting the follow-up region division plan to improve follow-up efficiency. Attached Figure Description

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

[0016] Figure 1 This is a structural block diagram of a long-term intelligent follow-up system for pediatric hypospadias patients provided in one embodiment of the present invention; Figure 2 This is a weighted graph diagram of an address clustering cluster provided in one embodiment of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a long-term intelligent follow-up system for pediatric hypospadias patients proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] The following description, in conjunction with the accompanying drawings, details a specific solution for a long-term intelligent follow-up system for pediatric hypospadias patients provided by this invention.

[0020] Please see Figure 1 The diagram illustrates a structural block diagram of a long-term intelligent follow-up system for pediatric hypospadias patients provided by an embodiment of the present invention. The system includes: a data collection module 101, a follow-up necessity analysis module 102, a follow-up rejection analysis module 103, and a follow-up planning module 104.

[0021] The data collection module 101 is used to obtain the follow-up address, operation time, historical follow-up time and recovery score of the patient to be followed up.

[0022] In this embodiment of the invention, multi-dimensional data such as age, weight, surgical difficulty level, and follow-up address of pediatric hypospadias patients are collected to characterize the patients' basic follow-up information. The multi-dimensional data of each patient are standardized by decimal scaling to make them have similar scale and distribution. The follow-up address data are obtained through GIS technology.

[0023] To better consider the need for patient follow-up, the historical follow-up time, surgery time, and recovery score of each patient are obtained to facilitate postoperative analysis. Medical staff will use a scoring system to assess the patient's recovery as a recovery score, which typically includes symptom scores, functional scores, and quality of life scores. It should be noted that the data collection and extraction methods used in this embodiment are well-known to those skilled in the art and are not intended to limit the scope of the invention. In the specific implementation of this application, patient-related data must be collected, used, and processed in accordance with relevant national and regional laws, regulations, and standards.

[0024] The follow-up necessity analysis module 102 is used to obtain recovery assessment indicators for each patient based on the interval between the time of surgery and the level of surgical difficulty, as well as the recovery score; and to determine the follow-up necessity for each patient based on the recovery assessment indicators and the interval of historical follow-up time.

[0025] Recovery in pediatric hypospadias patients is a dynamic process, with varying outcomes at each stage. By tracking recovery trends, it's possible to clearly understand whether the patient is continuously improving, has reached a plateau, or is showing signs of deterioration. Individual differences in patient outcomes lead to variations in recovery. Therefore, each patient's surgical history and historical recovery scores are used to assess the potential extent of recovery during follow-up.

[0026] Preferably, in this embodiment of the invention, the method for obtaining the recovery evaluation indicators includes: For any given patient, the interval between the current follow-up time and the patient's surgery time is taken as the postoperative duration, reflecting the postoperative recovery time. A longer duration indicates a longer recovery period and potentially better recovery. Furthermore, the postoperative duration is multiplied by the negative correlation value of the patient's surgical difficulty level to obtain the patient's postoperative recovery index. Higher surgical difficulty leads to a more complex wound healing process, a greater likelihood of complications such as wound infection and dehiscence, and a poorer recovery.

[0027] It should be noted that negative correlation mapping is a technique well known to those skilled in the art, and can take the form of inverse proportion or negative exponentiation, etc., without any restrictions.

[0028] Finally, by combining the patient's postoperative duration, postoperative recovery indicators, and recovery score, the patient's recovery assessment index is obtained. In this embodiment of the invention, the product of the patient's postoperative duration, postoperative recovery indicators, and recovery score is used as the patient's recovery assessment index. The higher the recovery score, the better the patient's historical recovery.

[0029] Recovery from hypospadias in children is a long-term process. Each follow-up visit by medical staff continuously tracks the patient's recovery trajectory, providing data support for subsequent treatment plans. When the patient's recovery at each follow-up visit is unsatisfactory, it is even more crucial for doctors to closely monitor the progression of the disease and take appropriate treatment measures as early as possible. Therefore, it is necessary to analyze the necessity of each follow-up visit and further determine the necessity of follow-up by comparing the recovery assessment with the historical follow-up time.

[0030] Preferably, in this embodiment of the invention, the method for obtaining the necessity of follow-up includes: For any given patient, the minimum of the patient's historical follow-up time and the current follow-up time is taken as the necessity of the follow-up time for that patient. The necessity is analyzed based on the historical follow-up interval. The time interval between the most recent historical follow-up and the current moment is considered. Children are in a stage of rapid physical growth and development. If there is no follow-up for a long time, medical staff will not be able to detect recovery problems caused by physical development in time. Therefore, the necessity of follow-up increases with the increase of the follow-up interval.

[0031] Further, the patient's recovery assessment indicators were negatively correlated and multiplied by the follow-up time necessity to obtain the patient's initial necessity. The worse the recovery assessment, the greater the time necessity, indicating that the next follow-up is more necessary.

[0032] By comparing the importance of all patients that the hospital plans to follow up with in the current follow-up, the relative necessity of each patient in the current follow-up list can be obtained. This is to prevent the omission of some patients from the follow-up list when there is insufficient follow-up time, and to make them available for the next follow-up.

[0033] Therefore, all patients to be followed up were sorted in ascending order of initial necessity to obtain a follow-up importance sequence, and relative analysis was performed based on the sorting order. The product of each patient's sequence number in the follow-up importance sequence and the initial necessity was normalized to obtain the follow-up necessity of each patient. The necessity of follow-up was then weighted by the sequence number; the larger the sequence number, the more important the patient's initial necessity for follow-up in the hospital's current follow-up list, and the more necessary the follow-up.

[0034] This completes the necessary analysis for patient follow-up.

[0035] The follow-up dropout analysis module 103 is used to cluster patients based on their follow-up addresses to obtain address clusters; within each address cluster, the optimal path for each address cluster is obtained based on the follow-up address distance and follow-up necessity between every two patients; the impact of each patient being dropped on the optimal path length of the address cluster is analyzed sequentially to obtain the possible dropout index for each patient; based on the follow-up necessity and possible dropout index for each patient, as well as the distance distribution between the patient and other address clusters and the recovery assessment index of patients in other address clusters, the follow-up dropout index for each patient is determined.

[0036] Since patients' addresses are usually scattered, medical staff will divide the area into regions to improve follow-up efficiency. By coordinating follow-up in relatively concentrated areas, the efficiency of follow-up can be improved. Therefore, the first step is to cluster patients based on their follow-up addresses to obtain address clusters.

[0037] In this embodiment of the invention, the elbow method is used to obtain the number of clusters for the follow-up addresses of all patients. The distance between the follow-up addresses of patients is used as a metric to perform K-means clustering algorithm on the patients, resulting in a number of address clusters. It should be noted that the elbow method for obtaining the number of clusters and the K-means clustering algorithm are well-known techniques to those skilled in the art, and will not be described in detail here.

[0038] Healthcare professionals typically estimate the required follow-up time based on the patient data for the day. Therefore, they usually need to conduct follow-ups within a limited timeframe. Since each patient's condition and recovery progress varies, those with poorer recovery or more severe conditions should be prioritized for follow-up to prevent missed appointments due to insufficient time. Patients who are not followed up can be scheduled for a later date. Therefore, a preliminary analysis of the optimal path is conducted within each clustered region, considering both the necessity of follow-up and distance.

[0039] Preferably, in this embodiment of the invention, the method for obtaining the optimal path includes: First, for any address cluster, calculate the distance between the follow-up addresses of every two patients in that cluster, using this distance as the distance weight for each pair of patients. The closer the distance, the better the connection selection between patients. Next, calculate the mean of the follow-up necessity between every two patients in that address cluster and perform a negative correlation mapping to obtain the necessity weight for each pair of patients. The greater the follow-up necessity, the better the connection selection between patients balances necessity.

[0040] Furthermore, by combining the distance weight and the necessary weight of each pair of patients in the address cluster, the edge weight of each pair of patients is obtained. In this embodiment of the invention, the product of the distance weight and the necessary weight is used as the edge weight between patients, providing weighted edge data for the subsequent construction of the graph.

[0041] Obtain the weighted graph of patients within this address cluster. The weights of the edges connecting nodes in the weighted graph are the edge weights. Please refer to [link to relevant documentation]. Figure 2 This diagram illustrates a weighted graph of an address cluster provided by an embodiment of the present invention, where A, B, C, D, E, F, and G are nodes in the weighted graph, that is, patients in the same address cluster. The values ​​on the connecting edges between nodes represent edge weights, such as the edge weight between patient A and patient B being 14, the edge weight between patient A and patient C being 8, and so on.

[0042] Therefore, when planning follow-up routes, not only geographical distance factors were considered, but also the necessity of patient follow-up was comprehensively taken into account. This ensured that patient groups requiring more follow-up were given greater priority in route planning and visited as early as possible. Finally, Dijkstra's algorithm was applied to the weighted graph to obtain the shortest path as the optimal path for that address cluster. The optimal follow-up route is reflected by the non-repeating shortest path. It should be noted that weighted graph construction and Dijkstra's algorithm are techniques well-known to those skilled in the art and will not be elaborated upon here.

[0043] Considering the possibility of inappropriate clustering or limited follow-up time, some patients may need to be dropped during follow-up in this area. Therefore, further analysis is needed to determine the impact of each patient's dropout on follow-up. Preferably, in this embodiment of the invention, the method for obtaining possible dropout indicators includes: For any address cluster, each patient in that cluster is treated as an analysis patient, and each patient is analyzed. The number of connections between the patient and other patients in the weighted graph of that address cluster is statistically analyzed, and this number is used as the importance of the analysis patient's associations. In the weighted graph, the more edges connected to the analysis patient, the stronger the connection between the patient's follow-up address and the follow-up addresses of other patients, indicating that the patient is a key node. The greater the impact on path continuity if the patient is discarded, the higher the importance and the lower the degree to which the patient can be discarded.

[0044] After excluding patients from the analysis, the length of the shortest path for the remaining patients in the address cluster is obtained as the path discard length for the analyzed patients. After removing the analyzed patients, the shortest path is obtained again to reflect the change in path length after discarding. Furthermore, the difference between the path discard length of the analyzed patients and the optimal path length of the address cluster is used as the discard impact degree of the analyzed patients. The smaller the length difference, the smaller the impact of the analyzed patients on reducing the optimal path length, and the smaller the degree of discarding.

[0045] Ultimately, the product of the negative correlation mapping value of the patient's association importance and the discard impact is used as the indicator for the patient's possible discard. Considering the overall association distribution and the change in path length after discard, the higher the association importance and the lower the discard impact, the smaller the degree of discardability, and therefore the smaller the possible discard indicator.

[0046] This concludes the initial discard analysis.

[0047] Further consideration is needed. Patients with high follow-up necessity may be located in non-critical locations along the optimal path, meaning that some patients' follow-up addresses deviate from the locations of most other patients. However, since these patients have poor recovery, discarding them might cause them to miss the optimal period for disease progression. Therefore, the decision-making level for discarding each patient's follow-up needs to be adjusted to avoid discarding patients with high follow-up necessity but significant impact on the optimal path.

[0048] Meanwhile, considering the possibility that patients with relatively offset positions can be assigned to other clusters for follow-up, when the follow-up processing time in other concentrated areas is low, patients with offset positions can be processed to improve the time processing efficiency of regional follow-up planning. Therefore, while analyzing the necessity of patient follow-up in clusters, the distribution of patients among other address clusters is combined to adjust the discard index results and obtain the final follow-up discard index, which reflects the degree to which patients are discarded in the current address cluster.

[0049] Preferably, in this embodiment of the invention, the method for obtaining the follow-up discard index includes: First, the total length of the patient follow-up addresses corresponding to the optimal path in each address cluster is used as the processing time indicator for each address cluster. The sum of the actual distances between follow-up addresses on the optimal path reflects the necessary working time for processing follow-ups in the concentrated area corresponding to this cluster.

[0050] Furthermore, for any patient in an address cluster, if the processing time index of other address clusters outside the patient's address cluster is less than that of the patient's address cluster, the corresponding other address clusters are recorded as the patient's reference clusters. Outside the patient's cluster, all other clusters with shorter processing times are also recorded as reference clusters, and the possibility of additional processing for the patient is analyzed.

[0051] Based on the distance distribution between the patient and each reference cluster, and the treatment time indicators and recovery assessment of the reference clusters, non-treatment indicators for the patient and each reference cluster are obtained, reflecting the likelihood that the patient can be followed up and treated in other concentrated areas. Preferably, in this embodiment of the invention, the method for obtaining non-treatment indicators includes: First, for any reference cluster of the patient, the distance between the center point of the follow-up addresses of all patients in the reference cluster and the patient's follow-up address is taken as the divisibility distance between the patient and the reference cluster. The closer the distance, the higher the probability that the corresponding concentrated area of ​​the reference cluster can handle the patient's follow-up in terms of distance.

[0052] Then, the sum of the recovery assessment indicators of all patients in the reference cluster is calculated and negative correlation mapping is performed to obtain the time required for the reference cluster. By comprehensively assessing the recovery assessment of patients in the reference cluster, the working time that patient follow-up may take is reflected. When the overall recovery assessment indicator is larger, that is, the time required is smaller, it indicates that the patient recovery in the reference cluster is greater, the working time required for the reference cluster is smaller, and the possibility of handling the patient follow-up in terms of time is higher.

[0053] Finally, the divisibility distance between the patient and the reference cluster, as well as the time requirement and processing time index of the reference cluster, are multiplied together to obtain the non-processing index of the patient and the reference cluster. The larger the divisibility distance, the lower the distance processing may be; the larger the time requirement, the lower the time processing may be; and the larger the processing time index, the higher the overall follow-up time of the cluster and the lower the processing possibility, hence the larger the non-processing index.

[0054] Further combining the patient's non-treatment indicators with all reference clusters, and the patient's follow-up necessity, a discardable indicator for the patient is obtained. Combining all analyzed reference clusters and necessity, this reflects the discardability after adjustment analysis. In this embodiment of the invention, the sum of the non-treatment indicators of the patient and all reference clusters is used as the patient's discard fit indicator. Combining the non-treatment indicators of all reference clusters reflects the patient's misfit to other clusters. The higher the sum, that is, the larger the discard fit indicator, the lower the patient's fit treatment with other clusters, and the smaller the degree of discard.

[0055] Further, the product of the patient's discard fit index and follow-up necessity was negatively correlated and normalized to obtain the patient's discardable index. The higher the follow-up necessity, the higher the patient's need for follow-up as soon as possible, and the lower the degree of discard in the path. Therefore, the greater the follow-up necessity and the larger the discard fit index, the smaller the patient's discardable index.

[0056] It should be noted that normalization is a technique well known to those skilled in the art. The choice of normalization can be linear normalization or standard normalization, etc. Specific normalization methods are not limited or elaborated here.

[0057] Finally, the product of the patient's discardable index and the discardable index is used as the patient's follow-up discard index. By further analyzing and adjusting the discardable index, the follow-up discard index is obtained, which reflects the greater the patient's likelihood of being discarded by the current cluster.

[0058] The follow-up planning module 104 is used to cluster patients based on their follow-up dropout criteria and follow-up addresses to obtain updated clusters; and to perform follow-up planning based on the optimal path in each updated cluster.

[0059] Finally, by analyzing the follow-up dropout index obtained under the regional centralized follow-up plan, the clustering situation can be adjusted to improve the reliability of clustering in patient follow-up and make the follow-up path planning more efficient. In this embodiment of the invention, the method for obtaining the updated clusters after adjustment includes: The follow-up locations are weighted using each patient's follow-up rejection index as a weight. Clustering is then performed based on these weighted follow-up locations to obtain updated clusters. This weighted re-clustering, followed by the application of Dijkstra's algorithm to find the optimal path for each updated cluster, ensures a more rational allocation of processing resources among the updated clusters. Finally, medical staff can follow up with patients based on the optimal path for each updated cluster, improving the efficiency of long-term follow-up for pediatric urethral patients. The number of queues before follow-up can be set based on the number of updated clusters and is not limited here.

[0060] In summary, this invention assesses the necessity of patient follow-up by considering the patient's surgical history and historical recovery status, allowing subsequent planning analyses to simultaneously consider the importance of follow-up and improving the rationality of the current follow-up plan. When planning follow-up paths based on geographical location clustering, it analyzes the patient's follow-up necessity and their participation in the optimal path to obtain a potential rejection index for each patient. This index characterizes the degree to which each patient can be rejected under the follow-up planning of a single cluster region, improving the optimal path planning for local areas. Furthermore, by combining the patient's distribution with other clusters, it reflects the degree to which the patient can participate in the planning of other cluster regions. Finally, it obtains the degree to which each patient can be rejected by the regional planning when clustered solely based on address, i.e., the follow-up rejection index. This index, incorporated into cluster planning analysis, provides more reliable data support for cluster-based follow-up planning. In cluster-based follow-up planning, this invention analyzes the patient's follow-up necessity and the impact of their participation in cluster planning, adjusting the division of follow-up regions to improve follow-up efficiency.

[0061] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0062] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A long-term intelligent follow-up system for pediatric hypospadias patients, characterized in that, The system includes: The data collection module is used to obtain the follow-up address, surgery time, historical follow-up time and recovery score of patients to be followed up; The follow-up necessity analysis module is used to obtain recovery assessment indicators for each patient based on the time interval from surgery, the level of surgical difficulty, and the recovery score; and to determine the necessity of follow-up for each patient based on the recovery assessment indicators and the interval of historical follow-up time. The follow-up dropout analysis module is used to cluster patients based on their follow-up addresses to obtain address clusters. Within each address cluster, the optimal path for each address cluster is obtained based on the follow-up address distance and follow-up necessity between every two patients. The impact of each patient's dropout on the optimal path length of the address cluster is analyzed sequentially to obtain the possible dropout index for each patient. Based on each patient's follow-up necessity and dropout possible index, as well as the distance distribution between the patient and other address clusters and the recovery assessment index of patients in other address clusters, the follow-up dropout index for each patient is determined. The follow-up planning module is used to cluster patients based on follow-up dropout indicators and follow-up addresses to obtain updated clusters; and to perform follow-up planning based on the optimal path in each updated cluster.

2. The long-term intelligent follow-up system for pediatric hypospadias patients according to claim 1, characterized in that, The methods for obtaining the recovery assessment indicators include: For any given patient, the interval between the current follow-up time and the patient's surgery time is taken as the postoperative duration for that patient. The postoperative recovery index of the patient is obtained by multiplying the negative correlation mapping value of the surgical difficulty level of the patient with the postoperative duration; By combining the patient's postoperative duration, postoperative recovery indicators, and recovery score, recovery assessment indicators for the patient were obtained.

3. The long-term intelligent follow-up system for pediatric hypospadias patients according to claim 1, characterized in that, The methods for determining the necessity of follow-up include: For any given patient, the minimum of the patient's historical follow-up time and current follow-up time is taken as the patient's follow-up time necessity. The initial necessity of the patient is obtained by multiplying the negatively correlated value of the patient's recovery assessment index by the necessity of follow-up time. All patients to be followed up are sorted in ascending order of initial necessity to obtain the important follow-up sequence; The follow-up necessity of each patient is obtained by normalizing the product of the patient's sequence number in the important follow-up sequence and the initial necessity.

4. The long-term intelligent follow-up system for pediatric hypospadias patients according to claim 1, characterized in that, The method for obtaining the optimal path includes: For any address cluster, calculate the distance between the follow-up addresses of every two patients in the address cluster, and use it as the distance weight between every two patients; calculate the mean of the follow-up necessity between every two patients in the address cluster, and then perform negative correlation mapping to obtain the necessity weight between every two patients. By combining the distance weight and the required weight of each pair of patients in the address cluster, the edge weight of each pair of patients is obtained; Obtain a weighted graph of patients in the address cluster, where the weight of the edges connecting nodes is the edge weight. Use Dixtra's algorithm on the weighted graph to obtain the shortest path as the optimal path for the address cluster.

5. The long-term intelligent follow-up system for pediatric hypospadias patients according to claim 4, characterized in that, The methods for obtaining the possible discard indicators include: For any address cluster, each patient in the address cluster is taken as the patient to be analyzed; the number of connections between patients in the weighted graph of the address cluster is statistically analyzed, and this number is taken as the association importance of the patient to be analyzed. After excluding the patients in the analysis, the length of the shortest path for the remaining patients in the address cluster is obtained as the path discard length for the analyzed patients; the difference between the path discard length for the analyzed patients and the optimal path length for the address cluster is taken as the discard impact degree for the analyzed patients. The product of the negative correlation mapping of the patient's association importance and the discard impact is used as a possible indicator for discarding patients.

6. The long-term intelligent follow-up system for pediatric hypospadias patients according to claim 1, characterized in that, The methods for obtaining the follow-up dropout indicators include: The total length of the patient follow-up addresses corresponding to the optimal path in each address cluster is used as the processing time indicator for each address cluster. For any patient in an address cluster, if the processing time index of other address clusters outside the patient's address cluster is less than that of the patient's address cluster, the corresponding other address clusters will be recorded as the patient's reference cluster. Based on the distance distribution between the patient and each reference cluster, as well as the treatment time index and recovery assessment of the reference clusters, the non-treatment index of the patient and each reference cluster was obtained. By combining the non-treatment indicators of this patient with all reference clusters, and considering the necessity of follow-up for this patient, the discardable indicators for this patient were obtained. The product of the patient's discardable indicators and the possible discard indicators is used as the patient's follow-up discard indicator.

7. The long-term intelligent follow-up system for pediatric hypospadias patients according to claim 6, characterized in that, The methods for obtaining the non-processing indicators include: For any reference cluster of the patient, the distance between the center point of the follow-up addresses of all patients in the reference cluster and the follow-up address of the patient is taken as the divisibility distance between the patient and the reference cluster. Calculate the sum of recovery assessment indicators for all patients in the reference cluster and perform negative correlation mapping to obtain the time required for the reference cluster; The non-treatment index of the patient and the reference cluster is obtained by multiplying the divisibility distance between the patient and the reference cluster, the time required for the reference cluster, and the processing time index.

8. The long-term intelligent follow-up system for pediatric hypospadias patients according to claim 6, characterized in that, The method for obtaining the discardable indicators includes: The sum of the non-treatment metrics of the patient and all reference clusters is used as the criteria for discarding the fit for that patient. The product of the patient's discard fit index and the necessity of follow-up was negatively correlated and normalized to obtain the patient's discardable index.

9. The long-term intelligent follow-up system for pediatric hypospadias patients according to claim 1, characterized in that, The method for obtaining the updated clusters includes: The follow-up dropout index for each patient is used as a weight to weight the follow-up address. Clustering is then performed based on the weighted follow-up address of the patient to obtain the updated cluster.

10. The long-term intelligent follow-up system for pediatric hypospadias patients according to claim 1, characterized in that, The method for obtaining the address clusters includes: The number of clusters was obtained by using the elbow method for the follow-up addresses of all patients; the distance between the follow-up addresses of patients was used as a metric to perform K-means clustering algorithm on the patients to obtain the number of address clusters.

Images