Data correction method and device and computer equipment
By acquiring and analyzing the window length data characteristics of the glucose monitoring system, abnormal sensor data can be identified and corrected, solving the problem of inaccurate monitoring caused by pressure and realizing the integration of low-power, small-volume devices.
Patent Information
- Application Number
- CN202411029529.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2024-11-15
AI Technical Summary
Existing continuous glucose monitoring systems exhibit abnormal fluctuations in sensor data when subjected to pressure, leading to inaccurate monitoring results. Furthermore, existing correction methods consume significant power and are difficult to integrate into small-sized devices.
By acquiring analyte concentration data within the current time and a preset time period, a first window length of data is formed. Based on the data characteristics, the data to be corrected is determined, the state type is set, and the output analyte concentration data is determined according to the trend characteristics. The single analyte sensor is used for identification and correction.
It reduces system integration difficulty and power consumption, improves monitoring accuracy and efficiency, and avoids reliance on additional physical sensors.
Smart Images

Figure CN118969246A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of continuous analyte monitoring, and in particular to a data correction method, device and computer equipment. Background Art
[0002] The invasive continuous glucose monitoring system (CGM) is a specific analyte monitoring system that detects the user's interstitial fluid glucose concentration by inserting sensor electrodes into the user's interstitial fluid. Because there is a high correlation between the interstitial fluid glucose concentration and the blood glucose concentration (mainly from capillary diffusion and the dynamic balance of glucose in the blood), the user's dynamic blood glucose information can be obtained.
[0003] Under special circumstances, the glucose concentration data detected by CGM will fluctuate abnormally. For example, when the site of action of CGM is under continuous pressure (typically, when the user is sleeping), the interstitial fluid glucose concentration deviates from the blood glucose concentration due to local ischemia of human tissue, causing the data measured by the sensor to be unable to truly reflect the blood glucose level of the human body. Typically, when under pressure, the sensor shows a large and deep depression in the data performance. This type of abnormal deviation from blood glucose expression may trigger false prompts such as hypoglycemia alarms when the deviation is more serious, seriously affecting the accuracy of the sensor monitoring results.
[0004] Existing methods for abnormal identification and correction of continuous glucose monitoring systems use data streams provided by additional physical sensors (typically, pressure sensors, gyroscopes, etc.) to perform abnormal analysis and switch signal processing algorithms accordingly. This type of method has the problem of high system power consumption and difficulty in integration on small-volume devices. Summary of the invention
[0005] In this embodiment, a data correction method, apparatus and computer device are provided to solve the problem in the related art that the system has high power consumption and is difficult to integrate on a small-volume device.
[0006] In a first aspect, a data correction method is provided in this embodiment, the method comprising:
[0007] Acquire analyte concentration data corresponding to the current moment and a first time period to form first window length data; the first time period is a moment within a first preset time length before the current moment;
[0008] Based on the data characteristics of the first window length data, determining whether the original analyte concentration data corresponding to the current moment is data to be corrected;
[0009] If the original analyte concentration data is data to be corrected, setting the state type at the current moment based on the trend characteristics of the first window length data;
[0010] The output analyte concentration data at the current time is determined based on the state type.
[0011] In some embodiments, judging whether the original analyte concentration data corresponding to the current moment is data to be corrected based on the data feature of the first window length data includes:
[0012] Extracting at least two types of data features from the first window length data;
[0013] Based on the confidence and weight values of each type of data features, jointly calculate each type of data features;
[0014] Based on the joint calculation result, it is determined whether the original analyte concentration data corresponding to the current moment is data to be corrected.
[0015] In some embodiments, the combined calculation of various types of data features based on the confidence and weight values of the various types of data features includes:
[0016] Calculate feature thresholds for each type of feature based on the current user's historical analyte concentration data;
[0017] Setting the confidence level of the data feature based on the magnitude relationship between the feature threshold and the corresponding data feature;
[0018] Based on the confidence of each type of data feature and a preset weight value, each type of data feature is jointly calculated; the weight value corresponds to each type of data feature.
[0019] In some embodiments, setting the state type at the current moment based on the trend characteristics of the first window length data includes:
[0020] Calculating the trend characteristics of the first window length data;
[0021] When the trend feature is a sharp upward peak, the state type at the current moment is marked as the beginning of fluctuation;
[0022] When the trend feature is a downward peak, the state type at the current moment is marked as fluctuation recovery.
[0023] In some embodiments, determining the output analyte concentration data at the current moment based on the state type includes:
[0024] If the state type mark is marked as the fluctuation start, the predicted data is used as the output analyte concentration data at the current moment;
[0025] If the state type mark is recorded as the fluctuation recovery, the difference between the original analyte concentration data and the predicted data is calculated; and the output analyte concentration data at the current moment is determined based on the difference.
[0026] In some embodiments, determining the output analyte concentration data at the current moment based on the difference comprises:
[0027] When the difference is less than a comparison threshold, the original analyte concentration data is used as the output analyte concentration data at the current moment;
[0028] When the difference is not less than the comparison threshold, the predicted data is used as the output analyte concentration data at the current moment.
[0029] In some of the embodiments, before the step of using the predicted data as the output analyte concentration data at the current moment if the state type mark is marked as the fluctuation start, the method further includes:
[0030] Obtaining second window length data based on analyte concentration data corresponding to a second time period, wherein the second time period is a moment within a second preset time length before the current moment;
[0031] The prediction data is generated based on the second window length data.
[0032] In some embodiments, determining the output analyte concentration data at the current moment based on the state type further includes:
[0033] If the original analyte concentration data is not data to be corrected, the state type at the current moment is a reset state;
[0034] In the reset state, the original analyte concentration data is used as the output analyte concentration data at the current moment.
[0035] In a second aspect, a data correction device is provided in this embodiment, the device comprising:
[0036] A raw data acquisition module, used to acquire analyte concentration data at a current moment and within a preset time length before the current moment, to form first window length data;
[0037] A module for determining data to be corrected, used for determining whether the original analyte concentration data corresponding to the current moment is data to be corrected based on the data characteristics of the first window length data;
[0038] A state type marking module, configured to set the state type at the current moment based on the trend characteristics at the current moment if the original analyte concentration data is data to be corrected;
[0039] A data output module is used to determine the output analyte concentration data at the current moment based on the state type.
[0040] In a third aspect, the present application further provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the data correction method described in the first aspect is implemented.
[0041] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the data correction method described in the first aspect is implemented.
[0042] Compared with the related art, the data correction method, apparatus and computer equipment provided in the present embodiment form first window length data by acquiring analyte concentration data at the current moment and within a preset time length before the current moment; based on the data characteristics of the first window length data, determine whether the original analyte concentration data corresponding to the current moment is the data to be corrected; if the original analyte concentration data is the data to be corrected, set the state type at the current moment based on the trend characteristics of the first window length data; determine the output analyte concentration data at the current moment based on the state type, solve the problem that the system power consumption is large and difficult to integrate on a small-volume device, and realize the use of a single analyte sensor to identify and correct the data to be corrected, which helps to reduce the difficulty of system integration and reduce system power consumption.
[0043] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0045] Figure 1 A hardware structure block diagram of a terminal of a data correction method in one embodiment;
[0046] Figure 2 A schematic diagram of a flow chart of a data correction method in an embodiment;
[0047] Figure 3 A schematic flow chart of a data correction method in a preferred embodiment;
[0048] Figure 4 Acquiring a raw signal for an analyte monitoring system in one embodiment;
[0049] Figure 5 A correction signal obtained after the analyte monitoring system in one embodiment identifies and corrects the data to be corrected;
[0050] Figure 6 A schematic diagram showing a comparison between a measured curve and a predicted curve of analyte concentration in a preferred embodiment;
[0051] Figure 7 A schematic diagram of a label for collecting signals when the analyte sensor is under pressure in a preferred embodiment;
[0052] Figure 8 Schematic diagram of the structure of a data correction device in one embodiment. DETAILED DESCRIPTION
[0053] In order to more clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments.
[0054] Unless otherwise defined, the technical terms or scientific terms involved in this application shall have the general meaning understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the", "these" and the like in this application do not represent quantitative restrictions, and they can be singular or plural. The terms "include", "comprise", "have" and any variants thereof involved in this application are intended to cover non-exclusive inclusions; for example, a process, method and system, product or device comprising a series of steps or modules (units) is not limited to the listed steps or modules (units), but may include unlisted steps or modules (units), or may include other steps or modules (units) inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether directly or indirectly. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, "A and / or B" can mean: A exists alone, A and B exist at the same time, and B exists alone. Usually, the character " / " indicates that the objects associated with each other are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.
[0055] The method embodiment provided in this embodiment can be executed in a terminal, a computer or a similar computing device. For example, running on a terminal, Figure 1 FIG. 1 is a hardware structure diagram of a terminal of the data correction method of this embodiment. Figure 1 As shown, the terminal may include one or more ( Figure 1 Only one is shown in the figure) processor 102 and memory 104 for storing data, wherein processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA. The above terminal may also include a transmission device 106 and an input and output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.
[0056] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the data correction method in this embodiment. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0057] The transmission device 106 is used to receive or send data via a network. The above network includes a wireless network provided by the communication provider of the terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (Radio Frequency, referred to as RF) module, which is used to communicate with the Internet wirelessly.
[0058] In this embodiment, a data correction method is provided, which is applied to an analyte monitoring system. An analyte monitoring system is a system for collecting, processing, and analyzing analytes in a specific object (such as chemical components in gases, liquids, and solids). Specifically, it can be a biological sample monitoring system or an environmental monitoring system, such as a wearable biological sample monitoring system, preferably a continuous glucose monitoring system (Continuous Glucose System, CGM), wherein the analyte concentration data used in the data correction method is preferably glucose concentration data or blood sugar concentration data.
[0059] Figure 2 is a flow chart of the data correction method of this embodiment, such as Figure 2 As shown, the process includes the following steps:
[0060] Step S210, acquiring analyte concentration data at the current moment and within a first time period to form first window length data; the first time period is a moment within a first preset time length before the current moment.
[0061] For details, see Figure 3 , the current time is time t, and the first preset time can be set accordingly according to the requirements of calculation accuracy, calculation data volume, etc. For example, the first preset time is set to T 1 -1, then the first period is from time t-1 to t-(T 1 -1) time, from which we can get t to t-(T 1 -1) Total T 1 The first window length data of the time length. Wherein, the analyte concentration data is the gas concentration or liquid concentration in the target object, preferably the glucose concentration data, or the blood sugar concentration data.
[0062] Step S220: Based on the data characteristics of the first window length data, it is determined whether the original analyte concentration data corresponding to the current moment is data to be corrected.
[0063] Specifically, the data to be corrected includes abnormal data, and may also include redundant data, distorted data, etc. At least two types of data features are extracted from the first window length data, and the data features include the amplitude of high-frequency data, principal component analysis residuals, data prediction errors, and high-frequency data control charts. In another embodiment, a certain type of data feature can also be extracted from the first window length data, and analysis is performed based on this type of single data feature to improve the calculation speed. A threshold judgment is performed on the extracted data feature, and the original analyte concentration data corresponding to the current moment is analyzed based on the judgment result of the data feature to determine whether it is the data to be corrected.
[0064] Step S230: If the original analyte concentration data is data to be corrected, the state type at the current moment is set based on the trend characteristics of the first window length data.
[0065] Specifically, when the data to be corrected appears, the data detected by the sensor may show a large and deep concave, and may also show a convex phenomenon in other types of anomalies. The trend characteristics of the current moment can be analyzed to determine the abnormal stage at the current moment, so as to process the data to be corrected more accurately.
[0066] Among them, the trend feature can be obtained by calculating the peak direction of the first window length data. For example, if the peak is upward, the slope of the subsequent data of the peak is negative, that is, the data shows a downward trend; or if the peak is downward, the slope of the subsequent data of the peak is positive, and the data shows an upward trend. When the trend feature is a peak upward, the state type is set to the start of fluctuation, and when the trend feature is a peak downward, the state type is set to the end of fluctuation; or, when the trend feature is a peak downward, the state type is set to the start of fluctuation, and when the trend feature is a peak upward, the state type is set to the end of fluctuation. For different types of fluctuations, the corresponding relationship between trend features and state types can be set differently.
[0067] Step S240: determining the output analyte concentration data at the current moment based on the state type.
[0068] Specifically, step S240 can process both abnormal and non-abnormal situations, that is, the state type includes the start of fluctuation, the end of fluctuation, and the reset state of non-fluctuation. Among them, the reset state is the default state at all times. When step S220 does not detect fluctuation, the default reset state is maintained at the current moment. When step S220 detects fluctuation, the fluctuation is distinguished based on step S230, and the reset state is replaced by the start of fluctuation or the end of fluctuation. The processing methods corresponding to different state types are as follows: if the current moment corresponds to the reset state, the original analyte concentration data is used as the output analyte concentration data; if the current moment corresponds to the start of fluctuation, the predicted data is used as the output analyte concentration data; if the current moment corresponds to the end of fluctuation, the difference between the predicted data and the original analyte concentration data is analyzed. If the difference is large, the predicted data is used as the output analyte concentration data. If the difference is small, the state type is restored to the reset state, and the original analyte concentration data is used as the output analyte concentration data. The large or small difference can be judged by the threshold. Figure 4 and Figure 5 The comparison between the original signal and the corrected signal collected by this embodiment is shown. Figure 4 The circled signals are the identified data to be corrected. The corrected data is processed by the above method to obtain the following Figure 5 The corrected data shown shows that abnormal fluctuations have been accurately eliminated.
[0069] Step S240 can also only process the situation where fluctuations have occurred, that is, the state type only includes the start of fluctuations and the end of fluctuations. When step S220 detects fluctuations, the corresponding current moment is marked as the start of fluctuations and the end of fluctuations. If it is marked as the start of fluctuations, the original analyte concentration data is replaced, and the predicted data is used as the output analyte concentration data to complete the correction of the data at the current moment. If it is marked as the end of fluctuations, the difference between the predicted data and the original analyte concentration data is analyzed; if the difference is large, the original analyte concentration data is replaced with the predicted data, and the predicted data is used as the output analyte concentration data; if the difference is small, there is no need to replace the original analyte concentration data, and the original analyte concentration data is used as the output analyte concentration data. For non-fluctuating situations, the original analyte concentration data can be directly used as the data at the current moment after detection in step S220, without the need to correct the original data.
[0070] The output analyte concentration data is used for output display or for other data analysis.
[0071] In this embodiment, the data correction method forms first window length data by acquiring analyte concentration data at the current moment and within a preset time length before the current moment; based on the data characteristics of the first window length data, it is determined whether the original analyte concentration data corresponding to the current moment is the data to be corrected; if the original analyte concentration data is the data to be corrected, the state type at the current moment is set based on the trend characteristics of the first window length data; based on the state type, the output analyte concentration data at the current moment is determined, so as to solve the problem that the system power consumption is large and difficult to integrate on a small-volume device, and realize the use of a single analyte sensor to identify and correct the data to be corrected, which helps to reduce the difficulty of system integration and reduce system power consumption.
[0072] In one embodiment, see Figure 3 Based on step S220, based on the data characteristics of the first window length data, determining whether the original analyte concentration data corresponding to the current moment is data to be corrected specifically includes the following steps:
[0073] Step S310: extract at least two types of data features from the first window length data.
[0074] Step S320, based on the confidence and weight values of each type of data features, jointly calculate the various types of data features.
[0075] Step S330 , judging whether the original analyte concentration data corresponding to the current moment is data to be corrected based on the joint calculation result.
[0076] Specifically, each type of feature has a corresponding set of confidence and weight values. Based on the confidence and weight values, a weighted sum calculation is performed on each type of data feature. If the calculation result exceeds the preset fluctuation threshold, the original analyte concentration data corresponding to the current moment is judged as the data to be corrected. If the calculation result does not exceed the preset fluctuation threshold (for example, the fluctuation threshold is 0.5), the original analyte concentration data corresponding to the current moment is not judged as the data to be corrected. Among them, the weight value is preset; the confidence is calculated based on the data characteristics.
[0077] In this embodiment, a multi-dimensional judgment is made on the situation to be corrected through joint calculation of multiple data features, thereby improving the accuracy of the judgment on the situation to be corrected.
[0078] In one embodiment, see Figure 3 Based on the above step S320, based on the confidence and weight values of each type of data features, various types of data features are jointly calculated, including:
[0079] Step S321, calculating feature thresholds of various features based on the current user's analyte concentration history data.
[0080] Specifically, historical data of the analyte concentration of the current user for N days is obtained, and a stable characteristic value of the analyte concentration of the user is calculated to serve as a characteristic threshold value of each characteristic.
[0081] Step S322, setting the confidence level of the data feature based on the magnitude relationship between the feature threshold and the corresponding data feature.
[0082] Specifically, data features exceeding the feature threshold are recorded as 1, and data features not exceeding the feature threshold are recorded as 0.
[0083] Step S323, based on the confidence of each type of data feature and the preset weight value, the various types of data features are jointly calculated; the weight value corresponds to the various types of data features.
[0084] In this embodiment, personalized feature thresholds are set based on the user's historical data, so that the determination to be corrected is more accurate for different individuals.
[0085] In one embodiment, based on the above step S230, setting the state type at the current moment based on the trend characteristics of the first window length data includes:
[0086] Step S231, calculating the trend characteristics of the first window length data.
[0087] Step S232, when the trend feature is a sharp upward trend, the state type at the current moment is marked as the beginning of fluctuation.
[0088] Step S233, when the trend feature is a downward peak, the current state type is marked as fluctuation recovery.
[0089] For details, see Figure 6 When the analyte sensor is compressed, the sensor will show a large and deep depression in the data performance. This type of abnormality that deviates from the normal expression of analyte concentration may trigger error prompts such as low value alarms when it accumulates to a sufficient level. The trend characteristics can be used to analyze the stage of fluctuation in which the anomaly is located. Since the slope of the subsequent data after the spike is negative, that is, the data is on a downward trend, it can be said that the current moment is at the beginning of compression; the slope of the subsequent data after the spike is positive, and the data is on an upward trend, it can be said that the current moment is at the end of compression, thereby further dividing and marking the fluctuation situation. Figure 6 If a peak is detected at point A and its trend is downward, point A is marked as the start of fluctuation. If a peak is detected at point B and its trend is upward, point B is marked as the recovery of fluctuation. Other fluctuations similar to those caused by compression can also be identified by the direction of the peak in the data.
[0090] In this embodiment, by analyzing the characteristics of two key nodes in the analyte concentration waveform during the fluctuation, the fluctuation stage at the current moment is accurately marked, thereby improving the monitoring system's perception of the correction data.
[0091] In one embodiment, based on the above step S240, determining the output analyte concentration data at the current moment based on the state type includes:
[0092] Step S410: If the state type mark is marked as fluctuation start, the predicted data is used as the output analyte concentration data at the current moment.
[0093] Step S420: If the state type is marked as fluctuation recovery, the difference between the original analyte concentration data and the predicted data is calculated; and the output analyte concentration data at the current moment is determined based on the difference.
[0094] Specifically, the forecast data can be obtained through dynamic forecast calculation, or can be set in advance through fixed forecast values.
[0095] See also Figure 6 After the fluctuation begins, the current moment must be a deviation from the normal data, so it is corrected by the predicted value. After the fluctuation recovers, the recovery process cannot be judged by the current moment, so by comparing the original analyte concentration data with the predicted data, it is determined whether the current moment is a deviation from the normal data.
[0096] In this embodiment, by classifying and discussing the fluctuation start and the fluctuation recovery, a corresponding correction scheme is given to ensure the correction quality of the data to be corrected.
[0097] In one embodiment, based on the above step S420, determining the output analyte concentration data at the current moment based on the difference value includes:
[0098] Step S421: when the difference is less than the comparison threshold, the original analyte concentration data is used as the output analyte concentration data at the current moment.
[0099] Step S422: when the difference is not less than the comparison threshold, the predicted data is used as the output analyte concentration data at the current moment.
[0100] See also Figure 6 When the difference is less than the comparison threshold, it means that the measured curve is close enough to the predicted curve, then the prediction can be stopped and the measured curve is used as the output; otherwise, the predicted curve is output.
[0101] In this embodiment, by comparing the difference values, data closer to the actual situation is used as the output data during fluctuation recovery, thereby improving the output accuracy during fluctuation recovery.
[0102] In one embodiment, before step S410, the method further includes:
[0103] Step S430, obtaining second window length data based on the analyte concentration data corresponding to the second time period, where the second time period is a moment within a second preset time length before the current moment.
[0104] For details, see Figure 3 , the current time is time t, the second preset time length can be the same as the first preset time length, or can be different. Preferably, the second preset time length is greater than the first preset time length. When the second preset time length is set to T 2 -1, and the second period is from time t-1 to t-(T 2 -1) time, from t to t-(T 2 -1) The analyte concentration data within the time window form the second window data.
[0105] Step S440: Generate prediction data based on the second window length data.
[0106] Specifically, data prediction methods include but are not limited to prediction methods based on autoregression, deep learning, machine learning, etc.
[0107] In this embodiment, dynamic prediction is performed through the second window length data, and the predicted data is updated in real time according to the current moment, thereby improving the quality of correction of the data to be corrected.
[0108] In one embodiment, based on the above step S240, the output analyte concentration data at the current moment is determined based on the state type, and further includes:
[0109] If the original analyte concentration data is not data to be corrected, the state type at the current moment is a reset state; in the reset state, the original analyte concentration data is used as the output analyte concentration data at the current moment.
[0110] Specifically, the default flag bit at the current moment is 0, i.e., reset state; when it is identified as fluctuation by step S220, the flag bit will be marked as +1, i.e., fluctuation starts, or the flag bit will be marked as -1, i.e., fluctuation resumes. After receiving the analyte concentration data value at the current moment, its state type is judged based on the flag bit. If the flag bit is 0, no data correction processing is performed, i.e., the original analyte concentration data is used as the output analyte concentration data at the current moment.
[0111] The present embodiment is described and illustrated below through the first preferred embodiment.
[0112] Figure 3 FIG. 1 is a flow chart of data correction in the first preferred embodiment, as shown in FIG. Figure 3 As shown, the method includes the following processes:
[0113] 1. Fluctuation Detection
[0114] S1. Receive the analyte concentration data Sample at the current time t collected by the continuous glucose monitoring system (CGM) t , and combined with the analyte concentration data corresponding to the first time period, to form the first window length data of length T1; the first time period is the moment within the first preset time length before the current moment. Among them, the analyte concentration data Sample t is the glucose concentration data.
[0115] S2, extract at least two types of data features from the first window length data; based on the confidence and weight value of each type of data features, jointly calculate the various types of data features; if the joint calculation result exceeds the fluctuation threshold, the original analyte concentration data Sample t The data is determined to be corrected; if the combined calculation result does not exceed the fluctuation threshold, the original analyte concentration data Sample t Determined as data not to be corrected.
[0116] S3, when it is determined that the data is not to be corrected, the original analyte concentration data Sample tThe corresponding current time t retains the reset state, that is, the flag bit is 0. When it is determined to be data to be corrected, the state type of the current time t is set based on the trend characteristics of the first window length data; if the trend characteristic is a sharp upward peak, the state type of the current time t is switched from the reset state to the start of fluctuation, that is, the flag bit is +1; if the trend characteristic is a sharp downward peak, the state type of the current time t is switched from the reset state to the fluctuation recovery, that is, the flag bit is -1.
[0117] 2. Volatility Correction
[0118] S4. Determine the output analyte concentration data at the current time t based on the flag bit:
[0119] S4.1, when the flag is in reset state 0, the original analyte concentration data Sample t As output the analyte concentration data is displayed.
[0120] S4.2. When the flag is in the fluctuation start state +1, the analyte concentration data corresponding to the second period is read to obtain a length of T 2 The second window length data is the second time period before the current moment within the second preset time length. Generate the predicted data PredictedSample based on the second window length data t , predict the data PredictedSample t As the output analyte concentration data at the current time t.
[0121] S4.3. When the flag is in the fluctuation recovery state -1, the analyte concentration data corresponding to the second period is read to obtain a length of T 2 The second window length data is the second time period before the current moment within the second preset time length. Generate the predicted data PredictedSample based on the second window length data t By comparing the original analyte concentration data Sample t and predicted data PredictedSample t When the deviation between the two is less than the threshold A, the predicted value alternative is exited and the actual data value Sample is used. t Display as output analyte concentration data; otherwise, continue to use the predicted data PredictedSample t to display.
[0122] This preferred embodiment does not rely on the data stream provided by other physical sensors, but only relies on the current signal fed back by the analyte sensor in the CGM to achieve the identification and correction of fluctuations. Compared with the solution that needs to rely on additional physical sensors, this method has the advantages of small system integration volume, no additional peripheral devices, and low system power consumption.
[0123] The present embodiment is described and illustrated by means of the second preferred embodiment. When the fluctuation specifically corresponds to abnormal pressure, the data processing process is as follows:
[0124] Based on the current analyte concentration data Sample t , select the time window length as 20, and generate the first window length data. Select principal component analysis as the first feature extraction method, reorganize the first window length data after retaining 90% of the principal components, and calculate the residual energy between the original data and the reorganized data after reorganization. When its residual energy exceeds the threshold, it is judged as abnormal pressure. The characteristic markers when under pressure are as follows Figure 7 As shown. The residual energy threshold can be determined by the chi-square distribution value of historical data. When the feature is an upward spike, it is judged as the beginning of compression and the compression start flag is set; when the feature is a downward spike, it is judged as the beginning of compression and the compression stop flag is set. After the compression start is detected, the predicted data is used to replace the current data value Sample t ; When the pressure state is corrected to pressure recovery, the real-time data value Sample t and the predicted data value PredictedSample t When the deviation A is less than 20%, the real-time data value Sample is restored. t Otherwise, continue to use the predicted data value PredictedSample t .
[0125] This preferred embodiment does not rely on the data stream provided by other physical sensors, but only relies on the current signal fed back by the analyte sensor in the CGM to achieve the recognition and correction of abnormal pressure. Compared with the solution that needs to rely on additional physical sensors, this method has the advantages of small system integration volume, no additional peripheral devices, and low system power consumption.
[0126] It should be noted that the steps shown in the above process or the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0127] In this embodiment, a data correction device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. The terms "module", "unit", "subunit" and the like used below can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware is also possible and conceivable.
[0128] Figure 8is a structural block diagram of the data correction device of this embodiment, such as Figure 8 As shown, the device includes: an original data acquisition module 10, a to-be-corrected data judgment module 20, a state type marking module 30 and a data output module 40.
[0129] The raw data acquisition module 10 is used to acquire the analyte concentration data at the current moment and within a preset time length before the current moment to form first window length data;
[0130] The to-be-corrected data judging module 20 is used to judge whether the original analyte concentration data corresponding to the current moment is the to-be-corrected data based on the data characteristics of the first window length data;
[0131] The state type marking module 30 is used to set the state type at the current moment based on the trend characteristics at the current moment if the original analyte concentration data is data to be corrected;
[0132] The data output module 40 is used to determine the output analyte concentration data at the current moment based on the state type.
[0133] In one embodiment, the module 20 for determining the data to be corrected is further used to extract at least two types of data features from the first window length data; perform joint calculations on the various types of data features based on the confidence and weight values of the various types of data features; and determine whether the original analyte concentration data corresponding to the current moment is the data to be corrected based on the joint calculation result.
[0134] In one of the embodiments, the data judgment module 20 to be corrected is also used to calculate the characteristic threshold of each type of feature based on the historical data of the analyte concentration of the current user; set the confidence of the data feature based on the size relationship between the characteristic threshold and the corresponding data feature; based on the confidence of each type of data feature and the preset weight value, jointly calculate the various types of data features; the weight value corresponds to the various types of data features.
[0135] In one embodiment, the state type marking module 30 is also used to calculate the trend characteristics of the first window length data; when the trend characteristic is a sharp upward peak, the state type at the current moment is marked as the beginning of fluctuation; when the trend characteristic is a sharp downward peak, the state type at the current moment is marked as fluctuation recovery.
[0136] In one embodiment, the data output module 40 is also used to use the predicted data as the output analyte concentration data at the current moment if the state type mark is recorded as the start of fluctuation; if the state type mark is recorded as the recovery of fluctuation, calculate the difference between the original analyte concentration data and the predicted data; and determine the output analyte concentration data at the current moment based on the difference.
[0137] In one embodiment, the data output module 40 is further configured to use the original analyte concentration data as the output analyte concentration data at the current moment when the difference is less than the comparison threshold; and use the predicted data as the output analyte concentration data at the current moment when the difference is not less than the comparison threshold.
[0138] In one embodiment, the data output module 40 is further used to obtain second window length data based on the analyte concentration data corresponding to the second time period, where the second time period is a moment within a second preset time length before the current moment; and generate prediction data based on the second window length data.
[0139] In one embodiment, the data output module 40 is further configured to, if the original analyte concentration data is not the data to be corrected, then the state type at the current moment is a reset state; in the reset state, the original analyte concentration data is used as the output analyte concentration data at the current moment.
[0140] It should be noted that the above modules can be functional modules or program modules, and can be implemented by software or hardware. For modules implemented by hardware, the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.
[0141] In this embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.
[0142] Optionally, the computer device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.
[0143] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation modes, and will not be repeated in this embodiment.
[0144] In addition, in combination with the data correction method provided in the above embodiments, a storage medium may be provided in this embodiment to implement the data correction method. The storage medium stores a computer program, and when the computer program is executed by a processor, any one of the data correction methods in the above embodiments is implemented.
[0145] It should be understood that the specific embodiments described herein are only used to explain the application, rather than to limit it. Based on the embodiments provided in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the protection scope of this application.
[0146] Obviously, the drawings are only some examples or embodiments of the present application. For ordinary technicians in the field, the present application can also be applied to other similar situations based on these drawings without creative work. In addition, it is understandable that although the work done in this development process may be complicated and lengthy, for ordinary technicians in the field, certain changes in design, manufacturing or production based on the technical content disclosed in this application are only conventional technical means and should not be regarded as insufficient content disclosed in this application.
[0147] The term "embodiment" in this application refers to a specific feature, structure or characteristic described in conjunction with the embodiment that can be included in at least one embodiment of the present application. The appearance of this phrase in various locations in the specification does not necessarily mean the same embodiment, nor does it mean that it is mutually exclusive with other embodiments and is independent or optional. It is clearly or implicitly understood by those of ordinary skill in the art that the embodiments described in this application can be combined with other embodiments without conflict.
[0148] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of patent protection. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the scope of protection of the present application. Therefore, the scope of protection of the present application shall be subject to the attached claims.
Claims
1. A data correction method, characterized in that: Applied to an analyte monitoring system, the method comprises: Acquire analyte concentration data corresponding to the current moment and a first time period to form first window length data; the first time period is a moment within a first preset time length before the current moment; Based on the data characteristics of the first window length data, determining whether the original analyte concentration data corresponding to the current moment is data to be corrected; If the original analyte concentration data is data to be corrected, setting the state type at the current moment based on the trend characteristics of the first window length data; The output analyte concentration data at the current time is determined based on the state type.
2. The data correction method according to claim 1, characterized in that: The determining, based on the data feature of the first window length data, whether the original analyte concentration data corresponding to the current moment is data to be corrected includes: Extracting at least two types of data features from the first window length data; Based on the confidence and weight values of each type of data features, jointly calculate each type of data features; Based on the joint calculation result, it is determined whether the original analyte concentration data corresponding to the current moment is data to be corrected.
3. The data correction method according to claim 2, characterized in that: The combined calculation of the various data features based on the confidence and weight values of the various data features includes: Calculate feature thresholds for each type of feature based on the current user's historical analyte concentration data; Setting the confidence level of the data feature based on the magnitude relationship between the feature threshold and the corresponding data feature; Based on the confidence of each type of data feature and a preset weight value, each type of data feature is jointly calculated; the weight value corresponds to each type of data feature.
4. The data correction method according to claim 1, characterized in that: The setting of the state type at the current moment based on the trend characteristics of the first window length data includes: Calculating the trend characteristics of the first window length data; When the trend feature is a sharp upward peak, the state type at the current moment is marked as the beginning of fluctuation; When the trend feature is a downward peak, the state type at the current moment is marked as fluctuation recovery.
5. The data correction method according to claim 4, characterized in that: The step of determining the output analyte concentration data at the current moment based on the state type comprises: If the state type mark is marked as the fluctuation start, the predicted data is used as the output analyte concentration data at the current moment; If the state type mark is recorded as the fluctuation recovery, the difference between the original analyte concentration data and the predicted data is calculated; and the output analyte concentration data at the current moment is determined based on the difference.
6. The data correction method according to claim 5, characterized in that: The step of determining the output analyte concentration data at the current moment based on the difference comprises: When the difference is less than a comparison threshold, the original analyte concentration data is used as the output analyte concentration data at the current moment; When the difference is not less than the comparison threshold, the predicted data is used as the output analyte concentration data at the current moment.
7. The data correction method according to claim 5, characterized in that: Before the step of using the predicted data as the output analyte concentration data at the current moment if the state type mark is marked as the fluctuation start, the method further includes: Obtaining second window length data based on analyte concentration data corresponding to a second time period, wherein the second time period is a moment within a second preset time length before the current moment; The prediction data is generated based on the second window length data.
8. The data correction method according to claim 1, characterized in that: The step of determining the output analyte concentration data at the current moment based on the state type further includes: If the original analyte concentration data is not data to be corrected, the state type at the current moment is a reset state; In the reset state, the original analyte concentration data is used as the output analyte concentration data at the current moment.
9. A data correction device, characterized in that: The device comprises: A raw data acquisition module, used to acquire analyte concentration data at a current moment and within a preset time length before the current moment, to form first window length data; A module for determining data to be corrected, used for determining whether the original analyte concentration data corresponding to the current moment is data to be corrected based on the data characteristics of the first window length data; A state type marking module, configured to set the state type at the current moment based on the trend characteristics at the current moment if the original analyte concentration data is data to be corrected; A data output module is used to determine the output analyte concentration data at the current moment based on the state type.
10. A computer device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps of the data correction method according to any one of claims 1 to 8.
Citation Information
Cited By
Pressure recognition method, data correction method, apparatus, and computer device
WO2026026813A1
