Retrospective analysis method and system for calculating physiological data threshold based on mathematical inflection point

Abstract

This invention belongs to the field of medical data analysis and provides a retrospective analysis method and system for calculating physiological data thresholds based on mathematical inflection points. The method includes: acquiring the patient's historical physiological data and the patient's outcome data; using the patient's historical physiological data as the independent variable and the patient's outcome data as the dependent variable, constructing a regression analysis model using a machine learning classification algorithm; performing second-order differentiation on the regression model to obtain the point where the second derivative is zero, which is the threshold; and predicting the occurrence of postoperative abnormalities based on the threshold. This invention calculates the corresponding threshold using mathematical means, providing reliable mathematical support and theoretical basis for the origin of the threshold. Furthermore, it allows for automated calculation of the corresponding threshold by designing computer programs, ensuring the uniqueness of the results for each calculation and eliminating interference from human factors.

Description

Technical Field

[0001] This invention belongs to the field of medical data analysis, and in particular relates to a retrospective analysis method and system for calculating physiological data thresholds based on mathematical inflection points. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Retrospective cohort studies group participants based on characteristics or exposures at a past point in time, and then further group them based on mortality or disease occurrences of each member from that point to a later point in time, or by tracing existing records back to the study point. Based on this, the causal relationship between the exposure factor and the medical research outcome is evaluated. Retrospective analysis is a common analytical tool in the medical field, used to analyze the relationship between data and outcomes based on historical data.

[0004] The existing retrospective analysis process involves organizing and summarizing historical data, matching it with outcome data, then using mathematical modeling to plot the probability of the outcome against physiological data. This is followed by analysis of clinical characteristics to determine a threshold, and finally calculating the area under the ROC curve (AUC) and other relevant data. However, this threshold-finding stage relies heavily on researchers' comprehensive judgment combined with clinical results, requiring a high level of expertise from the researchers. This can lead to inaccuracies due to subjective factors or differences in individual researchers' knowledge levels.

[0005] At the same time, since traditional methods are not end-to-end processing flows, different personnel are required to participate in the same queue task from data processing to data analysis. Due to the special nature of the work of medical personnel, this process often takes several days or even a month. In addition, experienced doctors are required to analyze the data based on clinical data, which leads to low efficiency in data analysis and causes delays in data analysis. Summary of the Invention

[0006] To address the technical issues in retrospective analysis of outcomes mentioned above, such as the need for manual threshold determination, this invention provides a retrospective analysis method and system based on mathematical inflection points to calculate physiological data thresholds. This method uses mathematical means to calculate the corresponding thresholds, providing reliable mathematical support and theoretical basis for their derivation. Furthermore, it allows for automated threshold calculation via computer programs, ensuring the uniqueness of the results for each calculation and eliminating interference from human factors.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] The first aspect of the present invention provides a retrospective analysis method for calculating physiological data thresholds based on mathematical inflection points.

[0009] Retrospective analysis methods based on mathematical inflection points to calculate physiological data thresholds include:

[0010] Obtain the patient's historical physiological data, as well as the patient's outcome data;

[0011] Using the patient's historical physiological data as the independent variable and the patient's outcome data as the dependent variable, a regression analysis model was constructed using a machine learning classification algorithm;

[0012] Taking the second derivative of the regression model, the point where the second derivative is zero is the threshold.

[0013] Based on the threshold, the occurrence of postoperative abnormalities in patients can be predicted.

[0014] Furthermore, the patient's historical physiological data includes non-invasive high blood pressure data, non-invasive low blood pressure data, invasive high blood pressure data, and invasive low blood pressure data, and the patient's outcome data includes delayed cerebral infarction.

[0015] Furthermore, based on non-invasive high blood pressure data, non-invasive low blood pressure data, invasive high blood pressure data, and invasive low blood pressure data, the patient's invasive mean arterial pressure and non-invasive mean arterial pressure were calculated.

[0016] Furthermore, the invasive mean arterial pressure and non-invasive mean arterial pressure of the patient are smoothed and rounded.

[0017] Furthermore, the invasive mean arterial pressure and non-invasive mean arterial pressure of the patient after smoothing and rounding are fused to obtain fused arterial pressure data.

[0018] Furthermore, using the minimum value in the fused arterial pressure data as the independent variable and the patient's outcome data corresponding to the minimum value as the dependent variable, a regression analysis model was constructed using a machine learning algorithm.

[0019] Furthermore, the fusion includes: selecting the invasive mean arterial pressure when both invasive and non-invasive mean arterial pressure are present; selecting the invasive mean arterial pressure when only the invasive mean arterial pressure is present; and selecting the non-invasive mean arterial pressure when only the non-invasive mean arterial pressure is present.

[0020] A second aspect of the present invention provides a retrospective analysis system for calculating physiological data thresholds based on mathematical inflection points.

[0021] A retrospective analysis system based on mathematical inflection points to calculate physiological data thresholds includes:

[0022] The data acquisition module is configured to acquire the patient's historical physiological data and the patient's outcome data.

[0023] The model building module is configured to use the patient's historical physiological data as the independent variable and the patient's outcome data as the dependent variable, and to use a machine learning classification algorithm to build a regression analysis model.

[0024] The differentiation module is configured to perform second-order differentiation on the regression model to obtain the point where the second derivative is zero, which is the threshold.

[0025] The prediction module is configured to predict the occurrence of postoperative abnormalities in patients based on the threshold.

[0026] A third aspect of the present invention provides a computer-readable storage medium.

[0027] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the retrospective analysis method for calculating physiological data thresholds based on mathematical inflection points as described in the first aspect above.

[0028] A fourth aspect of the present invention provides a computer device.

[0029] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps in the retrospective analysis method for calculating physiological data thresholds based on mathematical inflection points as described in the first aspect above.

[0030] Compared with the prior art, the beneficial effects of the present invention are:

[0031] This invention provides theoretical and technical support for retrospective analysis of medical outcomes by using mathematical inflection points as physiological data thresholds. It transforms the retrospective analysis problem into an end-to-end processing flow, improving the work efficiency of researchers in the medical field and reducing inaccuracies caused by differences in researchers' skill levels and experience.

[0032] The inflection point technology solution provided by this invention differs from the clinical inflection point solution originally provided by hospital experts. This invention applies the inflection point technology in mathematics to this type of retrospective analysis task and achieves results that are close to or even exceed those of the clinical inflection point, providing new ideas and directions for this type of experiment.

[0033] The inflection point calculated by this invention is quite close to the clinical inflection point results. It obtains more accurate and quantitative results from a more professional mathematical perspective, making the technical solution of this invention more scientific. In contrast, the determination of clinical results relies on the support of doctors' experience data, which places very high demands on doctors. The method of determining the inflection point through mathematical formulas is more scientific, improves calculation speed, and makes the data more interpretable and quantifiable. Attached Figure Description

[0034] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0035] Figure 1 This is a flowchart of a retrospective analysis method for calculating physiological data thresholds based on mathematical inflection points, as shown in Embodiment 1 of the present invention.

[0036] Figure 2 This is an image result diagram shown in Embodiment 1 of the present invention;

[0037] Figure 3 This is a structural diagram of a retrospective analysis system for calculating physiological data thresholds based on mathematical inflection points, as shown in Embodiment 2 of the present invention. Detailed Implementation

[0038] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0039] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0040] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0041] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and systems according to various embodiments of this disclosure. It should be noted that each block in a flowchart or block diagram may represent a module, segment, or portion of code, which may include one or more executable instructions for implementing the logical functions specified in the various embodiments. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutively represented blocks may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, may be implemented using a dedicated hardware-based system that performs the specified functions or operations, or using a combination of dedicated hardware and computer instructions.

[0042] Example 1

[0043] like Figure 1 As shown, this embodiment provides a retrospective analysis method for calculating physiological data thresholds based on mathematical inflection points. This embodiment uses the application of this method to a server as an example for illustration. It is understood that this method can also be applied to terminals, and can also be applied to systems including terminals, servers, and other components, and can be implemented through interaction between the terminal and the server. The server can be an independent physical server, a server cluster composed of multiple physical servers, or a distributed system. It can also be a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network servers, cloud communication, middleware services, domain name services, CDN security services, and big data and artificial intelligence platforms. The terminal can be a smartphone, tablet, laptop, desktop computer, etc., but is not limited to these. The terminal and server can be directly or indirectly connected via wired or wireless communication, which is not limited herein. In this embodiment, the method includes the following steps:

[0044] Obtain the patient's historical physiological data, as well as the patient's outcome data;

[0045] Using the patient's historical physiological data as the independent variable and the patient's outcome data as the dependent variable, a regression analysis model was constructed using a machine learning classification algorithm;

[0046] Taking the second derivative of the regression model, the point where the second derivative is zero is the threshold.

[0047] Based on the threshold, the occurrence of postoperative abnormalities in patients can be predicted.

[0048] This invention is applicable to various retrospective analysis tasks with outcomes. This embodiment uses a retrospective analysis task of intraoperative blood pressure data (or brain oxygenation, heart rate, and bis data) cohort with delayed-onset stroke (DCI) as an example, but it is not limited to tasks with the DCI outcome. Other outcomes such as stroke, total myocardial necrosis (TNI), endovascular treatment (EVT) for neurosurgical aneurysm hemorrhage, gastric diathermy-fluid therapy (GDFT), and aneurysmal subarachnoid hemorrhage (aSAH) are also directly applicable. This embodiment uses intraoperative blood pressure data from 539 patients undergoing craniotomy at a tertiary hospital. Blood pressure data was recorded every 10 seconds, or 6 data points per minute. Among them, 369 patients had negative data and 170 patients had positive data.

[0049] The system collects input data including non-invasive systolic blood pressure (non-invasive systolic), non-invasive diastolic blood pressure (non-invasive diastolic), invasive systolic blood pressure (invasive systolic), and invasive diastolic blood pressure (invasive diastolic), along with the corresponding outcome indicators. A positive result is recorded as 1, and a negative result as 0. These outcome data labels are determined by professional physicians. Non-invasive systolic, non-invasive diastolic, invasive systolic, and invasive diastolic blood pressure data are collected using specialized hospital equipment.

[0050] Note: Positive refers to individuals who develop DCI within 30 days post-surgery, while negative refers to individuals who do not develop DCI within 30 days post-surgery.

[0051] When the outcome is delayed cerebral infarction and the input physiological data is blood pressure data, the specific implementation scheme of this embodiment can be implemented with reference to the following:

[0052] (1) Data preprocessing

[0053] (1-1) Load the data for each patient: non-invasive high blood pressure and non-invasive low blood pressure, invasive high blood pressure and invasive low blood pressure;

[0054] (1-2) Calculate the MAP (mean arterial pressure) for each patient, which is divided into invasive_MAP (mean arterial pressure with invasive surgery) and non_invasive_MAP (mean arterial pressure without invasive surgery). The specific formulas are shown in formulas (1) and (2) respectively:

[0055] invasive_MAP=(invas_sys+2*invas_dia) / 3 (1)

[0056] non_invasive_MAP=(non_invas_sys+2*non_invas_dia) / 3 (2)

[0057] (1-3) Perform a Moving Average (MA) operation on the calculated invasive_MAP and non_invasive_MAP data to smooth them, and then obtain invasive_MAP_MA and non_invasive_MAP_MA; see formula (3) for the specific formula:

[0058]

[0059] Where N is the window period, which is set to 30 in this task, that is, a data length of 5 minutes.

[0060] (1-4) Round the invasive_MAP_MA and non_invasive_MAP_MA obtained in the previous step. The rounding method is to round to the nearest whole number.

[0061] (1-5) Based on the rounded results, the data are fused according to the following principles: if non-invasive blood pressure exists but invasive blood pressure does not at a certain time point, non-invasive blood pressure is selected; if both non-invasive and invasive blood pressure exist, invasive blood pressure is selected; if only invasive blood pressure exists at a certain time point, invasive blood pressure is selected. The final fused data is denoted as valid_MAP_MA, and the specific formula is as follows:

[0062]

[0063] Among them, condition 1 is: at time i, invasive pressure and non-invasive pressure exist simultaneously; condition 2 is: at time i, only non-invasive pressure exists; condition 3 is: at time i, only invasive pressure exists.

[0064] (2) Mathematical modeling

[0065] (2-1) Here, the modeling can be done using machine learning classification methods such as logistic regression (LR), random forest (RF), support vector machine (SVM), decision tree (DT), GBDT, XGBoost, and K-Means, all of which can be used for this modeling task;

[0066] (2-2) In this task, the model chosen is the logistic regression model (LR). The logistic function (Sigmoid function) maps data features to a probability value in the interval of 0 to 1 (the probability that a sample belongs to the positive example). The classification of the data is determined by comparing it with a threshold of 0.5. The specific formula is as follows:

[0067]

[0068] z = W T x+w0 (6)

[0069] (2-3) Take the minimum value of the valid_MAP_MA data in formula (4) for each patient to obtain the min_valid_MAP_MA data. The specific formula is as follows:

[0070] min_valid_MAP_MA k =min(valid_MAP_MA k (7)

[0071] Where k is the kth patient.

[0072] (2-4) Match all patient outcome data with min_valid_MAP_MA according to patient ID;

[0073] (2-5) Perform mathematical modeling -- LR model establishment: input the independent variable min_valid_MAP_MA and the dependent variable outcome into the model, train it, and obtain model M;

[0074] (2-6) Fill the independent variable min_valid_MAP_MA with arithmetic interpolation according to the maximum and minimum values ​​to obtain the continuous sequence min_valid_MAP_MA_inter, and obtain the corresponding probability prediction_outcome sequence of outcome through model M;

[0075] (3) Inflection point analysis

[0076] (3-1) The mathematical inflection point used here is: Suppose that the function y = f(x) is continuous in a certain neighborhood of the point x0. If (x0, f(x0)) is the boundary between the concavity and convexity of the curve y = f(x), then (x0, f(x0)) is called the inflection point of the curve y = f(x), in other words, the point where the second derivative is zero.

[0077] (3-2) Fit the results min_valid_MAP_MA_inter and predict_outcome obtained after modeling in step (2-6) to the curve y=f(x), then solve for the corresponding inflection point (x0, f(x0)), and record the corresponding x0 as cut_off;

[0078] (3-3) Plot the inflection point results on the image and mark the points.

[0079] (4) Parameter Calculation Module

[0080] (4-1) Calculate the difference. Calculate the value below the cut-off for each patient, denoted as dMAP. The specific formula is as follows:

[0081] dMAP=cut_off-valid_MAP_MA (8)

[0082] (4-2) Calculate the AUC_MAP parameter and calculate the area under the curve below the cut_off:

[0083]

[0084] (4-3) Calculate the duration below the cut_off value, denoted as td: the cumulative duration when AUC_MAP > 0. If all MAPs are greater than cut_off, then td = 0.

[0085] (4-4) Calculate the average MAP decrease, denoted as avMAP, using the following formula:

[0086] avMAP = AUC_MAP / td (10)

[0087] (4-5) Calculate the time-weighted MAP, denoted as twMAP, using the following formula:

[0088] twMAP = AUC_MAP / st (11)

[0089] Where st is the operation duration, which is the difference between the time point of the last valid data point and the time point of the first valid data point.

[0090] By calculating parameters such as avMAP and twMAP, it can be used to predict whether patients will experience adverse events in organs such as the brain, heart, and kidneys after surgery, and to guide the intervention of adverse events after surgery.

[0091] This embodiment uses blood pressure data and DCI outcome data from 539 craniotomy patients at a tertiary hospital for mathematical modeling and inflection point analysis, while simultaneously calculating subsequent parameter results. In this task, the inflection point calculated in this embodiment is 68.5 mmHg, which is very close to the clinically observed value of 70 mmHg, proving the effectiveness of the method described in this embodiment. Specific image results are shown below. Figure 2 As shown.

[0092] This embodiment selects the first 20 surgical patients, and the parameters calculated based on the desired inflection point cut_off are shown in Table 1 below:

[0093] Table 1. Parameters of the top 20 surgical patients

[0094]

[0095]

[0096] Example 2

[0097] This embodiment provides a retrospective analysis system for calculating physiological data thresholds based on mathematical inflection points.

[0098] like Figure 3 As shown, a retrospective analysis system for calculating physiological data thresholds based on mathematical inflection points includes:

[0099] The data acquisition module is configured to acquire the patient's historical physiological data and the patient's outcome data.

[0100] The model building module is configured to use the patient's historical physiological data as the independent variable and the patient's outcome data as the dependent variable, and to use a machine learning classification algorithm to build a regression analysis model.

[0101] The differentiation module is configured to perform second-order differentiation on the regression model to obtain the point where the second derivative is zero, which is the threshold.

[0102] The prediction module is configured to predict the occurrence of postoperative abnormalities in patients based on the threshold.

[0103] It should be noted that the data acquisition module, model building module, differentiation module, and prediction module described above are the same examples and application scenarios implemented in the steps of Embodiment 1, but are not limited to the content disclosed in Embodiment 1. It should also be noted that the above modules, as part of the system, can be executed in a computer system such as a set of computer-executable instructions.

[0104] Example 3

[0105] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the retrospective analysis method for calculating physiological data thresholds based on mathematical inflection points as described in Embodiment 1 above.

[0106] Example 4

[0107] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the retrospective analysis method for calculating physiological data thresholds based on mathematical inflection points as described in Embodiment 1 above.

[0108] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0109] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0110] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0111] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0112] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0113] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A retrospective analysis method for calculating physiological data thresholds based on mathematical inflection points, characterized in that, include: Obtain the patient's historical physiological data, as well as the patient's outcome data; The patient's historical physiological data includes non-invasive high blood pressure data, non-invasive low blood pressure data, invasive high blood pressure data, and invasive low blood pressure data. The patient's outcome data includes delayed cerebral infarction. Based on non-invasive high blood pressure data, non-invasive low blood pressure data, invasive high blood pressure data, and invasive low blood pressure data, calculate the patient's invasive mean arterial pressure and non-invasive mean arterial pressure. The invasive mean arterial pressure and non-invasive mean arterial pressure of the patients were smoothed and rounded. The invasive mean arterial pressure and non-invasive mean arterial pressure of the patient after smoothing and rounding are fused to obtain the fused arterial pressure data. Using the minimum value in the fused arterial pressure data as the independent variable and the patient's outcome data corresponding to the minimum value as the dependent variable, a regression analysis model was constructed using machine learning algorithms; The minimum value in the arterial pressure data after fusion of independent variables is filled by arithmetic interpolation based on the maximum and minimum values ​​to obtain a continuous sequence. Then, the probability sequence of the corresponding patient's outcome data is obtained through a regression analysis model. Let the function At point If it is continuous within a certain domain, if ( () is a curve The point where the concave and convex surfaces meet is called (the point where the concave and convex surfaces meet). () is a curve The inflection point is the point where the second derivative is zero; The continuous sequence and probability sequence obtained after modeling are fitted to a curve. Then solve for the corresponding inflection point ( ) results, and the corresponding Recorded as Plot the inflection point results on the image and mark the points; these points are the thresholds. Based on the threshold, the occurrence of postoperative abnormalities in patients can be predicted.

2. The retrospective analysis method for calculating physiological data thresholds based on mathematical inflection points according to claim 1, characterized in that, The fusion includes: selecting the invasive mean arterial pressure when both invasive and non-invasive mean arterial pressure are present; selecting the invasive mean arterial pressure when only the invasive mean arterial pressure is present; and selecting the non-invasive mean arterial pressure when only the non-invasive mean arterial pressure is present.

3. A retrospective analysis system for calculating physiological data thresholds based on mathematical inflection points, characterized in that, include: The data acquisition module is configured to acquire the patient's historical physiological data and the patient's outcome data; the patient's historical physiological data includes non-invasive high blood pressure data, non-invasive low blood pressure data, invasive high blood pressure data and invasive low blood pressure data, and the patient's outcome data includes delayed cerebral infarction. After acquiring the patient's historical physiological data and outcome data, the process further includes: calculating the patient's invasive mean arterial pressure and non-invasive mean arterial pressure based on non-invasive systolic pressure data, non-invasive diastolic pressure data, invasive systolic pressure data, and invasive diastolic pressure data; smoothing and rounding the patient's invasive mean arterial pressure and non-invasive mean arterial pressure; and fusing the smoothed and rounded invasive mean arterial pressure and non-invasive mean arterial pressure to obtain fused arterial pressure data. The model building module is configured to: use the minimum value in the fused arterial pressure data as the independent variable and the patient's outcome data corresponding to the minimum value as the dependent variable, and use a machine learning algorithm to build a regression analysis model; fill the minimum value in the fused arterial pressure data with arithmetic interpolation according to the maximum and minimum values ​​to obtain a continuous sequence, and obtain the probability sequence of the corresponding patient's outcome data through the regression analysis model; The differentiation module is configured to: set the function At point If it is continuous within a certain domain, if ( () is a curve The point where the concave and convex surfaces meet is called (the point where the concave and convex surfaces meet). () is a curve The inflection point is the point where the second derivative is zero; the continuous sequence and probability sequence obtained after modeling are fitted to form a curve. Then solve for the corresponding inflection point ( ) results, and the corresponding Recorded as Plot the inflection point results on the image and mark the points; these points are the thresholds. The prediction module is configured to predict the occurrence of postoperative abnormalities in patients based on the threshold.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the steps in the retrospective analysis method for calculating physiological data thresholds based on mathematical inflection points as described in any one of claims 1-2.

5. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the retrospective analysis method for calculating physiological data thresholds based on mathematical inflection points as described in any one of claims 1-2.

Images