A method and system for constructing an HBV-ACLF hierarchical model based on multi-dimensional biochemical response after artificial liver treatment, and a medium

Abstract

The application discloses a kind of based on artificial liver therapy post-multiple dimensional biochemical response HBV-ACLF stratification model construction method, system and medium, relating to medical analysis technical field, its technical solution main point is: obtaining first artificial liver treatment time and 72 hours TBil, INR and Cr data, and test unit and actual detection time, pre-processing is carried out, and the data after pre-processing is obtained;According to the data after pre-processing, the decline rate RT of TBil, the decline rate RI of INR and the change state of Cr are calculated;According to RT, RI and the change state of Cr and corresponding threshold, response vector V (T, I, C) is generated, N=T+I+C is calculated, and response field is output.The application establishes a tool capable of automatically, standard, traceable to identify the multiple dimensional biochemical recovery mode of HBV-ACLF patient artificial liver therapy post.

Description

Technical Field

[0001] This invention relates to the field of medical analysis technology, and more specifically, to a method, system, and medium for constructing a hierarchical model of HBV-ACLF based on multidimensional biochemical responses after artificial liver treatment. Background Technology

[0002] In patients with hepatitis B virus-related acute-on-chronic liver failure (HBV-ACLF), after receiving plasma exchange (PE)-based artificial liver support, the liver's excretion, synthesis, and extrahepatic organ status may exhibit asynchronous changes. Changes in total bilirubin (TBil) primarily reflect the recovery of circulating bilirubin clearance and bile excretion; changes in the international normalized ratio (INR) reflect liver synthetic function and coagulation factor status, and may also be affected by plasma supplementation; changes in creatinine (Cr) can reflect renal function, circulatory perfusion, and responses in some extrahepatic organs.

[0003] Current electronic medical record and laboratory systems typically display baseline and follow-up results separately for each individual indicator. Clinicians must manually locate the initial artificial liver treatment time, match multiple 72-hour test results, calculate rates of change, and comprehensively assess different physiological dimensions. However, existing technologies have the following shortcomings: 1. Incomplete information on single-indicator responses. A decrease in TBil alone cannot identify patients whose TBil has improved but whose INR or creatinine has worsened.

[0004] 2. Complex multi-organ scores depend on many variables. SOFA, CLIF-C, ACLF, and other similar scores require respiratory, circulatory, neurological, urine output, or organ support data, which are often missing in conventional databases.

[0005] 3. The detection times for multiple indicators are not entirely consistent. If the time of the first artificial liver treatment is not uniformly used as the anchor, TBil, INR, and Cr may come from different time periods, leading to biases in response judgment.

[0006] 4. Heavy workload for manual calculation. It is necessary to calculate the rate of decrease in TBil, the rate of decrease in INR, and the change in creatinine separately, and to make judgments based on multiple combined conditions, which is prone to omissions or errors.

[0007] 5. Lack of standardized response vectors. Existing systems cannot convert hepatic excretion, hepatic synthesis, and renal responses into a unified binary vector and 0-3 response domains.

[0008] 6. Inadequate handling of missing data. When an indicator is missing, traditional manual judgment may still yield incomplete conclusions, failing to clearly distinguish between complete and partial phenotypes.

[0009] 7. Lack of traceability. Clinical records typically do not retain information such as the actual testing time of the indicator, selection rules, rate of change, response assignment, missing status, and system version.

[0010] 8. Simple risk probability is insufficient to demonstrate recovery patterns across different physiological dimensions, which is not conducive to research stratification, post-treatment trajectory management, and cross-center standardization.

[0011] In view of this, the present invention proposes a method, system and medium for constructing a hierarchical model of HBV-ACLF based on multidimensional biochemical response after artificial liver treatment. Summary of the Invention

[0012] The purpose of this invention is to provide a method, system, and medium for constructing a hierarchical model of HBV-ACLF based on multidimensional biochemical responses after artificial liver treatment, thereby solving the above-mentioned problems.

[0013] The above-mentioned technical objective of the present invention is achieved through the following technical solution: The first aspect of this invention provides a method for constructing a hierarchical model of HBV-ACLF based on multidimensional biochemical responses after artificial liver treatment, comprising the following steps: The time of the first artificial liver treatment and the TBil, INR and Cr data at 72 hours were obtained, along with the testing unit and the actual testing time. The data were preprocessed to obtain the preprocessed data. The decrease rate of TBil (RT), the decrease rate of INR (RI), and the change state of Cr are calculated based on the preprocessed data. Based on the changes in RT, RI, and Cr and the corresponding thresholds, a response vector V(T, I, C) is generated, N = T + I + C is calculated, and the response domain is output.

[0014] In conjunction with the first aspect, the present invention is further configured such that: the preprocessing includes unifying units, processing duplicate data, identifying missing data, abnormal ranges, and time window matching.

[0015] In conjunction with the first aspect, the present invention is further configured such that: if the preprocessing identifies missing TBil, INR, and Cr data, the final output is the computable domain and the missing domain.

[0016] In conjunction with the first aspect, the present invention is further configured such that the calculation method for the TBil decrease rate RT is: ; in, The rate of decline of TBil; Data for TBil, the first artificial liver treatment; TBil data at 72 hours post-treatment; The calculation method for the INR decline rate RI is as follows: ; in, The rate of decrease in INR; This is the INR data for the first artificial liver treatment; The data represents INR 72 hours after treatment.

[0017] In conjunction with the first aspect, the present invention is further configured such that: the method for generating the response vector V(T, I, C) based on the changes in RT, RI, and Cr and the corresponding thresholds is as follows: If RT ≥ 15%, then T = 1; otherwise, T = 0. If RI ≥ 10%, then I = 1; otherwise, I = 0. If C72 ≤ C0, then C = 1; otherwise, C = 0. The response vector V = (T, I, C).

[0018] In conjunction with the first aspect, the present invention is further configured to include generating hierarchical labels based on the N value: if N≥2, it is a multidimensional response; if N≤1, it is an insufficient response.

[0019] A second aspect of the present invention also provides an apparatus / device / system for HBV-ACLF stratification based on multidimensional biochemical response after artificial liver treatment, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0020] A third aspect of the present invention also provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0021] A fourth aspect of the present invention also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.

[0022] In summary, the present invention has the following beneficial effects: This invention solves the problem of judgment bias caused by inconsistent detection times of multiple biochemical indicators by using the time of the first artificial liver treatment as the anchor point and matching the 72-hour results of three indicators respectively. It ensures that all assessments are based on a unified 72-hour time window after treatment, making the responses of different dimensions comparable in time. This invention overcomes the limitations of relying solely on a single indicator (such as TBil) to assess treatment response by mapping changes in TBil, INR, and Cr to a three-dimensional response vector and 0-3 response domains, thus achieving a comprehensive profile of the patient's multi-organ function recovery status. It transforms continuous test values ​​into interpretable binary responses (T, I, C), making the rules transparent and the judgment criteria directly traceable. It provides a standardized tool for identifying asynchronous recovery trajectories such as "TBil improvement but INR or Cr not improving synchronously."

[0023] This invention generates a concise and unified hierarchical standard (N=0,1,2,3) by “synthesizing a three-dimensional response vector (V) and calculating the number of response domains (N)”, making complex biochemical recovery patterns quantifiable and comparable.

[0024] This invention employs a standardized quality control and output logic for missing and anomaly data. When an indicator is missing, the system only outputs a partial response vector, without generating a complete "multidimensional response / insufficient response" label, thus avoiding erroneous or misleading conclusions due to incomplete information. Outliers (such as baseline values ​​≤ 0) and inconsistent units are explicitly handled, and a verification mark is generated to ensure the reliability of the output results. All data processing steps, judgment criteria, and system versions are logged, achieving full-process traceability.

[0025] This invention utilizes a system that "designs auxiliary indicators to display without interfering with the core stratification architecture" to calculate and display the 72-hour MELD score to reflect the overall disease severity, but explicitly excludes it from the calculation of the core response vector V and the number of domains N. This enriches the presentation of clinical information while maintaining the transparency and interpretability of the core stratification rules (based on three independent physiological dimensions). Attached Figure Description

[0026] Figure 1 This is a flowchart of a method for constructing a hierarchical model of HBV-ACLF based on multidimensional biochemical response after artificial liver treatment in an embodiment of the present invention; Figure 2 This is the determination process for the three response domains of TBil, INR, and creatinine in the embodiments of the present invention; Figure 3 These are the number of different response domains and the 28-day and 90-day event rates in the embodiments of this invention. Detailed Implementation

[0027] 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.

[0028] In the embodiments provided in this application, it should be understood that the disclosed methods and systems can be implemented in other ways. The system embodiments described below are merely illustrative. For example, the division of units and modules is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or modules can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the various components shown or discussed can be through some interfaces, devices, or modules, and can be electrical, mechanical, or other forms.

[0029] In addition, each functional unit in the various embodiments of this application can be integrated into a single processor, or each unit can be a separate device, or two or more units can be integrated into a single device; each functional unit in the various embodiments of this application can be implemented in hardware or in the form of hardware plus software functional units.

[0030] Those skilled in the art will understand that all or part of the steps of the following method embodiments can be implemented by program instructions and related hardware. The aforementioned program instructions can be stored in a computer-readable storage medium. When the program instructions are executed, they perform the steps of the following method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0031] It should be understood that the use of terms such as "system," "device," "unit," and / or "module" in this application is merely one method of distinguishing different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0032] Furthermore, 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 application, "a plurality of" or "several" means two or more, unless otherwise explicitly specified.

[0033] If a flowchart is used in this application, it is used to illustrate the operations performed by the system according to embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0034] It should also be noted that, in this document, terms such as “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the article or device that includes the aforementioned element.

[0035] Example 1

[0036] In this embodiment, data from 190 patients undergoing PE-based artificial liver support in the COSSH-HBV-ACLF operation were used to map changes in TBil, INR, and Cr before and 72 hours after treatment to hepatic excretion response, hepatic synthesis response, and renal function stabilization response, respectively, forming a three-dimensional response vector (T, I, C). The vector was then automatically stratified based on the number N of satisfied response domains. This technique does not rely on complex machine learning models and emphasizes time anchoring, data quality control, transparent rules, and traceable output. Specifically, as follows: A method for constructing a hierarchical model of HBV-ACLF based on multidimensional biochemical responses after artificial liver treatment, such as Figure 1 and Figure 2 As shown, it includes the following steps: S100. Obtain the time of the first artificial liver treatment and the TBil, INR and Cr data for 72 hours, as well as the testing unit and the actual testing time, perform preprocessing, and obtain the preprocessed data; S200. Calculate the rate of decrease of TBil (RT), the rate of decrease of INR (RI), and the change state of Cr based on the preprocessed data; S300. Based on the changes in RT, RI, and Cr and the corresponding thresholds, generate a response vector V(T, I, C), calculate N=T+I+C, and output the response domain.

[0037] In step S100, the preprocessing includes unifying units, processing duplicate data, identifying missing data, anomaly ranges, and time window matching.

[0038] When the baseline TBil or INR ≤ 0, the corresponding rate of decline is not calculated.

[0039] When multiple similar test results exist within a 72-hour time window, the result closest to the target time point should be selected first; if the distances are the same, the earlier result should be selected.

[0040] When the units of the same indicator are inconsistent, they should be converted to a unified unit first, and the original values ​​and conversion rules should be retained.

[0041] When data anomalies or conflicts remain unresolved, complete hierarchical labels are not output, and reasons for review are generated instead.

[0042] The actual detection time, input source, rate of change, and response results of all indicators are written into the log.

[0043] If the preprocessing identifies missing (one or more) TBil, INR, and Cr data, then in step 300, the output is the computable domain and the missing domain.

[0044] In this embodiment, the time of the first artificial liver treatment is taken as T0, and the results of the three indicators closest to 72 hours are selected in the preset time window; the actual detection time of each indicator is saved.

[0045] In step S200, the TBil decrease rate RT is calculated as follows: ; in, The rate of decline of TBil; Data for TBil, the first artificial liver treatment; TBil data at 72 hours post-treatment; The calculation method for the INR decline rate RI is as follows: ; in, The rate of decrease in INR; This is the INR data for the first artificial liver treatment; The data represents INR 72 hours after treatment.

[0046] In step S300, the method for generating the response vector V(T, I, C) based on the changes in RT, RI, and Cr and the corresponding thresholds is as follows: If RT ≥ 15%, then T = 1; otherwise, T = 0. If RI ≥ 10%, then I = 1; otherwise, I = 0. If C72 ≤ C0, then C = 1; otherwise, C = 0. The response vector V = (T, I, C).

[0047] In step S300, the method further includes generating hierarchical labels based on the N value: if N≥2, it is a multidimensional response; if N≤1, it is an insufficient response.

[0048] In addition, the 72-hour MELD can be calculated based on 72-hour TBil, INR, and Cr and used as an auxiliary display to illustrate the overall disease severity after treatment. The 72-hour MELD is not involved in the calculation of the three-dimensional response vector and N value to maintain the transparency of the stratification rules and the independent interpretation of each physiological domain.

[0049] Finally, output the original value, rate of change, response assignment, response vector, N, hierarchical labels, auxiliary MELD, and data quality status.

[0050] The different response domains and outcomes are shown in Table 1 below: Table 1. Different Response Domains and Outcomes

[0051] After defining N≥2 as a multidimensional response, the 28-day mortality or liver transplantation event rate was 5.0%, while it was 45.2% for those with N≤1; the 90-day mortality rates were 10.5% and 64.5%, respectively. These results are intended to illustrate that the response vector can distinguish different biochemical recovery trajectories and do not represent that this threshold has been validated externally in clinical trials.

[0052] The results of de-identifying patient-level response vectors are shown in Table 2: Table 2 De-identified patient-level response vectors

[0053] The INR of object F did not decrease to 10%, but TBil and Cr reached the response, and the system output V=(1,0,1), N=2 and multidimensional response; although the INR of object E improved, TBil and Cr did not reach the response, and the output V=(0,1,0), N=1 and insufficient response.

[0054] Model validation: In the same complete case, multidimensional response stratification was compared with single TBil response stratification.

[0055] Perform descriptive analysis and Fisher's exact test; do not present exploratory thresholds as externally validated clinical scores.

[0056] The software displays the event rate of development data as reference information, but does not automatically change the patient's response judgment.

[0057] The comparison results with the single TBil response are shown in Table 3 and Figure 3 As shown: Table 3 Comparison results with single TBil response

[0058] Multidimensional response models can identify patients whose TBil has improved but whose INR or Cr has not improved synchronously, thus expressing the recovery status of different physiological domains in a unified vector.

[0059] Example 2

[0060] The present invention also provides an apparatus / device / system for HBV-ACLF stratification based on multidimensional biochemical response after artificial liver treatment, comprising a memory, a processor, and a computer program stored in the memory, characterized in that the processor executes the computer program to implement the steps of the above method.

[0061] The present invention also provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0062] The present invention also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.

[0063] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0064] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0065] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0066] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A method for constructing a hierarchical model of HBV-ACLF based on multidimensional biochemical responses after artificial liver treatment, characterized in that, Includes the following steps: The time of the first artificial liver treatment and the TBil, INR and Cr data at 72 hours were obtained, along with the testing unit and the actual testing time. The data were preprocessed to obtain the preprocessed data. The decrease rate of TBil (RT), the decrease rate of INR (RI), and the change state of Cr are calculated based on the preprocessed data. Based on the changes in RT, RI, and Cr and the corresponding thresholds, a response vector V(T, I, C) is generated, N = T + I + C is calculated, and the response domain is output.

2. The method according to claim 1, characterized in that, The preprocessing includes standardizing units, processing duplicate data, identifying missing data, anomaly ranges, and time window matching.

3. The method according to claim 2, characterized in that, If the preprocessing identifies missing TBil, INR, and Cr data, the final output will be the computable domain and the missing domain.

4. The method according to claim 1, characterized in that, The method for calculating the TBil decrease rate RT is as follows: ； in, The rate of decline of TBil; Data for TBil, the first artificial liver treatment; TBil data at 72 hours post-treatment; The calculation method for the INR decline rate RI is as follows: ； in, The rate of decrease in INR; This is the INR data for the first artificial liver treatment; The data represents INR 72 hours after treatment.

5. The method according to claim 4, characterized in that, The method for generating the response vector V(T, I, C) based on the changes in RT, RI, and Cr and the corresponding thresholds is as follows: If RT ≥ 15%, then T = 1; otherwise, T = 0. If RI ≥ 10%, then I = 1; otherwise, I = 0. If C72 ≤ C0, then C = 1; otherwise, C = 0. The response vector V = (T, I, C).

6. The method according to claim 1, characterized in that, It also includes generating hierarchical labels based on the N value: if N≥2, it is a multidimensional response; if N≤1, it is an insufficient response.

7. A device / equipment / system for HBV-ACLF stratification based on multidimensional biochemical response after artificial liver treatment, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1-6.

8. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method described in any one of claims 1-6.

9. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method described in any one of claims 1-6.

Images