Physiological data analysis method and system for patient risk assessment

Abstract

The invention discloses a physiological data analysis method and system for patient risk assessment, and the method comprises the steps: collecting the multi-dimensional physiological data of a patient, carrying out the feature quantification of the multi-dimensional physiological data, carrying out the normalization processing of the multi-dimensional physiological data after the feature quantification, and obtaining the multi-dimensional physiological features of the patient; based on the multi-dimensional physiological features of the patient, constructing an optimization objective function of a comprehensive risk assessment model, calculating the optimization objective function, obtaining a comprehensive risk assessment score of the patient, adopting a quantization feature mapping mode to consider a quantum optimization objective direction, and constructing a variable component sub-circuit with parameters, so that the comprehensive risk assessment score of the patient is minimized; converting a quantum optimization result into original multi-dimensional physiological data by using a reverse normalization processing mode, and recalculating a comprehensive risk assessment score of the patient; through a quantum optimization strategy and nonlinear modeling, the efficiency and accuracy of physiological data analysis are remarkably improved, and the method can be widely applied to the fields of medical risk assessment and health management.

Description

Technical Field

[0001] The present invention relates to the field of data analysis and quantum computing technology, and in particular to a physiological data analysis method and system for patient risk assessment. Background Art

[0002] In recent years, with the rapid development of medical technology and information technology, the application of multidimensional physiological data in patient health management and disease risk assessment has received widespread attention. These multidimensional physiological data can not only reflect the physiological changes of the human body in different health states, but also become an important basis for disease prevention, diagnosis and treatment. At the same time, the advancement of modern computing technology has made it possible to collect, process and analyze massive physiological data, providing strong support for medical research and clinical practice. In addition, the analysis methods based on physiological data have also undergone iterative upgrades from traditional statistical methods to machine learning models, especially in deep learning, feature extraction and modeling technology, which has further promoted the development of related fields.

[0003] Although these methods and technologies have achieved certain results in feature extraction accuracy, analysis efficiency and accuracy of risk prediction. However, when faced with complex multidimensional time series data and nonlinear interactions, some shortcomings are still exposed, especially in terms of real-time performance, analysis depth and personalized adaptability. First, traditional methods are usually based on univariate or simple multivariate analysis, which makes it difficult to fully capture the complex nonlinear interaction characteristics between multidimensional physiological data, limiting the accuracy of risk assessment. Although machine learning models have been applied in recent years, their deep relationship modeling between multidimensional features is still insufficient, especially in the field of nonlinear interaction modeling. Second, in terms of large-scale data processing and real-time optimization, traditional methods generally rely on models with high computational complexity in order to achieve the accuracy of risk assessment, resulting in low computational efficiency and difficulty in adapting to the needs of real-time risk assessment. Third, traditional methods often lack personalized design capabilities. For example, there are individual differences in the physiological characteristics and health status of different patients, and fixed thresholds may not be applicable to everyone, which limits the application of technology in actual clinical scenarios. In addition, existing quantum computing applications are mostly focused on specific optimization problems, and research on the adaptability of medical data analysis is still in its infancy.

[0004] In summary, the existence of these problems not only constitutes an obstacle to the practical application of technology, but also challenges the scientificity and effectiveness of patient health management. Therefore, how to solve the above bottleneck problems has become the most promising research focus in this field. Summary of the invention

[0005] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] In view of the above existing problems, the present invention is proposed. Therefore, the present invention provides a physiological data analysis method for patient risk assessment to solve the problems mentioned in the background technology.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, the present invention provides a physiological data analysis method for patient risk assessment, comprising:

[0009] Collecting multidimensional physiological data of the patient, quantifying the characteristics of the multidimensional physiological data, and normalizing the multidimensional physiological data after the characteristic quantification to obtain the multidimensional physiological characteristics of the patient;

[0010] Based on the multidimensional physiological characteristics of the patient, an optimization objective function of a comprehensive risk assessment model is constructed, the optimization objective function is calculated, and a comprehensive risk assessment score of the patient is obtained. At the same time, a quantized feature mapping method is adopted to consider the quantum optimization target direction, and a variational quantum circuit with parameters is constructed to minimize the comprehensive risk assessment score of the patient;

[0011] The quantum optimization results are converted into original multidimensional physiological data by using the anti-normalization method, and the patient's comprehensive risk assessment score is recalculated to form a physiological data analysis method for patient risk assessment.

[0012] As a preferred solution of the physiological data analysis method for patient risk assessment described in the present invention, wherein: collecting multi-dimensional physiological data of the patient and quantifying the characteristics of the multi-dimensional physiological data include:

[0013] Collect the patient's multidimensional physiological data X = {x1, x2, ..., x n}, x n For the nth physiological data time series, take out any physiological data time series, divide the physiological data time series into a number of overlapping windows, and set the sliding step of the window;

[0014] For each divided window, the dynamic trend slope of the physiological data is calculated by the linear fitting method. At the same time, for the time series of the extracted physiological data, the difference between the maximum and minimum values ​​of the time series is calculated to obtain the fluctuation range of the physiological data.

[0015] As a preferred embodiment of the physiological data analysis method for patient risk assessment described in the present invention, it also includes:

[0016] Move each window by the set sliding step size, and repeat the calculation process of dynamic trend slope and fluctuation range until each window of the entire time series is calculated.

[0017] As a preferred embodiment of the physiological data analysis method for patient risk assessment described in the present invention, the optimization objective function of the comprehensive risk assessment model is constructed based on the multidimensional physiological characteristics of the patient, and the optimization objective function is calculated to obtain the comprehensive risk assessment score of the patient, including:

[0018] Nonlinear interaction weights are introduced to assign weights to each patient's multidimensional physiological characteristics according to their contribution to risk, thus constructing the optimization objective function of the comprehensive risk assessment model.

[0019] By calculating the optimization objective function, a comprehensive risk assessment score for each patient is obtained.

[0020] As a preferred solution of the physiological data analysis method for patient risk assessment described in the present invention, wherein: a quantized feature mapping method is simultaneously used to consider the quantum optimization target direction, including:

[0021] Before the quantum feature mapping method, the multidimensional physiological characteristics of the patient in the optimization objective function are quantum-state mapped, and the multidimensional physiological characteristics of each patient are mapped into a quantum state;

[0022] Based on the optimization objective function of the comprehensive risk assessment model, a quantum Hamiltonian is constructed to determine the quantum optimization target direction as the minimization of the comprehensive risk assessment score.

[0023] As a preferred embodiment of the physiological data analysis method for patient risk assessment of the present invention, wherein: by constructing a variational quantum circuit with parameters, the comprehensive risk assessment score of the patient is minimized, including:

[0024] The variational quantum circuit adopts a classical-quantum computing strategy, and optimizes the parameters in the variational quantum circuit by alternating and iteratively combining classical computing and quantum computing, gradually bringing the parameters closer to the optimal solution of the optimization objective function, thereby minimizing the comprehensive risk assessment score of the patient.

[0025] As a preferred solution of the physiological data analysis method for patient risk assessment described in the present invention, the quantum optimization results are converted into original multidimensional physiological data by using an anti-normalization processing method, and the comprehensive risk assessment score of the patient is recalculated, including:

[0026] Extracting optimized multidimensional physiological data from the optimization objective function of minimizing the comprehensive risk assessment score, and converting the optimized multidimensional physiological data into a data form having the same representation as the original multidimensional physiological data;

[0027] Using the optimized multidimensional physiological data represented in the same data form as the original multidimensional physiological data as the input of the optimization objective function to obtain a comprehensive risk assessment score of the patient;

[0028] The patient's current risk level is divided according to the comprehensive risk assessment score.

[0029] In a second aspect, the present invention provides a physiological data analysis system for patient risk assessment, comprising:

[0030] A multidimensional physiological data processing module is configured to collect multidimensional physiological data of a patient, perform feature quantification on the multidimensional physiological data, and perform normalization processing on the multidimensional physiological data after feature quantification to obtain multidimensional physiological features of the patient;

[0031] A quantum computing optimization module is configured to construct an optimization objective function of a comprehensive risk assessment model based on the multidimensional physiological characteristics of the patient, calculate the optimization objective function, obtain a comprehensive risk assessment score of the patient, and simultaneously consider the quantum optimization target direction by using a quantized feature mapping method, and minimize the comprehensive risk assessment score of the patient by constructing a variational quantum circuit with parameters;

[0032] The quantum decoding and risk level assessment module is configured to utilize an inverse normalization process to convert the quantum optimization results into original multi-dimensional physiological data, and recalculate the patient's comprehensive risk assessment score to form a physiological data analysis method for patient risk assessment.

[0033] In a third aspect, the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: the processor implements any step of the above method when executing the computer program.

[0034] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: the computer program implements any step of the above method when executed by a processor.

[0035] Compared with the prior art, the invention has the following beneficial effects:

[0036] 1. Aiming at the nonlinear and complex relationship between physiological data, the present invention introduces nonlinear interaction weights into the optimization objective function, which enhances the model's ability to depict the potential impact between multidimensional physiological characteristics and can more realistically reflect the contribution of multidimensional physiological characteristics to risk assessment;

[0037] 2. Through quantized feature mapping, the physiological data optimization problem is transformed into a quantum computing problem. Combining variational quantum circuits with classical-quantum hybrid optimization strategies greatly improves the efficiency of solving high-dimensional complex problems. It breaks through the calculation method of traditional optimization algorithms and brings new inspiration to quantum computing in the processing of medical data.

[0038] 3. By constructing variational quantum circuits, the solution time of complex optimization problems is greatly shortened, which can reduce the demand for computing resources when processing multi-dimensional physiological data and real-time computing scenarios, and make up for the shortcomings of traditional methods in terms of efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0040] Figure 1 This is an overall flow chart of a physiological data analysis method for patient risk assessment according to an embodiment of the present invention;

[0041] Figure 2 A comparison diagram of patient characteristic distribution and classification boundaries of a physiological data analysis method for patient risk assessment according to an embodiment of the present invention;

[0042] Figure 3 A comparison diagram of classification performance (ROC curve) of a physiological data analysis method for patient risk assessment according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0044] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0045] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0046] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0047] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0048] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0049] Example 1

[0050] Reference Figure 1 , which is the first embodiment of the present invention, provides a physiological data analysis method for patient risk assessment, comprising:

[0051] S1. Collecting multidimensional physiological data of the patient, quantifying the characteristics of the multidimensional physiological data, and normalizing the multidimensional physiological data after the characteristics quantification to obtain the multidimensional physiological characteristics of the patient;

[0052] Specifically, the patient's multidimensional physiological data includes the patient's heart rate, blood oxygen, blood pressure, and parameters that can affect the patient's physiological data;

[0053] It needs to be explained that due to the limitation of traditional feature extraction that it only focuses on statistical characteristics (such as mean, variance, etc.) and cannot reflect the complex dynamic changes in time series, it is necessary to use sliding windows to enhance the features of multidimensional physiological data;

[0054] Specifically, a sliding window processing method is used to quantify the features of multi-dimensional physiological data;

[0055] Furthermore, the sliding window is processed as follows:

[0056] Collect the patient's multidimensional physiological data X = {x1, x2, ..., x n}, x n For the time series of the nth physiological data, take out any time series of physiological data, and divide the taken out time series of physiological data into a number of overlapping windows;

[0057] Specifically, each window includes a window length and a sliding step size;

[0058] For each divided window, the dynamic trend slope of the physiological data is calculated by the linear fitting method. At the same time, for the time series of the extracted physiological data, the difference between the maximum and minimum values ​​of the time series is calculated to obtain the fluctuation range of the physiological data.

[0059] Specifically, the linear fitting method formula is expressed as:

[0060]

[0061] Among them, t j Represented as the time index of each data point in window j; y j Represented as the corresponding physiological data value in window j; and are the mean of the time index and data value in the window respectively; w is the total number of windows; k n is the dynamic trend slope k of the nth window;

[0062] It should be explained that the dynamic trend slope reflects the changing trend and rate of change of the patient's physiological data during the time period;

[0063] Furthermore, if the obtained dynamic trend slope k is greater than 0, it means that the change trend and change rate of the physiological data show an upward trend; if the obtained dynamic trend slope k is equal to 0, it means that the change trend and change rate of the physiological data are stable, and no obvious upward and downward trends can be seen; if the obtained dynamic trend slope k is less than 0, it means that the change trend and change rate of the physiological data show a downward trend;

[0064] Specifically, the difference between the maximum and minimum values ​​of the time series is calculated to obtain the fluctuation range r of the physiological data. n , the mathematical expression is:

[0065] r n =max(x n)-min(x n )

[0066] It needs to be explained that r n It indicates the degree of dispersion of physiological data in the time series, reflecting the volatility of physiological state;

[0067] It should be noted that the local trend and short-term variation characteristics implicit in the time series are extracted through the dynamic trend slope and the fluctuation range of physiological data. These characteristics can well supplement the deficiencies of static statistical characteristics (such as mean, variance, etc.) and provide a more comprehensive dynamic description of the patient's multidimensional physiological data;

[0068] Furthermore, each window is moved with a sliding step, and the calculation process of the dynamic trend slope and fluctuation range is repeated until all windows of the time series are calculated;

[0069] Furthermore, the multidimensional physiological data after feature enhancement is normalized to obtain the multidimensional physiological characteristics of the patient;

[0070] Specifically, the mathematical expression of normalization is:

[0071]

[0072] It should be noted that, since all the multidimensional physiological characteristics of patients are converted to the range of [0,1] through normalization, the multidimensional physiological characteristics of each patient can be expressed as the amplitude of the quantum state;

[0073] S2. Based on the multidimensional physiological characteristics of the patient, an optimization objective function of the comprehensive risk assessment model is constructed, and the optimization objective function is calculated to obtain the patient's comprehensive risk assessment score. At the same time, the quantized feature mapping method is used to consider the quantum optimization target direction, and the patient's comprehensive risk assessment score is minimized by constructing a variational quantum circuit with parameters.

[0074] Furthermore, nonlinear interaction weights are introduced to assign weights to each patient's multidimensional physiological characteristics according to their contribution to risk, and to construct the optimization objective function of the comprehensive risk assessment model;

[0075] Specifically, the optimization objective function R(X) is expressed as:

[0076]

[0077] Among them, ω n Expressed as the risk weight corresponding to the patient's multidimensional physiological characteristics; ω nm represents the risk weight between two adjacent patients’ multidimensional physiological characteristics; N is the total number of patients’ multidimensional physiological characteristics;

[0078] It should be noted that the optimization objective function is constructed in order to make it the core problem to be solved in subsequent quantum computing;

[0079] Furthermore, by calculating and optimizing the objective function, the comprehensive risk assessment score of each patient is obtained, that is, the output result of R(X);

[0080] Furthermore, before the quantum feature mapping method, the multidimensional physiological characteristics of the patient in the optimization objective function are quantum-state mapped, and the multidimensional physiological characteristics of each patient are mapped into a quantum state;

[0081] Specifically, the quantum state |ψ n >The mapping process is expressed as:

[0082] |ψ n > = cos(πx′ n )|0>+sin(πx′ n )|1>

[0083] Here, |0> and |1> represent the ground state and excited state in the quantum state, respectively; cos(·) and sin(·) are represented as amplitude mapping functions, defining x′ n The proportion of ground state and excited state;

[0084] It should be noted that the previous normalization process ensures that the input value range of the quantum state mapping is within [0,1], so that it can correspond one-to-one with the amplitude of the quantum state; in addition, the risk weight of the patient's multidimensional physiological characteristics and the corresponding patient's multidimensional physiological characteristics can also be introduced in the mapping process to form a multi-bit quantum state, thereby optimizing the mapping process, but this optimization mapping process does not belong to the research direction of the present invention, and will not be described in detail here;

[0085] Furthermore, based on the optimization objective function of the comprehensive risk assessment model, a quantum Hamiltonian is constructed to determine the quantum optimization target direction as the minimization of the comprehensive risk assessment score;

[0086] Specifically, we set the initial state |ψ0>: the uniform superposition state of all quantum states, and construct the quantum Hamiltonian H C It is expressed as:

[0087]

[0088] Among them, Z n Represents x′ n The Pauli-Z operator, X m represents the Pauli-X operator, which is used to generate the search path;

[0089] It needs to be explained that the Pauli-X operator is also called the quantum NOT gate, which is similar to the NOT operation in classical logic when acting on a quantum state; the Pauli-Z operator is also called the phase flip gate, which is used to adjust the phase of a quantum state;

[0090] It should be noted that since the construction of the quantum Hamiltonian makes full use of the interaction between the risk weight in the optimization objective function and the multidimensional physiological characteristics of the patient, the optimization problem is converted from a traditional mathematical model to a quantum physics problem; in addition, in order to guide the quantum to efficiently find the optimal solution in the optimization space, the search path is generated in the optimization space by applying the quantum gate operation to the quantum state;

[0091] It needs to be explained that when the problem scale increases or the model complexity is high, classical computing methods may be difficult to solve efficiently due to computing resource and time constraints; by combining the parallelism and superposition characteristics of quantum computing methods, the optimal search path can be achieved to significantly reduce the computational cost of complex optimization problems;

[0092] Furthermore, a parameterized variational quantum circuit is constructed, which adopts a classical-quantum computing strategy;

[0093] Furthermore, the classical-quantum computing strategy mainly optimizes the parameters in the variational quantum circuit by alternating classical computing and quantum computing, gradually moving the parameters closer to the optimal solution of the optimization objective function, thereby minimizing the patient's comprehensive risk assessment score;

[0094] Specifically, the process of alternating and iterative optimization of variational quantum circuit parameters is as follows:

[0095] The calculation is performed by the variational quantum circuit to obtain |ψ0(θ)>=U(θ)|ψ0>, where U(θ) is a parameterized quantum circuit, indicating that it contains multiple quantum gates with parameters;

[0096] In each iteration, the quantum circuit generates a quantum state |ψ0(θ)> based on the current parameter θ and measures the expected value of the quantum state <ψ0(θ)|H C |ψ0(θ)>, where the measurement process can be described as: running the circuit U(θ) through a quantum computer to obtain |ψ0(θ)>; measuring the quantum state in the quantum Hamiltonian H C The expected value under , as the current value of the optimization objective function;

[0097] Then, the classical optimizer receives the quantum computation result, that is, measures the quantum state in the quantum Hamiltonian H CThe expected value under the condition of θ is updated, and the updated parameter θ is transmitted back to the quantum circuit. The quantum circuit regenerates the quantum state |ψ0(θ)> according to the updated parameter, and enters the next round of iteration until the current value of the optimization objective function converges, that is, the minimization condition of the comprehensive risk assessment score is reached, and the iteration stops;

[0098] It should be noted that at this time, the multidimensional physiological data in the optimization objective function corresponding to the minimization condition of the comprehensive risk assessment score also changed;

[0099] S3. Using the anti-normalization processing method, the quantum optimization results are converted into original multi-dimensional physiological data, and the comprehensive risk assessment score of the patient is recalculated to form a physiological data analysis method for patient risk assessment;

[0100] It should be noted that the quantum optimization result refers to the optimization objective function of the minimized comprehensive risk assessment score;

[0101] Further, the optimized multidimensional physiological data is extracted from the optimization objective function of the minimized comprehensive risk assessment score, and the optimized multidimensional physiological data is converted into a data form having the same representation as the original multidimensional physiological data;

[0102] Specifically, extract optimized multidimensional physiological data Denormalize it to the actual value, the formula is:

[0103]

[0104] Furthermore, the optimized multidimensional physiological data (i.e., the actual value obtained by the above formula) expressed in the same data form as the original multidimensional physiological data is used as the input of the optimization objective function R(X) to obtain the comprehensive risk assessment score of the patient;

[0105] Furthermore, the patient's current risk level is divided according to the comprehensive risk assessment score;

[0106] The risk levels are as follows:

[0107] Low risk: R(X * ) <T1;

[0108] Medium risk: T1≤R(X * ) <T2;

[0109] High risk: R(X * )≥T2;

[0110] Among them, R(X *) represents the comprehensive risk assessment score of the patient; T1 is the comprehensive risk assessment score of the patient in the past year, and T2 is the comprehensive risk assessment score of the patient in the past year + the standard deviation of the comprehensive risk assessment score of the patient in the past year;

[0111] It should be noted that the standard deviation of the patient's comprehensive risk assessment score in the past year reflects the extent of change in the patient's unhealthy status in the past year.

[0112] Furthermore, this embodiment also provides a physiological data analysis system for patient risk assessment, comprising:

[0113] A multidimensional physiological data processing module is configured to collect multidimensional physiological data of a patient, perform feature quantification on the multidimensional physiological data, and perform normalization processing on the multidimensional physiological data after feature quantification to obtain multidimensional physiological features of the patient;

[0114] A quantum computing optimization module is configured to construct an optimization objective function of a comprehensive risk assessment model based on the multidimensional physiological characteristics of the patient, calculate the optimization objective function, obtain a comprehensive risk assessment score of the patient, and simultaneously consider the quantum optimization target direction by using a quantized feature mapping method, and minimize the comprehensive risk assessment score of the patient by constructing a variational quantum circuit with parameters;

[0115] The quantum decoding and risk level assessment module is configured to utilize an inverse normalization process to convert the quantum optimization results into original multi-dimensional physiological data, and recalculate the patient's comprehensive risk assessment score to form a physiological data analysis method for patient risk assessment.

[0116] This embodiment also provides a computer device, which is applicable to the case of a physiological data analysis method for patient risk assessment, including:

[0117] A memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the physiological data analysis method for patient risk assessment proposed in the above embodiment.

[0118] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0119] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the physiological data analysis method for patient risk assessment proposed in the above embodiment is implemented.

[0120] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be found in the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0121] Example 2

[0122] Reference Figure 2 and Figure 3 , which is the second embodiment of the present invention, and this embodiment provides a physiological data analysis method for patient risk assessment, including: In order to verify the beneficial effects of the scheme of the present invention, this experiment was conducted in the cardiovascular department of a tertiary-level A hospital, and the experimental environment was strictly controlled to ensure the accuracy and consistency of data collection;

[0123] The experimental environment is as follows: The device model uses a portable physiological monitoring device: Mindray BeneVision N17, which supports high-precision real-time monitoring of multiple parameters such as heart rate, blood oxygen, and blood pressure; data processing server: a high-performance server equipped with NVIDIA A100 GPU, used to run the quantum computing simulation framework (Qiskit0.36.0); programming environment: Python 3.9, including TensorFlow 2.8 and Qiskit quantum simulation modules; experimental duration: data collection for each patient lasted 30 minutes, with a sampling frequency of 1Hz, and a total of 1,800 sets of raw data were obtained;

[0124] Experimental subjects: 50 representative patients were selected, covering subjects of different age groups (25-65 years old), different underlying diseases (no underlying disease, hypertension, diabetes) and different health conditions (low risk, medium risk, high risk);

[0125] Experimental method: The method of the present invention (introducing nonlinear interaction weights and variational quantum circuits, minimizing the comprehensive risk assessment score through a quantum optimization framework) and the traditional method (using a linear regression model to model the patient's physiological data and calculate the risk assessment score) were used. The sliding window of the present invention was set to a length of 10 seconds for each window and a sliding step of 5 seconds.

[0126] Experimental results: Reference Figure 2 , you can see, Figure 2 (Left) shows the feature distribution (heart rate trend, blood oxygen fluctuation, blood pressure trend) of low-risk patients and high-risk patients; the features of low-risk patients are concentrated in the area close to the origin, while the features of high-risk patients are far away from the origin and distributed more widely, indicating that patients with different risk levels have obvious clustering characteristics in the feature space; Figure 2 (Right figure) shows the classification boundaries of the traditional method (linear classification) and the method of the present invention (non-linear classification): From the traditional method, it can be seen that the classification boundary is a linear plane, which is difficult to deal with the nonlinear relationship between high-risk patients and low-risk patients in feature distribution, and some high-risk patients are misclassified as low-risk; while the method of the present invention adopts a nonlinear classification model, the classification boundary is more complex, and it can accurately distinguish low-risk and high-risk patients, significantly reducing the phenomenon of misjudgment and missed judgment; at the same time, you can refer to Figure 3 , the triangle arrows and the circular dots represent the key dots of the traditional method and the method of the present invention, respectively. It can be seen that the traditional method has a certain growth trend in the low false positive rate (FPR<0.2) area, but the overall curve is relatively flat, indicating that the classification model has a weak adaptability to complex features, and after the false positive rate (FPR) increases, the true rate (TPR) increases slowly, indicating that high-risk patients may be missed; while the method of the present invention, in the low false positive rate area (FPR<0.2), the true rate rises rapidly, showing the outstanding ability of the method of the present invention in accurately identifying high-risk patients, and the curve is close to the upper left corner, indicating that the method of the present invention can more accurately distinguish high-risk from low-risk patients; in addition, the AUC difference between the two methods is obvious, which is enough to show the superiority of the method of the present invention in classification performance. Therefore, it is explained that the method of the present invention significantly improves the classification performance through nonlinear optimization;

[0127] In summary, the solution of the present invention can not only meet the processing requirements of quantum computing on physiological data, but also achieve the ability to accurately identify patients, which illustrates the feasibility of the solution of the present invention.

[0128] Those skilled in the art will appreciate that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program codes. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0129] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0130] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0131] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0132] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0133] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A physiological data analysis method for patient risk assessment, characterized in that: include: Collecting multidimensional physiological data of the patient, quantifying the characteristics of the multidimensional physiological data, and normalizing the multidimensional physiological data after the characteristic quantification to obtain the multidimensional physiological characteristics of the patient; Based on the multidimensional physiological characteristics of the patient, an optimization objective function of a comprehensive risk assessment model is constructed, the optimization objective function is calculated, and a comprehensive risk assessment score of the patient is obtained. At the same time, a quantized feature mapping method is adopted to consider the quantum optimization target direction, and a variational quantum circuit with parameters is constructed to minimize the comprehensive risk assessment score of the patient; The quantum optimization results are converted into original multidimensional physiological data by using the anti-normalization method, and the patient's comprehensive risk assessment score is recalculated to form a physiological data analysis method for patient risk assessment.

2. The physiological data analysis method for patient risk assessment according to claim 1, characterized in that: Collecting multi-dimensional physiological data of the patient and quantifying the characteristics of the multi-dimensional physiological data includes: Collect the patient's multidimensional physiological data X = {x1, x2, ..., x n }, x n For the nth physiological data time series, take out any physiological data time series, divide the physiological data time series into a number of overlapping windows, and set the sliding step of the window; For each divided window, the dynamic trend slope of the physiological data is calculated by the linear fitting method. At the same time, for the time series of the extracted physiological data, the difference between the maximum and minimum values ​​of the time series is calculated to obtain the fluctuation range of the physiological data.

3. The physiological data analysis method for patient risk assessment according to claim 2, characterized in that: Also includes: Move each window by the set sliding step size, and repeat the calculation process of dynamic trend slope and fluctuation range until each window of the entire time series is calculated.

4. The physiological data analysis method for patient risk assessment according to claim 1, characterized in that: Based on the multidimensional physiological characteristics of the patient, an optimization objective function of a comprehensive risk assessment model is constructed, and the optimization objective function is calculated to obtain a comprehensive risk assessment score of the patient, including: Nonlinear interaction weights are introduced to assign weights to each patient's multidimensional physiological characteristics according to their contribution to risk, thus constructing the optimization objective function of the comprehensive risk assessment model. By calculating the optimization objective function, a comprehensive risk assessment score for each patient is obtained.

5. The physiological data analysis method for patient risk assessment according to claim 4, characterized in that: At the same time, the quantized feature mapping method is used to consider the target direction of quantum optimization, including: Before the quantum feature mapping method, the multidimensional physiological characteristics of the patient in the optimization objective function are quantum-state mapped, and the multidimensional physiological characteristics of each patient are mapped into a quantum state; Based on the optimization objective function of the comprehensive risk assessment model, a quantum Hamiltonian is constructed to determine the quantum optimization target direction as the minimization of the comprehensive risk assessment score.

6. The physiological data analysis method for patient risk assessment according to claim 5, characterized in that: By constructing a variational quantum circuit with parameters, the comprehensive risk assessment score of the patient is minimized, including: The variational quantum circuit adopts a classical-quantum computing strategy, and optimizes the parameters in the variational quantum circuit by alternating and iteratively combining classical computing and quantum computing, gradually bringing the parameters closer to the optimal solution of the optimization objective function, thereby minimizing the comprehensive risk assessment score of the patient.

7. The physiological data analysis method for patient risk assessment according to claim 6, characterized in that: Using the anti-normalization method, the quantum optimization results are converted into original multi-dimensional physiological data, and the patient's comprehensive risk assessment score is recalculated, including: Extracting optimized multidimensional physiological data from the optimization objective function of minimizing the comprehensive risk assessment score, and converting the optimized multidimensional physiological data into a data form having the same representation as the original multidimensional physiological data; Using the optimized multidimensional physiological data represented in the same data form as the original multidimensional physiological data as the input of the optimization objective function to obtain a comprehensive risk assessment score of the patient; The patient's current risk level is divided according to the comprehensive risk assessment score.

8. A physiological data analysis system for patient risk assessment, based on the physiological data analysis method for patient risk assessment according to any one of claims 1 to 7, characterized in that: include: A multidimensional physiological data processing module is configured to collect multidimensional physiological data of a patient, perform feature quantification on the multidimensional physiological data, and perform normalization processing on the multidimensional physiological data after feature quantification to obtain multidimensional physiological features of the patient; A quantum computing optimization module is configured to construct an optimization objective function of a comprehensive risk assessment model based on the multidimensional physiological characteristics of the patient, calculate the optimization objective function, obtain a comprehensive risk assessment score of the patient, and simultaneously consider the quantum optimization target direction by using a quantized feature mapping method, and minimize the comprehensive risk assessment score of the patient by constructing a variational quantum circuit with parameters; The quantum decoding and risk level assessment module is configured to utilize an inverse normalization process to convert the quantum optimization results into original multi-dimensional physiological data, and recalculate the patient's comprehensive risk assessment score to form a physiological data analysis method for patient risk assessment.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images