Chronic disease hierarchical management method and system based on quantum simulation

Abstract

In view of the low processing efficiency of high-dimensional medical data, the insufficient grading accuracy, the poor individual difference adaptability and other pains in traditional chronic disease grading management, the application discloses a chronic disease grading management method and system based on quantum simulation; the method relies on the superposition and entanglement advantages of quantum bits, adopts special algorithms such as quantum phase estimation algorithm (QPE) and variational quantum eigensolver (VQE); the system comprises a data acquisition module, a quantum simulation calculation module, a grading decision module, a scheme generation module and a terminal interaction module, each module cooperatively operates, breaks the experiential limitation of traditional grading management, significantly improves the grading accuracy and management efficiency, provides new technical support for fine and personalized management of chronic diseases, helps to delay the progress of the disease, reduce the incidence of complications, and promotes the transformation of chronic disease management from 'experience-driven' to 'precision calculation-driven'.

Description

Technical Field

[0001] This invention relates to the field of chronic disease hierarchical management technology, and more specifically, to a method and system for chronic disease hierarchical management based on quantum simulation. Background Technology

[0002] In the practice of chronic disease hierarchical management, most existing technologies revolve around forming a stratification result as soon as possible based on the patient's current health status and arranging subsequent management accordingly. In practice, it is generally necessary to collect blood pressure, blood sugar, blood lipid, complication records, medication, follow-up visit results and family monitoring information, and then combine them with established stratification rules, scoring methods or data analysis results to determine the corresponding level, and use the level as the main basis for follow-up frequency, re-examination arrangements and intervention intensity. Taking the scenario of community health service institutions working with outpatient management platforms to conduct continuous follow-up on patients with diabetes and hypertension as an example, the management side needs to maintain continuous management in the long term under the premise of limited doctor manpower, follow-up capacity and patient terminal reach capabilities. On the other hand, it cannot deviate from the existing data collection links and daily service processes to add high-intensity manual follow-up for each patient. At the same time, the grading results must be able to be directly used for subsequent follow-up organization and resource allocation. However, in this usage scenario, a situation that can be directly verified often occurs in actual operation: patients with similar medical indicators are often classified into the same or similar levels by the system and receive similar management arrangements. Among them, some patients can continuously upload monitoring data, have follow-up visits on time, and implement the established plan relatively stably, while other patients repeatedly fail to report, report late, delay follow-up visits, do not respond to reminders, or interrupt the implementation. As a result, limited management resources are continuously invested in the subjects who are easier to maintain management continuity, while the subjects who really need additional investment to maintain management integrity are not identified in advance. The reason is that the existing classification results mainly reflect the disease status itself and cannot reflect the difference in the difficulty of investment in the subsequent management implementation process for patients of the same level. The technical problem this application aims to solve is: how to identify differences in management resistance among patients at the same level while forming disease severity levels during the process of chronic disease classification management, and generate classification results that can directly support subsequent follow-up organization and resource allocation. Summary of the Invention

[0003] To overcome the aforementioned deficiencies in the prior art, embodiments of the present invention provide a method and system for the graded management of chronic diseases based on quantum simulation. By jointly constructing a quantum simulation input sequence with patient condition evolution information and management execution information, the key position results reflecting the disease recovery process and management implementation resistance are solved. Based on this, a coupled graded result of disease level and management resistance level and a corresponding management scheme are formed to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a method for hierarchical management of chronic diseases based on quantum simulation, comprising: S1. Collect the target patient's genomic data, physiological monitoring data, clinical diagnosis and treatment data, lifestyle data, terminal response records, task completion records, follow-up visit records, device online records, and execution deviation records in the current round. Perform merging and alignment according to patient identifier, round identifier, and record time sequence, and output the current patient table. S2. Based on the current patient table and corresponding historical records, extract the disease record group and management record group, jointly encode the disease record group and management record group into a quantum simulation input sequence, and perform quantum simulation operation on different management input sequences corresponding to the same disease record group to solve the disease stabilization position, execution discontinuity position and backpropagation position corresponding to each management input sequence, and output the simulation result table. S3. Based on the simulation results table, perform a ranking comparison on the disease stability position, intermittent execution position and feedback position corresponding to each management input order, determine the current patient's disease level and management resistance level, and output the coupling grading table. S4. Based on the coupling grading table, generate follow-up order, re-examination arrangement, reminder arrangement, supplementary sampling arrangement and intervention arrangement according to the correspondence between disease level and management resistance level, and output management plan table; S5. Based on the management plan table, issue corresponding arrangements to the doctor, patient and follow-up sides, collect execution confirmation results, supplementary collection and return results and subsequent response results, execute write-back according to patient and round identifiers, and output the next round record table.

[0005] In a preferred embodiment, S1 includes: S1-1. Separate the genomic data, physiological monitoring data, clinical diagnosis and treatment data, lifestyle data, terminal response records, task completion records, follow-up visit records, device online records, and execution deviation records into raw record units with patient identifiers, record source identifiers, record start time, record end time, and record content items. Collect the data according to the patient identifiers and arrange them in the execution order according to the record start time, and output the patient raw record table. S1-2. Based on the start time and end time of the current round corresponding to the round identifier, perform round truncation on each original record unit in the patient's original record table. Record items that fall within the current round range are retained as round-in-round record items, and record items that cross the current round boundary are divided into round-in-round retained items. Output the current round record table. S1-3. For record items in different rounds that have the same time position in the current round record table, perform same-time merging according to the record source identifier and record time sequence. Perform forward inheritance and reverse verification for record items in rounds with continuous coverage relationship. For record items in rounds without continuous coverage relationship, retain the corresponding source position and generate the current patient table containing disease record items and management record items.

[0006] In a preferred embodiment, S2 includes: S2-1. Based on the current patient table and corresponding historical records, extract the disease record items and management record items of the current round and adjacent historical rounds according to the patient ID, round ID and record time sequence. Perform same-time correspondence for record items with the same record start time, and perform succession correspondence for record items whose record end time is connected to the start time of the next record. Preserve the correspondence results of continuous disease change direction and complete start and end of management function, and generate disease record group, management record group and group correspondence table. S2-2. Based on the group correspondence table, each management record group is divided into single action record items, arranged in ascending order of the start time of the record. For adjacent single action record items where the end time of the previous single action record item and the start time of the next single action record item are connected end to end, a connecting edge is established to generate the complete management input order from the first single action record item to the last single action record item. For each management input order, the number of action steps, the number of disease reversals, and the number of interruptions are calculated in sequence. The field-by-field comparison is performed according to the field arrangement order of the number of action steps, the number of disease reversals, and the number of interruptions. The management input order with the first lexicographical order is retained to generate a candidate order table.

[0007] In a preferred embodiment, S2 further includes: S2-3. Based on the medical record group, management record group and candidate order table, write the medical location item, management action item, action interval item and feedback association item into the coding position according to the fixed field order to generate the initial coding sequence. Perform consistency check on the position of the same field in each initial coding sequence, fill in the missing positions of the field, and retain the coding result of the field with the earlier order for the conflicting positions to generate the quantum simulation input sequence table. S2-4. For the quantum simulation input sequence list, perform quantum simulation deduction for each candidate management input order corresponding to the same disease record group. Solve the disease stabilization position, execution discontinuity position and return position step by step according to the order of management action occurrence. Compare the results of this round of deduction with the results of the previous round of deduction item by item according to the field order of disease stabilization position, execution discontinuity position and return position. Stop the deduction when the results of two consecutive rounds of comparison are completely consistent, retain the corresponding deduction results, and generate a simulation result table.

[0008] In a preferred embodiment, S3 includes: S3-1. Based on the simulation results table, extract the management input order, condition stabilization position, execution discontinuity position and feedback position corresponding to the same patient in the current round. Construct a sequential control group according to the order of action occurrence within the management input order. Perform a first-to-last closure check on the condition stabilization position, execution discontinuity position and feedback position in the same management input order. Retain the sequential control group where the first position can be pointed back to the last position and the last position can be reversed to the first position. Generate a sequential check table. S3-2. Based on the ranking check table, calculate the stabilization difference, discontinuity difference, and return difference for each ranking control group. The stabilization difference is composed of the position difference between adjacent stabilization positions, the discontinuity difference is composed of the position difference between adjacent discontinuity positions, and the return difference is composed of the position difference between adjacent return positions. Then, connect the stabilization difference, discontinuity difference, and return difference for each ranking control group into a grading judgment string according to a fixed field order, and perform field-by-field comparison according to the fixed field order to generate the first-rank table.

[0009] In a preferred embodiment, S3 further includes: S3-3. Based on the first-order table, perform disease-side merging and management-side merging on each grading judgment string. Among them, grading judgment strings with continuous forward shift of disease stability position and execution discontinuity position not earlier than the corresponding return position are merged into the same disease level candidate group. Grading judgment strings with continuous forward shift of execution discontinuity position and no expansion of the interval between adjacent return positions are merged into the same management resistance level candidate group. Perform bidirectional consistency verification on each disease level candidate group and each management resistance level candidate group, retain the candidate groups with the same forward merging result and reverse merging result, and generate a grading candidate table. S3-4. For the candidate grading table, link the candidate disease grade and the candidate management resistance grade according to the same management input order. For candidate groups that cannot be linked one by one, delete their corresponding management input order and re-link them. Stop linking when the disease grade order and management resistance grade order of the linking results are completely consistent in two consecutive rounds, retain the corresponding linking results, and generate a coupled grading table.

[0010] In a preferred embodiment, S4 includes: S4-1. Based on the coupling grading table, extract the disease level and management resistance level of the same patient in the current round. Concatenate each disease level and each management resistance level in pairs according to the field order to form coupling items. Perform same-round merging and same-patient merging on each coupling item to generate a coupling item table. S4-2. Based on the coupling item table, write follow-up items, re-examination items, reminder items, supplementary data collection items, and intervention items for each coupling item. Generate corresponding sequence codes according to the field order of disease severity first and management resistance level second, and arrange the follow-up items, re-examination items, reminder items, supplementary data collection items, and intervention items in the order of execution according to the sequence codes to generate a candidate plan table. S4-3. For the candidate plan table, perform a head-to-tail check on the follow-up items, re-examination items, reminder items, supplementary collection items, and intervention items corresponding to the same sequence number. For candidate plans that pass the check, retain the original order and write them in. For candidate plans that fail the check, delete the corresponding items and rearrange them. Stop the arrangement when the results of two consecutive rounds of arrangement are completely consistent, retain the corresponding results, and generate the management plan table.

[0011] In a preferred embodiment, S5 includes: S5-1. Based on the management plan table, extract the follow-up items, re-visit items, reminder items, supplementary data collection items and intervention items corresponding to the same patient in the current round. Then, attach each item to the doctor's side icon, the patient's side icon and the follow-up side icon respectively, arrange them in ascending order according to the planned time, and generate a lateral release table. S5-2. Based on the lateral release table, write follow-up visit and intervention items to the doctor's side, reminder and supplementary data collection items to the patient's side, and follow-up and supplementary data collection items to the follow-up side. Then, perform the same item correspondence for the returned content from each side according to the original release items to generate a lateral feedback table.

[0012] In a preferred embodiment, S5 further includes: S5-3. Based on the lateral feedback table, the execution confirmation results, supplementary collection and feedback results and subsequent response results are merged according to the patient identifier, round identifier and release item identifier. A connection is established for feedback results whose result time is connected to the corresponding release item plan time. A completion correspondence is established for feedback results whose result content is consistent with the corresponding release item content. A round feedback table is generated. S5-4. For the round feedback table, write the established and completed feedback results back to the starting record position of the next round according to the patient and round identifiers. Write the feedback results that have not been established or have not been completed back to the remaining record position of the current round according to the original publication item identifier, and generate the next round record table.

[0013] A chronic disease hierarchical management system based on quantum simulation, comprising a data acquisition module, a quantum simulation computing module, a hierarchical decision-making module, a treatment plan generation module, and a terminal interaction module: The data acquisition module is used to collect the target patient's genomic data, physiological monitoring data, clinical diagnosis and treatment data, lifestyle data, terminal response records, task completion records, follow-up visit records, device online records, and execution deviation records in the current round. It performs merging and alignment according to the patient identifier, round identifier, and record time sequence, and outputs the current patient table. The quantum simulation calculation module is used to extract the disease record group and management record group based on the current patient table and the corresponding historical records, jointly encode the disease record group and management record group into a quantum simulation input sequence, and perform quantum simulation operation on different management input orders corresponding to the same disease record group to solve the disease stabilization position, execution discontinuity position and backpropagation position corresponding to each management input order, and output the simulation result table. The hierarchical decision-making module is used to compare the order of the disease stabilization position, the intermittent execution position, and the feedback position corresponding to each management input order based on the simulation result table, determine the current patient's disease level and management resistance level, and output a coupled hierarchical table. The plan generation module is used to generate follow-up order, re-examination arrangement, reminder arrangement, supplementary sampling arrangement and intervention arrangement based on the correspondence between disease level and management resistance level, and output management plan table; The terminal interaction module is used to combine the management plan table to publish corresponding arrangements to the doctor, patient and follow-up side, collect execution confirmation results, supplementary collection and transmission results and subsequent response results, execute write-back according to patient identifier and round identifier, and output the next round record table.

[0014] The technical effects and advantages of this invention are as follows: This scheme, by jointly encoding the disease record group and the management record group and decoding the disease stability position, the intermittent execution position and the return position, can identify the differences in management resistance among patients of the same level while forming the disease level, so that the grading results can directly support the follow-up organization and resource allocation. Performing on the current round record and the adjacent historical round record to perform on the same-time correspondence, succession correspondence and continuous coverage verification can organize multi-source heterogeneous data under the unified round and time series caliber, thereby relatively reducing the impact of record misalignment and cross-round mixing on subsequent hierarchical calculations; By developing different management input sequences around the same disease record group and performing field-by-field comparisons based on the number of action steps, the number of disease reversals, and the number of interruptions, candidate management input sequences with fixed order criteria can be screened out, thereby relatively improving the consistency of subsequent quantum simulation inputs. Encoding the disease location item, management action item, action interval item, and feedback association item in a fixed field order, and uniformly handling position consistency, gap filling, and field conflict, can form a structurally stable quantum simulation input sequence, which is conducive to maintaining the comparability of the simulation results. By performing sequential comparison, difference merging, and bidirectional consistency verification on the disease status stabilization, intermittent status, and retrospective status, coupled classification results of disease status and management resistance level can be generated, thereby relatively improving the correspondence between classification results and subsequent management implementation process; By generating management plans based on the coupled grading results and writing the execution confirmation results, supplementary sampling and feedback results, and subsequent response results back to the next round of records, a complete chain of grading, execution, feedback, and re-grading can be formed, which helps to maintain the continuity of the chronic disease management process. Attached Figure Description

[0015] Figure 1 This is a flowchart outlining the method steps of the present invention; Figure 2 This is a schematic diagram of the system module structure of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Refer to the instruction manual appendix Figure 1-2 The present invention provides a quantum simulation-based method for hierarchical management of chronic diseases, comprising: S1. Collect the target patient's genomic data, physiological monitoring data, clinical diagnosis and treatment data, lifestyle data, terminal response records, task completion records, follow-up visit records, device online records, and execution deviation records in the current round. Perform merging and alignment according to patient identifier, round identifier, and record time sequence, and output the current patient table. This implementation method is used to organize multi-source heterogeneous patient records into a current patient table that can be directly used in subsequent disease record groups and management record groups. First, the record units are uniformly split, then the content within the current round is extracted, followed by simultaneous merging, continuation verification, and type classification. This implementation process includes the following steps: In S1-1, the input includes genomic data, physiological monitoring data, clinical diagnosis and treatment data, lifestyle data, terminal response records, task completion records, follow-up visit records, device online records, and execution deviation records. During processing, these nine types of records are divided into raw record units based on single test results, continuous monitoring intervals, single visit results, continuous statistical periods, single response events, single task events, single follow-up visit events, single online intervals, and single deviation events. Each raw record unit is then written with a patient identifier, record source identifier, record start time, record end time, and record content item. For instantaneous recordings, the start and end times are the same. For continuous recordings lacking an end time, the start time of the next record from the same source after the previous record is used as the end time. If no next record from the same source exists, the current acquisition end time is used as the end time. After writing, the records are grouped by patient identifier, arranged in ascending order of start time, and those with the same start time are arranged in a fixed order by record source identifier. The patient's original record table is output and written to S1-2 for reading. When a record content item is empty, the original record unit is retained and marked as an empty content record, and is not directly deleted. In S1-2, the inputs are the patient's original record table, the start time of the current round corresponding to the round identifier, and the end time of the current round. During processing, each original record unit in the patient's original record table is read, and the positional relationship between the record start time, record end time, and the start time and end time of the current round are compared: if both the start and end times fall within the current round's range, the complete record content is retained as a record item within the round; if the record start time is earlier than the current round's start time and the record end time falls within the current round's range, it is segmented by the current round's start time, and the latter part of the content is retained; if the record start time falls within the current round's range, the remaining part is retained; if the record start time falls within the current round's range, the remaining part is retained. For records within the previous round whose end time is later than the current round's end time, the record is segmented by the current round's end time and the preceding content is retained. For records whose start time is earlier than the current round's start time and whose end time is later than the current round's end time, the record is segmented at both ends by the current round's start time and end time and the intermediate content is retained. The current round's record table is then output and written to S1-3 for reading. Records whose start time is later than the end time are marked as invalid time-series records and removed from the current round's record table. When there are no records in the current round, a current round's record table containing only the patient identifier, round identifier, and empty record marker is generated. In S1-3, the input is the record table of the current round. During processing, the records are first grouped according to the start time of the record. Records with the same start time in the same round are identified as record items at the same position. Then, the record items are merged according to the record source identifier and the record time sequence. Among them, the contents that represent the patient's status in the genomic data, physiological monitoring data, clinical diagnosis and treatment data and lifestyle data are included in the condition record item. The contents that represent the management implementation process in the terminal response record, task completion record, follow-up visit record, equipment online record and execution deviation record are included in the management record item. Subsequently, for records within adjacent rounds, a continuous coverage relationship is determined. If the record termination time of a record in the previous round is consecutive to the record start time of a record in the next round, and there are no other records from the same source between them, a continuous coverage relationship is confirmed. Forward inheritance and reverse verification are then performed. Forward inheritance is used to transmit the patient identifier, round identifier, and source attribution. Reverse verification is used to check the source attribution, content type, and temporal continuity. If the two are consistent, the inheritance relationship is retained; if they are inconsistent, the inheritance relationship is deleted, and the independent source position is retained. For records within rounds without a continuous coverage relationship, only the corresponding source position is retained. Finally, the medical record items and management record items are written into the current patient table according to the patient identifier, round identifier, record source identifier, record start time, record end time, record content item, record type identifier, and source position identifier for S2 to read. When there are multiple records from the same source at the same position, the record item with the later record end time is retained first. If the record end times are the same, the record item with the earlier record content item field is retained. This implementation method splits the original record units, extracts the rounds, and merges them at the same time to form a current patient table with fixed fields, clear rounds, and clear boundaries between the disease side and the management side, thereby eliminating the impact of multi-source record temporal misalignment, cross-round overlap, and local missing data on subsequent calculations; In practical applications: Genetic testing results, seven-day blood glucose and blood pressure monitoring intervals, outpatient follow-up results, diet and exercise records, terminal reminder response records, task completion records, device online intervals, and execution deviation records for patients with diabetes and hypertension are each broken down into original record units. Then, the content within the current management round is extracted, and the blood glucose value, blood pressure value, and outpatient test value at the same moment are included in the disease record item. Terminal response, task completion, follow-up visit, device online status, and execution deviation are included in the management record item. Finally, the current patient table is generated for direct use in subsequent steps.

[0018] S2. Based on the current patient table and corresponding historical records, extract the disease record group and management record group, jointly encode the disease record group and management record group into a quantum simulation input sequence, and perform quantum simulation operation on different management input sequences corresponding to the same disease record group to solve the disease stabilization position, execution discontinuity position and backpropagation position corresponding to each management input sequence, and output the simulation result table. This implementation method is used to organize the current patient table and corresponding historical records into a simulation result table. The processing order is as follows: extract the patient record group and management record group, expand the management input order, encode it to form a quantum simulation input sequence table, and then perform quantum simulation deduction to solve for the patient's stable position, the execution discontinuity position, and the backpropagation position. This implementation process includes the following steps: In S2-1, the inputs are the current patient table and the corresponding historical records. First, the current round record items and adjacent historical round record items are extracted according to the patient identifier. Among them, the adjacent historical round record items are the completed round record items before the current round whose record termination time is closest to the start time of the current round. Then, they are sorted in ascending order by round identifier and record time sequence. Subsequently, the same-time correspondence is performed on record items with the same record start time, and the succession correspondence is performed on record items whose record termination time is connected to the start time of the next record. The direction of disease change and the integrity of the start and end of management action are calculated in the correspondence results. The direction of disease change is determined by the sign of the difference in the numerical values ​​of the same disease field in adjacent record items, and the integrity of the start and end of management action is determined by whether the start time, end time, and feedback association of the same management action exist simultaneously. The correspondence results with continuous disease change direction and complete management action start and end are retained, and disease record group, management record group, and intra-group correspondence table are generated for S2-2 to read. Record items with missing disease field values ​​are not included in the calculation of disease change direction, and management record items with missing action termination time or feedback association are not included in the judgment of the integrity of the start and end of management action. In S2-2, the input consists of a group correspondence table and management record groups. First, each management record group is divided into single-action record items according to the management action boundaries, and then arranged in ascending order of the record start time. Next, for adjacent single-action record items whose end time connects to the start time of the next single-action record item, a connecting edge is established. Starting from the first single-action record item and ending at the last single-action record item, the entire complete path is expanded to generate all management input sequences. For each management input sequence, the number of action steps, the number of disease reversals, and the number of interruptions are calculated sequentially. The number of action steps is the number of single-action record items within that management input sequence, and the number of disease reversals... The number of times the direction of adjacent disease changes in the corresponding disease record group is swapped between upward and downward or downward and upward is taken. The number of discontinuities is taken as the number of times that the first and last single action record items in the management input order do not meet the condition of being connected end to end. The field comparison is performed in the order of action steps, number of disease reversals, and number of discontinuities. The management input order with the first lexicographical order is retained to generate a candidate order table for S2-3 to read. Paths that cannot be connected from the first single action record item to the last single action record item are directly deleted. If the first three fields are completely consistent, the field comparison is continued in the order of the start time of the first single action record item, the end time of the last single action record item, and the action identifier. In S2-3, the inputs are the patient record group, the management record group, and the candidate order table. First, the patient location item is extracted from the patient record group, the management action item is extracted from the management record group, and the action interval item is extracted from the candidate order table. Then, the return-to-the-home related item is extracted according to the correspondence between the management action item and the return-to-the-home related item. Subsequently, the coding positions are written in a fixed field order, which is: patient location item, management action item, action interval item, and return-to-the-home related item. Within the same type of item, the fields are written in the field mapping order in the system configuration table to generate an initial coding sequence. Then, the position of the same field in each initial coding sequence is checked for consistency. For missing fields, a pre-configured empty code is written. For conflicting fields, the coding result of the field with the earlier order is retained to generate a quantum simulation input sequence table for S2-4 to read. If the same candidate order is missing a patient location item or a management action item, no corresponding quantum simulation input sequence is generated. If only the action interval item or the return-to-the-home related item is missing, it is filled with an empty code. In S2-4, the input is a quantum simulation input sequence list; quantum state inputs are established for each candidate management input order corresponding to the same disease record group, and quantum simulation is executed round by round according to the order of management actions; after each round of simulation, the disease stabilization position, the execution discontinuity position, and the feedback position are extracted. Among them, the disease stabilization position is the record position number corresponding to the disease field changing from continuous abnormality to continuous recovery; the execution discontinuity position is the action position number corresponding to the adjacent management actions not being connected end to end; and the feedback position is the first and corresponding management action item of the feedback correlation item. The record position number is established when establishing the association; the first round of simulation results are written to the temporary storage table. From the second round onwards, the simulation results of this round are compared with the simulation results of the previous round in the order of field order of disease stabilization, execution intermittent position, and return position. The simulation stops when the comparison results of two consecutive rounds are completely consistent, the corresponding simulation results are retained, the simulation result table is generated and read by S3; if any position cannot be solved in a round, the position is written as an empty position code and the simulation continues. If the same position is an empty position code in two consecutive rounds and the other two items are consistent, the result is retained and written to the simulation result table. This implementation method, through record group extraction, input sequence expansion, input sequence encoding, and quantum simulation deduction, forms a simulation result table with fixed fields and clear stopping conditions, providing a unified input for subsequent ranking comparison and coupling classification. In practical applications: for patients with diabetes and hypertension, blood glucose, blood pressure, and glycated hemoglobin records can be extracted from the current and previous rounds to form a condition record group. Medication adjustments, dietary interventions, exercise reminders, follow-up appointments, and their feedback records can be extracted to form a management record group. Then, the management input sequence is expanded and a unique candidate sequence is retained. The condition location item, management action item, action interval item, and feedback correlation item are encoded into a quantum simulation input sequence. After quantum simulation deduction, the condition stabilization position, execution discontinuity position, and feedback position are solved, generating a simulation result table for direct reading in subsequent steps.

[0019] S3. Based on the simulation results table, perform a ranking comparison on the disease stability position, intermittent execution position and feedback position corresponding to each management input order, determine the current patient's disease level and management resistance level, and output the coupling grading table. This implementation method is used to convert the simulation result table into a coupled grading table that can directly drive the generation of subsequent management plans. The processing sequence is as follows: First, a ranking control group is constructed under the same patient and the same round, and the first and last closure checks are completed. Then, the ranking control group forms a grading judgment string and the first ranking result is selected. Subsequently, the candidate groups of disease level and management resistance level are merged respectively. Finally, the corresponding connection is executed and the coupled grading table is output. This implementation process includes the following steps: In S3-1, the goal is to unify the positional results in the simulation results table into a comparable ranking control group, and to delete results that cannot form a complete ranking chain through a head-to-tail closure check. The input is the simulation results table. During processing, firstly, each management input order, each condition stabilization position, each execution discontinuity position, and each feedback position are extracted according to the patient identifier and the current round identifier. Then, they are arranged in ascending order of the action occurrence within the management input order to construct a ranking control group. Subsequently, a head-to-tail closure check is performed on the condition stabilization position, execution discontinuity position, and feedback position in the same management input order. The first position is the position of the first action in the management input order, and the last position is the position of the last action in the management input order. The first position can be referenced back to the last position, which means that the condition stabilization position, execution discontinuity position, and feedback position corresponding to the first position can be covered to the last position after continuous transmission along the action occurrence order. The last position can be reversed to the first position, which means that the condition stabilization position, execution discontinuity position, and feedback position corresponding to the last position can be referenced back to the first position after reverse tracing along the action occurrence order. Control groups that simultaneously satisfy the condition that the first position can be traced back to the last position and the last position can be traced back to the first position are retained and written into the ranking check table. The ranking check table includes at least the patient identifier, round identifier, management input order identifier, disease stabilization sequence, execution discontinuous sequence, return sequence, and closure check identifier, and is output for S3-2 to read. In case of abnormality or missing information, if any management input order is missing a disease stabilization, execution discontinuous, or return sequence, the first and last closure check is not performed, and the control groups whose closure check is not valid are directly deleted. In S3-2, the goal is to convert the position sequence in the ranking check table into a graded judgment string that can be compared field by field, and to select the first result of the ranking. The input is the ranking check table. During processing, the stabilization difference order, discontinuity difference order, and return difference order are calculated for each ranking control group. The stabilization difference order is calculated item by item by the position difference of adjacent stabilization positions according to the order of action occurrence. The discontinuity difference order is calculated item by item by the position difference of adjacent discontinuous positions according to the order of action occurrence. The return difference order is calculated item by item by the position difference of adjacent return positions according to the order of action occurrence. If a ranking control group contains only one position value, the corresponding difference order is written as a zero position difference sequence. After the difference order calculation is completed, the stabilization difference order, discontinuity difference order, and return difference order are concatenated into a graded judgment string according to a fixed field order, which is stabilization difference order, discontinuity difference order, and return difference order in that order. Subsequently, all grading judgment strings are compared field by field in a fixed field order. The string with the smaller preceding field is retained first. If the preceding fields are the same, the following fields are compared until the lexicographically ordered first result is obtained and written to the first-order table. The first-order table includes at least the patient identifier, round identifier, management input order identifier, and grading judgment string. The output is available for S3-3 to read. In case of anomalies or missing values, if there is a missing code in any difference order, the missing code is retained and included in the field comparison. For grading judgment strings with completely identical fields, the comparison is continued according to the field order of the management input order identifier, and the first result is retained. In S3-3, the purpose is to form candidate groups for disease severity level and candidate groups for management resistance level based on the first-order priority table, and retain candidate groups with consistent merging results through bidirectional consistency verification; the input is the first-order priority table; during processing, disease-side merging and management-side merging are first performed on each grading decision string. Among them, disease-side merging is based on the condition that the disease stabilization position is continuously moved forward and the execution of the discontinuous position is not earlier than the corresponding return position. Continuous forward movement means that the disease stabilization positions arranged according to the order of action are moved forward one by one or remain in their original positions without moving backward. Not earlier than the corresponding return position means that the position of the execution of the discontinuous position is greater than or equal to the position of the return position at the same position. Grading decision strings that meet this condition are merged into the same disease severity level candidate group. The merging condition for management resistance level is that the execution of discontinuous positions moves forward continuously and the interval between adjacent return positions does not expand. "The interval between adjacent return positions does not expand" means that the difference in position between adjacent return positions calculated according to the order of action occurrence remains constant. Classification decision strings that meet this condition are merged into the same management resistance level candidate group. After merging, a two-way consistency check is performed on each disease level candidate group and each management resistance level candidate group. Forward consistency check is performed in the order from the first to the last position of management input, and reverse consistency check is performed in the order from the last to the first position of management input. Candidate groups with the same forward consistency check result and reverse consistency check result are retained and written into the level candidate table. The level candidate table includes at least the patient identifier, round identifier, disease level candidate group identifier, management resistance level candidate group identifier, and management input order identifier. The output is read by S3-4. In case of abnormalities or missing information, classification decision strings that do not meet the merging condition are not written into the candidate group, and candidate groups with different forward consistency check results and reverse consistency check results are directly deleted. In S3-4, the goal is to establish a one-to-one correspondence between the disease level candidate groups and the management resistance level candidate groups in the grading candidate table, and to determine the unique coupled grading result through continuous rounds of connection consistency; the input is the grading candidate table; during processing, the disease level candidate groups and management resistance level candidate groups are first connected according to the same management input order. If there is only one disease level candidate group and one management resistance level candidate group under the same management input order, the connection relationship is directly established; if there are multiple disease level candidate groups or multiple management resistance level candidate groups under the same management input order, their corresponding management input order is deleted and the connection is re-executed. After each round of connection is completed, the connection result of that round is written to the connection temporary storage table. Starting from the second round, the connection result of this round is compared with the connection result of the previous round in order of disease severity level and management resistance level. Connection is stopped when the disease severity level and management resistance level order of the connection results of two consecutive rounds are completely consistent. The corresponding connection result is retained and written to the coupling grading table. The coupling grading table includes at least patient identifier, round identifier, disease severity level, management resistance level, and coupling grading identifier. The output is available for S4 to read. In case of abnormality or missing information, if the disease severity level candidate group or management resistance level candidate group is empty due to the deletion of the management input order, no coupling grading result is generated. If the result of any round during the continuous connection process is empty, the connection is terminated and an empty result identifier is written. This implementation method forms a coupled grading table with fixed fields, clear comparison rules, and unique results by using ranking comparison, difference calculation, grade merging, and corresponding linkage. This enables subsequent management plans to be generated based on a stable combination of disease severity and management resistance levels. In practical applications: For simulation results of patients with diabetes and hypertension, the disease stabilization position, intermittent execution position, and retrograde position under different management input sequences can be extracted first. A sequential control group is constructed and the results of the first and last closed verification are retained. Then, the stabilization difference sequence, intermittent difference sequence, and retrograde difference sequence are calculated to form a grading judgment string. Subsequently, the disease level candidate group and the management resistance level candidate group are merged respectively, and the corresponding attachments are executed under the same management input sequence. Finally, the coupled grading table corresponding to the current round of the patient is generated, which can be directly called when generating subsequent follow-up sequence, re-examination arrangement, reminder arrangement, supplementary sampling arrangement, and intervention arrangement.

[0020] S4. Based on the coupling grading table, generate follow-up order, re-examination arrangement, reminder arrangement, supplementary sampling arrangement and intervention arrangement according to the correspondence between disease level and management resistance level, and output management plan table; This implementation method is used to convert a coupled grading table into a management plan table that can be directly published and executed. The processing sequence is as follows: First, coupled items corresponding to the disease level and management resistance level are generated under the same patient and the same round. Then, follow-up items, revisit items, reminder items, supplementary data collection items, and intervention items are generated from the coupled items to form a candidate plan table. Finally, a unique management plan is retained through first-to-last correspondence verification and rearrangement. This implementation process includes the following steps: In S4-1, the goal is to organize the disease severity level and management resistance level in the coupling grading table into a coupling item table that can be directly called in subsequent schemes. The input is the coupling grading table. During processing, firstly, the disease severity level and management resistance level corresponding to the same patient in the current round are extracted according to the patient identifier and the current round identifier. Then, each disease severity level and each management resistance level are concatenated pairwise according to the field order to form a coupling item. The field order is: patient identifier, round identifier, disease severity level, and management resistance level. Subsequently, the generated coupling items are merged within the same round and merged within the same patient. Merging within the same round is used to retain coupling items with completely identical fields within the same round, and merging within the same patient is used to retain the unique coupling item for the same patient in the current round. After merging is completed, the results are written to the coupling item table, which includes at least the patient identifier, round identifier, disease severity level, management resistance level, and coupling item identifier. The output is read by S4-2. In case of anomalies or missing records, no coupling item is generated for records with empty disease severity level or management resistance level. If there are multiple coupling items with completely identical fields for the same patient in the same round, only one is retained. In S4-2, the purpose is to convert the grading results into a fixed-structured candidate table based on the coupling item table. The input is the coupling item table. During processing, follow-up items, re-examination items, reminder items, supplementary data collection items, and intervention items are written around each coupling item. The follow-up item includes at least the follow-up method, follow-up sequence, and follow-up content. The re-examination item includes at least the re-examination sequence and re-examination content. The reminder item includes at least the reminder recipient, reminder sequence, and reminder content. The supplementary data collection item includes at least the supplementary data collection recipient, supplementary data collection content, and supplementary data collection sequence. The intervention item includes at least the intervention recipient, intervention content, and intervention sequence. The specific content of each item is taken from the pre-configured grading scheme mapping table. The grading scheme mapping table is formed by the pre-set combination relationship of disease level and management resistance level. Subsequently, corresponding sequence codes are generated according to the field order of disease severity first and management resistance level second. The sequence code is formed by concatenating the values ​​of the disease severity field and the management resistance level field in sequence. Then, the follow-up items, re-visit items, reminder items, supplementary data collection items, and intervention items are arranged in the order of the sequence code to generate a candidate plan table. The candidate plan table includes at least the patient identifier, round identifier, sequence code, follow-up items, re-visit items, reminder items, supplementary data collection items, and intervention items. The output is available for S4-3 to read. In case of abnormality or missing items, no candidate plan is generated for coupled items that do not have a corresponding combination relationship in the graded plan mapping table. For multiple plan items corresponding to the same sequence code, they are arranged in the field order of follow-up items, re-visit items, reminder items, supplementary data collection items, and intervention items, and the first result is retained. In S4-3, the goal is to retain a management plan table with a closed structure and consistent execution order from the plan candidate table through head-to-tail correspondence verification and rearrangement. The input is the plan candidate table. During processing, head-to-tail correspondence verification is first performed on follow-up items, revisit items, reminder items, supplementary collection items, and intervention items corresponding to the same sequence number. The first item is the plan item that appears first under the same sequence number, and the last item is the plan item that appears last under the same sequence number. The head-to-tail correspondence verification ensures that the patient identifier, round identifier, sequence number, and execution order between plan items are consistent. Plan candidates that pass the head-to-tail correspondence verification are retained in their original order and written back. Candidate schemes that fail the corresponding verification are deleted and rearranged. After each round of arrangement, the arrangement results are written to a temporary table. Starting from the second round, the arrangement results of the current round are compared with the previous round's arrangement results item by item. Arrangement stops when the arrangement results of two consecutive rounds are completely consistent. The corresponding results are retained and written to the management scheme table. The management scheme table includes at least the patient identifier, round identifier, sequence code, follow-up item, re-visit item, reminder item, supplementary collection item, intervention item, and retention identifier. The output is available for S5 to read. In case of abnormality or missing items, candidate schemes with incomplete five categories of scheme items after deleting the corresponding items are directly deleted. Candidate schemes with empty arrangement results for two consecutive rounds are written with an empty scheme identifier. This implementation method generates a management scheme table with fixed fields, clear order, and unique results through coupling item generation, scheme candidate arrangement, and first-to-last correspondence verification, so that subsequent side release and feedback write-back have a unified input basis; In practical applications: For the coupled grading table of patients with diabetes and hypertension, the disease level and management resistance level corresponding to the current round can be extracted first and coupled items can be generated. Then, according to the grading scheme mapping table, follow-up items, re-visit items, reminder items, supplementary data collection items and intervention items are written and arranged by sequence code to form a scheme candidate table. Subsequently, the first and last correspondence of the five types of scheme items under the same sequence code is checked and rearranged to finally generate the management scheme table for the current round of the patient, which can be directly called when the doctor, patient and follow-up sides publish it.

[0021] S5. Based on the management plan table, issue corresponding arrangements to the doctor, patient and follow-up sides, collect execution confirmation results, supplementary collection and return results and subsequent response results, execute write-back according to patient and round identifiers, and output the next round record table; This implementation method converts a management plan table into a write-back record table for the next round. The processing sequence is as follows: First, the current round management plan is split and linked to the doctor's side, patient's side, and follow-up side. Then, corresponding arrangements are written according to the issuing objects, and feedback is collected. Subsequently, the feedback results are merged, assigned, and completed. Finally, the records are written to the starting record position of the next round and the remaining record position of the current round according to the write-back status, to ensure that the management execution results of the current round can enter the next round processing chain without ambiguity. This implementation process includes the following steps: In S5-1, the purpose is to organize the five types of plan items in the management plan table into a lateral release table according to the release target and the planned time sequence; the input is the management plan table; during processing, first extract the follow-up items, revisit items, reminder items, supplementary data collection items, and intervention items corresponding to the same patient in the current round according to the patient identifier and the current round identifier, and then attach each item to the doctor-side identifier, patient-side identifier, and follow-up-side identifier respectively. Among them, the doctor-side identifier is taken from the responsible doctor identifier corresponding to the current round of the patient, the patient-side identifier is taken from the patient's terminal account identifier, and the follow-up-side identifier is taken from the follow-up execution account identifier corresponding to the current round of the patient. Subsequently, the linked items are arranged in ascending order of planned time, and those with the same planned time are arranged in a fixed order according to the release item identifier, generating a lateral release table. The lateral release table includes at least the patient identifier, round identifier, release item identifier, doctor-side identifier, patient-side identifier, follow-up-side identifier, planned time, and release content, and is available for S5-2 to read. In case of abnormality or missing items, release items that are missing any side identifier are not deleted from the original plan item, but the release item identifier is retained and an empty side identifier is written. Release items with empty planned time are arranged after the release items of the same patient's current round according to the generation order in the management plan table. In S5-2, the objective is to complete lateral writing based on the lateral release table and form a lateral feedback table that corresponds one-to-one with the original release items; the input is the lateral release table; during processing, based on the lateral release table, follow-up and intervention items are written to the doctor's side, reminder and supplementary data collection items are written to the patient's side, and follow-up and supplementary data collection items are written to the follow-up side, while retaining the original release item identifier for each written item; subsequently, the returned content from each side is received, and the same-item correspondence is performed according to the original release item. The same-item correspondence first compares the release item identifier, then compares the patient identifier and the round identifier, and so on. When the results match, a corresponding relationship is established. If multiple responses are received for the same publication item, they are arranged in ascending order of result time and all are retained. After the correspondence is completed, a lateral feedback table is generated. The lateral feedback table includes at least the patient identifier, round identifier, publication item identifier, feedback side identifier, result time, result content, and result category, and is available for S5-3 to read. When handling anomalies or missing information, for publication items that have not received a response, the original publication item identifier is retained and an empty feedback identifier is written. If the response content cannot match the original publication item identifier, it is not written to the lateral feedback table. In S5-3, the goal is to organize the feedback results in the lateral feedback table into a round feedback table that can be directly written back. The input is the lateral feedback table. During processing, the execution confirmation results, supplementary collection feedback results, and subsequent response results are first merged according to the patient identifier, round identifier, and publication item identifier to form a set of feedback results under the same publication item. Subsequently, the merged feedback results are used to establish the receiving correspondence and the completion correspondence. The receiving correspondence is established when the result time and the planned time of the corresponding publication item are connected end to end. End to end means that the result time is later than or equal to the planned time and there are no other result times of the same publication item between them. The completion correspondence is established when the result content is consistent with the content of the corresponding publication item. The consistency judgment criteria are that the result content fields and the publication content fields are identical field by field. When the same feedback result satisfies both the acceptance and completion criteria, it is marked as an acceptance-completed result. When only one of them is satisfied, it is marked according to the corresponding status. After processing is completed, a round feedback table is generated. The round feedback table includes at least the patient identifier, round identifier, publication item identifier, result time, result content, acceptance identifier, and completion identifier, and is available for S5-4 to read. In case of abnormal or missing information, feedback results with empty result content are not included in the completion identifier, and feedback results with empty result time are not included in the acceptance identifier, but the original feedback record is retained and an incomplete identifier is written. In S5-4, the purpose is to write the feedback results into the starting record position of the next round and the remaining record position of the current round according to the round feedback table, forming a record table for the next round that can be read again later; the input is the round feedback table; during processing, each feedback result in the round feedback table is read first, and for the feedback results that have been established and completed, they are written back to the starting record position of the next round according to the patient identifier and round identifier. The starting record position of the next round is the first feedback writing position of the corresponding patient in the record table of the next round. The written content includes the patient identifier, the next round identifier, the original publication item identifier, the result time, and the result content. For feedback results that have not been established or have not been completed, write them back to the current round's remaining record position according to the original publication item identifier. The current round's remaining record position is taken from the remaining field position of the corresponding publication item in the current round's record table. The written content includes the patient identifier, current round identifier, original publication item identifier, result time, result content, and incomplete identifier. After all write-backs are completed, the next round's record table is generated. The next round's record table includes at least the patient identifier, round identifier, record source identifier, record start time, record end time, and record content item, and is available for subsequent S1 reading. In case of abnormal or missing data, if there are multiple feedback results at the same write position, they are written in ascending order of result time, without deletion. For feedback results whose next round identifier has not yet been generated, write them to the current round's remaining record position first, and then write them to the next round's start record position after the next round identifier is generated. This implementation method forms a complete write-back link from the management plan table to the record table of the next round through side release, same-item feedback, round merging and position write-back, so that the execution result of the current round can stably enter the processing flow of the next round according to the established state; In practical applications: For follow-up items, re-examination items, reminder items, supplementary data collection items, and intervention items corresponding to the current round for patients with diabetes and hypertension, the responsible doctor's account, patient terminal account, and follow-up execution account can be linked separately to form a lateral release table. Then, re-examination items and intervention items are written to the doctor's side, reminder items and supplementary data collection items are written to the patient's side, and follow-up items and supplementary data collection items are written to the follow-up side. Subsequently, the execution confirmation results, supplementary data collection feedback results, and subsequent response results returned from each side are collected. Feedback results that meet both the corresponding acceptance and completion conditions are written to the starting record position of the next round, and feedback results that do not meet any corresponding conditions are written to the remaining record position of the current round. Finally, the record table of the next round is generated for direct reading by subsequent steps.

[0022] Furthermore, the present invention also includes a chronic disease hierarchical management system based on quantum simulation, the system comprising a data acquisition module, a quantum simulation calculation module, a hierarchical decision-making module, a scheme generation module, and a terminal interaction module: The data acquisition module is used to collect the target patient's genomic data, physiological monitoring data, clinical diagnosis and treatment data, lifestyle data, terminal response records, task completion records, follow-up visit records, device online records, and execution deviation records in the current round. It performs merging and alignment according to the patient identifier, round identifier, and record time sequence, and outputs the current patient table. The quantum simulation calculation module is used to extract the disease record group and management record group based on the current patient table and the corresponding historical records, jointly encode the disease record group and management record group into a quantum simulation input sequence, and perform quantum simulation operation on different management input orders corresponding to the same disease record group to solve the disease stabilization position, execution discontinuity position and backpropagation position corresponding to each management input order, and output the simulation result table. The hierarchical decision-making module is used to compare the order of the disease stabilization position, the intermittent execution position, and the feedback position corresponding to each management input order based on the simulation result table, determine the current patient's disease level and management resistance level, and output a coupled hierarchical table. The plan generation module is used to generate follow-up order, re-examination arrangement, reminder arrangement, supplementary sampling arrangement and intervention arrangement based on the correspondence between disease level and management resistance level, and output management plan table; The terminal interaction module is used to combine the management plan table to publish corresponding arrangements to the doctor, patient and follow-up side, collect execution confirmation results, supplementary collection and transmission results and subsequent response results, execute write-back according to patient identifier and round identifier, and output the next round record table.

[0023] Working Principle: This scheme first organizes the patient's genomic data, physiological monitoring data, clinical diagnosis and treatment data, lifestyle data, and records of terminal response, task completion, follow-up visits, equipment online status, and execution deviations into a unified current patient table within the current management round. Then, it separates the patient condition record group reflecting changes in the condition and the management record group reflecting the management implementation process. Subsequently, different management input sequences are developed around the same patient condition record group, and the condition position, management action, action interval, and feedback correlation are encoded into quantum simulation input sequences. Quantum simulation is used to deduce the condition stabilization position, execution discontinuity position, and feedback position corresponding to each management input sequence. Next, based on these position results, a ranking comparison is performed to obtain the condition level and management resistance level. The two are then coupled to generate a management plan consisting of follow-up, re-visit, reminders, supplementary sampling, and intervention. Finally, the plan is distributed to the doctor, patient, and follow-up sides, and the execution confirmation, supplementary sampling feedback, and subsequent response are written back to the next round's record, forming a complete process of data organization, quantum simulation, coupling and grading, plan generation, and feedback writing. For example, in a community-based chronic disease management scenario, a patient with diabetes and hypertension will continuously generate various records, including blood glucose, blood pressure, follow-up visits, medication reminder responses, and follow-up completion status. The system first organizes these scattered records into the same management cycle, and then determines how different management input sequences will lead to disease recovery and execution interruption results for this patient under similar disease conditions. If the quantum simulation results show that although the patient's disease level is high, they can stabilize earlier and have continuous data transmission under a certain management input sequence, the system will provide corresponding follow-up visit arrangements, reminder arrangements, and supplementary data transmission arrangements. If the results show that the patient faces high management resistance, such as frequent execution interruptions or delayed data transmission, the system will increase the frequency of follow-up visits and supplementary data transmission requirements in the plan. In this way, doctors can see not only the severity of the disease, but also the difficulty of managing the patient, thus more accurately arranging subsequent management.

[0024] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for hierarchical management of chronic diseases based on quantum simulation, characterized in that, include: S1. Collect the target patient's genomic data, physiological monitoring data, clinical diagnosis and treatment data, lifestyle data, terminal response records, task completion records, follow-up visit records, device online records, and execution deviation records in the current round. Perform merging and alignment according to patient identifier, round identifier, and record time sequence, and output the current patient table. S2. Based on the current patient table and corresponding historical records, extract the disease record group and management record group, jointly encode the disease record group and management record group into a quantum simulation input sequence, and perform quantum simulation operation on different management input sequences corresponding to the same disease record group to solve the disease stabilization position, execution discontinuity position and backpropagation position corresponding to each management input sequence, and output the simulation result table. S3. Based on the simulation results table, perform a ranking comparison on the disease stability position, intermittent execution position and feedback position corresponding to each management input order, determine the current patient's disease level and management resistance level, and output the coupling grading table. S4. Based on the coupling grading table, generate follow-up order, re-examination arrangement, reminder arrangement, supplementary sampling arrangement and intervention arrangement according to the correspondence between disease level and management resistance level, and output management plan table; S5. Based on the management plan table, issue corresponding arrangements to the doctor, patient and follow-up sides, collect execution confirmation results, supplementary collection and return results and subsequent response results, execute write-back according to patient and round identifiers, and output the next round record table.

2. The method for hierarchical management of chronic diseases based on quantum simulation according to claim 1, characterized in that: S1 includes: S1-1. Separate the genomic data, physiological monitoring data, clinical diagnosis and treatment data, lifestyle data, terminal response records, task completion records, follow-up visit records, device online records, and execution deviation records into raw record units with patient identifiers, record source identifiers, record start time, record end time, and record content items. Collect the data according to the patient identifiers and arrange them in the execution order according to the record start time, and output the patient raw record table. S1-2. Based on the start time and end time of the current round corresponding to the round identifier, perform round truncation on each original record unit in the patient's original record table. Record items that fall within the current round range are retained as round-in-round record items, and record items that cross the current round boundary are divided into round-in-round retained items. Output the current round record table. S1-3. For record items in different rounds that have the same time position in the current round record table, perform same-time merging according to the record source identifier and record time sequence. Perform forward inheritance and reverse verification for record items in rounds with continuous coverage relationship. For record items in rounds without continuous coverage relationship, retain the corresponding source position and generate the current patient table containing disease record items and management record items.

3. The method for hierarchical management of chronic diseases based on quantum simulation according to claim 2, characterized in that: S2 includes: S2-1. Based on the current patient table and corresponding historical records, extract the disease record items and management record items of the current round and adjacent historical rounds according to the patient ID, round ID and record time sequence. Perform same-time correspondence for record items with the same record start time, and perform succession correspondence for record items whose record end time is connected to the start time of the next record. Preserve the correspondence results of continuous disease change direction and complete start and end of management function, and generate disease record group, management record group and group correspondence table. S2-2. Based on the group correspondence table, each management record group is divided into single action record items, arranged in ascending order of the start time of the record. For adjacent single action record items where the end time of the previous single action record item and the start time of the next single action record item are connected end to end, a connecting edge is established to generate the complete management input order from the first single action record item to the last single action record item. For each management input order, the number of action steps, the number of disease reversals, and the number of interruptions are calculated in sequence. The field-by-field comparison is performed according to the field arrangement order of the number of action steps, the number of disease reversals, and the number of interruptions. The management input order with the first lexicographical order is retained to generate a candidate order table.

4. The method for hierarchical management of chronic diseases based on quantum simulation according to claim 3, characterized in that: S2 also includes: S2-3. Based on the medical record group, management record group and candidate order table, write the medical location item, management action item, action interval item and feedback association item into the coding position according to the fixed field order to generate the initial coding sequence. Perform consistency check on the position of the same field in each initial coding sequence, fill in the missing positions of the field, and retain the coding result of the field with the earlier order for the conflicting positions to generate the quantum simulation input sequence table. S2-4. For the quantum simulation input sequence list, perform quantum simulation deduction for each candidate management input order corresponding to the same disease record group. Solve the disease stabilization position, execution discontinuity position and return position step by step according to the order of management action occurrence. Compare the results of this round of deduction with the results of the previous round of deduction item by item according to the field order of disease stabilization position, execution discontinuity position and return position. Stop the deduction when the results of two consecutive rounds of comparison are completely consistent, retain the corresponding deduction results, and generate a simulation result table.

5. The method for hierarchical management of chronic diseases based on quantum simulation according to claim 4, characterized in that: S3 includes: S3-1. Based on the simulation results table, extract the management input order, condition stabilization position, execution discontinuity position and feedback position corresponding to the same patient in the current round. Construct a sequential control group according to the order of action occurrence within the management input order. Perform a first-to-last closure check on the condition stabilization position, execution discontinuity position and feedback position in the same management input order. Retain the sequential control group where the first position can be pointed back to the last position and the last position can be reversed to the first position. Generate a sequential check table. S3-2. Based on the ranking check table, calculate the stabilization difference, discontinuity difference, and return difference for each ranking control group. The stabilization difference is composed of the position difference between adjacent stabilization positions, the discontinuity difference is composed of the position difference between adjacent discontinuity positions, and the return difference is composed of the position difference between adjacent return positions. Then, connect the stabilization difference, discontinuity difference, and return difference for each ranking control group into a grading judgment string according to a fixed field order, and perform field-by-field comparison according to the fixed field order to generate the first-rank table.

6. The method for hierarchical management of chronic diseases based on quantum simulation according to claim 5, characterized in that: S3 also includes: S3-3. Based on the first-order table, perform disease-side merging and management-side merging on each grading judgment string. Among them, grading judgment strings with continuous forward shift of disease stability position and execution discontinuity position not earlier than the corresponding return position are merged into the same disease level candidate group. Grading judgment strings with continuous forward shift of execution discontinuity position and no expansion of the interval between adjacent return positions are merged into the same management resistance level candidate group. Perform bidirectional consistency verification on each disease level candidate group and each management resistance level candidate group, retain the candidate groups with the same forward merging result and reverse merging result, and generate a grading candidate table. S3-4. For the candidate grading table, link the candidate disease grade and the candidate management resistance grade according to the same management input order. For candidate groups that cannot be linked one by one, delete their corresponding management input order and re-link them. Stop linking when the disease grade order and management resistance grade order of the linking results are completely consistent in two consecutive rounds, retain the corresponding linking results, and generate a coupled grading table.

7. The method for hierarchical management of chronic diseases based on quantum simulation according to claim 6, characterized in that: S4 includes: S4-1. Based on the coupling grading table, extract the disease level and management resistance level of the same patient in the current round. Concatenate each disease level and each management resistance level in pairs according to the field order to form coupling items. Perform same-round merging and same-patient merging on each coupling item to generate a coupling item table. S4-2. Based on the coupling item table, write follow-up items, re-examination items, reminder items, supplementary data collection items, and intervention items for each coupling item. Generate corresponding sequence codes according to the field order of disease severity first and management resistance level second, and arrange the follow-up items, re-examination items, reminder items, supplementary data collection items, and intervention items in the order of execution according to the sequence codes to generate a candidate plan table. S4-3. For the candidate plan table, perform a head-to-tail check on the follow-up items, re-examination items, reminder items, supplementary collection items, and intervention items corresponding to the same sequence number. For candidate plans that pass the check, retain the original order and write them in. For candidate plans that fail the check, delete the corresponding items and rearrange them. Stop the arrangement when the results of two consecutive rounds of arrangement are completely consistent, retain the corresponding results, and generate the management plan table.

8. The method for hierarchical management of chronic diseases based on quantum simulation according to claim 7, characterized in that: S5 includes: S5-1. Based on the management plan table, extract the follow-up items, re-visit items, reminder items, supplementary data collection items and intervention items corresponding to the same patient in the current round. Then, attach each item to the doctor's side icon, the patient's side icon and the follow-up side icon respectively, arrange them in ascending order according to the planned time, and generate a lateral release table. S5-2. Based on the lateral release table, write follow-up visit and intervention items to the doctor's side, reminder and supplementary data collection items to the patient's side, and follow-up and supplementary data collection items to the follow-up side. Then, perform the same item correspondence for the returned content from each side according to the original release items to generate a lateral feedback table.

9. The method for hierarchical management of chronic diseases based on quantum simulation according to claim 8, characterized in that: S5 also includes: S5-3. Based on the lateral feedback table, the execution confirmation results, supplementary collection and feedback results and subsequent response results are merged according to the patient identifier, round identifier and release item identifier. A connection is established for feedback results whose result time is connected to the corresponding release item plan time. A completion correspondence is established for feedback results whose result content is consistent with the corresponding release item content. A round feedback table is generated. S5-4. For the round feedback table, write the established and completed feedback results back to the starting record position of the next round according to the patient and round identifiers. Write the feedback results that have not been established or have not been completed back to the remaining record position of the current round according to the original publication item identifier, and generate the next round record table.

10. A quantum simulation-based chronic disease hierarchical management system, used to implement the quantum simulation-based chronic disease hierarchical management method according to any one of claims 1-9, the system comprising a data acquisition module, a quantum simulation calculation module, a hierarchical decision-making module, a scheme generation module, and a terminal interaction module, characterized in that: The data acquisition module is used to collect the target patient's genomic data, physiological monitoring data, clinical diagnosis and treatment data, lifestyle data, terminal response records, task completion records, follow-up visit records, device online records, and execution deviation records in the current round. It performs merging and alignment according to the patient identifier, round identifier, and record time sequence, and outputs the current patient table. The quantum simulation calculation module is used to extract the disease record group and management record group based on the current patient table and the corresponding historical records, jointly encode the disease record group and management record group into a quantum simulation input sequence, and perform quantum simulation operation on different management input orders corresponding to the same disease record group to solve the disease stabilization position, execution discontinuity position and backpropagation position corresponding to each management input order, and output the simulation result table. The hierarchical decision-making module is used to compare the order of the disease stabilization position, the intermittent execution position, and the feedback position corresponding to each management input order based on the simulation result table, determine the current patient's disease level and management resistance level, and output a coupled hierarchical table. The plan generation module is used to generate follow-up order, re-examination arrangement, reminder arrangement, supplementary sampling arrangement and intervention arrangement based on the correspondence between disease level and management resistance level, and output management plan table; The terminal interaction module is used to combine the management plan table to publish corresponding arrangements to the doctor, patient and follow-up side, collect execution confirmation results, supplementary collection and transmission results and subsequent response results, execute write-back according to patient identifier and round identifier, and output the next round record table.

Images