Diabetic foot risk assessment method and system based on big data

Abstract

The present invention discloses a diabetic foot risk assessment method and system based on big data, and relates to the field of intelligent medical technology. The method includes real-time acquisition of multi-dimensional physiological data, pre-processing the real-time acquired multi-dimensional physiological data; extracting physiological features based on the pre-processed multi-dimensional physiological data, and fusing them into multi-dimensional feature vectors; constructing a personalized health baseline model, inputting the multi-dimensional feature vectors into the personalized health baseline model, and predicting a personalized safety baseline; comparing the physiological features with the personalized safety baseline, and formulating a nonlinear penalty function P based on the deviation of each physiological feature from the corresponding personalized safety baseline, predicting the degree of influence of the deviation of each physiological feature on the overall health status of the patient, and evaluating the patient's current diabetic foot risk. Through personalized health baseline prediction, the accuracy and reliability of risk assessment are improved, especially in the early warning of diabetic foot and individualized health management, which has significant advantages.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medical technology, and in particular to a diabetic foot risk assessment method and system based on big data. Background Art

[0002] As one of the main complications of diabetes, diabetic foot has received a lot of attention in recent years. With the increase in the number of diabetic patients, the incidence of diabetic foot has also increased year by year. Diabetic foot is usually caused by lower limb neuropathy and peripheral vascular disease, which affects the blood circulation of the foot, leading to infection, ulcers and even gangrene. Traditional diabetic foot risk assessment methods mainly rely on the experience of doctors, regular foot examinations and self-management of patients. However, these methods have certain limitations, especially in the early warning of diabetic foot, which is often delayed and cannot timely and accurately assess the patient's risk. In addition, with the development of big data technology, more and more physiological data can be collected in real time, such as plantar pressure, plantar temperature, blood sugar level, blood flow velocity, etc. How to efficiently process these multi-dimensional physiological data and conduct personalized diabetic foot risk assessment based on this has become a hot topic in current research.

[0003] Although existing technologies are able to obtain multi-dimensional physiological data of patients, there are still many deficiencies in data processing and analysis. For example, most existing technologies rely on changes in a single physiological indicator to judge the health status of patients, and fail to effectively utilize the relationship between multi-dimensional data. At the same time, there is less consideration of individual differences, and many methods use fixed thresholds or models that cannot adapt to the individual needs of different patients, thereby reducing the accuracy and reliability of the assessment. Especially in the early warning of diabetic foot, existing technologies cannot effectively capture the subtle changes in multi-dimensional physiological data, as well as the differences in these changes between different individuals, resulting in large deviations in risk assessment results. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a diabetic foot risk assessment method based on big data to solve the problem that the prior art fails to effectively utilize multi-dimensional physiological data and individual differences for accurate risk assessment.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, the present invention provides a diabetic foot risk assessment method based on big data, which includes: real-time collection of multi-dimensional physiological data, and preprocessing of the real-time collected multi-dimensional physiological data; extracting physiological features based on the preprocessed multi-dimensional physiological data, and fusing them into a multi-dimensional feature vector; constructing a personalized health baseline model, inputting the multi-dimensional feature vector into the personalized health baseline model, and predicting a personalized safety baseline; comparing the physiological features with the personalized safety baseline, and formulating a nonlinear penalty function P based on the deviation of each physiological feature from the corresponding personalized safety baseline, predicting the degree of influence of the deviation of each physiological feature on the patient's overall health status, and evaluating the patient's current diabetic foot risk.

[0008] As a preferred embodiment of the big data-based diabetic foot risk assessment method described in the present invention, the multi-dimensional physiological data includes plantar pressure, plantar temperature, blood flow velocity, blood sugar level, blood pressure, heart rate and gait stability.

[0009] As a preferred embodiment of the diabetic foot risk assessment method based on big data of the present invention, wherein: the multi-dimensional physiological data collected in real time is preprocessed, and the specific steps are as follows;

[0010] Use statistically based outlier detection algorithms to identify and remove anomalous data points;

[0011] Use linear interpolation to synchronize high-frequency physiological data with low-frequency physiological data;

[0012] Standardize physiological data of different dimensions.

[0013] As a preferred solution of the diabetic foot risk assessment method based on big data of the present invention, wherein: the physiological features are extracted based on the pre-processed multi-dimensional physiological data and fused into a multi-dimensional feature vector, the specific steps are as follows:

[0014] The plantar pressure data were analyzed by sliding window method to extract plantar pressure features;

[0015] Through time series analysis of plantar temperature data, plantar temperature features are extracted;

[0016] By performing frequency domain analysis and time series analysis on blood flow velocity data, blood flow velocity characteristics and heart rate characteristics are extracted;

[0017] By performing statistical analysis on blood sugar data and blood pressure data, blood sugar characteristics and blood pressure characteristics are extracted;

[0018] Based on the nonlinear relationship and periodic changes of each feature, exponential decay, sine function and smooth adjustment method of feature deviation from the baseline value are used for feature fusion. The expression is: ; where F is a multi-dimensional feature vector, is the ith physiological characteristic, is the exponential decay parameter of the ith physiological characteristic, is the baseline value of the ith physiological characteristic, is the frequency adjustment parameter of the sine function, Indicates physiological characteristics periodic fluctuations, n is the total number of physiological characteristics, i is the index variable of the physiological characteristics, and e is the base of the natural logarithm.

[0019] As a preferred solution of the diabetic foot risk assessment method based on big data described in the present invention, wherein: the personalized health baseline model is constructed, the multi-dimensional feature vector is input into the personalized health baseline model, and the personalized safety baseline is predicted, and the specific steps are as follows:

[0020] The complex interaction between physiological characteristics in the multidimensional feature vector F is captured through nonlinear relationships. When the physiological characteristics deviate from the health baseline, the adjustment factors are used to adapt to the sensitivity differences of physiological characteristics of different individuals, and a personalized health baseline model is constructed to predict the personalized safety baseline. The expression is: ;in, represents the personalized safety baseline of the ith physiological feature, Represents the multidimensional feature vector F after a nonlinear function The comprehensive health status index obtained by combining is the baseline value of the multidimensional feature vector F in a healthy state, is the regulating factor.

[0021] As a preferred solution of the big data-based diabetic foot risk assessment method of the present invention, the influence of the deviation of each physiological characteristic on the overall health status of the patient is predicted, and the expression is: ;in, represents the deviation of the ith physiological feature The degree of impact on the patient's overall health status, Represents the overall deviation of the personalized safety baseline from the physiological characteristics.

[0022] As a preferred embodiment of the big data-based diabetic foot risk assessment method of the present invention, the specific steps of assessing the patient's current diabetic foot risk are as follows:

[0023] Define low-risk thresholds based on historical data and personalized security baselines and high risk threshold ;

[0024] when < When the patient's risk of diabetic foot is considered to be low;

[0025] when ≤ < When the patient's diabetic foot risk is considered to be at medium risk;

[0026] when ≥ The patient is considered to be at high risk for diabetic foot.

[0027] In a second aspect, the present invention provides a diabetic foot risk assessment system based on big data, comprising: a data acquisition module, a feature fusion module, a personalized safety baseline prediction module and a risk assessment module;

[0028] A data acquisition module, used for real-time acquisition of multi-dimensional physiological data and pre-processing of the real-time acquired multi-dimensional physiological data;

[0029] A feature fusion module is used to extract physiological features based on the preprocessed multi-dimensional physiological data and fuse them into a multi-dimensional feature vector;

[0030] The baseline prediction module is used to build a personalized health baseline model, input the multi-dimensional feature vector into the personalized health baseline model, and predict the personalized safety baseline;

[0031] The risk assessment module is used to compare physiological characteristics with personalized safety baselines, analyze the overall deviation of personalized safety baselines, and assess the patient's current risk of diabetic foot.

[0032] 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: when the computer program is executed by the processor, any step of the big data-based diabetic foot risk assessment method as described in the first aspect of the present invention is implemented.

[0033] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the big data-based diabetic foot risk assessment method as described in the first aspect of the present invention is implemented.

[0034] The beneficial effects of the present invention are: by real-time collection of multi-dimensional physiological data and preprocessing it to ensure the accuracy and consistency of the data; then, based on the preprocessed physiological data, a personalized health baseline model is constructed to capture the complex interactions between physiological characteristics, dynamically adjust the personalized health score, and then realize personalized diabetic foot risk assessment. The present invention not only improves the efficiency of data processing, but also improves the accuracy and reliability of risk assessment through personalized health baseline prediction, especially in the early warning of diabetic foot and personalized health management, which has significant advantages. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0036] Figure 1 This is a flow chart of the diabetic foot risk assessment method based on big data in Example 1.

[0037] Figure 2 This is a schematic diagram of the diabetic foot risk assessment system based on big data in Example 1. DETAILED DESCRIPTION

[0038] In order to make the above-mentioned objects, 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 accompanying drawings.

[0039] 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.

[0040] 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.

[0041] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a diabetic foot risk assessment method based on big data, comprising the following steps:

[0042] S1: real-time collection of multi-dimensional physiological data, and pre-processing of the real-time collected multi-dimensional physiological data;

[0043] S1.1: Multi-dimensional physiological data include plantar pressure, plantar temperature, blood flow velocity, blood sugar level, blood pressure, heart rate and gait stability.

[0044] Furthermore, these data can be collected in real time through smart devices, and key features can be extracted after preprocessing, such as the changing trends of pressure, temperature, and blood flow velocity, thereby providing a reliable basis for subsequent feature extraction and fusion, and thus achieving accurate diabetic foot risk assessment.

[0045] S1.2: Use statistically based outlier detection algorithms to identify and remove anomalous data points;

[0046] For example, when continuously collected plantar pressure data show extreme values ​​in a short period of time (such as the pressure value at a certain moment is much higher than the normal range, which may be caused by equipment errors or the patient's short-term abnormal posture), the algorithm can identify these abnormal data points and remove or replace them with reasonable interpolation values ​​to ensure the accuracy of subsequent analysis.

[0047] S1.3: Use linear interpolation to synchronize high-frequency physiological data with low-frequency physiological data;

[0048] For example, heart rate data is usually collected once a second, while blood glucose levels may be collected once an hour. In order to analyze these data on the same time axis, the linear interpolation method can insert reasonable estimates between adjacent time points of blood glucose data, so that high-frequency heart rate data and low-frequency blood glucose data are consistent in time, thus facilitating the fusion analysis of multidimensional data.

[0049] High-frequency physiological data usually refers to data collected every second, such as heart rate and plantar pressure, while low-frequency physiological data refers to data collected every hour or at longer time intervals, such as blood sugar level and blood pressure.

[0050] S1.4: Standardize physiological data of different dimensions.

[0051] For example, converting plantar pressure (kPa) and blood glucose level (mmol / L) to the same scale ensures that they have the same weight in the model and avoids bias caused by dimensional differences.

[0052] S2: Extract physiological features based on the preprocessed multi-dimensional physiological data and fuse them into a multi-dimensional feature vector;

[0053] S2.1: Analyze the plantar pressure data by sliding window method and extract plantar pressure features;

[0054] Specifically, the sliding window method extracts the changing trend of plantar pressure segment by segment, calculates the maximum value, average value and other features, and captures the pressure distribution within the gait stability cycle.

[0055] S2.2: Extract the sole temperature features by performing time series analysis on the sole temperature data;

[0056] Specifically, the plantar temperature data is continuously observed through time series analysis technology to extract its changing trend over time. The main features include the average temperature value, the rate of change, and the number and amplitude of abnormal temperature fluctuations.

[0057] S2.3: Extract blood flow velocity characteristics and heart rate characteristics by performing frequency domain analysis and time series analysis on blood flow velocity data;

[0058] Specifically, the blood flow velocity data is analyzed in time series to extract its average value and fluctuation range, and the short-term change trend of blood flow velocity is identified. At the same time, the data is converted into the frequency domain through frequency domain analysis (such as Fourier transform) to extract the main frequency components and periodic characteristics. This can reveal the periodic changes and abnormal fluctuations of blood flow velocity.

[0059] S2.4: Extract blood sugar features and blood pressure features by performing statistical analysis on blood sugar data and blood pressure data;

[0060] Specifically, basic statistical analysis is performed on blood sugar and blood pressure data to extract statistical features such as mean, variance, and extreme values ​​to describe the long-term trend and fluctuation of blood sugar and blood pressure. By analyzing the frequency of abnormal fluctuations in blood sugar, the patient's metabolic control ability can be evaluated; while the extreme value and fluctuation range of blood pressure can reflect the stability of blood pressure regulation.

[0061] S2.5: Based on the nonlinear relationship and periodic changes of each feature, feature fusion is performed using exponential decay, sine function, and smooth adjustment method of feature deviation from the baseline value. The expression is: ; where F is a multi-dimensional feature vector, is the ith physiological characteristic, is the exponential decay parameter of the ith physiological characteristic, is the baseline value of the ith physiological characteristic, is the frequency adjustment parameter of the sine function, Indicates physiological characteristics periodic fluctuations, n is the total number of physiological characteristics, i is the index variable of the physiological characteristics, and e is the base of the natural logarithm.

[0062] It should be noted that It is derived through statistical analysis of a large amount of historical data and adjusted based on the individual's long-term physiological data to reflect a personalized health benchmark.

[0063] It is determined by performing frequency domain analysis (such as Fourier transform) on the historical data of physiological characteristics to identify the main periodic fluctuation frequency of the characteristic. In addition, It can also be determined through the model parameter tuning process to make the model better adapt to the periodic changes of the characteristics.

[0064] S3: Build a personalized health baseline model, input the multi-dimensional feature vector into the personalized health baseline model, and predict the personalized safety baseline;

[0065] S3.1: The complex interaction between physiological characteristics in the multidimensional feature vector F is captured through nonlinear relationships. When the physiological characteristics deviate from the health baseline, the adjustment factors are used to adapt to the sensitivity differences of physiological characteristics of different individuals, build a personalized health baseline model, and predict the personalized safety baseline. The expression is: ;in, represents the personalized safety baseline of the ith physiological feature, Represents the multidimensional feature vector F after a nonlinear function The comprehensive health status index obtained by combining is the baseline value of the multidimensional feature vector F in a healthy state, yes The adjustment factor for the sensitivity of physiological characteristics to deviations from the healthy baseline is generally taken as (0,1).

[0066] It should be noted that the comprehensive health status index The expression is: ;in, It is the plantar pressure. is the foot temperature, is the blood flow velocity, is the blood sugar level, is the blood pressure value, is the heart rate value;

[0067] The above parameters are all included in the multidimensional feature vector F;

[0068] The baseline value of the multidimensional feature vector F in a healthy state , which is obtained by calculating the average value of each physiological characteristic;

[0069] S4: Compare the physiological characteristics with the personalized safety baseline. Based on the deviation of each physiological characteristic from the corresponding personalized safety baseline, formulate a nonlinear penalty function P to predict the impact of the deviation of each physiological characteristic on the patient's overall health status and assess the patient's current risk of diabetic foot.

[0070] S4.1: Calculate each physiological characteristic Personalized security baseline Deviation ;

[0071] S4.2: Based on each physiological characteristic Personalized security baseline Deviation , formulate a nonlinear penalty function P to predict the impact of the deviation of each physiological characteristic on the patient's overall health status, expressed as: ;in, represents the deviation of the ith physiological feature The degree of impact on the patient's overall health status, Represents the overall deviation of the personalized safety baseline from the physiological characteristics.

[0072] It should be noted that the nonlinear penalty function is used to measure the impact of the deviation between each physiological feature and its personalized safety baseline on the overall health status. Compared with the linear function, the nonlinear penalty function can handle different degrees of deviation more flexibly. When the deviation is small, the function may be insensitive to health effects, but as the deviation increases, the impact will increase rapidly.

[0073] S4.3: Define low-risk thresholds based on historical data and personalized security baselines and high risk threshold ;

[0074] when < When the patient's risk of diabetic foot is considered to be low;

[0075] when ≤ < When the patient's diabetic foot risk is considered to be at medium risk;

[0076] when ≥ The patient is considered to be at high risk for diabetic foot.

[0077] It should be noted that the threshold is defined based on historical data and personalized safety baselines. First, the normal fluctuation range of each characteristic in a healthy state is determined by analyzing the patient's long-term accumulated physiological characteristic data (such as blood sugar, blood pressure, blood flow velocity, etc.), and then combined with the individual's personalized safety baseline to identify the relationship between characteristic deviations and health risks. By statistically analyzing these data, the low-risk threshold is determined. and high risk threshold , so that it reflects the degree of deviation of physiological characteristics under different risk levels.

[0078] It should also be noted that low-risk and high-risk thresholds are dynamically set based on historical data to accurately distinguish the risk level of diabetic foot in patients. This personalized assessment method can flexibly adapt to the changes in physiological characteristics of different individuals and provide a reliable basis for the early prevention and intervention of diabetic foot.

[0079] This embodiment also provides a diabetic foot risk assessment system based on big data, including: a data acquisition module, a feature fusion module, a personalized safety baseline prediction module and a risk assessment module;

[0080] A data acquisition module, used for real-time acquisition of multi-dimensional physiological data and pre-processing of the real-time acquired multi-dimensional physiological data;

[0081] A feature fusion module is used to extract physiological features based on the preprocessed multi-dimensional physiological data and fuse them into a multi-dimensional feature vector;

[0082] The baseline prediction module is used to build a personalized health baseline model, input the multi-dimensional feature vector into the personalized health baseline model, and predict the personalized safety baseline;

[0083] The risk assessment module is used to compare physiological characteristics with personalized safety baselines, analyze the overall deviation of personalized safety baselines, and assess the patient's current risk of diabetic foot.

[0084] This embodiment also provides a computer device, including: 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 big data-based diabetic foot risk assessment method proposed in the above embodiment.

[0085] 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 through a system bus. Among them, 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 covered on 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.

[0086] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for risk assessment of diabetic foot based on big data as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, disk or optical disk.

[0087] In summary, the present invention collects multi-dimensional physiological data in real time and preprocesses it to ensure the accuracy and consistency of the data; then, a personalized health baseline model is constructed based on the preprocessed physiological data to capture the complex interactions between physiological characteristics, dynamically adjust the personalized health score, and then realize personalized diabetic foot risk assessment. The present invention not only improves the efficiency of data processing, but also improves the accuracy and reliability of risk assessment through personalized health baseline prediction, especially in the early warning of diabetic foot and personalized health management, which has significant advantages.

[0088] Example 2, referring to Table 1, is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of a diabetic foot risk assessment method based on big data is provided.

[0089] In order to verify the effectiveness of the big data-based diabetic foot risk assessment method proposed in this paper, a three-month experiment was designed and 10 diabetic patients were selected as research subjects. Each patient wore a smart insole, a smart bracelet, a smart blood glucose meter and a smart blood pressure monitor to collect multi-dimensional physiological data in real time, including plantar pressure, plantar temperature, blood flow velocity, blood sugar level, blood pressure, heart rate and gait stability. The data collection frequency was once an hour to ensure the continuity and integrity of the data.

[0090] First, the collected multi-dimensional physiological data are preprocessed. For outliers, a statistical outlier detection algorithm is used to automatically identify and delete abnormal data points to reduce the impact of extreme data on the model. Then, the linear interpolation method is used to synchronize high-frequency physiological data (such as heart rate and blood pressure) with low-frequency physiological data (such as plantar pressure and blood sugar level) to ensure the timeliness and consistency of the analysis. Finally, physiological data of different dimensions are standardized to eliminate confusion caused by dimensional differences.

[0091] After data preprocessing, the sliding window method is used to extract plantar pressure features, the plantar temperature features are analyzed in time series, the frequency domain and time series analysis are combined to extract blood flow velocity and heart rate features, and the change characteristics of blood sugar and blood pressure data are statistically analyzed. All extracted physiological features are fused into multi-dimensional feature vectors through feature fusion algorithms and input into the personalized health baseline model, which dynamically adjusts the personalized health score by capturing the complex nonlinear relationship between multi-dimensional physiological features, and finally predicts the patient's personalized safety baseline.

[0092] The experiment also used traditional diabetic foot risk assessment methods on patients, which mainly rely on regular physical examinations and analysis of some single physiological data. Doctors assess patients' diabetic foot risk based on abnormalities in these single indicators, such as high blood sugar or abnormally high sole temperature.

[0093] The details are shown in Table 1 below:

[0094] Table 1 Diabetic foot risk assessment experimental data table

[0095] Patient number Plantar pressure (kPa) Foot temperature (°C) Blood flow velocity (cm / s) Blood glucose level (mmol / L) Blood pressure (mmHg) Heart rate (bpm) Gait stability (%) Risk Assessment (Traditional Methods) Risk Assessment (Method of the Invention) Patient 1 67.3 29.1 42.5 6.8 118 / 76 74 83 Medium risk Low risk Patient 2 88.7 30.8 34.2 8.6 132 / 88 79 70 High risk Medium risk Patient 3 92.4 31.2 37.6 9.3 140 / 92 82 68 High risk High risk Patient 4 61.5 28.9 41.1 7.5 126 / 82 71 86 Medium risk Low risk Patient 5 94.8 32.1 32.9 10.1 146 / 98 85 64 High risk High risk Patient 6 74.2 30.5 38.7 7.9 125 / 80 76 78 Medium risk Low risk Patient 7 85.1 31.0 36.8 8.4 135 / 87 80 72 High risk Medium risk Patient 8 69.8 29.4 40.2 7.3 120 / 79 73 81 Medium risk Low risk Patient 9 95.6 32.5 33.1 9.8 144 / 95 84 66 High risk High risk Patient 10 62.7 28.7 42.9 6.9 119 / 77 75 84 Medium risk Low risk

[0096] Through the data analysis of the above table, it can be clearly seen that the method of the present invention has significant advantages in the risk assessment of diabetic foot. Compared with the traditional assessment method, the present invention can more accurately assess the risk of diabetic foot for each patient through real-time collection and comprehensive analysis of multi-dimensional physiological data. The traditional method relies on regular physical examinations and analysis of single physiological data, which makes it difficult to capture the complex associations and individual differences between physiological characteristics, and easily leads to deviations and misjudgments in the assessment. For example, Patient 1 was rated as "medium risk" under the traditional method, while the present invention was evaluated as "low risk" through comprehensive analysis of multiple dimensions such as plantar pressure, blood sugar level, and gait stability, which avoids excessive intervention due to abnormal single indicators. For another example, Patient 2 was rated as "high risk" in the traditional method, but the present invention was based on multi-dimensional feature vector analysis and found that his heart rate and gait stability were relatively stable, so it was evaluated as "medium risk", which is more in line with his actual health status.

[0097] Through the above comparison, the method of the present invention can not only improve the accuracy of the assessment through the comprehensive analysis of multi-dimensional physiological data, but also better reflect individual differences and provide personalized health management suggestions. Traditional methods often rely on a single or a few physiological indicators, which can easily lead to misjudgment. The method of the present invention constructs a personalized health baseline model and dynamically adjusts the impact of each feature on the assessment results, thereby more accurately assessing the risk of diabetic foot.

[0098] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A diabetic foot risk assessment method based on big data, characterized in that: include: Collect multi-dimensional physiological data in real time, and pre-process the multi-dimensional physiological data collected in real time; Extract physiological features based on the preprocessed multi-dimensional physiological data and fuse them into a multi-dimensional feature vector; Build a personalized health baseline model, input the multi-dimensional feature vector into the personalized health baseline model, and predict the personalized safety baseline; Compare the physiological characteristics with the personalized safety baseline, formulate a nonlinear penalty function P based on the deviation of each physiological characteristic from the corresponding personalized safety baseline, predict the impact of the deviation of each physiological characteristic on the patient's overall health status, and assess the patient's current risk of diabetic foot; The physiological features are extracted based on the preprocessed multi-dimensional physiological data and fused into a multi-dimensional feature vector. The specific steps are as follows: The plantar pressure data were analyzed by sliding window method to extract plantar pressure features; Through time series analysis of plantar temperature data, plantar temperature features are extracted; By performing frequency domain analysis and time series analysis on blood flow velocity data, blood flow velocity characteristics and heart rate characteristics are extracted; By statistically analyzing the blood sugar level data and blood pressure data, blood sugar characteristics and blood pressure characteristics are extracted; Based on the nonlinear relationship and periodic changes of each feature, exponential decay, sine function and smooth adjustment method of feature deviation from the baseline value are used for feature fusion. The expression is: ； Among them, F is a multi-dimensional feature vector, is the ith physiological characteristic, is the exponential decay parameter of the ith physiological characteristic, is the baseline value of the ith physiological characteristic, is the frequency adjustment parameter of the sine function, Indicates physiological characteristics The periodic fluctuation of, n is the total number of physiological characteristics, i is the index variable of the physiological characteristics, and e is the base of the natural logarithm; The personalized health baseline model is constructed, the multi-dimensional feature vector is input into the personalized health baseline model, and the personalized safety baseline is predicted. The specific steps are as follows: The complex interaction between physiological characteristics in the multi-dimensional feature vector F is captured through nonlinear relationships. When the physiological characteristics deviate from the health baseline, the adjustment factors are used to adapt to the sensitivity differences of physiological characteristics of different individuals, and a personalized health baseline model is constructed to predict the personalized safety baseline. The expression is: ； in, represents the personalized safety baseline of the ith physiological feature, Represents the multidimensional feature vector F after a nonlinear function The comprehensive health status index obtained by combining is the baseline value of the multidimensional feature vector F in a healthy state, is the regulating factor; Predict the impact of the deviation of each physiological characteristic on the patient's overall health status, expressed as: ； in, represents the deviation of the ith physiological feature The degree of impact on the patient's overall health status, Represents the overall deviation of the personalized safety baseline from the physiological characteristics.

2. The big data-based diabetic foot risk assessment method according to claim 1, characterized in that: The multi-dimensional physiological data includes plantar pressure, plantar temperature, blood flow velocity, blood sugar level, blood pressure, heart rate and gait stability.

3. The big data-based diabetic foot risk assessment method according to claim 2, characterized in that: The specific steps of preprocessing the multi-dimensional physiological data collected in real time are as follows: Use statistically based outlier detection algorithms to identify and remove anomalous data points; Use linear interpolation to synchronize high-frequency physiological data with low-frequency physiological data; Standardize physiological data of different dimensions.

4. The big data-based diabetic foot risk assessment method according to claim 3, characterized in that: The specific steps for assessing the patient's current risk of diabetic foot are as follows: Define low-risk thresholds based on historical data and personalized security baselines and high risk threshold ; when < When the patient's risk of diabetic foot is considered to be low; when ≤ < When the patient's diabetic foot risk is considered to be at medium risk; when ≥ The patient is considered to be at high risk for diabetic foot.

5. A diabetic foot risk assessment system based on big data, based on the diabetic foot risk assessment method based on big data according to any one of claims 1 to 4, characterized in that: include: Data collection module, feature fusion module, personalized security baseline prediction module and risk assessment module; A data acquisition module, used for real-time acquisition of multi-dimensional physiological data and pre-processing of the real-time acquired multi-dimensional physiological data; A feature fusion module is used to extract physiological features based on the preprocessed multi-dimensional physiological data and fuse them into a multi-dimensional feature vector; The baseline prediction module is used to build a personalized health baseline model, input the multi-dimensional feature vector into the personalized health baseline model, and predict the personalized safety baseline; The risk assessment module is used to compare physiological characteristics with personalized safety baselines, analyze the overall deviation of personalized safety baselines, and assess the patient's current risk of diabetic foot.

6. 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 diabetic foot risk assessment method based on big data described in any one of claims 1 to 4 are implemented.

7. 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 big data-based diabetic foot risk assessment method according to any one of claims 1 to 4 are implemented.

Images