Medical index risk early warning method and device, electronic equipment and storage medium

Abstract

The embodiment of the application provides a kind of medical index risk early warning method and device, electronic equipment and storage medium, belong to medical science and technology and artificial intelligence technical field.The method comprises: the feature extraction of target medical index data is carried out, and medical index feature is obtained;Medical index feature is identified to risk degree, and the risk degree of medical index is obtained;The target residual between medical index risk degree and observation medical index degree is calculated by residual layer;According to target residual, risk early warning is carried out to target medical index data, and risk early warning grade is obtained.The embodiment of the application can effectively identify the abnormal state of medical index by calculating the target residual between medical index risk degree and observation medical index degree, and potential health problems are discovered in time, and risk early warning is carried out to target medical index data according to target residual and risk early warning layer, which can realize the dynamic prediction of medical index, and help to improve the accuracy of medical index risk early warning.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical technology and artificial intelligence, in particular to a medical index risk early warning method and device, an electronic device and a storage medium. BACKGROUND

[0002] At present, the medical index of a user is usually monitored through a periodic data collection method of a smart watch, and the medical index is further analyzed and warned by a medical health detection system. For example, in the medical technology scene, a user monitors his own blood glucose and heart rate data through a smart watch, the smart watch automatically measures the blood glucose every 2 hours and records the heart rate data throughout the day, and the detection software built in the smart watch uses a static risk threshold to evaluate the health status of the user (for example, fasting blood glucose higher than 7.0 mmol / L is considered abnormal). However, this method relies on a single static risk threshold, and the collected medical index is not a continuous data set, so that the risk analysis of the medical index of the user is relatively single, resulting in a low accuracy of the medical index risk warning. Therefore, how to improve the accuracy of the medical index risk warning has become a problem to be solved. SUMMARY

[0003] The main purpose of the embodiments of the present application is to provide a medical index risk early warning method and device, an electronic device and a storage medium, which aims to improve the accuracy of the medical index risk early warning.

[0004] To achieve the above-mentioned purpose, the first aspect of the embodiments of the present application provides a medical index risk early warning method, which comprises:

[0005] Collecting target medical index data of a target user;

[0006] Obtaining a pre-trained risk early warning model; wherein the risk early warning model comprises an encoding layer, a decoding layer, a residual layer and a risk early warning layer;

[0007] Performing feature extraction on the target medical index data through the encoding layer to obtain medical index features;

[0008] Performing risk degree identification on the medical index features through the decoding layer to obtain a medical index risk degree of the target medical index data;

[0009] Obtaining an observed medical index degree of the target medical index data, and calculating a target residual error between the medical index risk degree and the observed medical index degree through the residual layer;

[0010] Performing risk early warning on the target medical index data according to the target residual error and the risk early warning layer to obtain a risk early warning level.

[0011] In some embodiments, the risk warning of the target medical indicator data according to the target residual and the risk warning layer comprises:

[0012] obtaining a risk warning threshold of the target medical indicator data;

[0013] risk warning of the target medical indicator data according to the risk warning threshold, the target residual and the risk warning layer, to obtain the risk warning level.

[0014] In some embodiments, the risk warning threshold of the target medical indicator data comprises:

[0015] obtaining the window health medical indicator data of the target user by using a preset sliding window;

[0016] obtaining window predicted health medical indicator data by predicting the window health medical indicator data through the risk warning model;

[0017] calculating the window health indicator residual between the window health medical indicator and the window predicted health medical indicator data, and combining the window health indicator residual based on a preset window time sequence to obtain a window health indicator residual sequence;

[0018] obtaining the residual mean and residual standard deviation of the window health indicator residual sequence, and determining the risk warning threshold of the target medical indicator data according to the multiple of the residual mean and the residual standard deviation.

[0019] In some embodiments, the risk degree identification of the medical indicator feature through the decoding layer to obtain the medical indicator risk degree of the target medical indicator data comprises:

[0020] mapping the medical indicator feature to a preset high-dimensional space through the decoding layer to obtain a high-dimensional indicator feature;

[0021] risk degree identification of the high-dimensional indicator feature according to a preset reconstruction weight matrix to obtain the medical indicator risk degree.

[0022] In some embodiments, the feature extraction of the target medical indicator data through the encoding layer to obtain the medical indicator feature comprises:

[0023] statistical distribution feature extraction of the target medical indicator data through the encoding layer to obtain an indicator statistical feature;

[0024] time series feature extraction of the indicator statistical feature to obtain an indicator time feature;

[0025] perform frequency domain feature extraction on the index time feature to obtain an index frequency domain feature;

[0026] The index statistical feature, the index time feature and the index frequency domain feature are fused to obtain the medical index feature.

[0027] In some embodiments, the target medical index data of the target user is collected, including:

[0028] A preset hardware device detects the initial medical index data of the target user;

[0029] The initial medical index data is subjected to outlier detection to obtain abnormal medical index data, and the abnormal medical index data is removed from standard medical index data to obtain normal medical index data;

[0030] The normal medical index data is subjected to smoothing processing to obtain smoothed medical index data;

[0031] The smoothed medical index data is subjected to data filling processing to obtain the target medical index data.

[0032] In some embodiments, after the target medical index data is subjected to risk early warning according to the target residual error and the risk early warning layer to obtain a risk early warning level, the method further includes:

[0033] The residual error sum of squares of each target residual error is obtained;

[0034] The risk contribution degree of each target medical index data is obtained according to each target residual error and the residual error sum of squares;

[0035] The target risk contribution degree with the largest contribution degree is screened out from the risk contribution degrees, and the target medical index data of the target risk contribution degree is taken as the risk cause of the risk early warning level.

[0036] To achieve the above object, a second aspect of the embodiments of the present application proposes a medical index risk early warning device, which comprises:

[0037] A medical index acquisition module is configured to collect target medical index data of a target user;

[0038] A model acquisition module is configured to acquire a pre-trained risk early warning model; wherein the risk early warning model comprises an encoding layer, a decoding layer, a residual error layer and a risk early warning layer;

[0039] A feature extraction module is configured to perform feature extraction on the target medical index data through the encoding layer to obtain a medical index feature;

[0040] a risk degree identification module, configured to identify a risk degree of the medical indicator feature through the decoding layer, to obtain a medical indicator risk degree of the target medical indicator data;

[0041] a residual calculation module, configured to obtain an observed medical indicator degree of the target medical indicator data, and calculate a target residual between the medical indicator risk degree and the observed medical indicator degree through the residual layer;

[0042] a risk early warning module, configured to perform risk early warning on the target medical indicator data according to the target residual and the risk early warning layer, to obtain a risk early warning level.

[0043] To achieve the above object, a third aspect of the embodiment of the present application provides an electronic device, which comprises a memory and a processor, the memory stores a computer program, and the processor implements the method of the first aspect when executing the computer program.

[0044] To achieve the above object, a fourth aspect of the embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the method of the first aspect.

[0045] The medical indicator risk early warning method and device, electronic device and storage medium provided by the present application collect target medical indicator data of a target user, obtain a pre-trained risk early warning model, wherein the risk early warning model comprises an encoding layer, a decoding layer, a residual layer and a risk early warning layer, the encoding layer is used to extract features of the target medical indicator data to obtain medical indicator features, the decoding layer is used to identify a risk degree of the medical indicator features to obtain a medical indicator risk degree of the target medical indicator data, the residual layer is used to calculate a target residual between the medical indicator risk degree and an observed medical indicator degree of the target medical indicator data, and the risk early warning layer is used to perform risk early warning on the target medical indicator data according to the target residual to obtain a risk early warning level. The encoding layer is used to extract features of the target medical indicator data to obtain medical indicator features, and the decoding layer is used to identify a risk degree of the medical indicator features, so that the key information of the medical indicator can be identified and captured, the health risk of an individual can be predicted according to the individual indicator features, and personalized health monitoring can be realized. The residual layer is used to calculate a target residual between the medical indicator risk degree and an observed medical indicator degree, so that the abnormal state of the medical indicator can be effectively identified, and potential health problems can be found in time. The risk early warning layer is used to perform risk early warning on the target medical indicator data according to the target residual, so that the dynamic prediction of the medical indicator can be realized, and the accuracy of the medical indicator risk early warning can be improved. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 is a flowchart of the medical index risk early warning method provided by the embodiment of the present application;

[0047] Figure 2 is Figure 1 is a flowchart of step S101 in

[0048] Figure 3 is Figure 1 is a flowchart of step S103 in

[0049] Figure 4 is Figure 1 is a flowchart of step S104 in

[0050] Figure 5 is Figure 1 is a flowchart of step S106 in

[0051] Figure 6 is Figure 5 is a flowchart of step S501 in

[0052] Figure 7 is another flowchart of the medical index risk early warning method provided by the embodiment of the present application;

[0053] Figure 8 is a structural schematic diagram of the medical index risk early warning device provided by the embodiment of the present application;

[0054] Figure 9 is a hardware structural schematic diagram of the electronic device provided by the embodiment of the present application. DETAILED DESCRIPTION

[0055] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0056] It should be noted that although the functional modules are divided in the device schematic diagram, and the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a manner different from the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification and claims and the above-described drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application, and are not intended to limit the present application.

[0058] Firstly, several terms involved in the present application are analyzed:

[0059] Artificial intelligence (AI): It is a new technical science of studying, developing theories, methods, technologies and application systems for simulating, extending and expanding human intelligence; artificial intelligence is a branch of computer science, and artificial intelligence attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a similar way to human intelligence. The research in this field includes robots, language recognition, image recognition, natural language processing and expert systems, etc. Artificial intelligence can simulate the information process of human consciousness and thinking. Artificial intelligence is also the theory, method, technology and application system of using digital computers or digital computer controlled machines to simulate, extend and expand human intelligence, to perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0060] Therefore, the embodiments of the present application provide a medical index risk early warning method and device, an electronic device and a storage medium, which aims to improve the accuracy of medical index risk early warning.

[0061] The medical index risk early warning method and device, the electronic device and the storage medium provided by the embodiments of the present application are specifically explained by the following embodiments. First, the medical index risk early warning method in the embodiments of the present application is described.

[0062] The embodiments of the present application can acquire and process related data based on artificial intelligence technology. Wherein, artificial intelligence (AI) is a theory, method, technology and application system of using digital computers or digital computer controlled machines to simulate, extend and expand human intelligence, to perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0063] The basic technology of artificial intelligence generally includes technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, processing / interaction system, mechatronics, etc. The software technology of artificial intelligence mainly includes computer vision technology, robot technology, biometric technology, speech processing technology, natural language processing technology and machine learning / deep learning, etc.

[0064] The medical index risk early warning method provided by the embodiments of the present application relates to the technical field of medical science and artificial intelligence. The medical index risk early warning method provided by the embodiments of the present application can be applied to a terminal, can be applied to a server end, and can also be software running in the terminal or the server end. In some embodiments, the terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc.; the server end can be configured as a separate physical server, can be configured as a server cluster or a distributed system formed by multiple physical servers, can also be configured as a cloud server providing basic cloud computing services such as cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN, and big data and artificial intelligence platform; and the software can be an application that implements the medical index risk early warning method, but is not limited to the above forms.

[0065] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld devices or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in a distributed computing environment in which tasks are performed by remote processing devices connected by a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0066] It should be noted that in each specific embodiment of the present application, when it is necessary to perform relevant processing according to the target medical index data of the target user and other data related to the user's identity or characteristics, the user's permission or consent will be obtained first, and the collection, use and processing of these data will comply with relevant laws, regulations and standards. In addition, when the embodiments of the present application need to obtain sensitive personal information of the user, the separate permission or separate consent of the user will be obtained through a pop-up window or a jump to a confirmation page, and after obtaining the separate permission or separate consent of the user, the necessary user-related data for the normal operation of the embodiments of the present application will be obtained.

[0067] Figure 1 is an optional flowchart of the medical index risk early warning method provided by the embodiments of the present application, Figure 1 The method in can include but is not limited to steps S101 to S106.

[0068] Step S101, collect target medical index data of a target user.

[0069] Step S102, obtain a pre-trained risk warning model; wherein the risk warning model comprises an encoding layer, a decoding layer, a residual layer, and a risk warning layer.

[0070] Step S103, perform feature extraction on the target medical index data through the encoding layer to obtain medical index features.

[0071] Step S104, perform risk degree identification on the medical index features through the decoding layer to obtain a medical index risk degree of the target medical index data.

[0072] Step S105, obtain an observed medical index degree of the target medical index data, and calculate a target residual between the medical index risk degree and the observed medical index degree through the residual layer.

[0073] Step S106, perform risk warning on the target medical index data according to the target residual and the risk warning layer to obtain a risk warning level.

[0074] The steps S101 to S106 shown in the embodiments of the present application collect target medical index data of a target user, obtain a pre-trained risk warning model, wherein the risk warning model comprises an encoding layer, a decoding layer, a residual layer, and a risk warning layer, perform feature extraction on the target medical index data through the encoding layer to obtain medical index features, perform risk degree identification on the medical index features through the decoding layer to obtain a medical index risk degree of the target medical index data, obtain an observed medical index degree of the target medical index data, and calculate a target residual between the medical index risk degree and the observed medical index degree through the residual layer, and perform risk warning on the target medical index data according to the target residual and the risk warning layer to obtain a risk warning level. The present application first performs feature extraction on the target medical index data through the encoding layer to obtain medical index features, and performs risk degree identification on the medical index features through the decoding layer, which can identify and capture key information of the medical index, predict health risks according to individual index features, and realize personalized health monitoring. Secondly, the present application calculates a target residual between the medical index risk degree and the observed medical index degree through the residual layer, which can effectively identify abnormal states of the medical index and discover potential health problems in time. Finally, the present application performs risk warning on the target medical index data according to the target residual and the risk warning layer, which can realize dynamic prediction of the medical index and help improve the accuracy of medical index risk warning.

[0075] In step S101 of some embodiments, specifically, the target medical index data refers to physiological parameters of the target user.

[0076] For example, the parameters of heart rate, blood pressure, blood oxygen saturation, body temperature, sleep condition, stress, and blood sugar.

[0077] Please refer to Figure 2 In some embodiments, step S101 includes, but is not limited to, steps S201 to S204:

[0078] In step S201, the preset hardware device detects the initial medical index data of the target user.

[0079] In step S202, the initial medical index data is subjected to outlier detection to obtain abnormal medical index data, and the abnormal medical index data is excluded from the standard medical index data to obtain normal medical index data.

[0080] In step S203, the normal medical index data is subjected to smoothing processing to obtain smoothed medical index data.

[0081] In step S204, the smoothed medical index data is subjected to data filling processing to obtain target medical index data.

[0082] In step S201 of some embodiments, specifically, the collection of initial medical index data can be realized by intelligent hardware devices. The intelligent hardware devices can be wearable devices (such as smart watches, smart bracelets) or portable monitoring instruments (such as sphygmomanometers, blood glucose meters), etc.

[0083] Specifically, the intelligent hardware device can continuously collect the physiological parameters of the target user within 24 hours, and has a certain local storage capacity so as to continue to work in the case of unstable network.

[0084] Specifically, the intelligent hardware device integrates various sensors, such as photoelectric heart rate sensors, blood pressure sensors, and blood oxygen sensors, etc. The photoelectric heart rate sensor can measure the light absorption change caused by blood flow to monitor the heart rate and heart rate variability by emitting a light beam to irradiate the skin; the blood pressure sensor can calculate the systolic pressure and diastolic pressure by detecting the pressure change generated when the blood reflows; the blood oxygen sensor can emit two light beams of different wavelengths, which penetrate the skin and blood vessels, are absorbed by hemoglobin, and measure the absorption amount of hemoglobin to different wavelengths of light, thereby determining the blood oxygen saturation of the user.

[0085] For example, in the medical technology scene, the target user wears a smart watch integrating the above-mentioned sensors, and at different time points in a day, the smart watch automatically wakes up the sensors, collects the initial medical index data at preset time intervals (such as once per minute), and processes the initial medical index data through the microprocessor in the smart watch, thereby calculating the physiological parameters of the user's heart rate, blood pressure, blood oxygen saturation, and body temperature, etc.

[0086] In this embodiment, by using high-precision sensors such as photoelectric heart rate sensors, blood pressure sensors, and blood oxygen sensors to collect target medical index data, the accuracy of the collected target medical index data can be ensured.

[0087] In an optional embodiment of the present application, the sensor data can also be calibrated and compensated by the built-in algorithm of the intelligent hardware device, to eliminate environmental interference and device errors, and further monitor the motion state of the target user by using sensors such as accelerometers and gyroscopes, and adjust the collection strategy of the target medical index data according to the motion state, to improve the data quality under the motion state.

[0088] In step S202 of some embodiments, specifically, the initial medical index data can be detected for abnormal values by statistical methods such as standard deviation and box plot.

[0089] Specifically, first, the average value and the standard deviation of the medical index of the target user within one day are obtained, and the average value of the medical index is compared with the standard deviation of the medical index. If the initial medical index exceeds the average value of the medical index and is far less than or far greater than the standard deviation of the medical index, the initial medical index is determined to be an abnormal index.

[0090] For example, in the process of collecting heart rate medical index, if the heart rate data point exceeds the average heart rate value by more than 3 standard deviations or less, we can mark the heart rate data point as an abnormal value and exclude it from the collected heart rate medical index data.

[0091] In step S203 of some embodiments, specifically, the normal medical index data can be smoothed by moving average or exponential smoothing method.

[0092] Specifically, based on the preset smoothing window size, for each normal medical index data collected, the average value of the medical index covered by the front and rear smoothing window size is calculated, and the medical average value is taken as the smoothed medical index data.

[0093] For example, if the smoothing window size is 5, then the heart rate smoothing value every 5s is calculated, and the heart rates of 72bpm, 70bpm, 68bpm, 76bpm, and 74bpm collected within 0-5s are averaged to obtain a heart rate of 72.4bpm as the smoothed medical index data.

[0094] In this embodiment, by smoothing the normal medical index data to obtain smoothed medical index data, random fluctuations or noise in the collected medical index data can be reduced, so that the trend and pattern of the medical index data are more obvious, and the data quality of the collected medical index data is improved.

[0095] In step S204 of some embodiments, specifically, the filling can be performed by linear interpolation, moving average or time series interpolation method.

[0096] For example, for the collected heart rate monitoring data, the heart rate data every 30 minutes in a day is recorded, but due to sensor failure, part of the data is missing, the heart rate collected at 0.5h is 68bpm, the heart rate data collected at 1h is actually, the heart rate collected at 1.5h is 77bpm, and the heart rate at 1h is calculated by linear interpolation method as 71.6bpm.

[0097] In an optional embodiment of the present application, the target medical index data can also be stored in the memory of the intelligent hardware device, and the target medical index data is transmitted to the mobile terminal or cloud server of the target user through wireless communication technology such as Bluetooth or Wi-Fi.

[0098] In step S102 of some embodiments, specifically, the risk warning model is a neural network model composed of an encoding layer, a decoding layer, a residual layer and a risk warning layer.

[0099] Further, the encoding layer is used to extract the features of the target medical index data; the decoding layer is used to identify the risk degree of the medical index features, realize the prediction of health risks according to the individual index features, and realize personalized health monitoring; the residual layer is used to calculate the target residual between the risk degree of the medical index and the observed medical index degree, and effectively identify the abnormal state of the medical index; the risk warning layer is used to perform risk warning on the target medical index data, and realize dynamic prediction of the medical index.

[0100] Please refer to Figure 3 In some embodiments, step S103 includes but is not limited to steps S301 to S304:

[0101] Step S301: The target medical index data is subjected to statistical distribution feature extraction by the encoding layer to obtain index statistical features.

[0102] Step S302: The index statistical features are subjected to time series feature extraction to obtain index time features.

[0103] Step S303: The index time features are subjected to frequency domain feature extraction to obtain index frequency domain features.

[0104] Step S304: The index statistical features, index time features and index frequency domain features are fused to obtain medical index features.

[0105] In step S301 of some embodiments, specifically, the index statistical features refer to the statistical data features of the target medical index data, and the index statistical features include but are not limited to mean, median, standard deviation, skewness and kurtosis features, etc.

[0106] Specifically, multiple hidden layers can be included in the encoding layer, and by adjusting the weight and bias parameters of the hidden layers, different dimensional features can be extracted.

[0107] Specifically, the target medical indicator data can be subjected to statistical distribution feature extraction through the following formula:

[0108] h1 = σ (W1X + b1)

[0109] Where h1 represents the indicator statistical feature, σ represents the activation function, W1 represents the statistical weight, X represents the target medical indicator data, and b1 represents the statistical bias parameter.

[0110] In step S302 of some embodiments, specifically, the indicator time feature refers to a feature extracted from the time series data of the target medical indicator data, such as autocorrelation, periodic component, trend, etc.

[0111] Specifically, the indicator time feature can be subjected to frequency domain feature extraction through the following formula:

[0112] h2 = σ (W2h1 + b2)

[0113] Where h2 represents the indicator time feature, σ represents the activation function, W2 represents the time weight, h1 represents the indicator statistical feature, and b2 represents the time bias parameter.

[0114] In step S303 of some embodiments, specifically, the indicator frequency domain feature refers to the amplitude and phase features of the target medical indicator data.

[0115] Specifically, the time series data is converted to the frequency domain by fast Fourier transform before extracting the frequency domain feature.

[0116] Specifically, the indicator time feature can be subjected to frequency domain feature extraction through the following formula:

[0117] h3 = σ (W3h2 + b3)

[0118] Where h3 represents the indicator frequency domain feature, σ represents the activation function, W3 represents the frequency domain weight, h2 represents the indicator time feature, and b3 represents the frequency domain bias parameter.

[0119] In step S304 of some embodiments, specifically, the indicator statistical feature, the indicator time feature, and the indicator frequency domain feature are spliced, and the spliced features are subjected to activation processing by a preset activation function (such as a Sigmoid function) to obtain a medical indicator feature represented in a low-dimensional space.

[0120] For example, in the medical technology scenario, if the target medical indicator data is real-time collected heart rate monitoring data, the encoding layer first learns the statistical indicator features of the daily average heart rate and the standard deviation of the heart rate; secondly, the daily periodicity of the heart rate can be identified through the extraction of the time indicator features, and the main amplitude and phase components of the heart rate change are further extracted to determine the frequency of the heart rate change; finally, these features are fused to form a medical indicator feature containing all the features.

[0121] In this embodiment, by fusing the indicator statistical features, the indicator time features and the indicator frequency domain features, multi-level indicator features can be extracted from the target medical indicator data to accurately identify and capture the main key information of the medical indicator, improve the accuracy of the medical indicator feature extraction, and facilitate subsequent improvement of the accuracy of the medical indicator risk warning.

[0122] Please refer to Figure 4 In some embodiments, step S104 includes but is not limited to steps S401 to S402:

[0123] In step S401, the medical indicator features are mapped to a pre-set high-dimensional space by the decoding layer to obtain high-dimensional indicator features.

[0124] In step S402, the high-dimensional indicator features are identified according to a pre-set reconstruction weight matrix to obtain the medical indicator risk degree.

[0125] In step S401 of some embodiments, specifically, the high-dimensional indicator features refer to the features of the medical indicator features mapped to the high-dimensional space.

[0126] Specifically, the decoding weight and the decoding bias parameter are obtained, and based on the decoding weight and the decoding bias parameter, the medical indicator features output by the encoding layer can be mapped to the high-dimensional space, and the interaction relationship between the medical indicator features is further learned through a nonlinear activation function (such as ReLU, sigmoid or tanh function).

[0127] For example, there is an interaction relationship between heart rate and blood oxygen saturation, which may cause an increase in heart rate when blood oxygen saturation decreases; there is an interaction relationship between blood pressure and heart rate, which may cause an increase in heart rate when there is a change in stress or pressure.

[0128] In step S402 of some embodiments, specifically, the reconstruction weight matrix is obtained by performing nonlinear activation on the real health indicator matrix and the observed medical indicator matrix of the target user. The real health indicator matrix is used to describe the health medical indicator value of the target user in a historical time period; the observed medical indicator matrix is used to describe the medical indicator value of the target user in an observation time period.

[0129] Specifically, the risk degree of the high-dimensional index feature is identified by reconstructing the weight matrix, and the preliminary medical index risk degree of the target user is preliminarily determined.

[0130] For example, in the medical technology scenario, the heart rate and the blood oxygen saturation, under normal circumstances, the decrease of the blood oxygen saturation may cause the increase of the heart rate as a compensation mechanism for the body to improve oxygen delivery. If it is monitored that the blood oxygen saturation reading is continuously lower than 90% in sleep, and the heart rate does not increase accordingly, it indicates that the target user's heart rate or blood oxygen saturation index is at risk.

[0131] In step S105 of some embodiments, specifically, the target residual is used to describe the difference between the medical index risk degree and the observed medical index degree, and the abnormal state of the medical index of the target user can be determined by the target residual.

[0132] Specifically, the observed medical index degree refers to the health degree of the real-time collected target medical index data.

[0133] Specifically, when the medical index risk degree is combined with the real medical index health degree in the historical time period, and is close to the observed medical index degree, it indicates that the prediction accuracy of the medical index risk degree is high, but when the medical index is at great risk, the medical index risk degree combined with the real medical index health degree is greatly different from the observed medical index degree, indicating that the prediction accuracy of the medical index risk degree is low. Based on this, the size of the target residual can reflect whether the target medical index data is at health risk.

[0134] Please refer to Figure 5 In some embodiments, step S106 includes but is not limited to steps S501 to S502:

[0135] Step S501, acquiring a risk warning threshold of the target medical index data.

[0136] Step S502, performing risk warning on the target medical index data according to the risk warning threshold, the target residual, and the risk warning layer, to obtain a risk warning level.

[0137] Please refer to Figure 6 In some embodiments, step S501 includes but is not limited to steps S601 to S604:

[0138] Step S601, acquiring window health medical index data of the target user by using a preset sliding window.

[0139] Step S602, performing index prediction on the window health medical index data by using a risk warning model to obtain window predicted health medical index data.

[0140] In step S603, a window health index residual is calculated between the window health medical index and the window predicted health medical index data, and the window health index residuals are combined based on a preset window time sequence to obtain a window health index residual sequence.

[0141] In step S604, a residual mean and a residual standard deviation of the window health index residual sequence are obtained, and a risk warning threshold of the target medical index data is determined according to a multiple of the residual mean and the residual standard deviation.

[0142] In step S601 of some embodiments, specifically, the preset sliding window is based on the collection frequency of the target medical index data and the expected time range.

[0143] For example, the health heart rate index of the target user collected every ten minutes is taken as the sliding window size, the health heart rate is collected once every minute, which is the step length of the sliding window movement, and the health heart rate index collected every ten minutes is taken as the window health medical index data.

[0144] In step S602 of some embodiments, specifically, the health degree of the window health medical index data is predicted by the risk warning model, and the window predicted health medical index data can be obtained.

[0145] For example, the heart rate and blood pressure data in each 10-minute window are input into the risk warning model, and the corresponding predicted heart rate and blood pressure data are output by the model.

[0146] In step S603 of some embodiments, specifically, for each sliding window, the actual window health medical index is subtracted from the window predicted health medical index data to obtain the window health index residual corresponding to each sliding window, which is used to describe the risk situation of the health medical index data collected by each sliding window.

[0147] Specifically, according to the time sequence of the window, the window health index residuals of all windows are combined to obtain the window health index residual sequence.

[0148] In step S604 of some embodiments, specifically, the statistics such as the mean and the standard deviation of the window health index residual sequence are calculated, and the risk warning threshold is determined based on the multiple of the mean and the standard deviation; wherein the multiple of the residual standard deviation can be determined based on the actual application scenario, and by selecting different multiples, the quantification of medical index abnormalities of different degrees can be realized.

[0149] For example, the risk warning threshold can be a mild risk warning threshold, a moderate risk warning threshold and a severe risk warning threshold. Since the distribution of the window health index residual sequence is approximately normal distribution, in order to simplify the calculation process, the residual mean can be determined as 0 and the residual standard deviation as 1. Then the mild risk warning threshold is the residual mean plus 1.5 times the residual standard deviation: 0+1x1.5=1.5; the moderate risk warning threshold is the residual mean plus 2 times the residual standard deviation: 0+1x2=2; and the severe risk warning threshold is the residual mean plus 2.5 times the residual standard deviation: 0+1x2.5=2.5.

[0150] In this embodiment, the risk warning threshold of the target medical index data is determined according to the residual mean and the multiple of the residual standard deviation, which can dynamically adjust the risk warning threshold according to the change trend of the medical index data of the target user, adapt to the change of the medical index at any time, match the corresponding risk warning threshold for different users, provide personalized risk prediction for the user, and help to improve the accuracy of medical index risk warning in the future.

[0151] In step S502 of some embodiments, specifically, the risk warning level is divided based on the risk warning threshold and the target residual.

[0152] Specifically, the risk warning level can also include a mild risk warning level, a moderate risk warning level and a severe risk warning level. For the mild risk warning level, the target user can be reminded that the medical index is abnormal, and the target user can adjust the medical index back to normal by improving diet or work and rest; for the moderate risk warning level, the target user can be reminded that the abnormality of the medical index is relatively serious, and further pathological examination in the hospital is needed; and for the severe risk warning level, the target user can be reminded that the abnormality of the index may exist serious illness, and needs to go to the hospital for disease diagnosis as soon as possible, and an emergency notification is automatically sent to the emergency contact person preset by the user, the location information of the user is provided to facilitate emergency rescue, and basic emergency treatment guidance is provided for the user to help the user or the emergency contact person take correct first aid measures before the arrival of professional medical personnel.

[0153] For example, if the residual of the heart rate index is not more than the risk warning threshold of 1.5, it indicates that all the medical indexes of the target user are healthy; if the residual of the heart rate index is more than the risk warning threshold of 1.5 but less than 2, it indicates that the medical index of the target user belongs to the mild risk warning level; if the residual of the heart rate index is more than the risk warning threshold of 2 but less than 2.5, it indicates that the medical index of the target user belongs to the moderate risk warning level; and if the residual of the heart rate index is more than the risk warning threshold of 2.5, it indicates that the medical index of the target user belongs to the severe risk warning level.

[0154] In this embodiment, the target medical index data is risk warned according to the risk warning threshold, the target residual error and the risk warning layer, and a risk warning level is obtained. The risk prediction of the medical index can be dynamically adjusted in combination with the risk warning threshold and the target residual error, the dynamic prediction of the medical index is realized, the risk degree of the medical index is accurately quantified through the multi-level risk division, and the accuracy of the risk warning of the medical index is significantly improved.

[0155] In an optional embodiment of the present application, the medical index data and trend of the target user can also be intuitively displayed in the user interface in the form of a chart.

[0156] Further, when the medical index is displayed in the user interface, the medical index data (such as heart rate, blood pressure, blood sugar, etc.) to be displayed is determined, and the appropriate chart type is selected according to the medical index data and the display purpose (such as time series data suitable for using a line chart, distribution data suitable for using a column chart or a pie chart), the medical index data is converted into a chart through a data visualization tool or library (such as D3.js, Chart.js, Matplotlib, etc.), and the target user can also make a chart according to his own needs in the user interface.

[0157] For example, in a mobile health application for diabetes patients, a simple dashboard is provided, and the patient can view health data including the current blood sugar level, activity amount and drug reminders, and view more detailed data by clicking different icons or buttons. In order to improve the ease of use, the user interface also provides large font, clear icons and simple navigation structure for the patient, so that the patient can easily use the application even in the case of poor vision or lack of technical knowledge.

[0158] Please refer to Figure 7 In some embodiments, after step S106, the medical index risk warning method further includes but is not limited to steps S701 to S703:

[0159] Step S701: Obtain the residual sum of squares of each target residual error.

[0160] Step S702: Obtain the risk contribution degree of each target medical index data according to each target residual error and the residual sum of squares.

[0161] Step S703: Select the target risk contribution degree with the largest contribution degree from the risk contribution degrees, and take the target medical index data of the target risk contribution degree as the risk reason for the risk warning level.

[0162] In step S701 of some embodiments, specifically, the residual sum of squares refers to the Euclidean norm representation of the target medical index of each target user.

[0163] For example, in the medical technology scenario, if the observed medical indicators of the target user are [72, 130 / 85, 5.5] (heart rate 72 bpm, blood pressure 130 / 85 mmHg, and blood glucose 5.5 mmol / L), the predicted medical indicators are [70, 120 / 80, 5.0], the target residual is [2, 10 / 5, 0.5], the square of 2, the square of 10 / 5, and the square of 0.5 are summed, and the square root of the sum is taken to obtain the residual square sum.

[0164] In step S702 of some embodiments, specifically, the risk contribution degree refers to the ratio of the target residual of the target medical indicator to the residual square sum, which is used to represent the contribution degree of each target medical indicator to the residual size.

[0165] For example, the risk contribution degree of heart rate is the ratio of the square of 2 to the residual square sum, which is 0.2; the risk contribution degree of blood pressure is the ratio of the square of 0 / 5 to the residual square sum, which is 0.6; and the risk contribution degree of blood glucose is the ratio of the square of 0.5 to the residual square sum, which is 0.2.

[0166] In step S703 of some embodiments, for example, the risk contribution degree of heart rate is 0.2, the risk contribution degree of blood pressure is 0.6, and the risk contribution degree of blood glucose is 0.2, the risk contribution degree of blood pressure is taken as the target risk contribution degree, and the blood pressure indicator is taken as the risk reason for the risk warning level.

[0167] In this embodiment, by selecting the target risk contribution degree with the largest contribution degree from the risk contribution degrees, and taking the target medical indicator data of the target risk contribution degree as the risk reason for the risk warning level, the residual and the risk contribution degree are combined to not only generate a specific risk warning, but also point out the potential reason for the risk warning, thereby providing more targeted health advice for the target user.

[0168] The embodiment of the application collects target medical index data of a target user, obtains a pre-trained risk early warning model, wherein the risk early warning model comprises an encoding layer, a decoding layer, a residual layer and a risk early warning layer, performs feature extraction on the target medical index data through the encoding layer to obtain medical index features, performs risk degree identification on the medical index features through the decoding layer to obtain a medical index risk degree of the target medical index data, obtains an observed medical index degree of the target medical index data, calculates a target residual between the medical index risk degree and the observed medical index degree through the residual layer, performs risk early warning on the target medical index data according to the target residual and the risk early warning layer, and obtains a risk early warning level. The embodiment of the application first performs feature extraction on the target medical index data through the encoding layer to obtain medical index features, and performs risk degree identification on the medical index features through the decoding layer, so that the key information of the medical index can be identified and captured, and the health risk of an individual can be predicted according to the individual index features to realize personalized health monitoring. Secondly, the target residual between the medical index risk degree and the observed medical index degree is calculated through the residual layer, so that the abnormal state of the medical index can be effectively identified, and potential health problems can be found in time. Finally, risk early warning is performed on the target medical index data according to the target residual and the risk early warning layer, so that dynamic prediction of the medical index can be realized, which helps to improve the accuracy of medical index risk early warning.

[0169] Please refer to Figure 8 The embodiment of the application also provides a medical index risk early warning device, which can realize the medical index risk early warning method. The device comprises:

[0170] a medical index acquisition module configured to collect target medical index data of a target user;

[0171] a model acquisition module configured to obtain a pre-trained risk early warning model, wherein the risk early warning model comprises an encoding layer, a decoding layer, a residual layer and a risk early warning layer;

[0172] a feature extraction module configured to perform feature extraction on the target medical index data through the encoding layer to obtain medical index features;

[0173] a risk degree identification module configured to perform risk degree identification on the medical index features through the decoding layer to obtain a medical index risk degree of the target medical index data;

[0174] a residual calculation module configured to obtain an observed medical index degree of the target medical index data, and calculate a target residual between the medical index risk degree and the observed medical index degree through the residual layer;

[0175] a risk early warning module configured to perform risk early warning on the target medical index data according to the target residual and the risk early warning layer, and obtain a risk early warning level.

[0176] The specific implementation of the medical index risk early warning device is basically the same as the specific embodiments of the medical index risk early warning method described above, and will not be repeated here.

[0177] The embodiments of the present application also provide an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor implements the medical index risk early warning method described above when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a vehicle-mounted computer, etc.

[0178] Please refer to Figure 9 , Figure 9 The hardware structure of the electronic device of another embodiment is illustrated, which includes:

[0179] The processor 901 can be implemented in a general-purpose CPU (Central Processing Unit), a microprocessor, an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits, etc., and is used to execute related programs to implement the technical solutions provided by the embodiments of the present application.

[0180] The memory 902 can be implemented in the form of a ROM (Read Only Memory), a static storage device, a dynamic storage device, or a RAM (Random Access Memory), etc. The memory 902 can store processing systems and other application programs. When the technical solutions provided by the embodiments of the present application are implemented by software or firmware, the related program codes are saved in the memory 902 and are called and executed by the processor 901 to implement the medical index risk early warning method of the embodiments of the present application.

[0181] The input / output interface 903 is used to realize information input and output.

[0182] The communication interface 904 is used to realize the communication interaction between the device and other devices. The communication can be realized by a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.).

[0183] The bus 905 transmits information between various components (such as the processor 901, the memory 902, the input / output interface 903, and the communication interface 904) of the device.

[0184] The processor 901, the memory 902, the input / output interface 903, and the communication interface 904 are connected to each other through the bus 905 to realize the communication connection between them inside the device.

[0185] The embodiment of the present application also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the medical index risk early warning method.

[0186] The memory, as a non-transitory computer readable storage medium, can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory remotely arranged relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0187] The medical index risk early warning method, the medical index risk early warning device, the electronic equipment and the storage medium provided by the embodiment of the present application collect target medical index data of a target user; a pre-trained risk early warning model is obtained; wherein the risk early warning model includes an encoding layer, a decoding layer, a residual layer and a risk early warning layer; the target medical index data is feature extracted through the encoding layer to obtain medical index features; the medical index features are risk degree identified through the decoding layer to obtain a medical index risk degree of the target medical index data; an observed medical index degree of the target medical index data is obtained, and a target residual between the medical index risk degree and the observed medical index degree is calculated through the residual layer; the target medical index data is risk early warned according to the target residual and the risk early warning layer to obtain a risk early warning level. The embodiment of the present application first extracts features of the target medical index data through the encoding layer to obtain medical index features, and identifies the risk degree of the medical index features through the decoding layer, which can identify and capture the key information of the medical index, and predict the health risk according to the individual index features to realize personalized health monitoring; secondly, the target residual between the medical index risk degree and the observed medical index degree is calculated through the residual layer, which can effectively identify the abnormal state of the medical index and discover potential health problems in time; finally, the target medical index data is risk early warned according to the target residual and the risk early warning layer, which can realize dynamic prediction of the medical index and help to improve the accuracy of the medical index risk early warning.

[0188] The embodiments described in the embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that, with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0189] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation to the embodiments of the present application, and can include more or fewer steps than the figures, or combine certain steps, or different steps.

[0190] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separated, that is, can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiments.

[0191] Those skilled in the art can understand that all or some steps in the above disclosed method, functions of the modules / units in the system and the device can be implemented as software, firmware, hardware and their appropriate combinations.

[0192] The terms "first", "second", "third", "fourth" and the like in the description of the present application and the above-mentioned figures (if any) are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not necessarily limit to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0193] It should be understood that in the present application, "at least one" means one or more, and "multiple" means two or more. "And / or" is used to describe the association between the associated objects, which means that there can be three relationships, for example, "A and / or B" can represent three cases: only A, only B, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects. "At least one of the following" or similar expressions means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b or c can mean a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0194] In several embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented by other manners. For example, the apparatus embodiments described above are merely illustrative, for example, the division of the above units is merely a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. The coupling or direct coupling or communication connection between the units or components shown or discussed can be indirect coupling or communication connection through some interfaces, apparatuses or units, and can be electrical, mechanical or other forms.

[0195] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they can be located in one place or distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0196] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0197] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part of the prior art that contributes to the technical solutions or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method of each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program storage media.

[0198] The preferred embodiments of the embodiments of the present application are described above with reference to the accompanying drawings, but this does not limit the scope of the embodiments of the present application. Any modifications, equivalent replacements and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall be within the scope of the embodiments of the present application.

Claims

1. A medical index risk early warning method, characterized in that, The method comprises: collecting target medical index data of a target user; obtaining a pre-trained risk warning model; wherein the risk warning model comprises an encoding layer, a decoding layer, a residual layer and a risk warning layer; extracting features of the target medical index data through the encoding layer to obtain medical index features; identifying a risk degree of the medical index features through the decoding layer to obtain a medical index risk degree of the target medical index data; obtaining an observed medical index degree of the target medical index data, and calculating a target residual between the medical index risk degree and the observed medical index degree through the residual layer; warning the target medical index data of a risk according to the target residual and the risk warning layer to obtain a risk warning level; the method comprises: extracting statistical distribution features of the target medical index data through the encoding layer to obtain index statistical features; extracting time sequence features of the index statistical features to obtain index time features; extracting frequency domain features of the index time features to obtain index frequency domain features; fusing the index statistical features, the index time features and the index frequency domain features to obtain the medical index features; the method comprises: obtaining a risk warning threshold of the target medical index data; warning the target medical index data of a risk according to the risk warning threshold, the target residual and the risk warning layer to obtain the risk warning level; the method comprises: obtaining window health medical index data of the target user by using a preset sliding window; predicting index of the window health medical index data through the risk warning model to obtain window predicted health medical index data; calculating a window health index residual between the window health medical index and the window predicted health medical index data; determining the risk warning threshold of the target medical index data according to the window health index residual.

2. The method of claim 1, wherein, the method comprises: combining the window health index residual based on a preset window time sequence to obtain a window health index residual sequence; obtaining a residual mean and a residual standard deviation of the window health index residual sequence, and determining the risk warning threshold of the target medical index data according to a multiple of the residual mean and the residual standard deviation.

3. The method of claim 1, wherein, the method comprises: mapping the medical index features to a preset high-dimensional space through the decoding layer to obtain high-dimensional index features; identifying a risk degree of the high-dimensional index features according to a preset reconstruction weight matrix to obtain the medical index risk degree.

4. The method of claim 1, wherein, The method comprises the following steps: Obtaining initial medical index data of the target user by a preset hardware device; Performing outlier detection on the initial medical index data to obtain abnormal medical index data, and removing the abnormal medical index data from standard medical index data to obtain normal medical index data; Performing smoothing processing on the normal medical index data to obtain smoothed medical index data; Performing data filling processing on the smoothed medical index data to obtain the target medical index data.

5. The method according to any one of claims 1 to 4, characterized in that, After the risk warning level is obtained by performing risk warning on the target medical index data according to the target residual error and the risk warning layer, the method further comprises the following steps: Obtaining the sum of squares of each target residual error; Obtaining the risk contribution degree of each target medical index data according to each target residual error and the sum of squares of the target residual error; Selecting a target risk contribution degree with the largest contribution degree from the risk contribution degrees, and taking the target medical index data of the target risk contribution degree as the risk cause of the risk warning level.

6. A medical index risk early warning device, characterized in that, The device comprises: A medical index acquisition module for collecting target medical index data of a target user; A model acquisition module for acquiring a pre-trained risk warning model; wherein the risk warning model comprises an encoding layer, a decoding layer, a residual error layer, and a risk warning layer; A feature extraction module for extracting features of the target medical index data through the encoding layer to obtain medical index features; A risk degree identification module for identifying the risk degree of the target medical index data through the decoding layer to obtain medical index risk degree; A residual error calculation module for obtaining an observed medical index degree of the target medical index data, and calculating a target residual error between the medical index risk degree and the observed medical index degree through the residual error layer; A risk warning module for performing risk warning on the target medical index data according to the target residual error and the risk warning layer to obtain a risk warning level. The feature extraction module further comprises: Extracting statistical distribution features of the target medical index data through the encoding layer to obtain index statistical features; Extracting time series features of the index statistical features to obtain index time features; Extracting frequency domain features of the index time features to obtain index frequency domain features; Fusing the index statistical features, the index time features, and the index frequency domain features to obtain the medical index features. The risk warning module further comprises: Obtaining a risk warning threshold of the target medical index data; Performing risk warning on the target medical index data according to the risk warning threshold, the target residual error, and the risk warning layer to obtain the risk warning level. The risk warning module further comprises: Obtaining window health medical index data of the target user by using a preset sliding window; Performing index prediction on the window health medical index data by using the risk warning model to obtain window predicted health medical index data; calculating a window health index residual between the window health medical index and the window predicted health medical index data; determining a risk early warning threshold of the target medical index data according to the window health index residual.

7. An electronic device, comprising: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the medical index risk early warning method in any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium storing a computer program, the computer-readable storage medium comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 7. The computer program is executed by the processor to implement the medical index risk early warning method in any one of claims 1 to 5. The computer program is executed by the processor to implement the medical index risk early warning method in any one of claims 1 to 5.

Images