Myocardial infarction metabolic rhythm intelligent typing and drug administration decision method and system

Abstract

The present application relates to the technical field of medical typing decision, in particular to a myocardial infarction metabolic rhythm intelligent typing and dosing decision method and system, comprising: synchronously detecting biological clock gene BMAL1 mRNA, USP2-BCAT2 complex and branched chain amino acid BCAAs through a multi-index joint detection test box; collecting BCAAs circadian rhythm curve through a wearable monitoring patch integrating a sweat BCAAs sensor and an electrocardio monitoring module; and utilizing a random forest model myocardial infarction metabolic rhythm typing intelligent model. The present application establishes a multi-dimensional dynamic typing system by integrating BMAL1 methylation, USP2-BCAT2 complex and BCAAs rhythm fluctuation, accurately distinguishes high-risk subtypes requiring metabolic intervention, early warns metabolic abnormal state through real-time monitoring of BCAAs concentration fluctuation by an intelligent patch, and automatically optimizes dosing timing based on individual biological clock phase, thereby solving the clinical pain points of insufficient typing precision, abnormal early warning lag and dosing timing mismatch.

Description

Technical Field

[0001] This invention relates to the field of medical classification decision technology, specifically to a method and system for intelligent classification and drug administration decision-making based on metabolic rhythm in myocardial infarction. Background Technology

[0002] In recent years, significant breakthroughs have been made in the study of the association between circadian rhythm regulation and cardiovascular metabolic diseases, particularly the interaction between clock genes (such as BMAL1 and CLOCK) and branched-chain amino acid (BCAA) metabolism, which has become a research hotspot. Traditional myocardial injury treatment mainly focuses on direct intervention of ischemia and hypoxia, while the latest research shows that circadian rhythm disorders lead to abnormal accumulation of BCAAs through the BMAL1-USP2-BCAT2 cascade, thereby inducing mitochondrial dysfunction and oxidative stress. This discovery provides a new target for the prevention and treatment of cardiovascular diseases. Although existing studies have explored the individual application of clock regulators (such as melatonin) or metabolic interventions (such as BCAT2 inhibitors), a precise subtyping system integrating the "circadian rhythm-metabolism" dual dimensions is lacking.

[0003] Traditional classification methods in the current technology cannot identify metabolic abnormalities of myocardial infarction driven by circadian rhythm disorders (such as the CIR / MIX subtype), resulting in some patients still experiencing progressive deterioration of cardiac function after receiving standard treatment; existing BCAA detection technologies (such as mass spectrometry) only provide static concentration data and cannot reflect the correlation between dynamic changes in BCAT2 activity and circadian rhythm, causing lag in BCAA detection and resulting in delayed metabolic warnings; clinical decision-making systems lack biological clock phase adaptation mechanisms, and fixed dosing regimens significantly reduce drug metabolism efficiency in patients with low BMAL1 expression. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for intelligent classification and drug administration decision-making based on metabolic rhythm in myocardial infarction, in order to solve the technical problems in the prior art that it is difficult to accurately classify myocardial infarction, metabolic early warning is lagging, and drug administration strategies lack biological clock phase adaptation mechanisms.

[0005] To solve the above-mentioned technical problems, the present invention specifically provides the following technical solution:

[0006] A method for intelligent classification and dosing decision-making based on metabolic rhythm in acute myocardial infarction includes the following steps:

[0007] The biological clock gene BMAL1 mRNA, the USP2-BCAT2 complex, and branched-chain amino acids BCAAs were detected simultaneously using a multi-index detection kit.

[0008] A wearable monitoring patch integrating a sweat BCAAs sensor and an electrocardiogram monitoring module is used to monitor and record the concentration changes of branched-chain amino acids (BCAAs) in blood or other body fluids over a complete circadian cycle, thus obtaining the BCAAs circadian rhythm curve.

[0009] Using a random forest model, an intelligent model for myocardial infarction metabolic rhythm typing was established, with the biological clock gene BMAL1 mRNA, USP2-BCAT2 complex and branched-chain amino acids BCAAs and BCAAs diurnal rhythm curves as inputs and myocardial infarction metabolic rhythm typing as output.

[0010] Based on the circadian rhythm curves of BCAAs collected by wearable monitoring patches and the USP2-BCAT2 complex, the fluctuation nodes of BCAAs were determined, and metabolic abnormalities were predicted at the fluctuation nodes of BCAAs for myocardial infarction metabolic rhythm classification.

[0011] By matching the metabolic rhythm typing of myocardial infarction with the phase of the individual's biological clock, a dynamic drug delivery timing strategy for myocardial infarction metabolic rhythm typing was determined.

[0012] The dosage was optimized for myocardial infarction metabolic rhythm typing through dose gradient experiments, and the optimal dosage for myocardial infarction metabolic rhythm typing was obtained.

[0013] As a preferred embodiment of the present invention, the metabolic rhythm classification of myocardial infarction includes CIR type, MET type, MIX type and NORM type.

[0014] As a preferred embodiment of the present invention, the BMAL1 mRNA of the biological clock gene is obtained by qPCR detection, the USP2-BCAT2 complex is obtained by time-resolved fluorescence immunoassay, and the branched-chain amino acids BCAAs are obtained by enzyme colorimetric test strips.

[0015] As a preferred embodiment of the present invention, the BCAAs fluctuation node is the time point at which the concentration ratio between branched-chain amino acids BCAAs and BCAT2 activity, BCAAs / BCAT2, exceeds a threshold. The total amount of BCAT2 is obtained by immunoprecipitation, and the BCAT2 activity is calculated by the concentration difference between the total amount of BCAT2 and the USP2-BCAT2 complex.

[0016] As a preferred embodiment of the present invention, the method for determining the dynamic drug delivery timing strategy for myocardial infarction metabolic rhythm typing includes:

[0017] Wearable devices were used to continuously collect body temperature data for 72 hours. The cosine fitting method was used to calculate the trough value of body temperature rhythm CBTmin = M + A · cos(2πt / τ + φ), and the reference phase φ of the individual's biological clock was located. Where M is the median of rhythm adjustment, A is the amplitude, t is the time variable, and τ is the period.

[0018] Phase characteristic analysis of the trough CBTmin of body temperature rhythm was performed to determine the optimal dosing strategy. Specifically, for the CIR type, the drug was administered 4 hours after the trough CBTmin; for the MET type, the drug was administered during the rising phase of body temperature; and for the MIX type, the drug was administered both 4 hours after the trough CBTmin and during the rising phase of body temperature.

[0019] Based on the optimal dosing strategy, dynamic time windows were used to calculate and monitor body temperature and branched-chain amino acid (BCAA) concentration levels in real time to determine the optimal dosing time for each myocardial infarction metabolic rhythm subtype.

[0020] As a preferred embodiment of the present invention, the method for optimizing the drug dosage for myocardial infarction metabolic rhythm typing through dose gradient experiments includes:

[0021] To classify the metabolic rhythm of myocardial infarction, low, medium, and high dose groups were set up, corresponding to low-dose administration, medium-dose administration, and high-dose administration, respectively. Myocardial BCAT2 activity, mitochondrial function, and abnormal liver enzyme rates were measured in the low, medium, and high dose groups.

[0022] The optimal dosage for myocardial infarction metabolic rhythm classification was determined based on myocardial BCAT2 activity, mitochondrial function, and abnormal liver enzyme rates.

[0023] As a preferred embodiment of the present invention, the smart pillbox receives the calculated optimal dosing time and optimal dosage via NFC, releases the pre-loaded drug according to the optimal dosage within the target time window (±15 minutes), and simultaneously activates the electrocardiogram monitoring module to record ST segment change data after medication.

[0024] As a preferred embodiment of the present invention, the present invention provides an intelligent metabolic rhythm classification and drug administration decision system for acute myocardial infarction, applied to an intelligent metabolic rhythm classification and drug administration decision method for acute myocardial infarction, the system comprising:

[0025] Multi-index detection kit for simultaneous detection of BMAL1 mRNA, USP2-BCAT2 complex and branched-chain amino acids BCAAs;

[0026] A wearable monitoring patch integrates a sweat BCAAs sensor and an electrocardiogram monitoring module to monitor and record the concentration changes of branched-chain amino acids (BCAAs) in blood or other body fluids over a complete circadian cycle, thereby obtaining a BCAAs circadian rhythm curve.

[0027] The data processing unit is used to establish an intelligent model for myocardial infarction metabolic rhythm typing using a random forest model. The model takes the circadian rhythm curves of the biological clock gene BMAL1 mRNA, USP2-BCAT2 complex and branched chain amino acids BCAAs as input and myocardial infarction metabolic rhythm typing as output.

[0028] The early warning monitoring unit is used to determine the BCAA fluctuation nodes based on the BCAA diurnal rhythm curves and USP2-BCAT2 complex collected by the wearable monitoring patch, and to provide early warning of metabolic abnormalities in myocardial infarction metabolic rhythm classification at the BCAA fluctuation nodes.

[0029] The dosing decision unit is used to determine the dynamic dosing timing strategy for myocardial infarction metabolic rhythm classification by matching individualized biological clock phases; and to optimize the dosing dose for myocardial infarction metabolic rhythm classification through dose gradient experiments to obtain the optimal dosing dose for myocardial infarction metabolic rhythm classification.

[0030] NFC-triggered pillboxes are used for intelligent dynamic drug delivery based on dynamic dosing timing strategies and optimal dosage.

[0031] As a preferred embodiment of the present invention, the drug administration decision unit includes a time decision module for formulating a dynamic drug administration timing strategy and a dose decision module for determining the optimal drug administration dose;

[0032] The time decision module uses wearable devices to continuously collect 72 hours of body temperature data, and uses cosine fitting method to calculate the trough value of body temperature rhythm CBTmin = M + A · cos(2πt / τ + φ), and locates the individual biological clock reference phase φ, where M is the median of rhythm adjustment, A is the amplitude, t is the time variable, and τ is the period;

[0033] Phase characteristic analysis of the trough CBTmin of body temperature rhythm was performed to determine the optimal dosing strategy. Specifically, for the CIR type, the drug was administered 4 hours after the trough CBTmin; for the MET type, the drug was administered during the rising phase of body temperature; and for the MIX type, the drug was administered both 4 hours after the trough CBTmin and during the rising phase of body temperature.

[0034] Based on the optimal dosing strategy, dynamic time windows were used to calculate and monitor body temperature and branched-chain amino acid (BCAA) concentration levels in real time to determine the optimal dosing time for each myocardial infarction metabolic rhythm subtype.

[0035] As a preferred embodiment of the present invention, the dose decision module sets low, medium and high dose groups for myocardial infarction metabolic rhythm classification, corresponding to low-dose administration, medium-dose administration and high-dose administration respectively, and detects myocardial BCAT2 activity, mitochondrial function and abnormal rate of liver enzymes in the low, medium and high dose groups;

[0036] The optimal dosage for myocardial infarction metabolic rhythm classification was determined based on myocardial BCAT2 activity, mitochondrial function, and abnormal liver enzyme rates.

[0037] As a preferred embodiment of the present invention, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, realize a method for intelligent classification and drug administration decision-making of metabolic rhythm in myocardial infarction.

[0038] Compared with the prior art, the present invention has the following advantages:

[0039] This invention integrates multi-dimensional data such as BMAL1 methylation, USP2-BCAT2 complex, and dynamic BCAAs curves to achieve, for the first time, metabolic rhythm typing (CIR / MET / MIX / NORM) of myocardial infarction, increasing the detection rate of high-risk patients from 52% to 91% compared to the traditional STEMI / NSTEMI typing (p<0.001).

[0040] This invention is based on the technology of continuous monitoring of sweat BCAAs by intelligent patches, which enables the detection of metabolic abnormalities earlier than traditional blood tests, and combines AI algorithms to adjust the dosing regimen in real time.

[0041] This invention enhances the efficacy of clock synchronizers by matching the patient's biological clock phase (e.g., administering the CIR type 4 hours after CBTmin). Attached Figure Description

[0042] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0043] Figure 1 This is a flowchart of the intelligent classification and drug administration decision-making method for myocardial infarction metabolic rhythm provided in an embodiment of the present invention;

[0044] Figure 2 This is a block diagram of an intelligent classification and drug administration decision system for myocardial infarction metabolic rhythm provided in an embodiment of the present invention. Detailed Implementation

[0045] 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.

[0046] like Figure 1 As shown, this invention provides a method for intelligent classification and drug administration decision-making based on metabolic rhythm in acute myocardial infarction, comprising the following steps:

[0047] The biological clock gene BMAL1 mRNA, the USP2-BCAT2 complex, and branched-chain amino acids BCAAs were detected simultaneously using a multi-index detection kit.

[0048] A wearable monitoring patch integrating a sweat BCAAs sensor and an electrocardiogram monitoring module is used to monitor and record the concentration changes of branched-chain amino acids (BCAAs) in blood or other body fluids over a complete circadian cycle, thus obtaining the BCAAs circadian rhythm curve.

[0049] Using a random forest model, an intelligent model for myocardial infarction metabolic rhythm typing was established, with the biological clock gene BMAL1 mRNA, USP2-BCAT2 complex and branched-chain amino acids BCAAs and BCAAs diurnal rhythm curves as inputs and myocardial infarction metabolic rhythm typing as output.

[0050] Based on the circadian rhythm curves of BCAAs collected by wearable monitoring patches and the USP2-BCAT2 complex, the fluctuation nodes of BCAAs were determined, and metabolic abnormalities were predicted at the fluctuation nodes of BCAAs for myocardial infarction metabolic rhythm classification.

[0051] By matching the metabolic rhythm typing of myocardial infarction with the phase of the individual's biological clock, a dynamic drug delivery timing strategy for myocardial infarction metabolic rhythm typing was determined.

[0052] The dosage was optimized for myocardial infarction metabolic rhythm typing through dose gradient experiments, and the optimal dosage for myocardial infarction metabolic rhythm typing was obtained.

[0053] In this invention, BMAL1 mRNA is aryl hydrocarbon receptor nuclear transporter-like protein 1 in brain and muscle tissue, which can be abbreviated as BMAL1. USP2-BCAT2 is ubiquitin-specific protease 2-branched amino acid transaminase 2 complex. The branched amino acids BCAAs include leucine, isoleucine, and valine. The mRNA is messenger RNA.

[0054] Myocardial infarction metabolic rhythm classification includes circadian rhythm disorder (CIR), metabolic disorder (MET), mixed disorder (MIX), and normal rhythm (NORM).

[0055] The BMAL1 mRNA of the biological clock gene was obtained by qPCR, the USP2-BCAT2 complex was obtained by time-resolved fluorescence immunoassay, and the branched-chain amino acids BCAAs were detected by enzyme colorimetric strips. The corresponding concentrations of each metabolic marker were obtained.

[0056] This invention collects and continuously records the concentration changes of branched-chain amino acids (BCAAs) in blood or other body fluids over a 24-hour period or a complete diurnal cycle, creating a BCAAs circadian rhythm curve. This curve shows how BCAA concentration fluctuates regularly over time, reflecting the body's internal metabolic rhythm, i.e., the dynamic concentration changes of branched-chain amino acids (leucine, isoleucine, valine) over 24 hours. The curve typically uses the horizontal axis to represent time (e.g., different time periods within a day or a 24-hour cycle) and the vertical axis to represent BCAA concentration or corresponding ratios (e.g., the BCAAs / BCAT2 ratio). For example, by continuously collecting BCAA data from sweat using a wearable device, a daily BCAA concentration fluctuation curve can be plotted, thus visually demonstrating its diurnal rhythm changes. Table 1 shows the BCAAs circadian rhythm curve data.

[0057] Table 1. BCAAs diurnal curve data (comparison between healthy patients and myocardial infarction patients)

[0058]

[0059] This invention uses a kit to detect the expression levels of BMAL1 / USP2 / BCAT2, and combines this with wearable devices to obtain 24-hour dynamic curves of BCAAs. It establishes a multidimensional dynamic classification model integrating BMAL1 methylation, the USP2-BCAT2 complex, and BCAA rhythm fluctuations, achieving automated identification of metabolic rhythm typing (CIR / MET / MIX / NORM) for myocardial infarction. Compared to the traditional STEMI / NSTEMI typing, the detection rate for high-risk patients is increased from 52% to 91% (p<0.001), as shown in Table 2. This technology can be applied in emergency departments for rapid screening of patients requiring metabolic intervention, avoiding treatment delays caused by missed diagnoses.

[0060] Based on a retrospective cohort analysis of 2,158 patients with coronary heart disease, this invention obtained the performance indicators of the intelligent model for myocardial infarction metabolic rhythm subtyping, with PPV as positive predictive value, NPV as negative predictive value, and AUC as the area under the receiver operating curve, as shown in Table 2, and the risk ratios for clinical outcomes of each subtype, as shown in Table 3.

[0061] Table 2 Comparison of detection rates of high-risk patients between metabolic rhythm typing and traditional typing

[0062]

[0063] Statistical notation: Data groups marked with * show p < 0.001 vs. conventional classification (McNemar test). High-risk definition: Composite endpoint of hospitalization for cardiac death / re-infarction / heart failure within 3 years of follow-up. Data source: UK Biobank sub-cohort (2010-2020), detection method: liquid chromatography-tandem mass spectrometry (LC-MS / MS) + continuous monitoring with wearable device.

[0064] Table 3. Hazard ratios of clinical outcomes for each subtype (Cox regression model)

[0065]

[0066] Statistical annotations: Event: Composite endpoint of cardiac death / re-infarction / heart failure hospitalization; Adjusted variables: age, sex, body mass index (BMI), history of diabetes, LDL-C level; p-value: p < 0.001 for data groups marked with ** (vs. NORM group). AUC calculation: using the pROC package in R (DeLong confidence interval). Survival analysis: Kaplan-Meier curve + Cox proportional hazards model (PH hypothesis test, p > 0.05).

[0067] As shown above, this application passed prospective cohort validation (N=2,158), and this typing system has the following advantages over traditional methods:

[0068] 1. The high-risk detection rate has improved: from 112 cases (52.3%) in the traditional classification to 196 cases (91.4%), of which:

[0069] Of the 84 new cases, 78 (92.9%) experienced a endpoint event during the follow-up period (χ²). 2 =63.2, p<0.001).

[0070] 2. The precision of intervention has been improved:

[0071] Treatment guided by subtype: 3-year event rate decreased by 58% (HR: 0.42, 95% CI: 0.31-0.57);

[0072] In contrast, the conventional treatment group showed a decrease of only 19% (HR: 0.81, 95% CI 0.65-1.02).

[0073] The BCAAs fluctuation point is the time point at which the concentration ratio of branched-chain amino acids BCAAs to BCAT2 activity exceeds a threshold. The total amount of BCAT2 is obtained by immunoprecipitation, and the BCAT2 activity is calculated by the concentration difference between the total amount of BCAT2 and the USP2-BCAT2 complex.

[0074] In the healthy group: a regular biphasic peak was observed (ZT4-8 after breakfast, ZT12 after lunch), with a trough at night (ZT16-24). In the myocardial infarction group: phase delay: the peak value was delayed from ZT8 to ZT12 (Δφ=4 hours). Amplitude reduction: fluctuation amplitude decreased by 62% (p<0.01). Nighttime accumulation: ZT12-24 consistently exceeded the threshold (BCAAs / BCAT2>2.0). Statistical annotation: p<0.05 for data groups marked with *, p<0.01 for data groups marked with ** vs. healthy group (ANOVA).

[0075] A BCAAs / BCAT2 ratio greater than 2.0 reflects the degree of imbalance in branched-chain amino acid metabolism.

[0076] Molecules (BCAAs): Total concentration of leucine, isoleucine, and valine in plasma (μmol / L); Denominator (BCAT2): Active units of branched-chain amino acid transaminase 2 in myocardial tissue or circulating blood (U / mg protein).

[0077] The methods for determining the threshold of 2.0 include: 1. Animal model validation: In mouse I / R experiments, the microvascular obstruction area was significantly increased in the group with a ratio ≥2.0 (35.2±6.8% vs 12.4±3.1% in the group <2.0, p<0.001). Animal experiments showed that when the ratio exceeded 2.0, the amount of mitochondrial ROS increased sharply by 3.2 times. 2. In vitro experiments confirmed that when BCAT2 activity was less than 50% of the BCAAs concentration, it led to the accumulation of α-keto acids and induced endothelial cell apoptosis. When BCAAs / BCAT2>2.0, BCAT2 substrate saturation overload triggered metabolic toxicity.

[0078] The USP2-BCAT2 complex, based on the following principles: 1. Total BCAT2 content, determined by immunoprecipitation (anti-BCAT2 antibody), is used as the denominator in the BCAAs / BCAT2 ratio (reflecting the size of the enzyme's active pool). 2. The USP2-BCAT2 complex itself is determined by time-resolved fluorescence resonance energy transfer (TR-FRET), specifically detecting the degree of binding between the two components, serving as a marker of BMAL1-USP2-BCAT2 pathway activation (independent of the free component).

[0079] The molecular mechanism is that only free BCAT2 that is not bound to USP2 has the activity to metabolize BCAAs. Therefore, BCAT2 in the BCAAs / BCAT2 ratio refers to the free active form, which is obtained by subtracting the concentration of the USP2-BCAT2 complex from the total BCAT2 concentration.

[0080] This invention utilizes a wearable patch to monitor BCAAs fluctuations in real time. When the BCAAs / BCAT2 ratio exceeds 2.0, an automatic alert is triggered, initiating intensive treatment. This avoids the lag inherent in static testing. Based on continuous monitoring of sweat BCAAs using a smart patch, it enables earlier detection of metabolic abnormalities compared to traditional blood tests, and combines AI algorithms to adjust medication regimens in real time. This technology is particularly suitable for continuous metabolic monitoring of patients in hospital wards, providing early warning of malignant cardiac events.

[0081] The circadian phase refers to a specific point in time within an organism's diurnal rhythm cycle, reflecting the synchronization between the peak expression of core circadian clock genes (such as BMAL1 and PER2) and the external environment (light / dark cycle).

[0082] The core body temperature minimum (CBTmin) refers to the lowest point of the human body's core body temperature in a 24-hour diurnal cycle. This time point is highly synchronized with the trough of expression of the core biological clock gene BMAL1 (correlation coefficient r=0.89, p<0.001).

[0083] Methods for determining dynamic dosing timing strategies for metabolic rhythm typing in myocardial infarction include:

[0084] Body temperature monitoring: Body temperature data was continuously collected for 72 hours using wearable devices. The trough value of the body temperature rhythm, CBTmin = M + A · cos(2πt / τ + φ), was calculated using the cosine fitting method, and the baseline phase φ of the individual's biological clock was located; (clinically acceptable error ±15 minutes).

[0085] Wherein, CBTmin = M + A · cos(2πt / τ + φ), is a standard cosine function model used to simulate and fit biological rhythm data with periodic fluctuations.

[0086] 1. M (Mesor): Meaning: Rhythm-adjusted median. It represents the average level or median of the fluctuations in body temperature rhythm over the entire cycle (e.g., 24 hours). It is not a simple arithmetic mean, but a more stable benchmark value obtained after fitting a rhythm model.

[0087] 2. A (Amplitude): Meaning: Amplitude. It represents half the amplitude of the body temperature rhythm fluctuation, that is, the maximum distance from the midline (M) to the peak or trough.

[0088] Its role in the model: It determines the peak height and trough depth of the cosine curve. The larger the amplitude, the more significant the diurnal temperature difference in an individual; a smaller amplitude (decreased amplitude) is a key characteristic of rhythm disorder.

[0089] 3. t (Time): Meaning: Time variable. It is a continuously changing point in time during the data acquisition process.

[0090] Its role in the model: the independent variable in the formula. It represents a series of consecutive time points (t1, t2, t3, ... t...). n Substituting these values ​​into the formula, the predicted body temperature value for the corresponding time can be calculated, thus fitting the entire curve.

[0091] 4. τ (tau): Meaning: Period. It represents the length of time for a complete cycle of a rhythm.

[0092] Role in the model: For the vast majority of people, the period (τ) of their circadian rhythm is very close to 24 hours. During the fitting process, τ is sometimes fixed at 24 hours to conform to known biological clock cycles.

[0093] 5. φ (phi): Meaning: Phase angle or peak phase. It is a parameter that determines the left and right shift of the cosine curve on the time axis, directly reflecting the internal time of the rhythm.

[0094] The φ value obtained through fitting can be used to accurately calculate the peak time (Acrophase) or trough time (Bathyphase, i.e., CBTmin) of the rhythm. The formula for calculating CBTmin is: CBTmin time = (π - φ) / (2π) * τ;

[0095] φ precisely quantifies the reference phase of an individual's biological clock relative to an external clock (such as local time), and is the core basis for developing personalized drug delivery strategies.

[0096] This invention provides an example for calculating the trough value of body temperature rhythm (CBTmin): Assuming that cosine fitting is performed on 72 hours of body temperature data of a patient, the following parameters are obtained:

[0097] M = 36.6 °C; A = 0.3 °C; τ = 24 h (fixed); φ = 3.0 radians;

[0098] Calculate the time when CBTmin occurs:

[0099] CBTmin time = (π - φ) / (2π) * τ = (3.1416 - 3.0) / (2 * 3.1416) * 24 ≈ (0.1416 / 6.2832) * 24 ≈ 0.54 hours. Dosing strategy: If the patient is classified as CIR, the optimal dosing time should be CBTmin + 4 hours, i.e., approximately 04:32 AM.

[0100] This timing precisely matches the stage when the body's internal biological clock begins to rise from its trough and genes such as BMAL1 begin to be actively expressed, thus maximizing the therapeutic effect of drugs (such as clock synchronizers).

[0101] Molecular detection can also be used: within a 4-hour interval before and after CBTmin, peripheral blood is used every 2 hours to measure the peak expression of BMAL1 gene and the melatonin secretion time (DLMO).

[0102] Phase characteristic analysis of the trough CBTmin of the body temperature rhythm was performed to determine the optimal dosing strategy, as shown in Table 4. Among them, the CIR type was administered 4 hours after the trough CBTmin (e.g., melatonin extended-release tablets), the MET type was administered during the rising phase of body temperature (e.g., BCAT2 activators), and the MIX type (mixed type) was administered 4 hours after the trough CBTmin and during the rising phase of body temperature (combining the above two strategies).

[0103] Table 4. Specific rules for adjusting dosing time (using phase shift strategy based on subtype)

[0104]

[0105] The dosing strategy is based on scientifically calculated phase-shifted dosing at the critical time point of CBTmin. The rationale is as follows: 1. Metabolic trough: At CBTmin, the activity of drug-metabolizing enzymes (such as CYP450) is only 25-30% of its peak, significantly reducing bioavailability. 2. Receptor sensitivity: Melatonin receptors MT1 / MT2 expression levels reach peak 4 hours after CBTmin.

[0106] Based on the optimal dosing strategy, dynamic time windows were used to calculate and monitor body temperature and branched-chain amino acid (BCAA) concentration levels in real time to determine the optimal dosing time for each myocardial infarction metabolic rhythm subtype.

[0107] Methods for optimizing drug dosage based on myocardial infarction metabolic rhythm typing through dose gradient experiments include:

[0108] To classify the metabolic rhythm of myocardial infarction, low, medium, and high dose groups were set up, corresponding to low-dose administration, medium-dose administration, and high-dose administration, respectively. Myocardial BCAT2 activity, mitochondrial function, and abnormal liver enzyme rates were measured in the low, medium, and high dose groups.

[0109] The optimal dosage for myocardial infarction metabolic rhythm classification was determined based on myocardial BCAT2 activity, mitochondrial function, and abnormal liver enzyme rates.

[0110] The smart pillbox receives the calculated optimal dosing time and optimal dosage via NFC. Within the target time window (±15 minutes), it releases the pre-loaded drug (such as melatonin extended-release tablets) according to the optimal dosage and simultaneously activates the electrocardiogram monitoring module to record the ST segment (the smooth line segment from the end of the QRS complex to the beginning of the T wave in the electrocardiogram, reflecting the early state of ventricular repolarization) changes after medication.

[0111] This invention is based on the REVIVE-AMI clinical trial, with a total trial size of n=215. The individualized dosing strategy was adjusted according to CBTmin+4h (CIR type) or CBTmin+6h (MET type), while the fixed dosing strategy was to administer the same drug (melatonin extended-release tablets or BCAT2 activator) uniformly at 21:00. Data was collected 3 months after the intervention. Based on the analysis of the collected data, the dosing timing was automatically optimized based on individualized circadian rhythm phase (the efficacy was improved by 2.3 times compared with the fixed regimen, and the NT-proBNP reduction rate was 58.3% vs 31.4%, p<0.01), as shown in Tables 5 and 6.

[0112] Table 5. Comparison of efficacy between individualized dosing and fixed-dose dosing

[0113]

[0114] Table 6. Subtype stratification analysis (relationship between BMAL1 expression time and drug administration time, BCAT2 activity detection data for each subtype)

[0115] This invention adjusts the dosing time based on the patient's trough of body temperature rhythm (reflecting the trough of endogenous BMAL1 activity) to maximize drug efficacy (solving the timing mismatch of fixed dosing). By matching the patient's biological clock phase (e.g., dosing 4 hours after CBTmin in CIR patients), the efficacy of clock synchronizers is improved by 2.3 times (NT-proBNP reduction rate 58.3% vs 31.4%, p<0.01). This strategy can be extended to cardiovascular disease prevention in high-risk groups with circadian rhythm disorders, such as shift workers and cross-time zone travelers.

[0116] This invention provides an example of optimizing the BCAT2 activator dosage for MET-type patients, and the testing method is as follows:

[0117] 1. Subtype screening: MET type (BCAT2 activity <30%+BCAAs / BCAT2≥2.0) was identified from 200 patients, totaling 45 cases.

[0118] 2. Dose gradient test:

[0119] Low-dose group (n=15): BCAT2 activator (phenylpropionic acid derivative 50mg / day);

[0120] Medium-dose group (n=15): 100 mg / day;

[0121] High-dose group (n=15): 200 mg / day;

[0122] 3. Monitoring indicators: 7 days after treatment:

[0123] Myocardial BCAT2 activity (enzyme-linked immunosorbent assay);

[0124] Mitochondrial function (OCR assay of myocardial biopsy).

[0125] 4. The test results are shown in Table 7.

[0126] Table 7. Detection results of the dose gradient test

[0127] Statistical notes: *vs 50mg p<0.05, the data group marked with * is statistically significant compared with the 50mg dose group; #vs other groups p<0.01, the data group marked with # is highly statistically significant compared with all other groups (i.e., the 50mg and 100mg groups).

[0128] The results of the dose gradient test show that 100mg is the optimal dose, balancing efficacy and safety (78% activity recovery and controllable hepatotoxicity).

[0129] This invention presents an optimized BCAT2 activator dosage regimen (100 mg / day) for MET-type patients, which restores 78% of enzyme activity while controlling hepatotoxicity to below 7% (vs. 23% in the 200 mg group, p<0.01). This achievement provides the first targeted therapy option with a clearly defined safety window for metabolic dysplastic myocardial infarction.

[0130] like Figure 2 As shown, this invention provides an intelligent metabolic rhythm typing and drug administration decision system for acute myocardial infarction, applied to an intelligent metabolic rhythm typing and drug administration decision method for acute myocardial infarction. The system includes:

[0131] Multi-index detection kit for simultaneous detection of BMAL1 mRNA, USP2-BCAT2 complex and branched-chain amino acids BCAAs;

[0132] A wearable monitoring patch integrates a sweat BCAAs sensor and an electrocardiogram monitoring module to monitor and record the concentration changes of branched-chain amino acids (BCAAs) in blood or other body fluids over a complete circadian cycle, thereby obtaining a BCAAs circadian rhythm curve.

[0133] The data processing unit is used to establish an intelligent model for myocardial infarction metabolic rhythm typing using a random forest model. The model takes the circadian rhythm curves of the biological clock gene BMAL1 mRNA, USP2-BCAT2 complex and branched chain amino acids BCAAs as input and myocardial infarction metabolic rhythm typing as output.

[0134] The early warning monitoring unit is used to determine the BCAA fluctuation nodes based on the BCAA diurnal rhythm curves and USP2-BCAT2 complex collected by the wearable monitoring patch, and to provide early warning of metabolic abnormalities in myocardial infarction metabolic rhythm classification at the BCAA fluctuation nodes.

[0135] The dosing decision unit is used to determine the dynamic dosing timing strategy for myocardial infarction metabolic rhythm classification by matching individualized biological clock phases; and to optimize the dosing dose for myocardial infarction metabolic rhythm classification through dose gradient experiments to obtain the optimal dosing dose for myocardial infarction metabolic rhythm classification.

[0136] NFC-triggered pillboxes are used for intelligent dynamic drug delivery based on dynamic dosing timing strategies and optimal dosage.

[0137] The dosing decision unit includes a time decision module for developing a dynamic dosing timing strategy and a dose decision module for determining the optimal dosing dose;

[0138] The time decision module uses wearable devices to continuously collect body temperature data for 72 hours, and uses the cosine fitting method to calculate the trough value of body temperature rhythm CBTmin = M + A · cos(2πt / τ + φ), and locates the individual biological clock reference phase φ;

[0139] Phase characteristic analysis of the trough CBTmin of body temperature rhythm was performed to determine the optimal dosing strategy. Specifically, for the CIR type, the drug was administered 4 hours after the trough CBTmin; for the MET type, the drug was administered during the rising phase of body temperature; and for the MIX type, the drug was administered both 4 hours after the trough CBTmin and during the rising phase of body temperature.

[0140] Based on the optimal dosing strategy, dynamic time windows were used to calculate and monitor body temperature and branched-chain amino acid (BCAA) concentration levels in real time to determine the optimal dosing time for each myocardial infarction metabolic rhythm subtype.

[0141] The dosage decision module sets low, medium, and high dose groups for myocardial infarction metabolic rhythm classification, corresponding to low-dose administration, medium-dose administration, and high-dose administration, respectively, and detects myocardial BCAT2 activity, mitochondrial function, and abnormal liver enzyme rates in the low, medium, and high dose groups;

[0142] The optimal dosage for myocardial infarction metabolic rhythm classification was determined based on myocardial BCAT2 activity, mitochondrial function, and abnormal liver enzyme rates.

[0143] This invention provides a computer-readable storage medium storing computer-executable instructions. When a processor executes the computer-executable instructions, it implements a method for intelligent classification and drug administration decision-making based on metabolic rhythm in acute myocardial infarction.

[0144] This invention integrates multi-dimensional data on BMAL1 methylation, USP2-BCAT2 complex, and dynamic BCAAs curves to achieve, for the first time, a metabolic rhythm classification (CIR / MET / MIX / NORM) for myocardial infarction. Compared with the traditional STEMI / NSTEMI classification, the detection rate of high-risk patients is increased from 52% to 91% (p<0.001).

[0145] This invention is based on the technology of continuous monitoring of sweat BCAAs by intelligent patches, which can detect metabolic abnormalities 3 hours earlier than traditional blood tests (peak delay 10.2 vs 182.4 minutes, p<0.001), and combine AI algorithms to adjust the dosing regimen in real time.

[0146] This invention improves the efficacy of clock synchronizers by matching the patient's biological clock phase (e.g., administration of CIR type 4 hours after CBTmin) by 2.3 times (reduction rate of N-terminal B-type natriuretic peptide precursor NT-proBNP 58.3% vs 31.4%, p<0.01).

[0147] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.

Claims

1. A method for intelligent classification of myocardial infarction metabolic rhythm and drug administration decision-making, characterized by: The method comprises the following steps: Synchronously detecting biological clock gene BMAL1 mRNA, USP2-BCAT2 complex and branched chain amino acid BCAAs through a multi-index joint detection kit; Monitoring and recording the concentration change of branched chain amino acid BCAAs in blood or other body fluids in a complete circadian cycle through a wearable monitoring patch integrating a sweat BCAAs sensor and an electrocardio monitoring module to obtain a BCAAs circadian rhythm curve; Establishing a myocardial infarction metabolic rhythm typing intelligent model taking biological clock gene BMAL1 mRNA, USP2-BCAT2 complex and branched chain amino acid BCAAs and the BCAAs circadian rhythm curve as inputs and taking myocardial infarction metabolic rhythm typing as output by using a random forest model; Determining a BCAAs fluctuation node according to the BCAAs circadian rhythm curve and the USP2-BCAT2 complex obtained by the wearable monitoring patch, and performing metabolic abnormality early warning of myocardial infarction metabolic rhythm typing at the BCAAs fluctuation node; Determining a dynamic drug administration timing strategy of myocardial infarction metabolic rhythm typing by matching individual biological clock phase with the myocardial infarction metabolic rhythm typing; Optimizing the drug administration dose of myocardial infarction metabolic rhythm typing through a dose gradient experiment to obtain an optimal drug administration dose of myocardial infarction metabolic rhythm typing; The BCAAs fluctuation node is a time point at which the concentration ratio BCAAs / BCAT2 between branched chain amino acid BCAAs and BCAT2 activity exceeds a threshold value, wherein the total amount of BCAT2 is obtained through immunoprecipitation detection, and the activity of BCAT2 is calculated through the concentration difference between the total amount of BCAT2 and the USP2-BCAT2 complex; The method for determining the dynamic drug administration timing strategy of myocardial infarction metabolic rhythm typing comprises: Continuously collecting 72-hour body temperature data by using a wearable device, calculating the body temperature rhythm valley CBTmin = M + A · cos(2πt / τ + φ) by using a cosine fitting method, and locating the individual biological clock reference phase φ, wherein M is a rhythm adjustment median value, A is an amplitude, t is a time variable, and τ is a period; Performing phase feature analysis on the body temperature rhythm valley CBTmin to determine an optimal drug administration strategy, wherein CIR type is administered 4 hours after the body temperature rhythm valley CBTmin, MET type is administered during the body temperature rising period, and MIX type is administered 4 hours after the body temperature rhythm valley CBTmin and during the body temperature rising period; According to the optimal drug administration strategy, real-time monitoring of body temperature and branched chain amino acid BCAAs concentration level is performed by using dynamic time window calculation to determine the optimal drug administration time of each myocardial infarction metabolic rhythm typing.

2. The intelligent myocardial infarction metabolic rhythm typing and dosing decision method according to claim 1, characterized in that: The myocardial infarction metabolic rhythm typing comprises CIR type, MET type, MIX type and NORM type.

3. The method of claim 1, wherein the method is characterized by: The biological clock gene BMAL1 mRNA is obtained through qPCR detection, the USP2-BCAT2 complex is obtained through time-resolved fluorescence immunoassay, and the branched chain amino acid BCAAs is obtained through enzyme color test paper detection.

4. The method of intelligent classification of metabolic rhythm of myocardial infarction and dosing decision according to claim 1, characterized in that: The method for optimizing the drug administration dose of myocardial infarction metabolic rhythm typing through a dose gradient experiment comprises: The low, medium and high dose groups are set for the metabolic rhythm classification of myocardial infarction, respectively corresponding to low dose administration, medium dose administration and high dose administration, and the myocardial BCAT2 activity, mitochondrial function and liver enzyme abnormality rate of the low, medium and high dose groups are detected; The optimal administration dose of the metabolic rhythm classification of myocardial infarction is determined according to the myocardial BCAT2 activity, mitochondrial function and liver enzyme abnormality rate.

5. The intelligent myocardial infarction metabolic rhythm typing and dosing decision method according to claim 1, characterized in that: The intelligent medicine box receives the optimal administration time and optimal administration dose calculated by NFC, releases the pre-loaded medicine according to the optimal administration dose within the target time window ± 15 minutes, and synchronously activates the electrocardio monitoring module to record the ST segment change data after administration.

6. A myocardial infarction metabolic rhythm intelligent classification and dosing decision system, characterized in that, The system is applied to the intelligent classification and administration decision method of the metabolic rhythm of myocardial infarction according to any one of claims 1-5, and comprises: a multi-index joint detection kit for synchronously detecting the biological clock gene BMAL1 mRNA, the USP2-BCAT2 complex and the branched chain amino acid BCAAs; A wearable monitoring patch integrates a sweat BCAAs sensor and an electrocardio monitoring module, and is used for monitoring and recording the concentration change of the branched chain amino acid BCAAs in blood or other body fluids in a complete day-night cycle to obtain a BCAAs day-night rhythm curve; A data processing unit is used for establishing an intelligent model of the metabolic rhythm classification of myocardial infarction by using a random forest model, taking the biological clock gene BMAL1 mRNA, the USP2-BCAT2 complex and the branched chain amino acid BCAAs and the BCAAs day-night rhythm curve as inputs, and taking the metabolic rhythm classification of myocardial infarction as output; An early warning monitoring unit is used for determining a BCAAs fluctuation node according to the BCAAs day-night rhythm curve and the USP2-BCAT2 complex obtained by the wearable monitoring patch, and performing metabolic abnormality early warning of the metabolic rhythm classification of myocardial infarction at the BCAAs fluctuation node; An administration decision unit is used for determining a dynamic administration timing strategy of the metabolic rhythm classification of myocardial infarction by matching individual biological clock phases for the metabolic rhythm classification of myocardial infarction, and optimizing the administration dose of the metabolic rhythm classification of myocardial infarction by a dose gradient experiment to obtain an optimal administration dose of the metabolic rhythm classification of myocardial infarction; An NFC triggered medicine box is used for intelligent dynamic administration according to the dynamic administration timing strategy and the optimal administration dose.

7. The myocardial infarction metabolic rhythm intelligent typing and dosing decision system according to claim 6, characterized in that: The administration decision unit comprises a time decision module for formulating the dynamic administration timing strategy and a dose decision module for determining the optimal administration dose; The time decision module continuously collects 72-hour body temperature data by using a wearable device, calculates a body temperature rhythm valley CBTmin = M + A · cos(2πt / τ + φ) by using a cosine fitting method, and locates an individual biological clock reference phase φ, wherein M is a rhythm adjustment median value, A is an amplitude, t is a time variable, and τ is a period. Phase feature analysis of CBTmin is performed to determine the optimal administration strategy, wherein: CIR type is administered 4 hours after CBTmin, MET type is administered during the temperature rise period, and MIX type is administered 4 hours after CBTmin and during the temperature rise period; According to the optimal administration strategy, the dynamic time window calculation is used to monitor the temperature and the concentration level of BCAAs in real time, and the optimal administration time of each myocardial infarction metabolic rhythm type is determined.

8. The myocardial infarction metabolic rhythm intelligent typing and dosing decision system of claim 6, wherein: The dose decision module sets low, medium and high dose groups for myocardial infarction metabolic rhythm types, which correspond to low dose administration, medium dose administration and high dose administration respectively, and detects myocardial BCAT2 activity, mitochondrial function and liver enzyme abnormality rate of the low, medium and high dose groups; The optimal administration dose of myocardial infarction metabolic rhythm type is determined according to myocardial BCAT2 activity, mitochondrial function and liver enzyme abnormality rate.

Images