Physiological index value calibration method, medium and system

Abstract

The disclosure describes a calibration method, medium and system for a physiological index value, the method comprising: acquiring first data of a target object, the first data comprising a physiological index value acquired by a first device, the first device not contacting subcutaneous liquid of the target object; retrieving second data of the target object, the second data comprising a physiological index value acquired by a second device, the second device contacting subcutaneous fluid of the target object; determining a first calibration model based on the first data, the first calibration model comprising a first reference baseline determined by the first data; determining a second calibration model based on the second data; updating the first calibration model with the second calibration model to re-determine the first reference baseline; and calibrating the physiological index value to be calibrated of the target object through the updated first calibration model. According to the invention, the accuracy of the physiological index value acquired by the first equipment can be improved.

Description

Technical Field

[0001] The present disclosure generally relates to the field of medical equipment, and more particularly to a method, medium and system for calibrating physiological indicator values. Background Art

[0002] Diabetes is a metabolic disease characterized by hyperglycemia, which is caused by a series of metabolic disorders in the body. Hyperglycemia refers to excessively high glucose concentrations in the blood. Long-term hyperglycemia can lead to chronic damage and dysfunction of various tissues. Measuring blood sugar can help diabetic patients develop a habit of self-management and reduce the possibility of various neurological diseases, cardiovascular and cerebrovascular diseases, microvascular diseases and complications caused by large fluctuations in blood sugar.

[0003] Current glucose monitoring systems generally include invasive glucose monitoring systems and non-invasive glucose monitoring systems. An invasive glucose monitoring system places at least a portion of the device under the skin of a target object (e.g., a diabetic patient) and contacts the subcutaneous fluid of the target object, thereby obtaining a more accurate glucose concentration value. Invasive glucose monitoring systems include continuous glucose monitoring systems that can continuously monitor changes in the glucose concentration value of the target object. Non-invasive glucose monitoring systems usually do not need to contact the subcutaneous fluid of the target object, and can indirectly obtain glucose concentration values ​​through infrared, electromagnetic, thermal capacity, ultrasound, and other methods.

[0004] However, in terms of obtaining glucose concentration values, the accuracy of non-invasive glucose monitoring systems is generally lower than that of invasive glucose monitoring systems, and errors may occur. Summary of the invention

[0005] The present disclosure is proposed in view of the above situation, and its purpose is to provide a physiological indicator value calibration method, medium and system that can improve the accuracy of the physiological indicator value obtained by the first device.

[0006] To this end, the first aspect of the present disclosure provides a method for calibrating physiological indicator values, including: acquiring first data of a target object, the first data including physiological indicator values ​​that change over time and are acquired by a first device within a first time window, and the first device does not contact the subcutaneous fluid of the target object; retrieving second data of the target object, the second data including physiological indicator values ​​that change over time and are acquired by a second device within a second time window corresponding to the first time window, the second device contacts the subcutaneous fluid of the target object, and the accuracy of the second device is higher than that of the first device; determining a first calibration model of the target object based on the first data, the first calibration model including a first reference baseline determined by the first data; determining a second calibration model of the target object based on the second data, the second calibration model including a second reference baseline determined by the second data; updating the first calibration model using information related to the second reference baseline in the second calibration model to redetermine the first reference baseline; and calibrating the physiological indicator value to be calibrated of the target object based on the redetermined first reference baseline through the updated first calibration model.

[0007] In the first aspect of the present disclosure, a reference baseline in a calibration model is determined based on the data of the target object itself. In this case, the degree of fit between the reference baseline in the calibration model and the physiological characteristics of the target object itself can be improved. At the same time, the calibration model is determined based on the first data and the second data, and when the first device is calibrated with a second device having a higher accuracy than the first device, the accuracy of the physiological indicator value obtained by the first device can be improved. In addition, after improving the accuracy of the physiological indicator value obtained by the first device, it is beneficial to reduce the frequency of use of the second device by the target object. In addition, after reducing the frequency of use of the second device by the target object, the comfort of the target object in monitoring the physiological indicator value can be improved for the situation where the first device does not contact the subcutaneous fluid of the target object and the second device contacts the subcutaneous fluid of the target object. In addition, calibrating the target object's to-be-calibrated physiological indicator value based on the calibration model determined by the physiological indicator value data that changes over time can improve the adaptability of the calibration model to the changes in the physiological condition of the target object, thereby further improving the accuracy of the physiological indicator value obtained by the first device.

[0008] In addition, in the calibration method involved in the first aspect of the present disclosure, optionally, the physiological indicator value is a glucose concentration value, the first device is a non-invasive glucose monitoring system, and the second device is a continuous glucose monitoring system. In this case, compared with the fingertip blood glucose monitoring system, a larger amount of second data can be obtained through the continuous glucose monitoring system, so that when the glucose concentration value of the first device is calibrated using the second calibration model determined by the second data, the accuracy of the physiological indicator value obtained by the first device can be improved. In addition, using a continuous glucose monitoring system to calibrate a non-invasive glucose monitoring system can improve the accuracy of the glucose concentration value obtained by the non-invasive glucose monitoring system, which can help reduce the frequency of use of the continuous glucose monitoring system by the target object, thereby improving the comfort of the target object in monitoring the glucose concentration value. In addition, when the cost of using a continuous glucose monitoring system is higher than the cost of using a non-invasive glucose monitoring system, the cost of monitoring the glucose concentration value of the target object can also be reduced.

[0009] In addition, in the calibration method involved in the first aspect of the present disclosure, optionally, the collection time period of the first data is equal to the collection time period of the second data; or the collection time period of the first data is before the collection time period of the second data; or the collection time period of the first data partially overlaps with the collection time period of the second data. In this case, when the collection time period of the first data is equal to the collection time period of the second data, there can be physiological indicator values ​​in the second data that are close to the time corresponding to the first data, so that when calibration is performed based on the physiological indicator values ​​at similar times, the degree of fit between the physiological indicator values ​​to be calibrated and the actual physiological indicator values ​​of the target object can be improved, thereby improving the accuracy of the physiological indicator values ​​acquired by the first device. In addition, when the collection time period of the first data is before the collection time period of the second data, it can be beneficial to reduce the frequency of use of the second device by the target object during the collection time period of the first data. In addition, when the collection time period of the first data partially overlaps with the collection time period of the second data, there can be a portion of physiological indicator values ​​close to the time corresponding to the first data in the second data, so that when calibration is performed based on a portion of physiological indicator values ​​at similar times, the degree of fit between the physiological indicator values ​​to be calibrated and the actual physiological indicator values ​​of the target object can be improved, thereby improving the accuracy of the physiological indicator values ​​acquired by the first device, and at the same time, it can also help to reduce the frequency of use of the second device by the target object.

[0010] In addition, in the calibration method involved in the first aspect of the present disclosure, optionally, the second reference baseline is determined based on the physiological index value in at least one reference cycle that meets the preset conditions in the second data; or the waveform characteristics and / or feature classification corresponding to the second data are determined based on the second data, and the second reference baseline is determined based on the waveform characteristics and / or feature classification corresponding to the second data; or the waveform characteristics corresponding to the second data are determined based on the second data, and the feature classification corresponding to the second data is determined based on the waveform characteristics corresponding to the second data, and the second reference baseline is determined based on the feature classification corresponding to the second data. In this case, by using the data in at least one reference cycle that meets the preset conditions, the error caused by a single data value can be reduced, thereby improving the accuracy of the second reference baseline. In addition, the second reference baseline is determined based on the waveform characteristics corresponding to the second data, and by analyzing the waveform characteristics corresponding to the second data, it is helpful to further understand the relationship between the second reference baseline and the change characteristics of the physiological index value of the target object, and the second reference baseline can be obtained more intuitively, and the interpretability of the second reference baseline can be improved. In addition, the second reference baseline is determined based on the feature classification corresponding to the second data, and by analyzing the feature classification corresponding to the second data, the adaptability of the calibration method to various types of data can be improved, and the robustness of the second calibration model can be improved. In addition, by determining the second reference baseline based on the waveform features and feature classification corresponding to the second data, the second reference baseline can be obtained more intuitively, the interpretability of the second reference baseline can be improved, and the adaptability of the calibration method to various types of data can be improved. In addition, by determining the feature classification corresponding to the second data based on the waveform features corresponding to the second data, and determining the second reference baseline based on the feature classification corresponding to the second data, the data can be classified into more intuitive information, which can further improve the interpretability of the second reference baseline.

[0011] In addition, in the calibration method involved in the first aspect of the present disclosure, optionally, the information related to the second reference baseline is the second reference baseline itself, and the re-determination includes: determining a correction judgment value and a correction amplitude value based on the offset between the previous first reference baseline and the second reference baseline, the correction judgment value is used to determine whether the previous first reference baseline needs to be corrected, and the correction amplitude value is used to characterize the correction amplitude of the previous first reference baseline; and correcting the previous first reference baseline based on the correction judgment value and the correction amplitude value. In this case, when correcting the previous first reference baseline, the first reference baseline and the second reference baseline can be taken into account at the same time, which can reduce the risk of the corrected first reference baseline deviating too much from the previous first reference baseline, thereby improving the fit between the first reference baseline and the first device. In addition, the fit between the first reference baseline and the actual physiological condition of the target object can also be improved.

[0012] In addition, in the calibration method involved in the first aspect of the present disclosure, optionally, the first reference baseline is determined based on the waveform features and / or feature classifications corresponding to the first data, and the information related to the second reference baseline includes the waveform features and / or feature classifications corresponding to the second data; the re-determination is to update the waveform features and / or feature classifications corresponding to the first data using the information related to the second reference baseline so that the first calibration model updates the first reference baseline using the updated waveform features and / or feature classifications corresponding to the first data. In this case, the first calibration model updates the first reference baseline using the updated waveform features and / or feature classifications corresponding to the first data, which can improve the fit between the first reference baseline and the first device.

[0013] In addition, in the calibration method involved in the first aspect of the present disclosure, optionally, the waveform characteristics corresponding to the second data include at least one of a peak value, a valley value, a rising rate, a falling rate, a rising amplitude, a falling amplitude, a fluctuation amplitude, a plateau period and a curve trend.

[0014] In addition, in the calibration method involved in the first aspect of the present disclosure, optionally, the feature classification corresponding to the second data includes physiological index value too high, physiological index value high, physiological index value normal, physiological index value low and physiological index value too low. In this case, the meticulousness of classifying the second data can be improved, so that determining the reference baseline based on finer-grained feature classification can improve the accuracy of the reference baseline.

[0015] In addition, in the calibration method involved in the first aspect of the present disclosure, optionally, calibrating the target object's to-be-calibrated physiological indicator value based on the re-determined first reference baseline includes: obtaining the first actual baseline corresponding to the first device and the preset time period, and the preset time period is related to the acquisition time of the physiological indicator value to be calibrated; determining the adjustment parameter of the physiological indicator value to be calibrated based on the offset between the first actual baseline and the updated first reference baseline; and adjusting the physiological indicator value to be calibrated based on the adjustment parameter. In this case, obtaining the first actual baseline corresponding to the first device and the preset time period, and the preset time period is related to the acquisition time of the physiological indicator value to be calibrated, can improve the fit between the calibrated physiological indicator value and the actual physiological indicator value of the target object, thereby improving the accuracy of the physiological indicator value acquired by the first device. In addition, adjusting the physiological indicator value to be calibrated based on the adjustment parameter can take into account the first actual baseline and the updated first reference baseline at the same time, can reduce the risk of the adjusted physiological indicator value deviating too much from the first actual baseline, thereby improving the fit between the adjusted physiological indicator value and the physiological indicator value of the target object acquired by the first device.

[0016] A second aspect of the present disclosure provides a computer-readable storage medium, wherein the computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the calibration method involved in the first aspect of the present disclosure is implemented.

[0017] The third aspect of the present disclosure provides a physiological indicator monitoring system, which does not contact the subcutaneous fluid of the target object, and includes a monitoring sensor and a processing device. The monitoring sensor is configured to generate a sensor signal, and the processing device is configured to receive a physiological indicator value related to the sensor signal and calibrate the physiological indicator value using the method involved in the first aspect of the present disclosure.

[0018] According to the present disclosure, a physiological indicator value calibration method, medium and system are provided, which can improve the accuracy of the physiological indicator value obtained by the first device. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The present disclosure will now be explained in further detail, by way of example only, with reference to the accompanying drawings.

[0020] Figure 1 It is a schematic diagram showing an application scenario of the calibration method involved in the example of the present disclosure.

[0021] Figure 2 is a flow chart showing a calibration method involved in examples of the present disclosure.

[0022] Figure 3 1 is a flowchart showing a first implementation method of redetermining a first reference baseline according to an example of the present disclosure.

[0023] Figure 4 1 is a flowchart showing a second embodiment of redetermining the first reference baseline according to the example of the present disclosure.

[0024] Figure 5 1 is a flowchart showing a third embodiment of redetermining the first reference baseline according to the example of the present disclosure.

[0025] Figure 6 1 is a flowchart showing a fourth embodiment of redetermining the first reference baseline according to the example of the present disclosure.

[0026] Figure 7 is a flow chart showing the calibration of the physiological indicator value to be calibrated of the target object based on the re-determined first reference baseline involved in the example of the present disclosure.

[0027] Figure 8 This is a flowchart showing a first embodiment of a calibration method according to the present disclosure example.

[0028] Fig. 9This is a flowchart showing a second embodiment of the calibration method according to the present disclosure example.

[0029] Fig.10 is a structural block diagram showing a physiological index monitoring system involved in the example of the present disclosure. DETAILED DESCRIPTION

[0030] Hereinafter, the preferred embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In the following description, the same symbols are assigned to the same components, and repeated descriptions are omitted. In addition, the accompanying drawings are only schematic diagrams, and the ratio of the dimensions of the components or the shapes of the components may be different from the actual ones.

[0031] It should be noted that the terms "including" and "having" and any variations thereof in the present disclosure, such as a process, method, system, product or device that includes or has a series of steps or units, are not necessarily limited to those steps or units clearly listed, but may include or have other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0032] The present disclosure provides a calibration method for physiological index values ​​(hereinafter referred to as calibration method or method), which is a method for calibrating a physiological index monitoring system. The calibration method provided by the present disclosure can improve the accuracy of the physiological index values ​​obtained by the physiological index monitoring system.

[0033] The calibration method of the physiological indicator value involved in the present disclosure may also be referred to as a verification method, a correction method, a determination method or a standardization method, etc. The calibration method involved in the present disclosure may be applicable to any application scenario where the physiological indicator monitoring system needs to be calibrated.

[0034] The physiological index value involved in the present disclosure can represent the physiological index level of the target object. In some examples, the physiological index value can include blood ketone concentration, acetylcholine concentration, amylase concentration, bilirubin concentration, cholesterol concentration, chorionic gonadotropin concentration, creatine kinase concentration, creatine concentration, creatinine concentration, DNA concentration, fructosamine concentration, glucose concentration, glutamine concentration, growth hormone concentration, hormone concentration, ketone body concentration, lactate concentration, peroxide concentration, prostate specific antigen concentration, prothrombin concentration, RNA concentration, thyroid stimulating hormone concentration, troponin concentration, blood pressure, heart rate, body temperature, respiratory rate, blood oxygen saturation and body weight. Thus, the present disclosure can be applied to the calibration of multiple physiological index values.

[0035] The physiological indicator values ​​involved in the present disclosure may come from a physiological indicator monitoring system. In some examples, the physiological indicator monitoring system is a system that can monitor the physiological indicator values ​​of the human body. In some examples, the physiological indicator monitoring system can continuously or intermittently monitor the physiological indicator values ​​of the human body, and record the data of the physiological indicator values ​​for later analysis. In some examples, the physiological indicator monitoring system may be a wearable smart device. In some examples, the physiological indicator monitoring system may include a glucose monitoring system, a blood pressure monitor, a thermometer, a weight scale, and a blood oxygen monitor, etc. As a result, the present disclosure can be applied to a variety of physiological indicator monitoring systems.

[0036] The interpretability involved in the examples disclosed herein means that the target object 30 can understand the calculation process of the corresponding parameters. Taking the determination of the second reference baseline based on the waveform features corresponding to the second data as an example, the target object 30 can know what waveform features the second reference baseline is calculated based on, and can know what waveform features correspond to what second reference baseline.

[0037] Hereinafter, the calibration method involved in the present disclosure will be described in detail with reference to the accompanying drawings.

[0038] Figure 1 It is a schematic diagram showing an application scenario of the calibration method involved in the example of the present disclosure.

[0039] In some examples, the calibration method disclosed herein can be applied to Figure 1 In the application scenario shown.

[0040] See also Figure 1 In some examples, the first device 10 and the second device 20 may be a physiological indicator monitoring system 100 (described later) for monitoring physiological indicator values. In some examples, the target object 30 may use the first device 10 and the second device 20 to obtain its own physiological indicator values. In some examples, the target object 30 may use the first device 10 and the second device 20 to obtain physiological indicator values ​​at the same time.

[0041] It should be noted that Figure 1 It is only a schematic diagram, and the application scenarios of the present disclosure are not limited thereto. In some examples, the target object 30 may not use the first device 10 and the second device 20 at the same time. For example, only the first device 10 may be used to obtain the physiological indicator value within a period of time. In some examples, only the second device 20 may be used to obtain the physiological indicator value within a period of time.

[0042] In some examples, when the physiological indicator value is a glucose concentration value, the first device 10 and the second device 20 may be a glucose monitoring system.

[0043] In some examples, the accuracy of the second device 20 may be higher than that of the first device 10. That is, the accuracy of the physiological indicator value obtained by the second device 20 may be higher than the accuracy of the physiological indicator value obtained by the first device 10. In some examples, the target object 30 may use the second device 20 to calibrate the first device 10. In this case, when the first device 10 is calibrated using the second device 20, the accuracy of the physiological indicator value obtained by the first device 10 can be improved, which can help reduce the frequency of use of the second device 20 by the target object 30.

[0044] However, the present disclosure is not limited thereto, and the calibration method involved in the present disclosure can be applied to any scenario where a high-precision physiological indicator monitoring system 100 is required to calibrate a low-precision physiological indicator monitoring system 100. In some examples, the calibration method involved in the present disclosure can be applied to multiple physiological indicator monitoring systems 100, such as 3, 4, 5, or 6. In some examples, a high-precision physiological indicator monitoring system 100 can be used to calibrate multiple low-precision physiological indicator monitoring systems 100. For example, one high-precision glucose monitoring system can be used to calibrate two low-precision glucose monitoring systems.

[0045] Figure 2 is a flow chart showing a calibration method involved in examples of the present disclosure.

[0046] See also Figure 2 In some examples, the calibration method may include acquiring first data of the target object 30 (step S100), retrieving second data of the target object 30 (step S200), determining a first calibration model of the target object 30 based on the first data (step S300), determining a second calibration model of the target object 30 based on the second data (step S400), updating the first calibration model using the second calibration model (step S500), and calibrating the physiological indicator value to be calibrated of the target object 30 by using the updated first calibration model (step S600). In this case, the degree of fit between the calibration model and the physiological characteristics of the target object 30 itself can be improved.

[0047] See also Figure 2 In some examples, in step S100, first data of the target object 30 may be acquired. In some examples, the first data may be acquired by the first device 10.

[0048] In some examples, the non-invasive physiological indicator monitoring system may not contact the subcutaneous fluid of the target object 30. In some examples, the physiological indicator monitoring system 100 that does not contact the subcutaneous fluid of the target object 30 may also be referred to as a non-invasive physiological indicator monitoring system.

[0049] In some examples, the first device 10 may be a physiological indicator monitoring system 100. In some examples, the first device 10 may not contact the subcutaneous fluid of the target object 30. In other words, the first device 10 may be a non-invasive physiological indicator monitoring system. Thus, the convenience of the target object 30 in obtaining the physiological indicator value can be improved.

[0050] In some examples, when the physiological indicator value is a glucose concentration value, the first device 10 may be a glucose monitoring system. In some examples, the first device 10 may be a non-invasive glucose monitoring system.

[0051] In some examples, the first data may include physiological indicator values ​​that vary over time and are acquired by the first device 10 within a first time window. In some examples, the first time window may be used to represent the acquisition time of the physiological indicator values ​​in the first data. For example, the first time window may be from 11:00 to 13:00 every day. For another example, the first time window may be from 8:00 to 9:00 every day.

[0052] Continue to see Figure 2 In some examples, in step S200, second data of the target object 30 may be retrieved. In some examples, the second data may be acquired by the second device 20.

[0053] In some examples, the invasive physiological indicator monitoring system can contact the subcutaneous fluid of the target object 30. In some examples, at least a portion of the invasive physiological indicator monitoring system can be subcutaneous to the target object 30. In some examples, the physiological indicator monitoring system 100 at least a portion of which can be subcutaneous to the target object 30 can also be referred to as an invasive physiological indicator monitoring system.

[0054] In some examples, the second device 20 may be a physiological indicator monitoring system 100. In some examples, the second device 20 may contact the subcutaneous fluid of the target object 30. In some examples, at least a portion of the second device 20 may be under the skin of the target object 30. In other words, the second device 20 may be an invasive physiological indicator monitoring system. As described above, the accuracy of the second device 20 may be higher than that of the first device 10. Thus, the accuracy of the physiological indicator value obtained by the target object 30 can be improved.

[0055] In some examples, at least a portion of the second device 20 can be placed in the subcutaneous fluid of the target object 30 obtained by the collection device. For example, the collection device can be a venous blood collection device. That is, the second device 20 can contact the venous blood of the target object 30.

[0056] In some examples, when the physiological indicator value is a glucose concentration value, the second device 20 may be a glucose monitoring system. In some examples, the second device 20 may be a continuous glucose monitoring system. In this case, compared with the fingertip blood glucose monitoring system, the continuous glucose monitoring system can obtain more second data, so that when the glucose concentration value of the first device 10 is calibrated using the second calibration model determined by the second data, the accuracy of the physiological indicator value obtained by the first device 10 can be improved.

[0057] In some examples, when the physiological indicator value is a glucose concentration value, the first device 10 can be a non-invasive glucose monitoring system, and the second device 20 can be an invasive glucose monitoring system. An invasive glucose monitoring system generally requires puncturing the skin to monitor the glucose concentration value, which often causes pain to the target object 30 and may also bring the risk of infection. In this case, after improving the accuracy of the physiological indicator value obtained by the first device 10, it can be helpful to reduce the frequency of use of the second device 20 by the target object 30, improve the comfort of the target object 30 in monitoring the physiological indicator value, and reduce the risk of infection of the target object 30.

[0058] In some examples, when the physiological indicator value is a glucose concentration value, the first device 10 may be a non-invasive glucose monitoring system, and the second device 20 may be a continuous glucose monitoring system. In this case, compared with the fingertip blood glucose monitoring system, a larger amount of second data can be obtained through the continuous glucose monitoring system, so that when the glucose concentration value of the first device 10 is calibrated using the second calibration model determined by the second data, the accuracy of the physiological indicator value obtained by the first device 10 can be improved. In addition, using a continuous glucose monitoring system to calibrate a non-invasive glucose monitoring system can improve the accuracy of the glucose concentration value obtained by the non-invasive glucose monitoring system, which can help reduce the frequency of use of the continuous glucose monitoring system by the target object 30, thereby improving the comfort of the target object 30 in monitoring the glucose concentration value. In addition, when the cost of using a continuous glucose monitoring system is higher than the cost of using a non-invasive glucose monitoring system, the cost of monitoring the glucose concentration value of the target object 30 can also be reduced.

[0059] In some examples, the second data may include physiological indicator values ​​that vary over time and are acquired by the second device 20 within a second time window. In some examples, the second time window may be used to represent the acquisition time of the physiological indicator values ​​in the second data. For example, the second time window may be from 11:00 to 13:00 every day. For another example, the second time window may be from 8:00 to 9:00 every day.

[0060] In some examples, the second time window may correspond to the first time window. For example, the first time window may be from 11:00 to 13:00 every day, and the second time window may be from 11:00 to 13:00 every day. In some examples, the first time window may be from 8:00 to 9:00 every day, and the second time window may be from 8:00 to 9:00 every day. In this case, by making the second time window correspond to the first time window, the second data can be correlated with the first data, so that the periodic characteristics of the physiological indicator value of the target object 30 itself changing with time can be obtained, and then calibration based on the first data and the second data can improve the accuracy of the physiological indicator value obtained by the first device 10.

[0061] In some examples, the first time window may be at least one, such as 1, 2, 3, 4, or 5, etc. In some examples, the second time window may be at least one, such as 1, 2, 3, 4, or 5, etc. In some examples, multiple second time windows may correspond one-to-one to multiple first time windows.

[0062] In some examples, the collection time period of the first data may be equal to the collection time period of the second data. For example, the first data and the second data may be collected simultaneously from January 1 to January 20, 2023. In this case, the second data may have physiological index values ​​close to the time corresponding to the first data, so that when calibrating based on the physiological index values ​​at similar times, the degree of fit between the physiological index values ​​to be calibrated and the actual physiological index values ​​of the target object 30 can be improved, thereby improving the accuracy of the physiological index values ​​acquired by the first device 10.

[0063] In some examples, the collection time period of the first data may be before the collection time period of the second data. For example, the first data may be collected from January 1 to January 10, 2023, and the second data may be collected from January 11 to January 20, 2023. In this case, it is possible to reduce the frequency of use of the second device 20 by the target object 30 during the collection time period of the first data.

[0064] In some examples, the collection time period of the first data may partially overlap with the collection time period of the second data. For example, the first data may be collected from January 1 to January 20, 2023, and the second data may be collected from January 10 to January 30, 2023. In this case, there may be a portion of physiological indicator values ​​in the second data that are close to the time corresponding to the first data, so that when calibration is performed based on a portion of physiological indicator values ​​at similar times, the degree of fit between the physiological indicator values ​​to be calibrated and the actual physiological indicator values ​​of the target object 30 can be improved, thereby improving the accuracy of the physiological indicator values ​​obtained by the first device 10, and at the same time, it can also help reduce the frequency of use of the second device 20 by the target object 30.

[0065] In some examples, the calibration model corresponding to the physiological indicator monitoring system 100 can be determined based on the physiological indicator value data of the target object 30 itself. In some examples, the calibration model can calibrate the physiological indicator monitoring system 100 using the physiological indicator value data obtained by the target object 30 through the physiological indicator monitoring system 100.

[0066] In some examples, the calibration model can determine the baseline of the physiological indicator monitoring system 100, and calibrate the physiological indicator values ​​obtained by the physiological indicator monitoring system 100 by adjusting the baseline of the physiological indicator monitoring system 100. In some examples, the calibration model can determine the characteristics of the target object 30, and calibrate the physiological indicator values ​​obtained by the physiological indicator monitoring system 100 by adjusting the characteristics.

[0067] In some examples, the calibration model may include a reference baseline corresponding to the physiological indicator monitoring system 100. In some examples, the reference baseline may represent a reference baseline value of the physiological indicator value acquired by the physiological indicator monitoring system 100. In some examples, the reference baseline may be a numerical range of the physiological indicator value. In some examples, the reference baseline may be determined based on data of the target object 30 (e.g., data of physiological indicator values ​​varying over time). In some examples, the reference baseline may vary with the characteristics of the physiological indicator values ​​of the target object 30 varying over time.

[0068] In some examples, in different calibration models, the algorithms for determining the reference baseline may be different. For example, the principles or network architectures of the algorithms may be different. Thus, the adaptability of the calibration method to different physiological indicator monitoring systems 100 can be improved.

[0069] In some examples, the algorithm for determining the reference baseline may be the same in different calibration models, in which case the consistency and comparability of the reference baseline can be improved, thereby improving the reliability of the calibration.

[0070] In some examples, the algorithm for determining the reference baseline may include an artificial intelligence algorithm. In some examples, the artificial intelligence algorithm may include a decision tree algorithm, a support vector machine algorithm, a neural network algorithm, a random forest algorithm, and the like.

[0071] In some examples, the calibration model may include an actual baseline corresponding to the physiological indicator monitoring system 100. In some examples, the actual baseline may represent an actual reference value of the physiological indicator value acquired by the physiological indicator monitoring system 100. In some examples, the actual baseline may be a numerical range of the physiological indicator value. In some examples, the actual baseline may be determined based on data of the target object 30 (e.g., data of physiological indicator values ​​changing over time). In some examples, the actual baseline may vary with the characteristics of the physiological indicator values ​​of the target object 30 changing over time.

[0072] In some examples, the algorithm for determining the actual baseline may include an artificial intelligence algorithm. In some examples, the calibration model may calibrate the physiological indicator monitoring system 100 using the updated reference baseline and the actual baseline of the physiological indicator monitoring system 100 .

[0073] In some examples, the database may be established based on the data of the target object 30 itself. In some examples, the database may be established based on the data of multiple target objects 30. In some examples, the calibration model may calibrate the physiological indicator monitoring system 100 using the database.

[0074] In some examples, the calibration model may include a conversion module. In some examples, the conversion module may convert a physical signal related to a physiological indicator value acquired by a monitoring sensor of the physiological indicator monitoring system 100 into a corresponding physiological indicator value.

[0075] In some examples, the conversion module may use an artificial intelligence algorithm. In some examples, the physical signal may include at least one of a magnetic signal, an optical signal, and an electrical signal. In some examples, the conversion module may associate the physiological indicator value with the acquisition time of the physiological indicator value. Thus, the physiological indicator value monitored by the physiological indicator monitoring system 100 can be obtained.

[0076] Continue to see Figure 2 In some examples, in step S300, a first calibration model of the target object 30 can be determined based on the first data. For example, the first calibration model of the target object 30 can be determined based on the first data using an artificial intelligence algorithm. In some examples, the first calibration model can be configured to calibrate the physiological indicator values ​​acquired by the first device 10. In this case, determining the first calibration model based on the data of the target object 30 itself can improve the pertinence to the target object 30. In addition, calibrating the physiological indicator values ​​to be calibrated of the target object 30 based on the calibration model determined based on the physiological indicator value data that changes over time can improve the adaptability of the calibration model to the changes in the physiological condition of the target object 30, thereby further improving the accuracy of the physiological indicator values ​​acquired by the first device 10.

[0077] In some examples, the first calibration model may include a first reference baseline, which may be determined by the first data. In some examples, the first reference baseline may represent a reference baseline value of a physiological indicator value acquired by the first device 10. In some examples, the first reference baseline may change as the physiological indicator value of the target object 30 changes over time.

[0078] In some examples, the first calibration model may include a first actual baseline, which may be determined by the first data. In some examples, the first actual baseline may represent an actual reference value of the physiological indicator value acquired by the first device 10 .

[0079] In some examples, the method for determining the first reference baseline may be the same as the method for determining the second reference baseline (described later). In this case, relative to the case where the methods for determining the reference baseline are different, the first reference baseline and the second reference baseline can be made consistent and comparable, and the accuracy of the physiological indicator value obtained by the first device 10 can be improved.

[0080] In some examples, the method for determining the first actual baseline may be different from the method for determining the first reference baseline. In this case, the adaptability to different application scenarios can be improved.

[0081] Continue to see Figure 2 In some examples, in step S400, a second calibration model of the target object 30 can be determined based on the second data. For example, the second calibration model of the target object 30 can be determined based on the second data using an artificial intelligence algorithm. In some examples, the second calibration model can be configured to calibrate the physiological indicator values ​​acquired by the second device 20. In this case, the second calibration model is determined based on the data of the target object 30 itself. When the first device 10 is calibrated using the second device 20 with higher accuracy than the first device 10, the targeting of the target object 30 can be improved, and the accuracy of the physiological indicator values ​​acquired by the first device 10 can be improved. In addition, calibration based on the physiological indicator value data that changes over time can improve the adaptability to changes in the physiological condition of the target object 30, thereby further improving the accuracy of the physiological indicator values ​​acquired by the first device 10.

[0082] In some examples, the second calibration model may include a second reference baseline, which may be determined by the second data. In some examples, the second reference baseline may represent a reference baseline value of a physiological indicator value acquired by the second device 20. In some examples, the second reference baseline may change as the physiological indicator value of the target object 30 changes over time.

[0083] In some examples, the second reference baseline can be determined based on the physiological indicator value within at least one reference cycle that meets the preset conditions in the second data. In this case, by using the data within at least one reference cycle that meets the preset conditions, the error caused by a single data value can be reduced, thereby improving the accuracy of the second reference baseline. In addition, determining the second reference baseline using the physiological indicator value within the reference cycle can improve the convenience of calculating the second reference baseline.

[0084] In some examples, the reference period may refer to a time period to which the physiological indicator value used to determine the baseline belongs. In some examples, the reference period may be 1, 2, 3, 4, or 5, etc.

[0085] In some examples, the reference period can be determined based on the characteristics of the physiological index value of the target object 30 changing over time. For example, when the physiological index value is a glucose concentration value, the reference period can correspond to monitoring the fasting glucose concentration value. In some examples, the reference period can include 6:00 to 7:00 and 20:00 to 21:00 every day. In this case, by determining the second reference baseline based on the characteristics of the physiological index value of the target object 30 changing over time, the accuracy of the second reference baseline can be improved.

[0086] In some examples, the preset condition may be any condition capable of determining the second reference baseline. In some examples, the preset condition may be a feature of the second data. For example, the preset condition may include a reference period, fluctuation amplitude, waveform feature, and feature classification. In some examples, the preset condition may be used to determine the reference period.

[0087] In some examples, the preset condition may include a preset amplitude. In some examples, the difference between the highest value and the lowest value of the physiological indicator value in the reference cycle in the second data may be calculated, and in response to the difference being less than the preset amplitude, the second reference baseline is determined based on the second data. In some examples, the preset amplitude may be an empirical value. In some examples, the empirical value may refer to a value preset for a parameter based on previous experience or experimental results.

[0088] In some examples, the calculated data within the reference period may be processed to obtain the second reference baseline. For example, the processing may include taking a statistical value, such as an average, a median, or a mode. For example, the average value of the calculated data may be taken as the value of the second reference baseline.

[0089] In some examples, the processing may include using an artificial intelligence algorithm to perform big data analysis using the calculated data within the reference period and an existing database to obtain a second reference baseline. In some examples, the artificial intelligence algorithm may include a decision tree algorithm, a support vector machine algorithm, a neural network algorithm, and a random forest algorithm. In some examples, the processing may include taking an empirical value. Thus, the second reference baseline can be determined based on the second data.

[0090] In some examples, the preset condition may include at least one of a waveform feature and a feature classification. In this case, by extracting features related to the target object 30 in the second data, the accuracy of the second reference baseline can be further improved.

[0091] In some examples, the preset condition may include a waveform feature. In some examples, the waveform feature may refer to a feature of a physiological indicator value that changes over time.

[0092] In some examples, the waveform feature corresponding to the second data can be determined based on the second data. In some examples, the waveform feature can include at least one of a peak value, a valley value, a rising rate, a falling rate, a rising amplitude, a falling amplitude, a fluctuation amplitude, a plateau, and a curve trend. In some examples, the waveform feature can include other features.

[0093] In some examples, the peak value may be the highest value of the second data in any time period. In some examples, the valley value may be the lowest value of the second data in any time period. In some examples, the rising rate may be the rising rate of the value of the second data in any time period. In some examples, the falling rate may be the falling rate of the value of the second data in any time period. In some examples, the rising amplitude may be the rising value of the value of the second data in any time period, such as the difference between the final value and the starting value. In some examples, the falling amplitude may be the falling value of the value of the second data in any time period, such as the difference between the starting value and the final value.

[0094] In some examples, the fluctuation amplitude may be the difference between the peak value and the valley value of the second data in any time period. In some examples, the stable period may be a time period in which the fluctuation amplitude is less than a preset amplitude. In some examples, the curve trend may refer to a change pattern of the physiological indicator value over time. In some examples, the curve trend may be obtained by mathematical methods such as exponential smoothing, linear regression or difference method.

[0095] In some examples, the second reference baseline can be determined based on the waveform features corresponding to the second data. In this case, by analyzing the waveform features corresponding to the second data, it is possible to further understand the relationship between the second reference baseline and the characteristics of changes in the physiological index values ​​of the target object 30, to obtain the second reference baseline more intuitively, and to improve the interpretability of the second reference baseline. In other words, it is possible to help the target object 30 understand the calculation process of the second reference baseline.

[0096] In some examples, the second reference baseline can be determined based on the physiological indicator value in at least one reference period in which the waveform characteristics of the second data meet the preset conditions. For example, a time period in the stable period that further meets the rising rate and the falling rate less than the preset rate can be selected as the reference period. For example, the preset rate can be 0.1 mmol / L / min.

[0097] In some examples, the physiological indicator values ​​in the second data within the reference period may be selected as calculation data, and the calculation data may be processed to obtain a second reference baseline. The method may refer to the relevant description in step S400.

[0098] In some examples, the waveform features corresponding to the second data may have a corresponding relationship with the second reference baseline. In some examples, the corresponding relationship may be a one-to-one correspondence. In some examples, the corresponding relationship may be predetermined. In some examples, the corresponding relationship may be a statistical value, such as an empirical value. In some examples, an artificial intelligence algorithm may be used to perform big data analysis based on the waveform features corresponding to the second data and an existing database to determine the second reference baseline. In some examples, the artificial intelligence algorithm may include a decision tree algorithm, a support vector machine algorithm, a neural network algorithm, a random forest algorithm, and the like.

[0099] In some examples, the preset condition may include a feature classification. In some examples, the feature classification may refer to a classification of features that match the target object 30 .

[0100] In some examples, the feature classification corresponding to the second data can be determined based on the second data. In some examples, the degree of matching between the second data and the feature classification of physiological indicator values ​​corresponding to different types of target objects 30 in an existing database can be determined. In some examples, the feature classification corresponding to the second data can be determined based on the degree of matching.

[0101] In some examples, the feature classification may include physiological indicator value too high, physiological indicator value high, physiological indicator value normal, physiological indicator value low, and physiological indicator value too low. In this case, the fineness of classifying the second data can be improved, so that determining the reference baseline based on finer-grained feature classification can improve the accuracy of the reference baseline.

[0102] In some examples, feature classifications may include other classifications.

[0103] In some examples, the physiological indicator value is too high and the physiological indicator value is too high can correspond to different degrees of hyperglycemia. In some examples, the physiological indicator value is too high can be a glucose concentration value greater than 7.5 mmol / L, and the physiological indicator value is too high can be a glucose concentration value between 7 mmol / L and 7.5 mmol / L.

[0104] In some examples, low physiological indicator values ​​and too low physiological indicator values ​​may correspond to different degrees of hypoglycemia events. In some examples, too low physiological indicator values ​​may be a glucose concentration value less than 2.8 mmol / L, and low physiological indicator values ​​may be a glucose concentration value between 2.8 mmol / L and 3 mmol / L.

[0105] In some examples, the feature classification may include the diabetes type of the target object 30. For example, type 1 diabetes, type 2 diabetes, gestational diabetes, and other types of diabetes. In some examples, the degree of matching between the second data and the diabetes types corresponding to different types of target objects 30 in an existing database may be determined. In some examples, the feature classification corresponding to the second data may be determined based on the degree of matching. Thus, the accuracy of calibration can be improved.

[0106] In some examples, the second reference baseline may be determined based on the feature classification corresponding to the second data. In this case, by analyzing the feature classification corresponding to the second data, the adaptability of the calibration method to various types of data can be improved, and the robustness of the second calibration model can be improved.

[0107] In some examples, the feature classification corresponding to the second data may have a corresponding relationship with the second reference baseline. In some examples, the corresponding relationship may be a one-to-one correspondence. In some examples, the corresponding relationship may be predetermined. In some examples, the corresponding relationship may be a statistical value, such as an empirical value. In some examples, an artificial intelligence algorithm may be used to perform big data analysis based on the feature classification corresponding to the second data and an existing database to determine the second reference baseline.

[0108] In some examples, the preset conditions may include waveform features and feature classifications.

[0109] In some examples, the waveform features and feature classification corresponding to the second data can be determined based on the second data. In some examples, the second reference baseline can be determined based on the waveform features and feature classification corresponding to the second data. In this case, the second reference baseline can be obtained more intuitively, the interpretability of the second reference baseline can be improved, and the adaptability of the calibration method to various types of data can also be improved.

[0110] In some examples, the waveform features and feature classifications corresponding to the second data may have a corresponding relationship with the second reference baseline. In some examples, the corresponding relationship may be a one-to-one correspondence. In some examples, the corresponding relationship may be predetermined. In some examples, the corresponding relationship may be a statistical value, such as an empirical value. In some examples, an artificial intelligence algorithm may be used to perform big data analysis based on the waveform features and feature classifications corresponding to the second data and an existing database to determine the second reference baseline.

[0111] In some examples, the waveform features corresponding to the second data can be determined based on the second data, and the feature classification corresponding to the second data can be determined based on the waveform features corresponding to the second data. The method for determining the waveform features corresponding to the second data is as described above. In this case, the data can be classified as more intuitive information, which can further improve the interpretability of the second reference baseline. As a result, the target object 30 can improve its understanding of the information of the second data.

[0112] In some examples, the waveform features and the feature classifications may have a corresponding relationship. In some examples, the corresponding relationship may be a one-to-one correspondence. In some examples, the corresponding relationship may be predetermined. For example, the feature classification corresponding to the second data may be determined based on the peak and valley values ​​in the waveform features. For example, when the peak value in the plateau is greater than 7.5 mmol / L and the valley value is greater than 7 mmol / L, it may be determined that the feature classification corresponding to the second data is a physiological indicator value that is too high. In some examples, an artificial intelligence algorithm may be used to perform big data analysis based on the waveform features corresponding to the second data and an existing database to determine the feature classification corresponding to the second data.

[0113] In some examples, the method for determining the second reference baseline based on the feature classification corresponding to the second data can refer to the relevant description in step S400.

[0114] Continue to see Figure 2 In some examples, in step S500, the first calibration model may be updated using information related to the second reference baseline in the second calibration model to redetermine the first reference baseline.

[0115] In some examples, the information related to the second reference baseline may be the second reference baseline itself. In some examples, the second reference baseline may be used as the updated first reference baseline, that is, the first reference baseline is replaced by the second reference baseline. Thus, the amount of calculation can be reduced.

[0116] In some examples, the information related to the second reference baseline may be information other than the second reference baseline itself. In some examples, the information other than the second reference baseline itself may include waveform features and / or feature classifications.

[0117] In some examples, when the first reference baseline is determined based on waveform features and / or feature classifications corresponding to the first data, the information associated with the second reference baseline may be used to update the corresponding information associated with the first reference baseline to update the first reference baseline.

[0118] In some examples, updating the first reference baseline may include substituting information related to the second reference baseline into the first calibration model to obtain a new first reference baseline (ie, an updated first reference baseline) from the first calibration model.

[0119] In some examples, the revised judgment value and the revised amplitude value may be determined based on the offset between the previous first reference baseline and the second reference baseline, and the first reference baseline may be revised.

[0120] Figure 3 1 is a flowchart showing a first implementation method of redetermining a first reference baseline according to an example of the present disclosure.

[0121] See also Figure 3 In some examples, re-determining the first reference baseline may include determining a correction judgment value and a correction amplitude value based on the offset between the previous first reference baseline and the second reference baseline (step S510), determining whether correction is required using the correction judgment value (step S520), correcting the previous first reference baseline based on the correction amplitude value in response to correction being required (step S530), and not correcting the previous first reference baseline in response to no correction being required (step S540). In this case, when correcting the previous first reference baseline, the first reference baseline and the second reference baseline can be considered at the same time, and the risk of the corrected first reference baseline deviating too much from the previous first reference baseline can be reduced, thereby improving the fit between the first reference baseline and the first device 10. In addition, the fit between the first reference baseline and the actual physiological condition of the target object 30 can also be improved.

[0122] See also Figure 3 In some examples, in step S510, a correction judgment value and a correction amplitude value may be determined based on an offset (also referred to as a first offset) between a previous first reference baseline and a second reference baseline. In some examples, the correction judgment value may be used to determine whether the previous first reference baseline needs to be corrected. In some examples, the correction amplitude value may be used to characterize a correction amplitude of the previous first reference baseline.

[0123] In some examples, the first reference baseline and the second reference baseline may be determined first, and the method may refer to the relevant description in step S400. In some examples, the first reference baseline and the second reference baseline may be a numerical value. In some examples, the first reference baseline and the second reference baseline may be a numerical range.

[0124] In some examples, when the reference baseline is a numerical value, the value of the second reference baseline minus the value of the first reference baseline can be used as the first offset. In some examples, when the reference baseline is a numerical range, the average value of the upper and lower limits of the second reference baseline minus the average value of the upper and lower limits of the first reference baseline can be used as the first offset. In this case, the first reference baseline and the second reference baseline can be considered at the same time, which can reduce the risk of the revised first reference baseline deviating too much from the previous first reference baseline, thereby improving the fit of the first reference baseline with the first device 10.

[0125] In some examples, the first offset may be a positive number. In some examples, the first offset may be a negative number.

[0126] In some examples, the calculation formula of the modified judgment value may be: modified judgment value=first preset proportionality coefficient×(value of the first reference baseline+value of the first offset).

[0127] In some examples, the calculation formula of the modified judgment value may be: modified judgment value=first preset proportionality coefficient×(average value of the upper limit and the lower limit of the first reference baseline+value of the first offset).

[0128] In some examples, the first preset proportionality factor may be 0.01, 0.05, 0.1, or 0.2, etc.

[0129] In some examples, a correction judgment value may be preset. For example, the correction judgment value may be an empirical value. In this case, the correction judgment value is used to determine whether correction is needed, which can improve the fit between the corrected first reference baseline and the previous first reference baseline, and further improve the fit between the corrected first reference baseline and the first device 10.

[0130] In some examples, the calculation formula of the correction amplitude value may be: correction amplitude value = second preset proportionality coefficient × value of the first offset. In some examples, the second preset proportionality coefficient may be 1, 0.9, 0.85, or 0.8, etc. In this case, correcting the previous first reference baseline based on the correction amplitude value can improve the fit between the corrected first reference baseline and the previous first reference baseline, and thus can improve the fit between the corrected first reference baseline and the first device 10.

[0131] In some examples, the previous first reference baseline may be revised based on the revised judgment value and the revised amplitude value.

[0132] In some examples, in step S520, the correction judgment value may be used to determine whether correction is required. Specifically, a comparison result of the correction judgment value and the absolute value of the first offset may be obtained, and whether the previous first reference baseline needs to be corrected may be determined based on the comparison result.

[0133] In some examples, when the absolute value of the first offset is not less than the revised judgment value, step S530 may be performed. In some examples, when the absolute value of the first offset is less than the revised judgment value, step S540 may be performed.

[0134] In some examples, in step S530, in response to the need for correction, the previous first reference baseline can be corrected based on the correction amplitude value. For example, the previous first reference baseline plus the value of the correction amplitude value can be used as the updated first reference baseline. In some examples, the upper limit and lower limit of the previous first reference baseline can be added to the numerical range of the correction amplitude value to serve as the updated first reference baseline.

[0135] In some examples, in step S540 , in response to no correction being required, the previous first reference baseline may not be corrected.

[0136] As described above, in some examples, the information related to the second reference baseline may be information other than the second reference baseline itself. In some examples, the information other than the second reference baseline itself may include waveform features and / or feature classifications.

[0137] In some examples, the first reference baseline may be determined based on the waveform features and / or feature classifications corresponding to the first data. In some examples, when the first reference baseline is determined based on the waveform features and / or feature classifications corresponding to the first data, the information associated with the second reference baseline may include waveform features and / or feature classifications. In some examples, the waveform features corresponding to the second data may correspond to the waveform features corresponding to the first data. In some examples, the feature classifications corresponding to the second data may correspond to the feature classifications corresponding to the first data.

[0138] In some examples, the re-determination may be to update the waveform features and / or feature classifications corresponding to the first data using information related to the second reference baseline. In some examples, the first calibration model may update the first reference baseline using the waveform features and / or feature classifications corresponding to the updated first data. In this case, the first calibration model updates the first reference baseline using the waveform features and / or feature classifications corresponding to the updated first data, which can improve the fit between the first reference baseline and the first device 10.

[0139] In some examples, the waveform features corresponding to the second data may be used to update the waveform features corresponding to the first data to update the first reference baseline. In some examples, the feature classification corresponding to the second data may be used to update the feature classification corresponding to the first data to update the first reference baseline. In some examples, the waveform features and feature classification corresponding to the second data may be used to update the waveform features and feature classification corresponding to the first data to update the first reference baseline. For example, the waveform features and / or feature classification corresponding to the second data may be substituted into the first calibration model to obtain a new first reference baseline (i.e., the updated first reference baseline) from the first calibration model.

[0140] Figure 41 is a flowchart showing a second embodiment of redetermining the first reference baseline according to the example of the present disclosure.

[0141] See also Figure 4 In some examples, re-determining the first reference baseline may include updating the waveform features corresponding to the first data using information related to the second reference baseline (step S501), and updating the first reference baseline using the waveform features corresponding to the updated first data (step S502). In this case, the first calibration model updates the first reference baseline using the waveform features corresponding to the updated first data, which can improve the fit between the first reference baseline and the first device 10.

[0142] In some examples, the first reference baseline can be determined based on waveform features corresponding to the first data.

[0143] In some examples, in step S501, the information related to the second reference baseline may include waveform features corresponding to the second data. In some examples, the waveform features corresponding to the second data may correspond to the waveform features corresponding to the first data. For example, if the waveform features corresponding to the first data may include a rising rate, a falling rate, and a plateau period, the waveform features corresponding to the second data may include a rising rate, a falling rate, and a plateau period.

[0144] In some examples, the waveform features corresponding to the second data may be used as the updated waveform features corresponding to the first data.

[0145] In some examples, in step S502, the first reference baseline may be updated using the waveform features corresponding to the updated first data. In some examples, the waveform features corresponding to the updated first data may have a corresponding relationship with the updated first reference baseline. In some examples, step S502 may refer to the method for determining the reference baseline using the waveform features in step S400. For example, an artificial intelligence algorithm may be used to perform big data analysis based on the waveform features corresponding to the updated first data and an existing database to update the first reference baseline.

[0146] In some examples, the feature classification corresponding to the first data may be re-determined based on the updated waveform features corresponding to the first data, and the first reference baseline may be updated based on the re-determined feature classification corresponding to the first data.

[0147] In some examples, step S502 may refer to the method for determining feature classification using waveform features in step S400. In some examples, step S502 may refer to the method for determining a reference baseline using feature classification in step S400. For example, the waveform features and feature classifications may have a corresponding relationship. In some examples, the feature classification corresponding to the updated first data may have a corresponding relationship with the updated first reference baseline. For example, an artificial intelligence algorithm may be used to perform big data analysis based on the feature classification corresponding to the updated first data and an existing database to update the first reference baseline.

[0148] Figure 5 1 is a flowchart showing a third embodiment of redetermining the first reference baseline according to the example of the present disclosure.

[0149] See also Figure 5 In some examples, re-determining the first reference baseline may include updating the feature classification corresponding to the first data using information related to the second reference baseline (step S503), and updating the first reference baseline using the updated feature classification corresponding to the first data (step S504). In this case, the first calibration model updates the first reference baseline using the updated feature classification corresponding to the first data, which can improve the fit between the first reference baseline and the first device 10.

[0150] In some examples, the first reference baseline may be determined based on a feature classification corresponding to the first data.

[0151] In some examples, in step S503, the information related to the second reference baseline may include a feature classification corresponding to the second data. In some examples, the feature classification corresponding to the first data may correspond to the feature classification corresponding to the second data. For example, the feature classification corresponding to the first data may be type 1 diabetes, and the feature classification corresponding to the second data may be type 1 diabetes. In some examples, the feature classification corresponding to the second data may be used as the updated feature classification corresponding to the first data. In this case, the first calibration model updates the first reference baseline using the updated feature classification corresponding to the first data, which can improve the fit of the first reference baseline with the first device 10.

[0152] In some examples, in step S504, the first reference baseline may be updated using the feature classification corresponding to the updated first data. In some examples, the feature classification corresponding to the updated first data may have a corresponding relationship with the updated first reference baseline. In some examples, step S504 may refer to the method for determining the reference baseline using feature classification in step S400. For example, an artificial intelligence algorithm may be used to perform big data analysis based on the feature classification corresponding to the updated first data and an existing database to update the first reference baseline.

[0153] Figure 6 1 is a flowchart showing a fourth embodiment of redetermining the first reference baseline according to the example of the present disclosure.

[0154] See also Figure 6 In some examples, re-determining the first reference baseline may include updating the waveform features and feature classifications corresponding to the first data using information related to the second reference baseline (step S505), and updating the first reference baseline using the updated waveform features and feature classifications corresponding to the first data (step S506). In this case, the first calibration model updates the first reference baseline using the updated waveform features and feature classifications corresponding to the first data, which can improve the fit between the first reference baseline and the first device 10.

[0155] In some examples, the first reference baseline may be determined based on the waveform features and feature classifications corresponding to the first data. In some examples, in step S505, the information associated with the second reference baseline may include the waveform features and feature classifications corresponding to the second data. In some examples, as described above, the waveform features and feature classifications corresponding to the first data may correspond to the waveform features and feature classifications corresponding to the second data, respectively.

[0156] In some examples, the waveform features corresponding to the second data may be used as the waveform features corresponding to the updated first data. In this case, the first calibration model updates the first reference baseline using the waveform features corresponding to the updated first data, which can improve the fit between the first reference baseline and the first device 10.

[0157] In some examples, the feature classification corresponding to the second data can be used as the feature classification corresponding to the updated first data. In this case, the first calibration model updates the first reference baseline using the feature classification corresponding to the updated first data, which can improve the fit between the first reference baseline and the first device 10.

[0158] In some examples, in step S506, the first reference baseline may be updated using the waveform features and feature classifications corresponding to the updated first data. In some examples, the waveform features and feature classifications corresponding to the updated first data may have a corresponding relationship with the updated first reference baseline. In some examples, step S506 may refer to the method for determining the reference baseline using waveform features and feature classifications in step S400. For example, an artificial intelligence algorithm may be used to perform big data analysis based on the waveform features and feature classifications corresponding to the updated first data and an existing database to update the first reference baseline.

[0159] Return to see Figure 2 In some examples, in step S600, the physiological indicator value to be calibrated of the target object 30 can be calibrated based on the re-determined first reference baseline through the updated first calibration model.

[0160] Figure 7 is a flow chart showing the calibration of the physiological indicator value to be calibrated of the target object 30 based on the re-determined first reference baseline involved in the example of the present disclosure.

[0161] See also Figure 7 In some examples, calibrating the physiological indicator value to be calibrated of the target object 30 based on the re-determined first reference baseline may include: obtaining the first actual baseline corresponding to the preset time period of the first device 10 (step S610), determining the adjustment parameter of the physiological indicator value to be calibrated based on the offset between the first actual baseline and the updated first reference baseline (step S620), and adjusting the physiological indicator value to be calibrated based on the adjustment parameter (step S630). In this case, the degree of fit between the calibrated physiological indicator value and the actual physiological indicator value of the target object 30 can be improved, thereby improving the accuracy of the physiological indicator value obtained by the first device 10.

[0162] In some examples, in step S610, a first actual baseline corresponding to the preset time period of the first device 10 may be obtained. In some examples, the first actual baseline may represent an actual reference value of the physiological indicator value obtained by the first device 10 in the preset time period. In some examples, the preset time period may be a time period to which the physiological indicator value to be calibrated belongs. In some examples, the first actual baseline may vary with the characteristics of the physiological indicator value of the target object 30 varying over time.

[0163] In some examples, the preset time period may be related to the acquisition time of the physiological indicator value to be calibrated. In this case, obtaining the first actual baseline corresponding to the preset time period of the first device 10, and the preset time period is related to the acquisition time of the physiological indicator value to be calibrated, can improve the fit between the calibrated physiological indicator value and the actual physiological indicator value of the target object 30.

[0164] In some examples, the preset time period may be a collection time period for the physiological indicator values ​​to be calibrated. In this case, the collected physiological indicator values ​​to be calibrated can be calibrated, and the degree of fit between the calibrated physiological indicator values ​​and the actual physiological indicator values ​​of the target object 30 can be improved.

[0165] In some examples, the preset time period may be before the collection time of the physiological indicator value to be calibrated. In this case, after calibration, it is helpful to reduce the frequency of use of the second device 20 by the target object 30, and improve the fit between the calibrated physiological indicator value and the actual physiological indicator value of the target object 30.

[0166] In some examples, the preset time period may be before the collection time period of the first data. For example, the preset time period may be from February 1 to February 10, 2023, and the collection time period of the first data may be from February 11 to February 20, 2023. In this case, the physiological indicator value before the collection time period of the first data can be calibrated, thereby improving the convenience of the target object 30 in collecting the physiological indicator value.

[0167] In some examples, the preset time period may be after the collection time period of the first data. For example, the collection time period of the first data may be from February 1 to February 10, 2023, and the preset time period may be from February 11 to February 20, 2023. In this case, the convenience of the target object 30 in collecting physiological indicator values ​​can be improved.

[0168] In some examples, the preset time period may partially overlap with the collection time period of the first data. For example, the collection time period of the first data may be from March 1 to March 20, 2023, and the preset time period may be from March 10 to March 30, 2023. In this case, the physiological indicator value after the collection time period of the first data can be calibrated, and the fit between the first reference baseline and the first actual baseline can be improved, thereby improving the fit between the calibrated physiological indicator value and the actual physiological indicator value of the target object 30, thereby improving the accuracy of the physiological indicator value acquired by the first device 10.

[0169] In some examples, the first actual baseline may be determined based on the physiological indicator value data of the first device 10 within a preset time period. In some examples, the first actual baseline may represent an actual reference value of the physiological indicator value acquired by the first device 10.

[0170] In some examples, the method for determining the first actual baseline may refer to step S400. For example, the first actual baseline may be determined based on the physiological indicator value in at least one reference period that satisfies the preset condition in the physiological indicator value data within the preset time period. In some examples, the preset condition may include a preset amplitude. In some examples, the preset condition may include a waveform feature corresponding to the physiological indicator value data within the preset time period.

[0171] In some examples, the method for determining the first actual baseline may be the same as the method for determining the first reference baseline. In this case, the first actual baseline and the first reference baseline can be made consistent and comparable.

[0172] In some examples, in step S620, an adjustment parameter of the physiological indicator value to be calibrated may be determined based on an offset between the first actual baseline and the updated first reference baseline.

[0173] In some examples, the first actual baseline and the first reference baseline may be a numerical value. In some examples, the first actual baseline and the first reference baseline may be a numerical range.

[0174] In some examples, when the baseline is a numerical value, the value of the updated first reference baseline minus the value of the first data baseline may be used as an offset (also referred to as a second offset).

[0175] In some examples, when the baseline is a numerical range, the average of the upper and lower limits of the updated first reference baseline minus the average of the upper and lower limits of the first actual baseline can be used as the second offset. In some examples, the second offset can be a positive number. In some examples, the second offset can be a negative number. In this case, the first actual baseline and the updated first reference baseline can be taken into account at the same time, and the risk of the first actual baseline deviating too much from the first reference baseline can be reduced, thereby improving the fit of the first actual baseline with the first device 10. In addition, the fit of the first actual baseline with the actual physiological condition of the target object 30 can also be improved.

[0176] In some examples, the adjustment parameter may include an adjustment direction and an adjustment amplitude. For example, when the second offset is a positive number, the adjustment direction may be to adjust the value of the physiological indicator to be calibrated upward. For example, when the second offset is a negative number, the adjustment direction may be to adjust the value of the physiological indicator to be calibrated downward. In some examples, the absolute value of the second offset may be used as the adjustment amplitude.

[0177] In some examples, in step S630, the physiological indicator value to be calibrated can be adjusted based on the adjustment parameter. For example, when the adjustment direction is to adjust the physiological indicator value to be calibrated upward, the physiological indicator value to be calibrated plus the adjustment amplitude can be used as the calibrated physiological indicator value. For example, when the adjustment direction is to adjust the physiological indicator value to be calibrated downward, the physiological indicator value to be calibrated minus the adjustment amplitude can be used as the calibrated physiological indicator value. In this case, the first actual baseline and the updated first reference baseline can be taken into account at the same time, and the risk of the adjusted physiological indicator value deviating too much from the first actual baseline can be reduced, thereby improving the fit between the adjusted physiological indicator value and the physiological indicator value of the target object 30 acquired by the first device 10.

[0178] In some examples, the calibrated physiological indicator value may be calculated using the updated first reference baseline. In some examples, the first reference baseline and the physiological indicator value may have a corresponding relationship.

[0179] In some examples, the first calibration model may include a conversion module as described above. In some examples, the conversion module may convert a physical signal related to a physiological indicator value acquired by a monitoring sensor of the first device 10 into a corresponding physiological indicator value. For example, the conversion formula may be: physiological indicator value = first reference baseline + (physical signal value × conversion coefficient). In some examples, the updated first reference baseline may be substituted into the conversion formula to calculate the calibrated physiological indicator value.

[0180] In some examples, when the baseline is a numerical range, the numerical range of the calibrated physiological indicator value can be calculated. For example, the calculation formula can be: physiological indicator value upper limit = reference baseline upper limit + (physical signal value × conversion coefficient), physiological indicator value lower limit = reference baseline lower limit + (physical signal value × conversion coefficient). In some examples, an artificial intelligence algorithm can be used to further obtain the calibration value of the physiological indicator value within the numerical range of the calibrated physiological indicator value based on the characteristics of the data in a preset time period and an existing database. Thus, the calibrated physiological indicator value can be obtained.

[0181] In some examples, the first calibration model may be updated in response to an update request. The method for updating the first calibration model may refer to the relevant description in step S500. In some examples, the update request may be initiated by the target object 30. In some examples, the update request may be automatically initiated by the first device 10. In some examples, the update request may be automatically initiated by the second device 20. In this case, updating the first calibration model can improve the adaptability of the calibration to changes in the physiological indicator values ​​of the target object 30, and can improve the accuracy of the physiological indicator values ​​obtained by the first device 10.

[0182] In some examples, the update request may be initiated by a third device. In some examples, the third device may be a smart device. For example, the third device may be a mobile phone or a smart watch. In some examples, the third device may include a cloud service center. In some examples, the third device may connect to the cloud service center.

[0183] In some examples, the first device 10 or the second device 20 may be connected to a third device. For example, the calibration model may be deployed on the third device.

[0184] In some examples, the update request may be automatically initiated. For example, the first calibration model may be updated in response to a difference between the first data and the second data being greater than a preset degree.

[0185] In some examples, the preset degree may be a degree indicating that the first data and the second data have a significant difference. In this case, updating the first calibration model can improve the adaptability of the calibration to changes in the physiological indicator values ​​of the target object 30, and can improve the accuracy of the physiological indicator values ​​obtained by the first device 10.

[0186] In some examples, a statistical method may be used to calculate the degree of difference between the first data and the second data. For example, the statistical method may include a T test, variance analysis, nonparametric test, or chi-square test.

[0187] In some examples, a statistical value between the first data and the second data may be calculated. For example, the statistical value may include a variance, a standard deviation, and a correlation coefficient. In some examples, when the standard deviation between the first data and the second data is greater than a preset standard deviation value, an update request may be automatically initiated. In some examples, the preset standard deviation value may be an empirical value.

[0188] In some examples, the waveform feature corresponding to the first data may be compared with the waveform feature corresponding to the second data. In some examples, when the waveform feature corresponding to the first data is greater than the waveform feature preset degree, an update request may be automatically initiated. In some examples, the waveform feature preset degree may be an empirical value. For example, the waveform feature preset degree may be a peak value difference of 1 mmol / l.

[0189] In some examples, the feature classification corresponding to the first data may be compared with the feature classification corresponding to the second data. In some examples, when the feature classification corresponding to the first data is different from the feature classification corresponding to the second data, an update request may be automatically initiated.

[0190] In some examples, the first reference baseline and the second reference baseline may be compared. In some examples, a correction judgment value may be determined based on the offset between the first reference baseline and the second reference baseline, and the correction judgment value may be used to determine whether an update request needs to be initiated. The method may refer to the relevant description in step S500.

[0191] In some examples, the first reference baseline and the first actual baseline may be compared. In some examples, the offset between the first actual baseline and the first reference baseline may be calculated, and the method may refer to the relevant description in step S600. In some examples, when the offset between the first actual baseline and the first reference baseline is greater than the update preset value, an update request may be initiated to update the first calibration model. In some examples, the update preset value may be an empirical value. For example, the update preset value may be 1 mmol / L.

[0192] In some examples, an update request may be initiated periodically. For example, the update period may be 1 day, 2 days, 3 days, 5 days, 7 days, 10 days, 14 days, 15 days, 21 days, 30 days, 60 days, 90 days, or 100 days. In some examples, an update request may be initiated irregularly. For example, when the difference between the first data and the second data is greater than a preset degree, an update request may be initiated.

[0193] Figure 8 This is a flowchart showing a first embodiment of a calibration method according to the present disclosure example.

[0194] See also Figure 8 In some examples, the calibration method may include determining a first calibration model at the first device 10 (step S810), determining a second calibration model at the second device 20 (step S820), the second device 20 sending information related to the second reference baseline in the second calibration model to the first device 10 (step S830), the first device 10 updating the first calibration model (step S840), and calibrating the physiological indicator value to be calibrated of the target object 30 based on the re-determined first reference baseline (step S850).

[0195] In some examples, in step S810, a first calibration model may be determined at the first device 10. In some examples, the method of determining the first calibration model may refer to the relevant descriptions in step S300 and step S400.

[0196] In some examples, in step S820, a second calibration model may be determined at the second device 20. In some examples, the method for determining the second calibration model may refer to the relevant description in step S400.

[0197] In some examples, in step S830, the second device 20 may send information related to the second reference baseline in the second calibration model to the first device 10. In some examples, step S830 may refer to the relevant description in step S500.

[0198] In some examples, in step S840, the first device 10 may update the first calibration model. In some examples, step S840 may refer to the relevant description in step S500.

[0199] In some examples, in step S850, the first device 10 may calibrate the physiological indicator value to be calibrated of the target object 30 based on the re-determined first reference baseline. In some examples, step S850 may refer to the relevant description in step S600.

[0200] Fig. 9 This is a flowchart showing a second embodiment of the calibration method according to the present disclosure example.

[0201] See also Fig. 9 In some examples, the calibration method may include determining a first calibration model (step S910), determining a second calibration model (step S920), obtaining information related to a second reference baseline in the second calibration model by a third device (step S930), updating the first calibration model using the third device (step S940), and calibrating the physiological indicator value to be calibrated of the target object 30 of the first device 10 based on the re-determined first reference baseline using the third device (step S950).

[0202] In some examples, the third device may be a smart device, such as a mobile phone or a smart watch.

[0203] In some examples, the third device may include a cloud service center. In some examples, the third device may be connected to the cloud service center.

[0204] In some examples, the third device may be the physiological indicator monitoring system 100 .

[0205] In some examples, the first device 10 can be connected to a cloud service center. In some examples, the second device 20 can be connected to a cloud service center.

[0206] In some examples, in step S910, the first device 10 may connect to the third device. In some examples, the first device 10 may send a physical signal acquired by a monitoring sensor to the third device. In some examples, the third device may determine a first calibration model based on the physical signal acquired by the monitoring sensor of the first device 10, wherein the first calibration model includes a conversion module that can convert the physical signal acquired by the monitoring sensor of the first device 10 into a corresponding physiological indicator value.

[0207] In some examples, the first device 10 may send the first data to the third device. In some examples, the third device may determine a first calibration model based on the first data. In some examples, the first calibration model may be determined at the first device 10, and the parameters in the first calibration model may be synchronized to the third device to perform the calibration method as described above. In some examples, the parameters in the first calibration model may include waveform features corresponding to the first data, feature classifications corresponding to the first data, a first reference baseline, and a first actual baseline, etc. In some examples, step S910 may refer to the relevant descriptions in steps S300 and S400.

[0208] In some examples, in step S920, the second device 20 may connect to the third device. In some examples, the second device 20 may send the physical signal acquired by the monitoring sensor to the third device. In some examples, the third device may determine the second calibration model based on the physical signal acquired by the monitoring sensor of the second device 20, wherein the second calibration model includes a conversion module that can convert the physical signal acquired by the monitoring sensor of the second device 20 into a corresponding physiological indicator value.

[0209] In some examples, the second device 20 may send the second data to the third device. In some examples, the third device may determine a second calibration model based on the second data. In some examples, the second calibration model may be determined at the second device 20, and the second calibration model may be synchronized to the third device to perform the calibration method as described above. In some examples, the parameters in the second calibration model may include waveform features corresponding to the second data, feature classifications corresponding to the second data, a second reference baseline, and the like. In some examples, in step S920, the second calibration model may be determined at the second device 20. In some examples, the second calibration model may be determined at the third device. In some examples, step S920 may refer to the relevant description in step S400.

[0210] In some examples, in step S930, the second device 20 may send information related to the second reference baseline in the second calibration model to the third device. In some examples, when the second calibration model is determined at the third device, the third device may obtain information related to the second reference baseline in the second calibration model. In some examples, step S930 may refer to the relevant description in step S500.

[0211] In some examples, in step S940, the first calibration model may be updated using the third device. In some examples, step S940 may refer to the relevant description in step S500.

[0212] In some examples, in step S950, the physiological indicator value to be calibrated of the target object 30 of the first device 10 can be calibrated based on the re-determined first reference baseline using the third device. In some examples, the third device can send the calibrated physiological indicator value to the first device 10. In some examples, the third device can send the adjusted first actual baseline to the first device 10, and the first device 10 can calibrate the physiological indicator value to be calibrated using the adjusted first actual baseline target object 30. In some examples, step S950 can refer to the relevant description in step S600.

[0213] Fig.10 is a block diagram showing the structure of the physiological index monitoring system 100 involved in the example of the present disclosure. Fig.10The example of the present disclosure also discloses a physiological indicator monitoring system 100, which may not contact the subcutaneous fluid of the target object 30. The physiological indicator monitoring system 100 may include a monitoring sensor 101 and a processing device 102. The monitoring sensor 101 may be configured to generate a sensor signal, and the processing device 102 may be configured to receive a physiological indicator value related to the sensor signal and calibrate the physiological indicator value using the above-mentioned calibration method.

[0214] The examples of the present disclosure also disclose a computer-readable storage medium, which can store at least one instruction, and when the at least one instruction is executed by a processor, one or more steps in the above-mentioned calibration method can be implemented.

[0215] In the calibration method involved in the present disclosure, the first device 10 does not contact the subcutaneous fluid of the target object 30, the second device 20 contacts the subcutaneous fluid of the target object 30, the accuracy of the second device 20 is higher than the accuracy of the first device 10, the first calibration model of the target object 30 is determined based on the first data acquired by the first device 10 in the first time window, the second calibration model of the target object 30 is determined based on the second data acquired in the second time window corresponding to the first time window, the first calibration model is updated using the second calibration model, and the physiological index value to be calibrated of the target object 30 is calibrated using the updated first calibration model. In this case, the fit between the reference baseline in the calibration model and the physiological characteristics of the target object 30 itself can be improved, and at the same time, the calibration model is determined based on the first data and the second data, and when the first device 10 is calibrated using the second device 20 with higher accuracy than the first device 10, the accuracy of the physiological index value acquired by the first device 10 can be improved. In addition, after improving the accuracy of the physiological index value acquired by the first device 10, it is helpful to reduce the frequency of use of the second device 20 by the target object 30. In addition, after reducing the frequency of use of the second device 20 by the target object 30, the comfort of the target object 30 in monitoring the physiological index value can be improved for the situation that the first device 10 does not contact the subcutaneous fluid of the target object 30 and the second device 20 contacts the subcutaneous fluid of the target object 30. In addition, calibrating the physiological index value to be calibrated of the target object 30 based on the calibration model determined based on the physiological index value data that changes over time can improve the adaptability of the calibration model to the changes in the physiological condition of the target object 30, thereby further improving the accuracy of the physiological index value obtained by the first device 10.

[0216] Although the present disclosure is specifically described above in conjunction with the accompanying drawings and examples, it is to be understood that the above description does not limit the present disclosure in any form. Those skilled in the art may modify and change the present disclosure as needed without departing from the essential spirit and scope of the present disclosure, and these modifications and changes all fall within the scope of the present disclosure.

Claims

1. A method for calibrating a physiological index value, characterized in that: include: Acquire first data of the target object, the first data comprising a physiological indicator value varying with time acquired by a first device within a first time window, the first device not contacting a subcutaneous fluid of the target object; Retrieving second data of the target object, the second data comprising a physiological indicator value varying over time acquired by a second device within a second time window corresponding to the first time window, the second device being in contact with subcutaneous fluid of the target object, the second device having a higher accuracy than the first device; determining a first calibration model of the target object based on the first data, the first calibration model comprising a first reference baseline determined by the first data; determining a second calibration model of the target object based on the second data, the second calibration model comprising a second reference baseline determined by the second data; updating the first calibration model using information related to the second reference baseline in the second calibration model to redetermine the first reference baseline; and The physiological indicator value to be calibrated of the target object is calibrated based on the re-determined first reference baseline using the updated first calibration model.

2. The calibration method according to claim 1, characterized in that: The physiological indicator value is a glucose concentration value, the first device is a non-invasive glucose monitoring system, and the second device is a continuous glucose monitoring system.

3. The calibration method according to claim 1, characterized in that: The collection time period of the first data is equal to the collection time period of the second data; or The collection time period of the first data is before the collection time period of the second data; or A collection time period of the first data partially overlaps with a collection time period of the second data.

4. The calibration method according to claim 1, characterized in that: Determine the second reference baseline based on the physiological indicator value in at least one reference period that meets the preset condition in the second data; or Determine the waveform feature and / or feature classification corresponding to the second data based on the second data, and determine the second reference baseline based on the waveform feature and / or feature classification corresponding to the second data; or A waveform feature corresponding to the second data is determined based on the second data, a feature classification corresponding to the second data is determined based on the waveform feature corresponding to the second data, and the second reference baseline is determined based on the feature classification corresponding to the second data.

5. The calibration method according to claim 1, characterized in that: The information related to the second reference baseline is the second reference baseline itself. The redetermination includes: Determining a correction judgment value and a correction amplitude value based on the offset between the previous first reference baseline and the second reference baseline, the correction judgment value being used to judge whether the previous first reference baseline needs to be corrected, and the correction amplitude value being used to characterize the correction amplitude of the previous first reference baseline; and The previous first reference baseline is corrected based on the correction judgment value and the correction amplitude value.

6. The calibration method according to claim 1, characterized in that: The first reference baseline is determined based on the waveform features and / or feature classification corresponding to the first data, and the information related to the second reference baseline includes the waveform features and / or feature classification corresponding to the second data; The re-determination is to update the waveform features and / or feature classification corresponding to the first data using information related to the second reference baseline so that the first calibration model updates the first reference baseline using the updated waveform features and / or feature classification corresponding to the first data.

7. The calibration method according to claim 4, characterized in that: The waveform features corresponding to the second data include at least one of a peak value, a valley value, a rising rate, a falling rate, a rising amplitude, a falling amplitude, a fluctuation amplitude, a stable period and a curve trend.

8. The calibration method according to claim 4, characterized in that: The characteristic classifications corresponding to the second data include physiological index values ​​that are too high, physiological index values ​​that are relatively high, physiological index values ​​that are normal, physiological index values ​​that are relatively low, and physiological index values ​​that are too low.

9. The calibration method according to claim 1, characterized in that: Calibrating the physiological indicator value to be calibrated of the target object based on the re-determined first reference baseline includes: Acquire a first actual baseline corresponding to a preset time period of the first device, wherein the preset time period is related to a collection time of the physiological indicator value to be calibrated; determining an adjustment parameter of the physiological indicator value to be calibrated based on an offset between the first actual baseline and the updated first reference baseline; and The physiological indicator value to be calibrated is adjusted based on the adjustment parameter.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the calibration method according to any one of claims 1 to 9 is implemented.

11. A physiological index monitoring system, wherein the physiological index monitoring system does not contact the subcutaneous fluid of the target object, characterized in that: The invention comprises a monitoring sensor and a processing device, wherein the monitoring sensor is configured to generate a sensor signal, and the processing device is configured to receive a physiological indicator value related to the sensor signal and calibrate the physiological indicator value using the calibration method described in any one of claims 1 to 9.

Images