A method and system for blood potassium management and monitoring based on multivariate inputs

Abstract

The application relates to the technical field of medical health data processing, and discloses a blood potassium management and monitoring method and system based on multivariate input, which comprises the following steps: firstly, obtaining demographic data, medical history data, test data, medication data and medication adjustment event data of a patient, and performing pretreatment; then, based on the pretreated data, determining whether to enter a potassium supplement path or a potassium reduction path, and performing risk stratification on the blood potassium state of the patient, and outputting a risk stratification result and a treatment strategy; then, according to the determined path and the risk stratification result, performing corresponding medication decision to generate a medication adjustment scheme, and based on the medication adjustment scheme, generating a review suggestion for the patient; finally, showing the analysis result containing the risk stratification result, the treatment strategy, the medication adjustment scheme and the review suggestion to a user end, and recording an operation log. The application meets the clinical demand for fine, individualized and whole-process management of blood potassium.

Description

Technical Field

[0001] This invention relates to the field of medical and health data processing technology, specifically to a method and system for managing and monitoring blood potassium based on multivariate input. Background Technology

[0002] Abnormal blood potassium levels (hypokalemia or hyperkalemia) are common electrolyte disturbances in clinical settings such as cardiovascular disease and kidney disease. They can lead to adverse outcomes such as arrhythmia and muscle weakness, and in severe cases, even endanger life. Clinically, blood potassium levels are influenced by a complex array of factors, and their management typically requires a comprehensive assessment considering factors such as the patient's renal function, comorbidities, drug interactions, and concurrent medications.

[0003] Existing solutions often utilize information systems in the form of low / high potassium alerts, potassium supplementation calculators, or high potassium risk predictions. However, these tools generally have limitations: they often only provide single-factor alerts or static dose estimates, lacking interpretable, personalized recommendations that account for individual differences. More importantly, existing tools typically only cover a single aspect of potassium management, failing to form a closed-loop system from dynamic monitoring and risk stratification to medication prescription recommendations. Furthermore, low and high potassium states can clinically transition between each other.

[0004] Therefore, existing solutions cannot meet the clinical needs for refined, individualized, and comprehensive management of blood potassium levels. Summary of the Invention

[0005] This invention provides a method and system for managing and monitoring blood potassium based on multivariate input, in order to solve the problem that existing solutions cannot meet the clinical needs for refined, individualized, and full-process management of blood potassium.

[0006] In a first aspect, the present invention provides a method for managing and monitoring blood potassium based on multivariate input, the method comprising: Acquire patients' demographic data, medical history data, laboratory data, medication data, and medication adjustment event data, and perform preprocessing. Based on the preprocessed data, it is determined whether to enter the potassium supplementation or potassium reduction pathway, and the patient's blood potassium status is risk-stratified, and the risk stratification results and treatment strategies are output. Based on the determined path and the risk stratification results, corresponding medication decisions are made to generate a medication adjustment plan; Based on the medication adjustment plan, follow-up examination recommendations for the patient are generated; The system displays the analysis results, including the risk stratification results, the processing strategy, the medication adjustment plan, and the follow-up recommendations, to the user and records the operation log.

[0007] The aforementioned approach achieves end-to-end, individualized management of abnormal blood potassium levels by integrating multi-source data, automatic risk stratification, two-way medication decision-making (potassium supplementation and potassium reduction), and dynamic follow-up recommendations. This method can perform comprehensive calculations based on various input variables (such as demographic data, medical history data, laboratory data, and medication adjustment events) to output interpretable and specific medication adjustment plans and follow-up schedules. This addresses the lack of individualized recommendations and closed-loop linkage in single-point tools in related technologies, contributing to improved timeliness, safety, and standardization of clinical decision-making, and creating traceable management records.

[0008] In one optional implementation, when it is determined that the potassium supplementation pathway has been entered, the step of executing the corresponding medication decision to generate a medication adjustment plan includes: Identify a set containing multiple pre-defined influencing factors; The pre-configured scoring function is invoked to quantify and score each factor in the set, and a comprehensive score is calculated. Based on the comprehensive score and patient classification, the pre-configured dose mapping table is queried to obtain the baseline potassium supplementation plan; Based on at least one of the patient's current potassium supplementation status, the timeliness of key test data, or special clinical conditions, the baseline potassium supplementation regimen is modified to generate the final medication adjustment plan.

[0009] The above scheme achieves the quantification and individualization of potassium supplementation schemes through multi-factor scoring and mapping table lookup, making potassium supplementation decisions more interpretable and accurate.

[0010] In one alternative implementation, the influencing factors include renal function stratification, current serum potassium level, and the type of medication adjustment event for drugs that affect serum potassium.

[0011] The above scheme clarifies the core influencing factors of potassium supplementation calculation, ensuring the comprehensiveness and clinical relevance of model input.

[0012] In one alternative implementation, the specific clinical condition includes the patient being on dialysis or having severe renal impairment.

[0013] The above-mentioned plan has been modified for specific clinical conditions, enhancing its safety and applicability in complex cases.

[0014] In one optional implementation, when it is determined that the potassium-lowering pathway has been entered, the step of executing the corresponding medication decision to generate a medication adjustment plan includes: Determine if the patient is already using potassium-lowering medication; If no potassium is used and the current blood potassium level reaches the pre-configured starting threshold, then the initial potassium-lowering drug regimen is output as the drug adjustment regimen. If already used, the system will output recommendations for maintaining, increasing, decreasing, or discontinuing the medication, based on the comparison between the current serum potassium level and the target range.

[0015] The above approach provides a clear logic for the use of potassium-lowering drugs and supports dynamic adjustment of the treatment plan based on serum potassium levels.

[0016] In one alternative implementation, generating follow-up recommendations for the patient includes: Based on the risk stratification results and the medication adjustment plan, at least one predefined review rule is triggered. When multiple review rules are triggered, the review rules for the same test item will be merged, and the final review cycle will be determined from the suggested cycles of each rule based on the preset clinical cycle set, so as to generate a unified review recommendation.

[0017] The above scheme enables intelligent merging and periodic adjudication of follow-up recommendations under multiple rule triggers, making the follow-up plan more in line with the actual clinical pace.

[0018] In one optional implementation, the acquisition and preprocessing of the patient's demographic data, medical history data, laboratory data, medication data, and medication adjustment event data includes: The acquired data is subjected to unit standardization, timestamp validity verification, and outlier identification. Derived variables are calculated based on the test data, including renal function stratification and / or serum potassium change trends; The medication adjustment event data is identified based on the medication data; the medication adjustment event data includes the initiation, dosage increase, dosage decrease, or discontinuation of medication.

[0019] The above scheme ensures the quality and consistency of input data through standardized data preprocessing, laying the foundation for accurate subsequent analysis.

[0020] In one alternative implementation, the rules used for risk stratification of the patient's blood potassium status include: Compare the blood potassium level with preset critical low threshold, low threshold, high threshold and critical high threshold; The risk level is determined by comprehensively considering the patient's treatment context, renal function status, and specific medication status.

[0021] The above scheme combines multi-dimensional information for risk stratification, which can more accurately identify high-risk patients and suggest appropriate treatment.

[0022] In one optional implementation, displaying the analysis results to the user, including the risk stratification results, the treatment strategy, the medication adjustment plan, and the follow-up recommendations, includes: The system displays the basis for generating the medication adjustment plan and the triggering rules for the follow-up recommendations to healthcare users, and provides an interactive interface for manually confirming or modifying the analysis results. Present simplified risk stratification results, treatment strategies, medication adjustment plans, and follow-up reminders to patient users.

[0023] The above solution differentiates the content displayed to different user terminals, which not only meets the professional decision-making needs of medical staff, but also makes it easier for patients to understand and implement.

[0024] Secondly, the present invention provides a blood potassium management and monitoring system based on multivariate input, for implementing the method described above, the system comprising: The data acquisition module is used to acquire patients' demographic data, medical history data, laboratory data, medication data, and medication adjustment event data; The data preprocessing module is used to preprocess the demographic data, medical history data, test data, medication data, and medication adjustment event data. The decision analysis engine is used to determine whether to enter the potassium supplementation or potassium reduction path based on the preprocessed data, and to perform risk stratification of the patient's blood potassium status, outputting the risk stratification results and treatment strategies. The medication recommendation generation module is used to make corresponding medication decisions based on the determined path to generate a medication adjustment plan; The follow-up examination suggestion generation module is used to generate follow-up examination suggestions for patients based on the medication adjustment plan; The interactive display and review module is used to display the analysis results, including the risk stratification results, the processing strategies, the medication adjustment plans, and the review suggestions, to the user. The logging and auditing module is used to record operation logs.

[0025] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform a method for managing and monitoring blood potassium based on multivariate input as described in the first aspect or any corresponding embodiment.

[0026] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute a method for managing and monitoring blood potassium based on multivariate inputs as described in the first aspect or any corresponding embodiment.

[0027] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to execute a multivariate input-based method for managing and monitoring blood potassium, as described in the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0028] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0029] Figure 1 This is a schematic diagram of an application scenario according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the first process of a method for managing and monitoring blood potassium based on multivariate input according to an embodiment of the present invention; Figure 3 This is a second flowchart illustrating a method for managing and monitoring blood potassium based on multivariate input according to an embodiment of the present invention; Figure 4 This is a structural block diagram of a blood potassium management and monitoring system based on multivariate input according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0031] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0032] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0033] As an optional application scenario of this invention, such as Figure 1 As shown, the system architecture of this multivariate input-based blood potassium management and monitoring system may include at least one terminal device and at least one server. Figure 1 The system architecture shown in the example includes a computer 101, a mobile terminal 102, and a server 103, and terminal devices such as the computer 101 and the mobile terminal 102 are connected to the server 103 through a network 110.

[0034] Specifically, the terminal device can be a smartphone, tablet, laptop, PDA, desktop computer, game console, smart TV, smart wearable device, in-vehicle terminal, VR (Virtual Reality) device, AR (Augmented Reality) device, etc. Server 103 can be a standalone physical server, a server cluster, a distributed system, or a cloud server providing cloud services. Network 110 can be a wired or wireless network, examples of which include, but are not limited to, the Internet, corporate intranet, local area network, wide area network, mobile communication network, and combinations thereof.

[0035] According to an embodiment of the present invention, a method for managing and monitoring blood potassium based on multivariate input is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0036] This embodiment provides a method for managing and monitoring blood potassium based on multivariate input, which can be used on the aforementioned mobile terminals, such as mobile phones, tablets, etc. (the executing entity is described in conjunction with the actual situation). Figure 2 This is a first flowchart of a method for managing and monitoring blood potassium based on multivariate input according to an embodiment of the present invention, as shown below. Figure 2 As shown, the process includes the following steps: Step S201: Obtain the patient's demographic data, medical history data, laboratory data, medication data, and medication adjustment event data, and perform preprocessing.

[0037] Furthermore, this embodiment actively acquires multi-source heterogeneous information related to patient potassium management from data sources such as Hospital Information System (HIS), Laboratory Information System (LIS), or patient follow-up platform. This mainly includes: demographic data (at least including gender and age), medical history data (at least including comorbidities), laboratory data (at least including serum potassium, creatinine, estimated glomerular filtration rate (eGFR) and corresponding test timestamps), medication data (currently used drug names, dosage, and administration), and medication adjustment event data (such as recent diuretic, renin-angiotensin-aldosterone inhibitor, and other drug initiation, dosage increase, reduction, or discontinuation events and their occurrence times). After acquiring the raw data, preprocessing is necessary to ensure the accuracy and reliability of subsequent analysis. Preprocessing may include data cleaning and standardization, such as unifying units for test results, verifying the validity of timestamps, and identifying and handling missing or extreme outliers. In addition, this embodiment will also calculate key derived variables based on the raw data, such as calculating eGFR and stratifying renal function based on creatinine and demographic information, or analyzing recent trends in serum potassium levels. This series of preprocessing operations transforms messy, raw medical data into structured, standardized, high-quality input.

[0038] Step S202: Based on the preprocessed data, determine whether to enter the potassium supplementation path or the potassium reduction path, perform risk stratification of the patient's blood potassium status, and output the risk stratification results and treatment strategies.

[0039] Furthermore, firstly, based on preprocessed data, particularly the current serum potassium level, this embodiment performs initial logical judgments to determine the general direction of management: if serum potassium is below the preset clinical target range, it determines to enter the potassium supplementation path, aiming to increase serum potassium; if serum potassium is above the target range, it enters the potassium-lowering path, aiming to reduce serum potassium. Simultaneously, this embodiment also uses a set of configurable rules to rapidly and automatically assess the risk level of the patient's current serum potassium status. The main basis for risk stratification includes: the absolute value of serum potassium level, preset multi-level thresholds (such as critically low, low, high, and critically high), the patient's treatment scenario (in-hospital or out-of-hospital), and renal function status. For example, if serum potassium is below the critically low value or above the critically high value, regardless of other factors, the system will directly determine it as high risk and recommend immediate emergency assessment; if serum potassium is in the low or high range, it may be determined as medium risk based on a comprehensive assessment of factors such as renal function. The results of risk stratification (e.g., low, medium, and high risk levels) and corresponding treatment recommendations (e.g., immediate emergency care, prompt outpatient visit, and routine follow-up) will be clearly output. This step quantifies complex clinical situations into clear risk levels, providing not only intuitive early warnings for medical staff but also core decision-making basis for determining the specific management path to follow.

[0040] Step S203: Based on the determined path and the risk stratification result, execute the corresponding medication decision to generate a medication adjustment plan.

[0041] Furthermore, the input for this step is explicitly the risk stratification result of the determined path (potassium supplementation or potassium reduction). If the potassium supplementation path is entered, the system typically invokes an interpretable quantitative model (such as a multi-factor scoring model). This model selects a series of preset influencing factors (such as renal function stratification, current degree of potassium deficiency, and recent adjustment events of medications affecting potassium levels), quantifies and scores each factor using pre-configured rules, and calculates a comprehensive score to characterize the overall potassium supplementation requirement. Subsequently, based on this score and patient classification (such as whether the patient has heart failure), the corresponding dose mapping table is queried to generate a baseline potassium supplementation plan. This embodiment also makes corrections based on the urgency of the risk stratification prompts (such as high risk potentially indicating a preference for intravenous potassium supplementation) and the patient's special circumstances (such as existing potassium supplementation medications or the timeliness of testing), ultimately outputting a medication adjustment plan that includes specific drugs, dosages, and frequencies. If the potassium reduction path is entered, the system mainly relies on a rule base based on clinical guidelines for logical reasoning, matching and outputting recommendations for starting, increasing, decreasing, or discontinuing medication based on the current potassium level, whether potassium-lowering medications are already in use, and renal function status. This step transforms clinical knowledge and rules into a structured computational process, generating medication recommendations that are precisely matched to the patient's specific pathway and risk status.

[0042] Step S204: Based on the medication adjustment plan, generate follow-up recommendations for the patient.

[0043] Furthermore, this embodiment intelligently triggers corresponding follow-up rules based on multiple pieces of information, including medication adjustment plans (such as adding or adjusting a drug that affects serum potassium), risk stratification results, and the patient's current renal function status. For example, for a hypokalemic patient who has just increased their diuretic dosage, the system will trigger a shorter follow-up cycle from the perspective of adverse drug reaction monitoring; simultaneously, their current hypokalemic state itself will also trigger a follow-up cycle based on critical values. When multiple follow-up rules are triggered simultaneously, the system will intelligently adjudicate and merge them. For example, it will merge rules targeting the same test items (such as serum potassium and renal function) and select the shortest cycle or the cycle that best fits the clinically commonly used follow-up rhythm (such as 1 day, 3 days, 1 week, etc.) from the suggested cycles of each rule, ultimately generating a unified, reasonable, and executable follow-up plan. This ensures that the follow-up recommendations are dynamic, individualized, and closely related to treatment interventions.

[0044] Step S205: Display the analysis results, including the risk stratification results, the treatment strategy, the medication adjustment plan, and the follow-up recommendation, to the user terminal, and record the operation log.

[0045] Furthermore, this embodiment integrates all the analysis results generated in the aforementioned steps, including risk level (i.e., risk stratification results), treatment recommendations (i.e., treatment strategies, such as immediate medical attention), specific medication adjustment plans and their generation basis, and detailed follow-up recommendations, into a complete report, which is displayed to the user through an interactive interface. The presentation is designed differently for different users: for healthcare professionals, the system provides detailed decision support information, including the calculation basis of the plan, the triggered rules, etc., and provides an interactive interface for manual confirmation, modification, or rejection, fully respecting the final decision-making power of clinicians; for patients, it provides simplified and easy-to-understand treatment guidance and follow-up reminders. Regardless of whether the results are modified, the system will fully record the input data, rule hit chain, initial output results, and any manual operations (including modification content and reasons) of this calculation through the log and audit modules, thereby achieving full-process traceability and quality control. This ensures both the flexibility of clinical applications and meets the needs of medical quality management and continuous improvement.

[0046] This embodiment provides a multivariate input-based method for managing and monitoring blood potassium levels. By integrating multi-source data, automatic risk stratification, two-way medication decision-making (potassium supplementation and potassium reduction), and dynamic follow-up recommendations, it achieves end-to-end, individualized management of abnormal blood potassium levels. This method can perform comprehensive calculations based on multiple input variables (such as test data, medication adjustment events, etc.) to output interpretable and specific medication adjustment plans and follow-up schedules. This solves the problem of single-point tools lacking individualized recommendations and closed-loop linkage in related technologies, helping to improve the timeliness, safety, and standardization of clinical decision-making, and creating traceable management records.

[0047] This embodiment provides a method for managing and monitoring blood potassium based on multivariate input, which can be used on the aforementioned mobile terminals, such as mobile phones and tablets. Figure 3 This is a second flowchart of a method for managing and monitoring blood potassium based on multivariate input according to an embodiment of the present invention, as shown below. Figure 3 As shown, the process includes the following steps: Step S301: Obtain the patient's demographic data, medical history data, laboratory data, medication data, and medication adjustment event data, and perform preprocessing.

[0048] In one optional implementation, step S301 includes: The acquired data is subjected to unit standardization, timestamp validity verification, and outlier identification. Based on the test data, derived variables were calculated, including renal function stratification and / or serum potassium change trends; The medication data is used to identify medication adjustment events; these events include the initiation, dosage increase, dosage decrease, or discontinuation of the medication.

[0049] Furthermore, this step is the initial stage of system operation, aiming to collect key information required for potassium management from multi-source heterogeneous data interfaces such as hospital information systems, electronic medical records, or follow-up platforms. The specific data categories acquired include at least demographic data (including at least gender and age), medical history data (including at least comorbidities), laboratory data (such as potassium, creatinine, estimated glomerular filtration rate (eGFR), and timestamps), medication data (such as potassium supplements, potassium-lowering drugs, diuretics, etc.), and medication adjustment events (such as drug initiation, dosage increase, dosage decrease, or discontinuation events, and timestamps). This raw data often has issues such as inconsistent units, invalid timestamps, or outliers, therefore rigorous preprocessing is necessary. Preprocessing includes data cleaning and standardization, such as standardizing units for all test results, verifying the validity of timestamps, and identifying and handling missing or extreme outliers (such as critical values). In addition, the system will calculate key derived variables based on the laboratory data, such as stratifying renal function according to indicators like creatinine, or analyzing recent trends in potassium levels. Simultaneously, the system identifies specific medication adjustment events from continuous medication records. This series of operations transforms messy, unstructured raw data into high-quality, standardized structured data, providing reliable input for subsequent intelligent analysis and decision-making.

[0050] Furthermore, the input variables can be categorized as follows (as shown in Table 1; specific fields can be mapped according to hospital data standards or system interfaces): Table 1 Categories of Input Variables

[0051] Data preprocessing and verification include, but are not limited to: unit standardization, timestamp validity verification, critical / extreme value identification, missing value handling and prompting, test result validity determination, and medication dosage standardization (conversion between different formulations / specifications).

[0052] Step S302: Based on the preprocessed data, determine whether to enter the potassium supplementation path or the potassium reduction path.

[0053] Furthermore, this step, based on the risk level output from step S302 and the specific serum potassium level, determines which specific management path the patient should currently enter. The core decision-making principle is: if the patient's serum potassium level is below the preset target range (i.e., in a state of hypokalemia or hyponormal potassium), the system guides the patient into a potassium supplementation path, aiming to assess and develop a treatment plan to raise serum potassium levels. Conversely, if the patient's serum potassium level is above the preset target range (i.e., in a state of hyperkalemia), the system guides the patient into a potassium-lowering path, aiming to assess and develop a treatment plan to lower serum potassium levels. This determination is a prerequisite for subsequent targeted and differentiated medication decisions, ensuring the correct direction of the entire management process and avoiding misleading hypokalemic patients to potassium-lowering treatment or hyperkalemic patients to potassium supplementation treatment.

[0054] In step S303, when determining the path, the patient's blood potassium status is also risk-stratified, and the risk stratification results and treatment strategies are output.

[0055] In one optional implementation, step S303 includes: The rules used to risk-stratify patients' blood potassium status include: Compare the blood potassium level with preset critical low threshold, low threshold, high threshold and critical high threshold; The risk level is determined by comprehensively considering the patient's treatment context, renal function status, and specific medication status.

[0056] Furthermore, this step automates and quantifies the patient's immediate risk assessment based on a set of configurable rules. The main logic of risk stratification is as follows: First, the patient's current serum potassium level is compared with preset multi-level thresholds, including critically low, low, high, and critically high thresholds. For example, if the serum potassium level is below the critically low threshold or above the critically high threshold, it will be directly classified as high risk regardless of other factors. Second, the system will comprehensively assess the patient's treatment context (e.g., in-hospital emergency or out-of-hospital follow-up), renal function status (e.g., the range of eGFR), and specific medication status (e.g., whether the patient is using medications that easily cause fluctuations in serum potassium). For example, for patients already using potassium-lowering drugs, the system may use a more sensitive threshold to trigger a risk warning. Finally, the system will output a clear risk level (e.g., low, medium, high risk) and corresponding treatment recommendations (e.g., immediate emergency assessment, prompt outpatient visit, or routine follow-up). This step transforms complex clinical situations into clear risk signals, providing a decisive basis for the selection of subsequent treatment pathways.

[0057] Step S304: Based on the determined path and risk stratification results, execute the corresponding medication decision to generate a medication adjustment plan.

[0058] In one optional implementation, step S304 includes: When it is determined that the potassium supplementation pathway is to be entered, a set of multiple preset influencing factors is identified; these influencing factors include renal function stratification, current serum potassium level, and medication adjustment event type of drugs that affect serum potassium. Call the pre-configured scoring function to quantify and score each factor in the set, and calculate the overall score; Based on the comprehensive score and patient classification, the pre-configured dose mapping table is queried to obtain the baseline potassium supplementation plan; The baseline potassium supplementation regimen is modified based on at least one of the following: the patient's current potassium supplementation status, the timeliness of key laboratory data, or a specific clinical condition, to generate a final medication adjustment plan. This specific clinical condition includes the patient being on dialysis or having severe renal impairment.

[0059] Further, this step first identifies a set of pre-defined influencing factors, including at least: the patient's renal function stratification, the current range of serum potassium levels, and the type of recent medication adjustment events affecting serum potassium. Then, the system uses a pre-configured scoring function to quantify each factor in the set; for example, moderate renal impairment is assigned a positive score, while low serum potassium is assigned a negative score. The scores of all factors are summed to obtain a comprehensive score, which characterizes the degree to which the patient's serum potassium deviates from the target and the expected rate of regression. Next, the system selects a corresponding pre-configured dose mapping table based on the patient's classification (e.g., whether the patient has heart failure) to map the comprehensive score to a baseline potassium supplementation plan. This plan includes key information such as the recommended daily potassium supplementation amount and dosing frequency. Finally, the system performs a secondary modification of the baseline plan based on whether the patient is currently taking potassium supplements, whether key serum potassium test results have exceeded the effective time window, or whether special clinical conditions such as dialysis are present, thereby generating a final safe, feasible, and individualized oral potassium supplementation adjustment plan.

[0060] Specifically, in the context of oral potassium supplementation, this embodiment quantifies the intensity of interference affecting changes in serum potassium levels based on multiple factors and maps the comprehensive calculation results into an executable individualized potassium supplementation plan. First, the system defines a set of factors F affecting serum potassium, which includes at least the following factors: renal function stratification (e.g., based on intervals of estimated glomerular filtration rate (eGFR), current serum potassium level stratification (based on the interval in which the serum potassium value falls), and medication adjustment events affecting serum potassium (e.g., the initiation, dosage increase, reduction, or discontinuation of aldosterone receptor antagonists, loop diuretics, thiazide or thiazide-like diuretics). Furthermore, dialysis status or other status labels affecting potassium metabolism can be used as optional extended factors. Each factor f_i corresponds to a configurable scoring function Score_i(·). The interval boundaries, event type sets, and specific scores or weight values ​​of each factor are stored as parameters in a rule base (e.g., in a database table or configuration file), thus supporting updates and adjustments without modifying the core code.

[0061] The system quantifies the overall impact by calculating a comprehensive score S, which is the sum of the scores assigned to each factor, i.e., S = Σ Score_i(f_i). This score S characterizes the combined effect of the rate of natural regression of serum potassium or the degree of deviation from the target without intervention, along with the direction and intensity of the effects of various confounding factors. Simultaneously, the system determines the target range G for serum potassium management based on patient classification (e.g., whether the patient has heart failure). Different target ranges or mapping tables can be used for different populations, and the target range and classification rules themselves are configurable.

[0062] To facilitate the conversion from scoring to specific medication, the system maintains a set of mapping tables M_p stratified by population group p. These mapping tables are implemented as parameterized matrices, mapping the comprehensive score S (or its stratified value) to specific potassium supplementation regimen parameters, including but not limited to: the recommended daily potassium supplementation D_total, the dosing frequency N, the single dose D_once, and the recommended course of treatment or duration. The mapping tables can be updated according to the latest clinical consensus.

[0063] After generating a baseline potassium supplementation plan, the system will make secondary adjustments for special circumstances. For example, if the patient is already taking potassium supplements, the system reads the current total potassium supplementation amount (D_current) and the medication start time (T_start), and combines this with the most recent blood potassium retest time (T_K) and the preset effective time window (W_lab) to determine whether the follow-up test after medication has been covered, thus deciding whether to adjust the plan. If the patient's key test data (such as blood potassium) has exceeded the preset window (retesting window), a retest is recommended or a conservative strategy is adopted to maintain the existing plan, without making a decision to increase the dosage. If the system identifies missing key data such as eGFR, it outputs a prompt that information needs to be supplemented or a retest is recommended, and a conservative strategy is adopted. For patients on dialysis or with severe renal impairment, a stricter upper limit for the total potassium supplementation amount, a shorter retest cycle, or direct medical consultation will be triggered. Ultimately, the individualized potassium supplementation plan is determined by the baseline plan and the results of the secondary adjustments. The system will output explanatory evidence, including the key triggering factors, score breakdown details, and the specific rule items that were hit.

[0064] Finally, the system supports formulation conversion and prescription splitting. By defining the potassium content or concentration coefficient C_j for each potassium supplement j, the recommended daily potassium supplementation total D_total is converted into the dosage unit of that specific formulation. Simultaneously, the system provides feasible prescription splitting rules: the total D_total is split into single doses D_once according to the dosing frequency N, and rounded or rounded according to preset rules. When the split single dose exceeds the safety limit or does not conform to the existing formulation specifications, the system automatically adjusts the dosing frequency or selects an alternative splitting scheme to ensure the safety and feasibility of the prescription.

[0065] In one optional implementation, step S304 includes: When it is determined that the patient has entered this potassium-lowering pathway, it is determined whether the patient has already been using potassium-lowering medication; If no potassium is used and the current blood potassium level has reached the pre-configured starting threshold, the starting protocol for potassium-lowering drugs will be output as the adjustment protocol for that drug. If already used, the system will output recommendations for maintaining, increasing, decreasing, or discontinuing the medication based on the comparison between the current serum potassium level and the target range.

[0066] Further, this step first determines whether the patient is currently using potassium-lowering medication. If the patient is not using potassium-lowering medication and their current serum potassium level has reached the pre-configured medication initiation threshold in the rule base, the system will output a potassium-lowering medication initiation plan as a medication adjustment plan. This plan typically includes the type of medication, initial dose, and dosing frequency. If the patient is already using potassium-lowering medication, the system will decide on the next action based on a comparison between the current serum potassium level and the target treatment range: if the serum potassium has dropped to within the target range, it is recommended to maintain the current dose; if it is still above the target range, it may recommend increasing the dose; if it is below the lower limit of the target range, it may recommend reducing the dose or discontinuing the medication. The entire decision-making logic relies on a configurable medication rule base, which includes the indications, contraindications, dose steps, and discontinuation conditions for different potassium-lowering medications, thereby ensuring the accuracy and safety of the recommendations.

[0067] Step S305: Based on the medication adjustment plan, generate follow-up recommendations for the patient.

[0068] In one optional implementation, step S305 includes: Based on the risk stratification results and the medication adjustment plan, at least one predefined review rule is triggered. When multiple such review rules are triggered, the review rules for the same test item will be merged, and the final review cycle will be determined from the recommended cycles of each rule based on the preset clinical cycle set, so as to generate a unified review recommendation.

[0069] Furthermore, this step aims to establish a closed-loop monitoring system to ensure timely evaluation of the effectiveness and safety of treatment interventions. This embodiment does not provide fixed follow-up times, but rather intelligently triggers one or more predefined follow-up rules based on the generated medication adjustment plan (e.g., the addition of a diuretic or potassium-lowering drug) and risk stratification results. These rules may be based on different categories such as adverse drug reaction monitoring, critical value re-examination, or routine follow-up. When multiple rules are triggered simultaneously (e.g., a rule requiring potassium monitoring due to the addition of a diuretic, and a rule requiring potassium re-examination due to the current hypokalemia), the system will intelligently adjudicate and merge them. Specifically, it will merge rules targeting the same test items (e.g., potassium, renal function) and, based on a preset set of cycles aligned with clinical work rhythms (e.g., {1 day, 3 days, 1 week, 2 weeks…}), determine a final, executable follow-up cycle from the suggested cycles of each rule. This generated unified follow-up recommendation is dynamic, individualized, and closely related to treatment, effectively guiding subsequent follow-up arrangements.

[0070] Specifically, this embodiment outputs follow-up items and cycles based on risk level, medication adjustment trends, and the validity of test results. The system categorizes follow-up rules into several types, such as: critical value follow-up, adverse drug reaction monitoring follow-up, and routine follow-up follow-up. Each rule outputs specific items, cycles, triggering conditions, and applicable populations. When multiple rules are triggered simultaneously, the system merges them according to comprehensive rules: similar items can be merged for the same major category of follow-up items; for the merged follow-up cycle, it can be configured to take the shortest value among the rules, or split and output according to different items; if the calculated recommended follow-up cycle conflicts with the patient's most recent examination time, the system will calculate the remaining days and select the nearest executable cycle from a preset array of commonly used clinical cycles, thus ensuring that the suggested follow-up rhythm aligns with actual clinical work arrangements; if the system determines that the current follow-up item is within its validity period according to existing rules, it can output a suggestion that immediate follow-up is not necessary or to follow up in the next cycle.

[0071] Step S306: Display the analysis results, including the medication adjustment plan and the follow-up recommendation, to the user terminal and record the operation log.

[0072] In one optional implementation, step S306 includes: Show healthcare users the basis for generating the medication adjustment plan, the triggering rules for the follow-up recommendation, and provide an interactive interface for manual confirmation or modification of the analysis results; Present simplified risk stratification results, treatment strategies, medication adjustment plans, and follow-up reminders to patient users.

[0073] Furthermore, this step is the final stage of system-user interaction and quality control. The system integrates the analysis results from all the preceding steps—including risk level, treatment recommendations, specific medication adjustment plans and their detailed generation basis, and individualized follow-up recommendations—into a complete decision support report, which is then displayed to different user terminals through an interactive interface. The displayed content is targeted: for healthcare users (such as doctors and pharmacists), the system provides detailed professional information, including the calculation logic behind the plan, the specific triggering rules, etc., and offers an interactive interface for manual confirmation, modification, or rejection of the analysis results, fully respecting and supporting clinical decision-making. For patient users, the system provides simplified, clear, and easy-to-understand treatment guidance and follow-up reminders.

[0074] In summary, this embodiment first acquires input data (basic information, tests, medications, etc.) and performs preprocessing and standardization. Based on serum potassium levels and patient status, it performs risk stratification, outputting stratification (low / medium / high risk) and treatment suggestions (emergency / early outpatient visit / routine follow-up, etc.). It then determines whether to enter a potassium supplementation or potassium-lowering pathway (or maintain observation) and executes the corresponding medication decision-making submodule. If a potassium supplementation pathway is entered, the potassium supplementation calculation module assigns scores and performs comprehensive calculations on factors affecting serum potassium, generating a baseline potassium supplementation plan based on target intervals for different populations. Secondary corrections are made for special cases such as existing potassium supplementation drugs, follow-up examinations exceeding the window, and dialysis. Dosage conversions are performed on preferred drugs to obtain the final conclusion. Then, follow-up recommendations (items and cycles) are generated, and when multiple rules are triggered, they are merged and prioritized, with the cycle aligning with commonly used clinical follow-up frequencies. Finally, the results are displayed to the healthcare / patient end and log audit information is recorded. After confirmation / modification by the healthcare end, the results are written back to the prescription or follow-up system.

[0075] The following two simple examples illustrate the content disclosed in the above embodiments: Example 1: Management and monitoring of blood potassium in a hypokalemic scenario during outpatient follow-up: (1) Input: A patient P belongs to the population classification P_A (e.g., heart failure with reduced ejection fraction; classification rules are configurable). The system obtains the most recent test data from the follow-up platform: serum potassium K=K1 (example is 3.2 mmol / L), estimated glomerular filtration rate eGFR=55 mL / min / 1.73m² (example is 45–60 mL / min / 1.73m² range), serum potassium test timestamp T_K is 2 days ago; at the same time, the system obtains the medication list and adjustment events for the day from the follow-up platform: the system plans to add loop diuretics (torasemide 10mg, once daily) on the day of operation, and continue to maintain the original dose of renin-angiotensin-aldosterone inhibitors and aldosterone receptor antagonists among the treatment drugs that may affect serum potassium, and is currently taking potassium chloride sustained-release tablets 0.5g, once daily for 14 days (this is just an example, the event category and reading window are configurable).

[0076] (2) Risk stratification: The system stratifies risks according to the diagnosis and treatment scenario, threshold set and trend rules. In the example, the blood potassium K is lower than the low potassium threshold K_low (4.0 mmol / L) but the critical value threshold K_crit_low (3.5 mmol / L) is not triggered. Therefore, the output risk level is medium risk (determined by parameter configuration). The treatment suggestion is to adjust according to the system suggestion and re-examine, and to be evaluated in the outpatient clinic in the near future.

[0077] (3) Path determination: The system determines that the patient is in a hypokalemic state and enters the potassium supplementation path. At the same time, a safety boundary is set according to the renal function stratification and contraindication rules (for example, severe renal function decline / dialysis status may trigger a stricter upper limit or transfer to the emergency room). It is determined that oral potassium supplementation can continue. The specific method is determined by the subsequent potassium supplementation calculation module.

[0078] (4) Potassium Supplementation Calculation: The system constructs a set of factors F, including renal function stratification, serum potassium stratification, and medication adjustment events related to potassium elevation / decrease. For each factor fi, the configurable scoring function Score_i(fi) is called to obtain the score decomposition {Score(eGFR), Score(K), Score(loop diuretic adjustment), Score(renin-angiotensin-aldosterone adjustment), Score(aldosterone receptor antagonist adjustment)...}, and the total score S=ΣScore_i(fi) is calculated. The patient's score decomposition is Score(eGFR)=0.5, Score(K)=-3, Score(loop diuretic adjustment)=-0.5, Score(renin-angiotensin-aldosterone adjustment)=0, Score(aldosterone receptor antagonist adjustment)=0, and the total score S=-3.

[0079] (5) Mapping and Conversion: The system selects the mapping table M_P_A based on the classification P_A, mapping the score S to the baseline potassium supplementation regimen (including daily total D_total, dosing frequency N, single dose D_once, and recommended duration). Different classification mapping tables correspond to different target ranges for serum potassium. For example, the target range for serum potassium in M_P_A is 4.0-4.5 mmol / L. When the total score S≥0, no potassium supplementation is given, and when the total score S<0, different potassium supplementation regimens are given for different intervals. This patient's total score is -3, which falls within the interval [-3.5, -2). According to the mapping table, a baseline regimen of 3g / day of potassium chloride sustained-release tablets is given.

[0080] (6) Special Circumstances Correction: The system identifies that the patient is already using potassium supplementation medication and reads the current total potassium supplementation amount D_current, medication start time T_start, and post-medication follow-up coverage. The patient's current potassium supplementation amount is 0.5g / day of potassium chloride extended-release tablets, taken for 14 days, and a follow-up potassium test has been performed after taking the potassium supplementation medication. The potassium test was conducted within the last 7 days, within the specified time window. Based on the above information, the baseline regimen is corrected, and the output potassium supplementation regimen is increased by the currently taken potassium supplementation amount, i.e., 3.5g / day of potassium chloride extended-release tablets, on the basis of the baseline potassium supplementation regimen.

[0081] (7) Subsequently, the formulation conversion unit is invoked to convert D_total into the dosage unit of the specific formulation (such as mL, g, mmol or number of tablets) based on the potassium content coefficient C_j, and the feasible prescription is split and rounded according to the preset step size Δ. If this patient continues to use potassium chloride sustained-release tablets, 3.5g / day is equivalent to 7 tablets / day. Considering that the clinical routine is to take the medication 2 or 3 times a day and the dose is usually the same each time, 7 tablets / day is not a commonly used clinical dose. Therefore, the system finally recommends that the total dose be adjusted to 3g / day, and split into 1g / time, 3 times a day.

[0082] (8) Follow-up Recommendations: The system generates follow-up items (at least electrolytes and renal function) and follow-up cycles based on risk level, medication adjustment range, and renal function stratification. The system aligns the recommended cycle with a preset clinically commonly used cycle array R (e.g., including 1 day, 3 days, 1 week, 2 weeks, 1 month, etc.). When multiple rules are triggered, the system will merge and prioritize them. Specifically, based on adverse drug reaction monitoring, this patient is advised to have their serum potassium checked after 1 week due to the addition of a diuretic and eGFR below 60 mL / min / 1.73 m²; simultaneously, based on the critical potassium value follow-up, because the serum potassium is 3.2 mmol / L and the eGFR is in the range of [45, 60) mL / min / 1.73 m², it is recommended to have their serum potassium checked after 5 days. After merging these two rules for the same test item (serum potassium), the system finally recommends a follow-up cycle of 5 days.

[0083] Example 2: Blood potassium management and monitoring in hospital-acquired hyperkalemia scenarios: (1) Input: A patient with coronary heart disease retrieves test results from the hospital information system (HIS) / laboratory information system (LIS) within the hospital. The reported serum potassium K=K2 (5.1mmol / L in the example) and the test timestamp is the current day; the estimated glomerular filtration rate eGFR=55mL / min / 1.73m². Information is read from the medication list in the HIS system, showing that the patient is not currently using potassium-lowering drugs.

[0084] (2) Risk stratification: Based on the treatment scenario (in-hospital) and serum potassium level, if K is in the hyperkalemia range (5.0-5.5 mmol / L) and renal function is poor, the system outputs a medium risk level (determined by parameter configuration) and recommends in-hospital oral medication as soon as possible. If serum potassium reaches the preset critical high value threshold K_crit_high (5.5 mmol / L), the system outputs in-hospital intravenous medication as soon as possible.

[0085] (3) Path determination: The system determines that the patient is in a hyperkalemic state and enters the potassium-lowering path.

[0086] (4) Potassium-lowering drug decision: The system retrieves the indications, contraindications, starting dose, dose ladder, and tapering / discontinuation conditions for potassium-lowering drugs from the rule base (all configurable). If no potassium-lowering drug is currently being used and the serum potassium level meets the trigger threshold, the system outputs the initial treatment plan; if a potassium-lowering drug is currently being used, the system outputs the maintenance, tapering, or discontinuation plan based on whether the serum potassium level is within the target range, below the target lower limit, and the trend of change. This patient is not using potassium-lowering drugs and meets the conditions for using sodium zirconium cyclosilicate. Based on the patient's serum potassium level, the system recommends an initial treatment plan of 10g, three times daily.

[0087] (5) Concurrent medication prompt: The system can simultaneously prompt suggestions for reviewing or adjusting related drugs that may cause hyperkalemia (this function can be configured to only prompt or automatically provide adjustment plans).

[0088] (6) Follow-up recommendations: The system generates follow-up recommendations for electrolytes and renal function, and the recommended cycle is usually shorter than when the patient is in a hypokalemic state or when the serum potassium is within the target range. When there is a recent increase in potassium-raising medication or an initial event, or when a trend of deterioration in renal function is identified, the system will further shorten the recommended follow-up cycle and align the follow-up rhythm with the preset clinically commonly used cycle array R (e.g., {1 day, 3 days, 1 week...}). In this patient, the serum potassium is between (5.0, 5.5] mmol / L and the renal function is poor. The system recommends that the serum potassium and renal function be rechecked again within 1 day.

[0089] In summary, this embodiment provides a multivariate input-based method for managing and monitoring blood potassium levels. By integrating multi-source data, automatic risk stratification, bidirectional medication decision-making (potassium supplementation and potassium reduction), and dynamic follow-up recommendations, it achieves end-to-end, individualized management of abnormal blood potassium levels. This method can perform comprehensive calculations based on multiple input variables (such as test data, medication adjustment events, etc.) to output interpretable and specific medication adjustment plans and follow-up schedules. This solves the problem of single-point tools lacking individualized recommendations and closed-loop linkage in related technologies, helping to improve the timeliness, safety, and standardization of clinical decision-making, and creating traceable management records.

[0090] This embodiment also provides a blood potassium management and monitoring system based on multivariate input. This system is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0091] This embodiment provides a blood potassium management and monitoring system based on multivariate input, such as... Figure 4 As shown, it includes: The data acquisition module 401 is used to acquire patients' demographic data, medical history data, laboratory data, medication data, and medication adjustment event data; The data preprocessing module 402 is used to preprocess the demographic data, medical history data, test data, medication data, and medication adjustment event data. The decision analysis engine 403 is used to determine whether to enter the potassium supplementation or potassium reduction path based on the preprocessed data, and to perform risk stratification of the patient's blood potassium status, outputting the risk stratification results and treatment strategies. The medication recommendation generation module 404 is used to generate a medication adjustment plan by making corresponding medication decisions based on the determined path and risk stratification results. The follow-up examination suggestion generation module 405 is used to generate follow-up examination suggestions for the patient based on the medication adjustment plan; The interactive display and review module 406 is used to display the analysis results, including the risk stratification results, the modified treatment strategy, the drug adjustment plan, and the review recommendation, to the user. The logging and auditing module 407 is used to record operation logs.

[0092] In some optional implementations, when it is determined that the potassium supplementation pathway has been entered, the medication recommendation generation module 404 is further configured to: Identify a set containing multiple pre-defined influencing factors; Call the pre-configured scoring function to quantify and score each factor in the set, and calculate the overall score; Based on the comprehensive score and patient classification, the pre-configured dose mapping table is queried to obtain the baseline potassium supplementation plan; Based on at least one of the patient's current potassium supplementation status, the timeliness of key test data, or special clinical conditions, the baseline potassium supplementation regimen is modified to generate the final medication adjustment plan.

[0093] In some alternative implementations, the influencing factors include renal function stratification, current serum potassium level, and the type of medication adjustment event for drugs that affect serum potassium.

[0094] In some alternative implementations, this specific clinical condition includes the patient being on dialysis or having severe renal impairment.

[0095] In some optional implementations, when it is determined that the potassium-lowering pathway has been entered, the medication recommendation generation module 404 is further configured to: Determine if the patient is already using potassium-lowering medication; If no potassium is used and the current blood potassium level has reached the pre-configured starting threshold, the starting protocol for potassium-lowering drugs will be output as the adjustment protocol for that drug. If already used, the system will output recommendations for maintaining, increasing, decreasing, or discontinuing the medication based on the comparison between the current serum potassium level and the target range.

[0096] In some alternative implementations, the review recommendation generation module 405 is further configured to: Based on the risk stratification results and the medication adjustment plan, at least one predefined review rule is triggered. When multiple such review rules are triggered, the review rules for the same test item will be merged, and the final review cycle will be determined from the recommended cycles of each rule based on the preset clinical cycle set, so as to generate a unified review recommendation.

[0097] In some alternative implementations, the data preprocessing module 402 is further configured to: The acquired data is subjected to unit standardization, timestamp validity verification, and outlier identification. Based on the test data, derived variables were calculated, including renal function stratification and / or serum potassium change trends; The medication data is used to identify medication adjustment events; these events include the initiation, dosage increase, dosage decrease, or discontinuation of the medication.

[0098] In some alternative implementations, the rules upon which the risk stratification of a patient's blood potassium status is based include: Compare the blood potassium level with preset critical low threshold, low threshold, high threshold and critical high threshold; The risk level is determined by comprehensively considering the patient's treatment context, renal function status, and specific medication status.

[0099] In some alternative implementations, the interactive display and approval module 406 is also used for: Show healthcare users the basis for generating the medication adjustment plan, the triggering rules for the follow-up recommendation, and provide an interactive interface for manual confirmation or modification of the analysis results; Present simplified risk stratification results, treatment strategies, medication adjustment plans, and follow-up reminders to patient users.

[0100] Furthermore, the data acquisition module 401 is also used to acquire patient demographic information, medical history information, laboratory test results (including at least serum potassium, creatinine, eGFR and timestamps), medication information (drug name, dosage and time), and dialysis / cardiopulmonary bypass status, etc.

[0101] The data preprocessing module 402 is also used to perform unit and timestamp verification, missing / outlier value processing, and derived variable calculation (such as blood potassium change prediction, renal function stratification, medication adjustment events (whether it has been or is planned to be added, discontinued, increased or decreased) etc.).

[0102] The decision analysis engine 403 is also used to call the rule base / scoring model to perform comprehensive calculations on the processed variables. The output includes at least: risk level and treatment recommendations based on blood potassium, potassium supplementation or potassium reduction recommendations (medication, dosage frequency, treatment course, etc.), follow-up items and cycles, and supports the merging of multiple situations and priority determination.

[0103] The medication recommendation generation module 404 is also used to generate recommendations on potassium supplementation and potassium reduction drugs, dosages, frequencies, treatment courses, or discontinuation / reduction of medication, and to provide explanatory evidence.

[0104] The follow-up recommendation generation module 405 is also used to generate and combine follow-up plans for electrolytes / renal function, etc., and output recommendations that fit the clinical cycle.

[0105] The interactive display and review module 406 is also used to display results and evidence on the medical staff / patient side, and supports manual confirmation / modification.

[0106] The log and audit module 407 is also used to record inputs, computational chains, outputs, manual operations, and parameter versions to achieve traceability and quality control.

[0107] Furthermore, the medication recommendation generation module 404 consists of two sub-modules: a potassium-lowering sub-module 4041 and a potassium-supplementing sub-module 4042. The system first determines the appropriate management path based on the indications, contraindications, and the patient's current medication status. The potassium-lowering sub-module 4041 generates recommendations for starting, increasing, decreasing, or discontinuing potassium-lowering medications for patients with elevated blood potassium levels. Its key execution points include: reading information such as whether the patient is currently using potassium-lowering medications, their blood potassium level, contraindications or allergy history, renal function status, and recent blood potassium trends. If the patient meets the contraindications or their blood potassium level is below the target lower limit, the module outputs a recommendation to discontinue or not activate potassium-lowering medications, and simultaneously triggers a corresponding follow-up recommendation. If the patient is not currently using potassium-lowering medications and their blood potassium level reaches the pre-configured high-potassium trigger threshold, the module outputs a potassium-lowering medication initiation plan. This plan outputs different types and dosages of potassium-lowering medications based on the specific blood potassium level and treatment scenario (e.g., in-hospital or out-of-hospital). If the patient is currently taking medication, the module outputs an adjustment plan based on whether their serum potassium level is within the target range, below the target lower limit, and the trend of change, suggesting maintaining the current dose, reducing the dose, or discontinuing the medication. All the thresholds, dose steps, dosing frequencies, and maximum or minimum dose limits mentioned above are managed through configurable parameter tables, allowing different hospitals to flexibly update them according to drug instructions and local clinical consensus.

[0108] The potassium supplementation submodule 4042 generates potassium supplementation recommendations for patients with hypokalemia or below-normal potassium levels. This module first performs a basic applicability assessment, which involves determining whether immediate medical attention or intravenous potassium supplementation (corresponding to high-risk situations) is necessary, or oral potassium supplementation is appropriate, based on the patient's serum potassium level and the aforementioned risk stratification results. Simultaneously, safety boundaries are set based on the patient's renal function stratification, urine output or dialysis status (as optional parameters), and drug contraindications to ensure the safety of the recommendations. Furthermore, the module considers recent medication adjustment events, such as the initiation, dosage increase, or discontinuation of diuretics, renin-angiotensin-aldosterone inhibitors, or aldosterone receptor antagonists, to determine the potential direction of interference with serum potassium changes (whether it leads to an increase or decrease in serum potassium). This determination serves as an important input for the subsequent potassium supplementation calculation module.

[0109] Furthermore, the interactive display and audit module 406 provides two interactive interfaces: one for healthcare professionals and one for patients. The healthcare professional interface displays the complete recommended treatment plan, the basis for triggering the plan, and allowed modifications, and records the specific reasons for healthcare professionals confirming or modifying the plan. The patient interface displays a simplified version of the treatment instructions and follow-up reminders. The log and audit module 407 is responsible for recording the complete calculation and operation process, including a summary of the input data, the rule-matching chain, details of the score decomposition, the version of the mapping table used, the final output results, and all manual operation records, thereby achieving full-process traceability and medical quality control.

[0110] Furthermore, the system also includes a risk stratification and early warning module 408, which is used to generate risk notifications and treatment suggestions (such as emergency room / as soon as possible outpatient / routine follow-up) in addition to the quality control of the decision analysis engine 403 and the medication recommendation generation module 404.

[0111] Furthermore, the risk stratification and early warning module 408 is also used to output a risk level R (such as low, medium, or high) and corresponding treatment suggestions, taking the treatment scenario, absolute blood potassium value, trend, renal function, and specific medication status as the main inputs. The following is an example of a configurable implementation: First, the subsequent treatment suggestions will differ depending on whether the patient is in or out of the hospital. Second, the system predefines key blood potassium thresholds, whose magnitudes satisfy the following relationship: Critically low threshold (K_crit_low) < Low threshold (K_low) < High threshold (K_high) < Critically high threshold (K_crit_high). Based on these thresholds, the stratification rules specifically include: For all population groups, if the current serum potassium level K ≤ K_crit_low or K ≥ K_crit_high, the risk level R is determined to be high, and immediate emergency or critical care assessment is recommended; if the serum potassium level K is within the interval [K_high, K_crit_high) or (K_crit_low, K_low], the risk level R is determined to be medium, and outpatient or emergency assessment should be arranged as soon as possible, with follow-up or medication adjustment; if the serum potassium level K is within the target interval, i.e., K∈(K_low, K_high), and no other high-risk trend rules are triggered, the risk level R is determined to be low, and routine follow-up and re-examination plans should be implemented. Furthermore, for specific medication populations, such as patients already using potassium-lowering drugs, the system can use a more sensitive high-value threshold K_high' or a stricter trend triggering rule for risk assessment. Ultimately, the risk stratification results will serve as an important basis for subsequent decisions regarding re-examination cycles and medication pathway selection, and can trigger corresponding early warning prompts.

[0112] The modules mentioned above can be deployed on HIS / EMR, mobile applications, or cloud platforms, and interact with testing systems, prescription systems, or follow-up systems via wired / wireless networks.

[0113] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0114] The following is a detailed reference. Figure 5 This diagram illustrates a suitable structural design for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 501, which can perform various appropriate actions and processes based on a program stored in read-only memory (ROM) 502 or a program loaded from memory 408 into random access memory (RAM) 503. RAM 503 also stores various programs and data required for the operation of the electronic device. The processor 501, ROM 502, and RAM 403 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0115] Typically, the following devices can be connected to I / O interface 505: input devices 506 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 507 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 508 including, for example, magnetic tapes, hard disks, etc.; and communication devices 509. Communication device 509 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 5 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0116] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 509, or installed from a memory 508, or installed from a ROM 502. When the computer program is executed by the processor 501, it performs the functions defined in the multivariate input-based blood potassium management and monitoring method of the embodiments of the present invention.

[0117] Figure 5 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0118] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, it implements the multivariate input-based blood potassium management and monitoring method shown in the above embodiments.

[0119] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0120] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and all such modifications and variations fall within the scope defined by the invention.

Claims

1. A method for managing and monitoring blood potassium based on multivariate input, characterized in that, The method includes: Acquire patients' demographic data, medical history data, laboratory data, medication data, and medication adjustment event data, and perform preprocessing. Based on the preprocessed data, it is determined whether to enter the potassium supplementation or potassium reduction pathway, and the patient's blood potassium status is risk-stratified, outputting the risk stratification results and treatment strategies. Based on the determined path and the risk stratification results, corresponding medication decisions are made to generate a medication adjustment plan; Based on the medication adjustment plan, follow-up examination recommendations for the patient are generated; The system displays the analysis results, including the risk stratification results, the processing strategy, the medication adjustment plan, and the follow-up recommendations, to the user and records the operation log.

2. The method according to claim 1, characterized in that, When it is determined that the potassium supplementation pathway has been entered, the step of executing the corresponding medication decision to generate a medication adjustment plan includes: Identify a set containing multiple pre-defined influencing factors; The pre-configured scoring function is invoked to quantify and score each factor in the set, and a comprehensive score is calculated. Based on the comprehensive score and patient classification, the pre-configured dose mapping table is queried to obtain the baseline potassium supplementation plan; Based on at least one of the patient's current potassium supplementation status, the timeliness of key test data, or special clinical conditions, the baseline potassium supplementation regimen is modified to generate the final medication adjustment plan.

3. The method according to claim 2, characterized in that, The influencing factors include renal function stratification, current serum potassium level, and the type of medication adjustment event for drugs that affect serum potassium.

4. The method according to claim 2, characterized in that, The specific clinical conditions include patients who are on dialysis or have severe renal impairment.

5. The method according to claim 1, characterized in that, When it is determined that the potassium-lowering pathway has been entered, the step of executing the corresponding medication decision to generate a medication adjustment plan includes: Determine if the patient is already using potassium-lowering medication; If no potassium is used and the current blood potassium level reaches the pre-configured starting threshold, then the initial potassium-lowering drug regimen is output as the drug adjustment regimen. If already used, the system will output recommendations for maintaining, increasing, decreasing, or discontinuing the medication, based on the comparison between the current serum potassium level and the target range.

6. The method according to claim 1, characterized in that, The generated follow-up recommendations for patients include: Based on the risk stratification results and the medication adjustment plan, at least one predefined review rule is triggered. When multiple review rules are triggered, the review rules for the same test item will be merged, and the final review cycle will be determined from the suggested cycles of each rule based on the preset clinical cycle set, so as to generate a unified review recommendation.

7. The method according to claim 1, characterized in that, The acquisition and preprocessing of patient demographic data, medical history data, laboratory data, medication data, and medication adjustment event data includes: The acquired data is subjected to unit standardization, timestamp validity verification, and outlier identification. Derived variables are calculated based on the test data, including renal function stratification and / or serum potassium change trends; The medication adjustment event data is identified based on the medication data; the medication adjustment event data includes the initiation, dosage increase, dosage decrease, or discontinuation of medication.

8. The method according to claim 1, characterized in that, The rules used to risk-stratify patients' blood potassium status include: Compare the blood potassium level with preset critical low threshold, low threshold, high threshold and critical high threshold; The risk level is determined by comprehensively considering the patient's treatment context, renal function status, and specific medication status.

9. The method according to claim 1, characterized in that, The analysis results displayed to the user include the risk stratification results, the treatment strategy, the medication adjustment plan, and the follow-up recommendations, including: The system displays the basis for generating the medication adjustment plan and the triggering rules for the follow-up recommendations to healthcare users, and provides an interactive interface for manually confirming or modifying the analysis results. Present simplified risk stratification results, treatment strategies, medication adjustment plans, and follow-up reminders to patient users.

10. A blood potassium management and monitoring system based on multivariate input, characterized in that, The system for implementing the method as described in any one of claims 1 to 9 includes: The data acquisition module is used to acquire patients' demographic data, medical history data, laboratory data, medication data, and medication adjustment event data; The data preprocessing module is used to preprocess the demographic data, medical history data, test data, medication data, and medication adjustment event data. The decision analysis engine is used to determine whether to enter the potassium supplementation or potassium reduction path based on the preprocessed data, and to perform risk stratification of the patient's blood potassium status, outputting the risk stratification results and treatment strategies. The medication recommendation generation module is used to execute corresponding medication decisions to generate medication adjustment plans based on the determined path and the risk stratification results. The follow-up examination suggestion generation module is used to generate follow-up examination suggestions for patients based on the medication adjustment plan; The interactive display and review module is used to display the analysis results, including the risk stratification results, the processing strategies, the medication adjustment plans, and the review suggestions, to the user. The logging and auditing module is used to record operation logs.

Images