CN117582222B - Informationized blood glucose monitoring system and informationized blood glucose monitoring method - Google Patents
Informationized blood glucose monitoring system and informationized blood glucose monitoring method Download PDFInfo
- Publication number
- CN117582222B CN117582222B CN202410072794.6A CN202410072794A CN117582222B CN 117582222 B CN117582222 B CN 117582222B CN 202410072794 A CN202410072794 A CN 202410072794A CN 117582222 B CN117582222 B CN 117582222B
- Authority
- CN
- China
- Prior art keywords
- blood glucose
- time sequence
- short
- time
- blood sugar
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000008280 blood Substances 0.000 title claims abstract description 570
- 210000004369 blood Anatomy 0.000 title claims abstract description 570
- WQZGKKKJIJFFOK-GASJEMHNSA-N Glucose Natural products OC[C@H]1OC(O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-GASJEMHNSA-N 0.000 title claims abstract description 376
- 239000008103 glucose Substances 0.000 title claims abstract description 376
- 238000012544 monitoring process Methods 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 title claims abstract description 38
- 239000013598 vector Substances 0.000 claims abstract description 308
- 238000005259 measurement Methods 0.000 claims abstract description 44
- 238000004458 analytical method Methods 0.000 claims abstract description 20
- 238000004364 calculation method Methods 0.000 claims description 20
- 230000004927 fusion Effects 0.000 claims description 18
- 238000012937 correction Methods 0.000 claims description 16
- 238000009826 distribution Methods 0.000 claims description 13
- 230000011218 segmentation Effects 0.000 claims description 12
- 230000009466 transformation Effects 0.000 claims description 9
- 238000005457 optimization Methods 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 206010012601 diabetes mellitus Diseases 0.000 abstract description 9
- 230000008859 change Effects 0.000 description 23
- 230000006870 function Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 10
- 238000004422 calculation algorithm Methods 0.000 description 5
- 230000005856 abnormality Effects 0.000 description 3
- 230000014509 gene expression Effects 0.000 description 3
- 208000013016 Hypoglycemia Diseases 0.000 description 2
- 238000012300 Sequence Analysis Methods 0.000 description 2
- 201000001421 hyperglycemia Diseases 0.000 description 2
- 230000002218 hypoglycaemic effect Effects 0.000 description 2
- NOESYZHRGYRDHS-UHFFFAOYSA-N insulin Chemical compound N1C(=O)C(NC(=O)C(CCC(N)=O)NC(=O)C(CCC(O)=O)NC(=O)C(C(C)C)NC(=O)C(NC(=O)CN)C(C)CC)CSSCC(C(NC(CO)C(=O)NC(CC(C)C)C(=O)NC(CC=2C=CC(O)=CC=2)C(=O)NC(CCC(N)=O)C(=O)NC(CC(C)C)C(=O)NC(CCC(O)=O)C(=O)NC(CC(N)=O)C(=O)NC(CC=2C=CC(O)=CC=2)C(=O)NC(CSSCC(NC(=O)C(C(C)C)NC(=O)C(CC(C)C)NC(=O)C(CC=2C=CC(O)=CC=2)NC(=O)C(CC(C)C)NC(=O)C(C)NC(=O)C(CCC(O)=O)NC(=O)C(C(C)C)NC(=O)C(CC(C)C)NC(=O)C(CC=2NC=NC=2)NC(=O)C(CO)NC(=O)CNC2=O)C(=O)NCC(=O)NC(CCC(O)=O)C(=O)NC(CCCNC(N)=N)C(=O)NCC(=O)NC(CC=3C=CC=CC=3)C(=O)NC(CC=3C=CC=CC=3)C(=O)NC(CC=3C=CC(O)=CC=3)C(=O)NC(C(C)O)C(=O)N3C(CCC3)C(=O)NC(CCCCN)C(=O)NC(C)C(O)=O)C(=O)NC(CC(N)=O)C(O)=O)=O)NC(=O)C(C(C)CC)NC(=O)C(CO)NC(=O)C(C(C)O)NC(=O)C1CSSCC2NC(=O)C(CC(C)C)NC(=O)C(NC(=O)C(CCC(N)=O)NC(=O)C(CC(N)=O)NC(=O)C(NC(=O)C(N)CC=1C=CC=CC=1)C(C)C)CC1=CN=CN1 NOESYZHRGYRDHS-UHFFFAOYSA-N 0.000 description 2
- 238000012806 monitoring device Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000000844 transformation Methods 0.000 description 2
- 102000004877 Insulin Human genes 0.000 description 1
- 108090001061 Insulin Proteins 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 210000001124 body fluid Anatomy 0.000 description 1
- 239000010839 body fluid Substances 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 235000005911 diet Nutrition 0.000 description 1
- 230000037213 diet Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 229940125396 insulin Drugs 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000004060 metabolic process Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- -1 then Substances 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/14532—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue for measuring glucose, e.g. by tissue impedance measurement
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/74—Details of notification to user or communication with user or patient ; user input means
- A61B5/746—Alarms related to a physiological condition, e.g. details of setting alarm thresholds or avoiding false alarms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Pathology (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- General Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Optics & Photonics (AREA)
- Physiology (AREA)
- Emergency Medicine (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses an informationized blood glucose monitoring system and an informationized blood glucose monitoring method, and relates to the field of blood glucose monitoring. Firstly, arranging blood glucose values at a plurality of preset time points according to a time dimension and carrying out local time sequence feature analysis to obtain a sequence of blood glucose local time sequence feature vectors, then calculating blood glucose short-time fluctuation semantic measurement coefficients between every two adjacent blood glucose local time sequence feature vectors in the sequence of blood glucose local time sequence feature vectors to obtain blood glucose short-time fluctuation feature vectors composed of a plurality of blood glucose short-time fluctuation semantic measurement coefficients, then fusing the sequence of blood glucose local time sequence feature vectors and the blood glucose short-time fluctuation feature vectors to obtain blood glucose multi-dimensional time sequence features, and finally determining a short-time predicted value of blood glucose based on the blood glucose multi-dimensional time sequence features and determining whether early warning prompts are generated. In this way, the accuracy and timeliness of monitoring blood glucose in a patient can be improved, thereby helping a diabetic to better monitor and manage blood glucose levels.
Description
Technical Field
The present application relates to the field of blood glucose monitoring, and more particularly, to an informative blood glucose monitoring system and method.
Background
Blood sugar is an important index of metabolism of human body, and abnormal blood sugar level can cause various diseases such as diabetes. Therefore, effective monitoring and prediction of blood glucose is an important means of ensuring patient health. At present, various blood sugar monitoring devices exist in the market, but most of the blood sugar monitoring devices need to be used for collecting blood samples for analysis by patients regularly, so that the patients need to collect and test blood sugar frequently, inconvenience and pain are brought to the patients, the problems that blood sugar data collection is discontinuous and the like can be caused, dynamic changes of blood sugar cannot be reflected in real time, and blood sugar prediction and management cannot be accurately carried out.
Accordingly, an optimized informative blood glucose monitoring system and method are desired.
Disclosure of Invention
In view of this, the application provides an informationized blood glucose monitoring system and method, which can predict and early warn the blood glucose value in a short time through the blood glucose dynamic change of a patient, and improve the accuracy and timeliness of blood glucose monitoring of the patient, so that the diabetes patient can be helped to monitor and manage the blood glucose level better.
According to an aspect of the present application, there is provided an informative blood glucose monitoring method comprising:
collecting blood glucose values of a monitored patient at a plurality of preset time points in a preset time period through a blood glucose sensor worn on the fingertip of the monitored patient;
Arranging the blood glucose values at a plurality of preset time points according to the time dimension to obtain a blood glucose time sequence input vector;
performing local time sequence feature analysis on the blood glucose time sequence input vector to obtain a sequence of blood glucose local time sequence feature vectors;
calculating blood sugar short-time fluctuation semantic measurement coefficients between every two adjacent blood sugar local time sequence feature vectors in the sequence of blood sugar local time sequence feature vectors to obtain blood sugar short-time fluctuation feature vectors composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients;
fusing the sequence of the local time sequence feature vector of the blood sugar with the short-time fluctuation feature vector of the blood sugar to obtain multi-dimensional time sequence features of the blood sugar;
and determining a short-time predicted value of the blood sugar based on the multi-dimensional time sequence characteristics of the blood sugar, and determining whether an early warning prompt is generated.
The calculating of the blood sugar short-time fluctuation semantic measurement coefficient between every two adjacent blood sugar local time sequence feature vectors in the sequence of the blood sugar local time sequence feature vectors to obtain the blood sugar short-time fluctuation feature vector composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients comprises the following steps:
calculating blood sugar short-time fluctuation semantic measurement coefficients between every two adjacent blood sugar local time sequence feature vectors in the sequence of blood sugar local time sequence feature vectors according to the following coefficient calculation formula to obtain the blood sugar short-time fluctuation feature vectors composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients; wherein, the coefficient calculation formula is:
；
Wherein,and->The characteristic values of each position of every two adjacent blood sugar local time sequence characteristic vectors in the sequence of the blood sugar local time sequence characteristic vectors are respectively +.>Represents a logarithmic function with base 2, +.>Is the scale of every two adjacent blood glucose local time sequence characteristic vectors in the sequence of blood glucose local time sequence characteristic vectors,/L>Is the characteristic value of each position in the blood sugar short-time fluctuation characteristic vector.
In the above-mentioned informationized blood glucose monitoring method, performing local time sequence feature analysis on the blood glucose time sequence input vector to obtain a sequence of blood glucose local time sequence feature vectors, including:
vector segmentation is carried out on the blood glucose time sequence input vector to obtain a sequence of blood glucose local time sequence input vectors;
and the sequence of the blood glucose local time sequence input vector is passed through a blood glucose time sequence feature extractor based on a one-dimensional convolution layer to obtain the sequence of the blood glucose local time sequence feature vector.
In the above informationized blood glucose monitoring method, fusing the sequence of the local time sequence feature vector of blood glucose and the short-time fluctuation feature vector of blood glucose to obtain the multi-dimensional time sequence feature of blood glucose comprises:
and after the sequence of the blood glucose local time sequence feature vector is spliced into a global blood glucose time sequence feature vector, fusing the blood glucose short-time fluctuation feature vector and the global blood glucose time sequence feature vector to obtain a blood glucose multi-dimensional time sequence feature vector serving as the blood glucose multi-dimensional time sequence feature.
In the above informationized blood glucose monitoring method, determining a short-time predicted value of blood glucose based on the multidimensional time sequence feature of blood glucose, and determining whether to generate an early warning prompt comprises:
performing feature distribution optimization on the blood glucose multi-dimensional time sequence feature vector to obtain an optimized blood glucose multi-dimensional time sequence feature vector;
the optimized blood sugar multidimensional time sequence feature vector passes through a blood sugar predictor based on a decoder to obtain a short-time predicted value of blood sugar;
and determining whether to generate an early warning prompt or not based on the comparison of the short-time predicted value of the blood sugar and the preset blood sugar threshold value.
In the above informationized blood glucose monitoring method, performing feature distribution optimization on the blood glucose multi-dimensional time sequence feature vector to obtain an optimized blood glucose multi-dimensional time sequence feature vector, including:
converting the blood sugar short-time fluctuation feature vector and the global blood sugar time sequence feature vector into the same length through linear transformation, and then carrying out fusion correction to obtain a correction feature vector;
and fusing the correction feature vector with the blood glucose multi-dimensional time sequence feature vector to obtain the optimized blood glucose multi-dimensional time sequence feature vector.
In the above information blood glucose monitoring method, the optimizing the blood glucose multidimensional time sequence feature vector to obtain the short-time predicted value of the blood glucose by a blood glucose predictor based on a decoder comprises the following steps:
And performing decoding regression on the optimized blood glucose multi-dimensional time sequence characteristic vector by using a plurality of fully connected layers of the blood glucose predictor based on the decoder according to the following decoding formula to obtain a short-time predicted value of the blood glucose, wherein the decoding formula is as follows:wherein->Is the optimized blood sugar multidimensional time sequence characteristic vector, < > and the method thereof>Is a short-term predictor of said blood glucose, < >>Is a weight matrix, < >>Is a bias vector, ++>Representing matrix multiplication +.>To activate the function.
According to another aspect of the present application, there is provided an informative blood glucose monitoring system comprising:
the data acquisition module is used for acquiring blood glucose values of a monitored patient at a plurality of preset time points in a preset time period through a blood glucose sensor worn on the fingertip of the monitored patient;
the vectorization module is used for arranging the blood glucose values of the plurality of preset time points according to the time dimension to obtain a blood glucose time sequence input vector;
the local time sequence feature analysis module is used for carrying out local time sequence feature analysis on the blood glucose time sequence input vector so as to obtain a sequence of blood glucose local time sequence feature vectors;
the coefficient calculation module is used for calculating blood sugar short-time fluctuation semantic measurement coefficients between every two adjacent blood sugar local time sequence feature vectors in the sequence of the blood sugar local time sequence feature vectors so as to obtain blood sugar short-time fluctuation feature vectors composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients;
The fusion module is used for fusing the sequence of the blood glucose local time sequence feature vector and the blood glucose short-time fluctuation feature vector to obtain a blood glucose multi-dimensional time sequence feature;
the prediction module is used for determining a short-time predicted value of the blood sugar based on the multi-dimensional time sequence characteristics of the blood sugar and determining whether an early warning prompt is generated;
wherein, the coefficient calculation module is used for:
calculating blood sugar short-time fluctuation semantic measurement coefficients between every two adjacent blood sugar local time sequence feature vectors in the sequence of blood sugar local time sequence feature vectors according to the following coefficient calculation formula to obtain the blood sugar short-time fluctuation feature vectors composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients; wherein, the coefficient calculation formula is:
；
wherein,and->The characteristic values of each position of every two adjacent blood sugar local time sequence characteristic vectors in the sequence of the blood sugar local time sequence characteristic vectors are respectively +.>Represents a logarithmic function with base 2, +.>Is the scale of every two adjacent blood glucose local time sequence characteristic vectors in the sequence of blood glucose local time sequence characteristic vectors,/L>Is the characteristic value of each position in the blood sugar short-time fluctuation characteristic vector.
In the above-mentioned informationized blood glucose monitoring system, the local time sequence feature analysis module includes:
The vector segmentation unit is used for carrying out vector segmentation on the blood glucose time sequence input vector so as to obtain a sequence of blood glucose local time sequence input vector;
and the blood glucose time sequence feature extraction unit is used for enabling the sequence of the blood glucose local time sequence input vectors to pass through a blood glucose time sequence feature extractor based on a one-dimensional convolution layer to obtain the sequence of the blood glucose local time sequence feature vectors.
According to the method, firstly, blood sugar values at a plurality of preset time points are arranged according to time dimension and are subjected to local time sequence feature analysis to obtain a sequence of blood sugar local time sequence feature vectors, then, blood sugar short-time fluctuation semantic measurement coefficients between every two adjacent blood sugar local time sequence feature vectors in the sequence of the blood sugar local time sequence feature vectors are calculated to obtain blood sugar short-time fluctuation feature vectors composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients, then, the sequence of the blood sugar local time sequence feature vectors and the blood sugar short-time fluctuation feature vectors are fused to obtain blood sugar multi-dimensional time sequence features, and finally, based on the blood sugar multi-dimensional time sequence features, the short-time predicted value of blood sugar is determined, and whether early warning prompt is generated or not is determined. In this way, the accuracy and timeliness of monitoring blood glucose in a patient can be improved, thereby helping a diabetic to better monitor and manage blood glucose levels.
Other features and aspects of the present application will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments, features and aspects of the present application and together with the description, serve to explain the principles of the present application.
Fig. 1 shows a flow chart of an informative blood glucose monitoring method according to an embodiment of the present application.
Fig. 2 shows a schematic architecture diagram of an informative blood glucose monitoring method according to an embodiment of the present application.
Fig. 3 shows a flowchart of substep S160 of the informationized blood glucose monitoring method according to an embodiment of the present application.
Fig. 4 shows a block diagram of an informationized blood glucose monitoring system according to an embodiment of the present application.
Fig. 5 shows an application scenario diagram of an informationized blood glucose monitoring method according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some, but not all embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present application without making any inventive effort, are also within the scope of the present application.
As used in this application and in the claims, the terms "a," "an," "the," and/or "the" are not specific to the singular, but may include the plural, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.
Various exemplary embodiments, features and aspects of the present application will be described in detail below with reference to the accompanying drawings. In the drawings, like reference numbers indicate identical or functionally similar elements. Although various aspects of the embodiments are illustrated in the accompanying drawings, the drawings are not necessarily drawn to scale unless specifically indicated.
In addition, numerous specific details are set forth in the following detailed description in order to provide a better understanding of the present application. It will be understood by those skilled in the art that the present application may be practiced without some of these specific details. In some instances, methods, means, elements, and circuits have not been described in detail as not to unnecessarily obscure the present application.
Aiming at the technical problems, the technical concept of the application is to wear the blood glucose sensor on the fingertip of a patient to collect blood glucose data from the skin body fluid of the patient in a noninvasive mode so as to monitor and collect the blood glucose value of the patient in real time, and introduce a data processing and analyzing algorithm at the rear end to perform time sequence analysis of the blood glucose value of the patient, so that short-time prediction and early warning of blood glucose are performed. Therefore, the blood sample is not required to be collected, inconvenience and pain of a patient are reduced, short-time prediction and early warning of the blood glucose level are carried out through the blood glucose dynamic change of the patient, the accuracy and timeliness of blood glucose monitoring of the patient are improved, and the diabetic can be helped to monitor and manage the blood glucose level better.
Fig. 1 shows a flow chart of an informative blood glucose monitoring method according to an embodiment of the present application. Fig. 2 shows a schematic architecture diagram of an informative blood glucose monitoring method according to an embodiment of the present application. As shown in fig. 1 and 2, the informationized blood glucose monitoring method according to the embodiment of the application includes the steps of: s110, acquiring blood glucose values of a monitored patient at a plurality of preset time points in a preset time period through a blood glucose sensor worn on the fingertip of the monitored patient; s120, arranging the blood glucose values of the plurality of preset time points according to a time dimension to obtain a blood glucose time sequence input vector; s130, carrying out local time sequence feature analysis on the blood glucose time sequence input vector to obtain a sequence of blood glucose local time sequence feature vectors; s140, calculating blood sugar short-time fluctuation semantic measurement coefficients between every two adjacent blood sugar local time sequence feature vectors in the sequence of the blood sugar local time sequence feature vectors to obtain blood sugar short-time fluctuation feature vectors composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients; s150, fusing the sequence of the local time sequence feature vector of the blood sugar and the short-time fluctuation feature vector of the blood sugar to obtain multi-dimensional time sequence features of the blood sugar; and S160, determining a short-time predicted value of the blood sugar based on the multi-dimensional time sequence characteristics of the blood sugar, and determining whether an early warning prompt is generated.
It should be understood that the step S110 serves to collect blood glucose values of the monitored patient at a plurality of predetermined time points within a predetermined period of time by means of a blood glucose sensor worn on the fingertip of the monitored patient; s120, arranging the collected blood glucose values at a plurality of preset time points according to a time dimension to obtain a blood glucose time sequence input vector; the sequence of the blood glucose local time sequence feature vector obtained through the local time sequence feature analysis in the step S130 can reflect the change trend and rule of blood glucose; s140, calculating blood sugar short-time fluctuation feature vectors composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients, wherein the blood sugar short-time fluctuation semantic measurement coefficients are obtained between every two adjacent blood sugar local time sequence feature vectors in the sequence of the blood sugar local time sequence feature vectors, and the blood sugar short-time fluctuation feature vectors can reflect the instantaneous fluctuation condition of blood sugar; in the S150 step, the sequence of the local time sequence feature vector of the blood sugar and the short-term fluctuation feature vector of the blood sugar are fused to obtain the multi-dimensional time sequence feature of the blood sugar, so that the long-term trend and the short-term fluctuation condition of the blood sugar can be comprehensively reflected; in step S160, according to the multidimensional time sequence feature of blood glucose, prediction of blood glucose can be performed, and whether an early warning prompt needs to be sent out is determined according to the prediction result, so as to remind a patient or doctor to take corresponding measures.
Specifically, in the technical scheme of the application, firstly, blood glucose values of a monitored patient at a plurality of preset time points in a preset time period are collected through a blood glucose sensor worn on the fingertip of the monitored patient. Next, it is considered that the blood glucose level has a dynamic change rule of time sequence in the time dimension, that is, there is a time sequence correlation between the blood glucose levels at the plurality of predetermined time points. Therefore, the blood glucose values at the plurality of predetermined time points need to be arranged according to the time dimension to obtain a blood glucose time sequence input vector so as to integrate time sequence distribution information of the blood glucose values in the time dimension.
Then, it is considered that the change in blood glucose level generally has a certain time-series property, and such time-series change in blood glucose has a fluctuation, resulting in that the blood glucose value of the patient exhibits different time-series change characteristics at different time spans. Therefore, in order to better perform time sequence analysis of blood glucose data of a patient, so as to more accurately describe the change condition of blood glucose and provide more reliable basis for subsequent prediction and early warning, in the technical scheme of the application, vector segmentation is performed on the blood glucose time sequence input vector so as to obtain a sequence of blood glucose local time sequence input vectors.
And then, the sequence of the blood glucose local time sequence input vector is subjected to feature mining in a blood glucose time sequence feature extractor based on a one-dimensional convolution layer so as to extract local time sequence feature information of the blood glucose value in each local time period in the time dimension respectively, thereby obtaining the sequence of the blood glucose local time sequence feature vector. Therefore, the local time sequence detail change characteristics of the blood sugar data can be captured, which is favorable for capturing the characteristic information such as the rising or falling trend, the fluctuation and the like of the blood sugar, so that the change condition of the blood sugar can be more accurately described, and a more reliable basis is provided for subsequent prediction and early warning.
Accordingly, in step S130, performing local time sequence feature analysis on the blood glucose time sequence input vector to obtain a sequence of blood glucose local time sequence feature vectors, including: vector segmentation is carried out on the blood glucose time sequence input vector to obtain a sequence of blood glucose local time sequence input vectors; and passing the sequence of blood glucose local time sequence input vectors through a blood glucose time sequence feature extractor based on a one-dimensional convolution layer to obtain the sequence of blood glucose local time sequence feature vectors.
It is worth mentioning that the function of vector segmentation is to segment the blood glucose time sequence input vector according to a certain window size, so as to obtain the sequence of the blood glucose local time sequence input vector. By slicing the blood glucose time sequence input vector, blood glucose data in the whole time period can be divided into a plurality of local time sequence subsections so as to better analyze local variation trend of blood glucose. The one-dimensional convolution layer is a neural network layer in deep learning and is used for extracting characteristics of input data. In blood glucose monitoring, a blood glucose time sequence feature extractor based on a one-dimensional convolution layer can extract time sequence features of blood glucose data by carrying out convolution operation on a sequence of blood glucose local time sequence input vectors. The one-dimensional convolution layer can capture the time sequence mode and the change trend in the blood sugar data, so that the sequence of the local time sequence feature vector of the blood sugar is obtained. Through vector segmentation and application of a one-dimensional convolution layer, time sequence characteristics of blood glucose data can be better analyzed, local variation trend of blood glucose is extracted, and more accurate information is provided for subsequent characteristic fusion and prediction.
It should be appreciated that the volatility of blood glucose levels is one of the important indicators of concern in diabetes management. The short-time fluctuation characteristic of the blood sugar of the patient reflects the change condition of the blood sugar in a short time, and can provide characteristic information about the fluctuation degree and the stability of the blood sugar, which is helpful for helping doctors and patients to better know the fluctuation condition of the blood sugar, so as to adjust the treatment scheme in time and prevent the occurrence of hypoglycemia or hyperglycemia. Therefore, in order to extract the short-time fluctuation feature of blood sugar, in the technical scheme of the application, the blood sugar short-time fluctuation semantic measurement coefficient between every two adjacent blood sugar local time sequence feature vectors in the sequence of the blood sugar local time sequence feature vectors is further calculated to obtain the blood sugar short-time fluctuation feature vector composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients. The short-time fluctuation degree of the blood sugar can be quantified by calculating the fluctuation semantic measurement coefficient between every two adjacent blood sugar local time sequence feature vectors. Therefore, the obtained blood sugar short-time fluctuation feature vector provides time sequence feature information of blood sugar fluctuation, and can be used for measuring instability and change trend of blood sugar. These features can help doctors and patients to better understand the fluctuation of blood sugar, adjust the treatment scheme in time, and prevent the occurrence of hypoglycemia or hyperglycemia.
Accordingly, in step S140, calculating the blood glucose short-time fluctuation semantic measurement coefficient between every two adjacent blood glucose local time sequence feature vectors in the sequence of blood glucose local time sequence feature vectors to obtain a blood glucose short-time fluctuation feature vector composed of a plurality of blood glucose short-time fluctuation semantic measurement coefficients, including: calculating blood sugar short-time fluctuation semantic measurement coefficients between every two adjacent blood sugar local time sequence feature vectors in the sequence of blood sugar local time sequence feature vectors according to the following coefficient calculation formula to obtain the blood sugar short-time fluctuation feature vectors composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients; wherein, the coefficient calculation formula is:
；
wherein,and->The characteristic values of each position of every two adjacent blood sugar local time sequence characteristic vectors in the sequence of the blood sugar local time sequence characteristic vectors are respectively +.>Represents a logarithmic function with base 2, +.>Is the scale of every two adjacent blood glucose local time sequence characteristic vectors in the sequence of blood glucose local time sequence characteristic vectors,/L>Is the characteristic value of each position in the blood sugar short-time fluctuation characteristic vector.
It should be understood that the blood glucose short-term fluctuation semantic measurement coefficient is a measurement index for measuring the short-term fluctuation of blood glucose, and is used for measuring the fluctuation degree between two adjacent blood glucose local time sequence feature vectors and reflecting the instantaneous change condition of blood glucose. Specifically, the blood glucose short-term fluctuation semantic measurement coefficient can be obtained by calculating the degree of difference between two adjacent blood glucose local time sequence feature vectors. These differences may include information on the slope of blood glucose, the amplitude of the fluctuations, the rate of change, etc. By calculating the short-term fluctuation semantic measurement coefficient of the blood sugar, the short-term fluctuation of the blood sugar can be quantified numerically, and then the instantaneous change condition of the blood sugar is provided. The calculation of the blood glucose short-time fluctuation semantic measurement coefficient can be performed based on a mathematical model or a statistical method, and the specific calculation method can be different according to the actual application. This metric can be used in aspects of feature extraction, predictive model training, etc. in blood glucose monitoring systems to provide more comprehensive and accurate blood glucose excursion information.
Further, taking into consideration that the change in the blood glucose in time series is a complicated process, the abnormality thereof includes both an abnormality of the overall time series change trend of the blood glucose and an abnormality of fluctuation in a short time. Therefore, in order to more accurately predict the blood glucose level in a short time and provide more accurate early warning, in the technical scheme of the application, after the sequence of the local blood glucose time sequence feature vector is spliced into the global blood glucose time sequence feature vector, the blood glucose short time fluctuation feature vector and the global blood glucose time sequence feature vector are fused to obtain the blood glucose multidimensional time sequence feature vector. It should be understood that the global blood glucose time sequence feature vector reflects time sequence global change features of blood glucose, and the blood glucose short-time fluctuation feature vector reflects short-time fluctuation feature information of blood glucose, so that by fusing the blood glucose short-time fluctuation feature vector and the global blood glucose time sequence feature vector, time sequence fluctuation and overall change trend of blood glucose can be combined to form a multi-dimensional blood glucose time sequence feature vector. The feature vector can comprehensively consider the local fluctuation and the overall change trend of the blood sugar, thereby providing more accurate and comprehensive blood sugar feature description and providing more accurate and reliable basis for subsequent prediction and early warning.
Accordingly, in step S150, fusing the sequence of the local timing feature vector of blood glucose and the short-term fluctuation feature vector of blood glucose to obtain a multi-dimensional timing feature of blood glucose, including: and after the sequence of the blood glucose local time sequence feature vector is spliced into a global blood glucose time sequence feature vector, fusing the blood glucose short-time fluctuation feature vector and the global blood glucose time sequence feature vector to obtain a blood glucose multi-dimensional time sequence feature vector serving as the blood glucose multi-dimensional time sequence feature.
It is worth mentioning that stitching is the operation of concatenating multiple vectors or sequences together in a certain order to form one longer vector or sequence. In step S150, the stitching operation is used to stitch the sequence of local timing feature vectors of blood glucose into a global timing feature vector of blood glucose. Specifically, the local time sequence feature vectors of blood sugar are connected together according to a certain sequence to form a global time sequence feature vector of blood sugar containing all local time sequence features. The purpose of this is to integrate the information of each local timing feature to provide a more comprehensive and complete blood glucose timing feature. After splicing, fusing the blood sugar short-time fluctuation feature vector and the global blood sugar time sequence feature vector to obtain the blood sugar multi-dimensional time sequence feature vector. By fusing the two different types of feature vectors, the local time sequence features and the short-time fluctuation features of the blood sugar can be comprehensively considered, so that more comprehensive and accurate multi-dimensional time sequence features of the blood sugar are provided. The splicing operation is used for integrating different characteristic information together so as to better perform subsequent tasks such as characteristic analysis, model training or prediction. By means of the splicing operation, feature information from different sources can be organically combined, and therefore richer and accurate feature representation is provided.
And then, the blood glucose multi-dimensional time sequence characteristic vector passes through a blood glucose predictor based on a decoder to obtain a short-time predicted value of the blood glucose. That is, the time series multidimensional feature information of the blood glucose level is subjected to decoding regression so as to perform short-time prediction of the blood glucose level of the patient based on the time series global variation trend and the short-time fluctuation condition of the blood glucose, and whether an early warning prompt is generated or not is determined based on comparison of the short-time predicted value of the blood glucose and a blood glucose preset threshold. Therefore, short-time prediction and early warning of the blood sugar value can be carried out based on the blood sugar dynamic change of the patient without collecting a blood sample, inconvenience and pain of the patient are reduced, and meanwhile, the accuracy and timeliness of blood sugar monitoring of the patient are improved, so that the diabetic can be helped to monitor and manage the blood sugar level better.
Accordingly, in step S160, as shown in fig. 3, based on the multidimensional time sequence feature of blood glucose, a short-term predicted value of blood glucose is determined, and whether an early warning prompt is generated is determined, including: s161, performing feature distribution optimization on the blood glucose multi-dimensional time sequence feature vector to obtain an optimized blood glucose multi-dimensional time sequence feature vector; s162, passing the optimized blood glucose multidimensional time sequence feature vector through a blood glucose predictor based on a decoder to obtain a short-time predicted value of blood glucose; and S163, determining whether an early warning prompt is generated or not based on the comparison of the short-time predicted value of the blood sugar and a preset blood sugar threshold.
It should be understood that the step S161 is to perform feature distribution optimization on the multi-dimensional time sequence feature vector of the blood glucose, and this step may adjust and optimize the multi-dimensional time sequence feature vector of the blood glucose through some optimization algorithms or methods, so that the feature vector better accords with the distribution feature of the blood glucose data, thereby improving the accuracy of the subsequent prediction model. In step S162, the prediction model uses the optimized multidimensional time sequence feature vector of blood glucose as input, and calculates and infers through a decoder to obtain a prediction result of future blood glucose values. The step S163 is to determine whether an early warning notice is generated based on the comparison of the short-term predicted value of blood glucose with a predetermined blood glucose threshold, and in this step, the output result of the prediction model is compared with the predetermined blood glucose threshold. If the predicted value exceeds or falls below the set threshold range, the system can send out an early warning prompt to remind a patient or doctor to take corresponding measures, such as adjusting diet, adjusting insulin dosage or seeking medical assistance, and the like. The method aims at providing prediction of future trend of the blood sugar by optimizing and predicting multi-dimensional time sequence characteristics of the blood sugar, and judging whether an early warning prompt needs to be sent according to the prediction result so as to help a patient or doctor to make corresponding decisions and actions.
In particular, in the above technical solution, each blood glucose local time series feature vector in the sequence of blood glucose local time series feature vectors expresses a local time domain time series correlation feature of the blood glucose value in a local time domain after global time domain division, whereby the global blood glucose time series feature vector expresses a local time domain time series correlation feature of the blood glucose value in a global time domain, and the blood glucose short time fluctuation feature vector expresses a local time domain blood glucose value time series correlation fluctuation metric feature in a global time domain, whereby the blood glucose short time fluctuation feature vector and the global blood glucose time series feature vector have blood glucose local time domain-local time domain correlation dimension expression differences in a global time domain.
In this way, when the blood glucose short-time fluctuation feature vector and the global blood glucose time sequence feature vector are fused, it is desirable to suppress the difference of high-dimensional geometric changes of the blood glucose short-time fluctuation feature vector and the global blood glucose time sequence feature vector, which are caused by the difference of associated dimension expressions between a local time domain and a local time domain of blood glucose in a global time domain, in a high-dimensional fusion feature space based on feature distribution directions corresponding to different geometric transformations, so that the fusion expression effect is improved.
Based on the above, the applicant of the present application firstly converts the blood glucose short-time fluctuation feature vector and the global blood glucose time sequence feature vector into the same length through linear transformation, and then further performs fusion correction to obtain a correction feature vector.
Accordingly, in step S161, performing feature distribution optimization on the blood glucose multi-dimensional time sequence feature vector to obtain an optimized blood glucose multi-dimensional time sequence feature vector, including: converting the blood sugar short-time fluctuation feature vector and the global blood sugar time sequence feature vector into the same length through linear transformation, and then carrying out fusion correction to obtain a correction feature vector; and fusing the correction feature vector with the blood glucose multi-dimensional time sequence feature vector to obtain the optimized blood glucose multi-dimensional time sequence feature vector.
The method for obtaining the corrected feature vector by performing fusion correction after converting the blood glucose short-time fluctuation feature vector and the global blood glucose time sequence feature vector into the same length through linear transformation comprises the following steps: converting the blood glucose short-time fluctuation feature vector and the global blood glucose time sequence feature vector into the same length through linear transformation by using the following fusion formula, and then carrying out fusion correction to obtain the correction feature vector; wherein, the fusion formula is:
；
Wherein,is saidBlood glucose short-term fluctuation feature vector->Is the global blood glucose timing feature vector, +.>、/>And->The blood sugar short-time fluctuation feature vector is +.>The global blood glucose time sequence feature vector +.>And the eigenvalues of said correction eigenvector, < >>And->Respectively 1-norm and 2-norm of the feature vector,/->Is the length of the feature vector, and +.>Is a weight superparameter,/->An exponential operation representing a numerical value, the exponential operation representing the calculation of a natural exponential function value that is a power of the numerical value.
In particular, to promote the perception and cognition of the spatial transformations corresponding to different feature distributions by features when fusion is performed, the blood glucose short-term fluctuation feature vector is based onAnd the global blood glucose timing feature vector +.>Rotation control of feature distribution of feature vectors in different directions is performed from vector scale dimensions to preserve rotational invariance of fusion features by relative rotation unwrapping, thereby avoiding high-dimensional geometric variation differences of feature distribution caused by geometric transformation in high-dimensional fusion feature space when fusion is performed. In this way, the correction feature vector and the blood sugar multi-dimensional time sequence feature vector are fused, so that the blood sugar short-time fluctuation feature vector +_of the blood sugar multi-dimensional time sequence feature vector can be improved >And the global blood glucose timing feature vector +.>Thereby improving the accuracy of the decoded values it obtains by the decoder. Therefore, the blood glucose value can be predicted and pre-warned in short time based on the time sequence global change and short time fluctuation of the blood glucose of the patient, so that inconvenience and pain of the patient are reduced, the accuracy and timeliness of monitoring the blood glucose of the patient are improved, and the diabetic can be helped to monitor and manage the blood glucose level better.
Further, in step S162, passing the optimized blood glucose multidimensional time series feature vector through a decoder-based blood glucose predictor to obtain a short-term predicted value of blood glucose, including: and performing decoding regression on the optimized blood glucose multi-dimensional time sequence characteristic vector by using a plurality of fully connected layers of the blood glucose predictor based on the decoder according to the following decoding formula to obtain a short-time predicted value of the blood glucose, wherein the decoding formula is as follows:wherein->Is the optimized blood sugar multi-dimensional time sequence characteristic vector，/>Is a short-term predictor of said blood glucose, < >>Is a weight matrix, < >>Is a bias vector, ++>Representing matrix multiplication +.>To activate the function.
In summary, according to the informationized blood glucose monitoring method disclosed by the embodiment of the application, the accuracy and timeliness of monitoring blood glucose of a patient can be improved, so that a diabetic patient can be helped to monitor and manage blood glucose level better.
Fig. 4 shows a block diagram of an informative blood glucose monitoring system 100 according to an embodiment of the present application. As shown in fig. 4, an informative blood glucose monitoring system 100 according to an embodiment of the present application includes: a data acquisition module 110 for acquiring blood glucose values of a monitored patient at a plurality of predetermined time points within a predetermined time period by a blood glucose sensor worn on a fingertip of the monitored patient; the vectorization module 120 is configured to arrange the blood glucose values at the plurality of predetermined time points according to a time dimension to obtain a blood glucose time sequence input vector; the local time sequence feature analysis module 130 is configured to perform local time sequence feature analysis on the blood glucose time sequence input vector to obtain a sequence of blood glucose local time sequence feature vectors; the coefficient calculation module 140 is configured to calculate a blood glucose short-time fluctuation semantic measurement coefficient between every two adjacent blood glucose local time sequence feature vectors in the sequence of blood glucose local time sequence feature vectors to obtain a blood glucose short-time fluctuation feature vector composed of a plurality of blood glucose short-time fluctuation semantic measurement coefficients; the fusion module 150 is configured to fuse the sequence of the local timing sequence feature vector of the blood glucose with the short-time fluctuation feature vector of the blood glucose to obtain a multi-dimensional timing sequence feature of the blood glucose; and a prediction module 160, configured to determine a short-term predicted value of the blood glucose based on the multidimensional time sequence feature of the blood glucose, and determine whether to generate an early warning prompt.
In one possible implementation, the local timing feature analysis module 130 includes: the vector segmentation unit is used for carrying out vector segmentation on the blood glucose time sequence input vector so as to obtain a sequence of blood glucose local time sequence input vector; and a blood glucose time sequence feature extraction unit, configured to obtain the sequence of blood glucose local time sequence feature vectors by using a blood glucose time sequence feature extractor based on a one-dimensional convolution layer.
In one possible implementation, the coefficient calculating module 140 is configured to: calculating blood sugar short-time fluctuation semantic measurement coefficients between every two adjacent blood sugar local time sequence feature vectors in the sequence of blood sugar local time sequence feature vectors according to the following coefficient calculation formula to obtain the blood sugar short-time fluctuation feature vectors composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients; wherein, the coefficient calculation formula is:
；
wherein,and->The characteristic values of each position of every two adjacent blood sugar local time sequence characteristic vectors in the sequence of the blood sugar local time sequence characteristic vectors are respectively +.>Represents a logarithmic function with base 2, +.>Is the scale of every two adjacent blood glucose local time sequence characteristic vectors in the sequence of blood glucose local time sequence characteristic vectors,/L >Is the characteristic value of each position in the blood sugar short-time fluctuation characteristic vector.
Here, it will be understood by those skilled in the art that the specific functions and operations of the respective units and modules in the above-described informationized blood glucose monitoring system 100 have been described in detail in the above description of the informationized blood glucose monitoring method with reference to fig. 1 through 3, and thus, repetitive descriptions thereof will be omitted.
As described above, the informationized blood glucose monitoring system 100 according to the embodiment of the present application may be implemented in various wireless terminals, such as a server or the like having an informationized blood glucose monitoring algorithm. In one possible implementation, the informative blood glucose monitoring system 100 according to embodiments of the present application may be integrated into the wireless terminal as a software module and/or hardware module. For example, the informative blood glucose monitoring system 100 may be a software module in the operating system of the wireless terminal, or may be an application developed for the wireless terminal; of course, the informative blood glucose monitoring system 100 could equally be one of a number of hardware modules of the wireless terminal.
Alternatively, in another example, the informative blood glucose monitoring system 100 and the wireless terminal may be separate devices, and the informative blood glucose monitoring system 100 may be connected to the wireless terminal through a wired and/or wireless network and transmit interactive information in a agreed data format.
Fig. 5 shows an application scenario diagram of an informationized blood glucose monitoring method according to an embodiment of the present application. As shown in fig. 5, in this application scenario, first, blood glucose values of a monitored patient at a plurality of predetermined time points (for example, D illustrated in fig. 5) within a predetermined period of time are collected, and then, the blood glucose values at the plurality of predetermined time points are input to a server (for example, S illustrated in fig. 5) in which an informationized blood glucose monitoring algorithm is deployed, wherein the server is capable of processing the blood glucose values at the plurality of predetermined time points using the informationized blood glucose monitoring algorithm to obtain short-time predicted values of blood glucose.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The embodiments of the present application have been described above, the foregoing description is exemplary, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (6)
1. An informative blood glucose monitoring method, comprising:
collecting blood glucose values of a monitored patient at a plurality of preset time points in a preset time period through a blood glucose sensor worn on the fingertip of the monitored patient;
arranging the blood glucose values at a plurality of preset time points according to the time dimension to obtain a blood glucose time sequence input vector;
performing local time sequence feature analysis on the blood glucose time sequence input vector to obtain a sequence of blood glucose local time sequence feature vectors;
calculating blood sugar short-time fluctuation semantic measurement coefficients between every two adjacent blood sugar local time sequence feature vectors in the sequence of blood sugar local time sequence feature vectors to obtain blood sugar short-time fluctuation feature vectors composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients;
Fusing the sequence of the local time sequence feature vector of the blood sugar with the short-time fluctuation feature vector of the blood sugar to obtain multi-dimensional time sequence features of the blood sugar;
determining a short-time predicted value of the blood sugar based on the multi-dimensional time sequence characteristics of the blood sugar, and determining whether an early warning prompt is generated;
the calculating of the blood sugar short-time fluctuation semantic measurement coefficient between every two adjacent blood sugar local time sequence feature vectors in the sequence of the blood sugar local time sequence feature vectors to obtain the blood sugar short-time fluctuation feature vector composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients comprises the following steps:
calculating blood sugar short-time fluctuation semantic measurement coefficients between every two adjacent blood sugar local time sequence feature vectors in the sequence of blood sugar local time sequence feature vectors according to the following coefficient calculation formula to obtain the blood sugar short-time fluctuation feature vectors composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients; wherein, the coefficient calculation formula is:
；
wherein,and->The characteristic values of each position of every two adjacent blood sugar local time sequence characteristic vectors in the sequence of the blood sugar local time sequence characteristic vectors are respectively +.>Represents a logarithmic function with base 2, +.>Is the scale of every two adjacent blood glucose local time sequence characteristic vectors in the sequence of blood glucose local time sequence characteristic vectors,/L >Is the characteristic value of each position in the blood sugar short-time fluctuation characteristic vector;
the local time sequence characteristic analysis is performed on the blood glucose time sequence input vector to obtain a sequence of the blood glucose local time sequence characteristic vector, and the method comprises the following steps:
vector segmentation is carried out on the blood glucose time sequence input vector to obtain a sequence of blood glucose local time sequence input vectors;
and the sequence of the blood glucose local time sequence input vector is passed through a blood glucose time sequence feature extractor based on a one-dimensional convolution layer to obtain the sequence of the blood glucose local time sequence feature vector.
2. The method of claim 1, wherein fusing the sequence of local timing feature vectors of blood glucose and the short-term fluctuation feature vectors of blood glucose to obtain the multi-dimensional timing feature of blood glucose comprises:
and after the sequence of the blood glucose local time sequence feature vector is spliced into a global blood glucose time sequence feature vector, fusing the blood glucose short-time fluctuation feature vector and the global blood glucose time sequence feature vector to obtain a blood glucose multi-dimensional time sequence feature vector serving as the blood glucose multi-dimensional time sequence feature.
3. The method of claim 2, wherein determining a short-term predictive value of blood glucose and determining whether to generate an early warning cue based on the multi-dimensional timing characteristics of blood glucose comprises:
Performing feature distribution optimization on the blood glucose multi-dimensional time sequence feature vector to obtain an optimized blood glucose multi-dimensional time sequence feature vector;
the optimized blood sugar multidimensional time sequence feature vector passes through a blood sugar predictor based on a decoder to obtain a short-time predicted value of blood sugar;
and determining whether to generate an early warning prompt or not based on the comparison of the short-time predicted value of the blood sugar and the preset blood sugar threshold value.
4. The method of claim 3, wherein optimizing the feature distribution of the glucose multi-dimensional time series feature vector to obtain an optimized glucose multi-dimensional time series feature vector comprises:
converting the blood sugar short-time fluctuation feature vector and the global blood sugar time sequence feature vector into the same length through linear transformation, and then carrying out fusion correction to obtain a correction feature vector;
and fusing the correction feature vector with the blood glucose multi-dimensional time sequence feature vector to obtain the optimized blood glucose multi-dimensional time sequence feature vector.
5. The method of claim 4, wherein passing the optimized glucose multi-dimensional time series feature vector through a decoder-based glucose predictor to obtain a short-term predicted value of glucose comprises:
And performing decoding regression on the optimized blood glucose multi-dimensional time sequence characteristic vector by using a plurality of fully connected layers of the blood glucose predictor based on the decoder according to the following decoding formula to obtain a short-time predicted value of the blood glucose, wherein the decoding formula is as follows:wherein->Is the optimized blood sugar multidimensional time sequence characteristic vector, < > and the method thereof>Is a short-term predictor of said blood glucose, < >>Is a weight matrix, < >>Is a bias vector, ++>Representing matrix multiplication +.>To activate the function.
6. An informative blood glucose monitoring system, comprising:
the data acquisition module is used for acquiring blood glucose values of a monitored patient at a plurality of preset time points in a preset time period through a blood glucose sensor worn on the fingertip of the monitored patient;
the vectorization module is used for arranging the blood glucose values of the plurality of preset time points according to the time dimension to obtain a blood glucose time sequence input vector;
the local time sequence feature analysis module is used for carrying out local time sequence feature analysis on the blood glucose time sequence input vector so as to obtain a sequence of blood glucose local time sequence feature vectors;
the coefficient calculation module is used for calculating blood sugar short-time fluctuation semantic measurement coefficients between every two adjacent blood sugar local time sequence feature vectors in the sequence of the blood sugar local time sequence feature vectors so as to obtain blood sugar short-time fluctuation feature vectors composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients;
The fusion module is used for fusing the sequence of the blood glucose local time sequence feature vector and the blood glucose short-time fluctuation feature vector to obtain a blood glucose multi-dimensional time sequence feature;
the prediction module is used for determining a short-time predicted value of the blood sugar based on the multi-dimensional time sequence characteristics of the blood sugar and determining whether an early warning prompt is generated;
wherein, the coefficient calculation module is used for:
calculating blood sugar short-time fluctuation semantic measurement coefficients between every two adjacent blood sugar local time sequence feature vectors in the sequence of blood sugar local time sequence feature vectors according to the following coefficient calculation formula to obtain the blood sugar short-time fluctuation feature vectors composed of a plurality of blood sugar short-time fluctuation semantic measurement coefficients; wherein, the coefficient calculation formula is:
；
wherein,and->The characteristic values of each position of every two adjacent blood sugar local time sequence characteristic vectors in the sequence of the blood sugar local time sequence characteristic vectors are respectively +.>Represents a logarithmic function with base 2, +.>Is the scale of every two adjacent blood glucose local time sequence characteristic vectors in the sequence of blood glucose local time sequence characteristic vectors,/L>Is the characteristic value of each position in the blood sugar short-time fluctuation characteristic vector;
the local time sequence characteristic analysis module comprises:
The vector segmentation unit is used for carrying out vector segmentation on the blood glucose time sequence input vector so as to obtain a sequence of blood glucose local time sequence input vector;
and the blood glucose time sequence feature extraction unit is used for enabling the sequence of the blood glucose local time sequence input vectors to pass through a blood glucose time sequence feature extractor based on a one-dimensional convolution layer to obtain the sequence of the blood glucose local time sequence feature vectors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410072794.6A CN117582222B (en) | 2024-01-18 | 2024-01-18 | Informationized blood glucose monitoring system and informationized blood glucose monitoring method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410072794.6A CN117582222B (en) | 2024-01-18 | 2024-01-18 | Informationized blood glucose monitoring system and informationized blood glucose monitoring method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117582222A CN117582222A (en) | 2024-02-23 |
| CN117582222B true CN117582222B (en) | 2024-03-29 |
Family
ID=89911932
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410072794.6A Active CN117582222B (en) | 2024-01-18 | 2024-01-18 | Informationized blood glucose monitoring system and informationized blood glucose monitoring method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117582222B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117598700B (en) * | 2024-01-23 | 2024-03-29 | 吉林大学 | Intelligent blood oxygen saturation detection system and method |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106980746A (en) * | 2016-12-16 | 2017-07-25 | 清华大学 | A kind of general Woundless blood sugar Forecasting Methodology based on Time-Series analysis |
| CN109171754A (en) * | 2018-07-27 | 2019-01-11 | 清华-伯克利深圳学院筹备办公室 | Training method, device, terminal and the storage medium of blood glucose prediction model |
| CN113080949A (en) * | 2021-03-30 | 2021-07-09 | 北京京东拓先科技有限公司 | Hypoglycemia early warning method and device and computer-readable storage medium |
| CN114207737A (en) * | 2019-05-31 | 2022-03-18 | 明智数据系统公司(d/b/a一滴公司) | System for biological monitoring and blood glucose prediction and associated methods |
| CN114239718A (en) * | 2021-12-15 | 2022-03-25 | 杭州电子科技大学 | High-precision long-term time sequence prediction method based on multivariate time sequence data analysis |
| CN114548158A (en) * | 2022-01-28 | 2022-05-27 | 广东工业大学 | Data processing method for blood sugar prediction |
| CN115062729A (en) * | 2022-07-15 | 2022-09-16 | 展讯通信（上海）有限公司 | Classification model and training method, classification method, device and medium thereof |
| CN115827988A (en) * | 2023-02-10 | 2023-03-21 | 成都桉尼维尔信息科技有限公司 | Self-media content popularity prediction method |
| CN116911419A (en) * | 2023-01-28 | 2023-10-20 | 杭州电子科技大学 | Long time sequence prediction method based on trend correlation feature learning |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IT1396465B1 (en) * | 2008-07-10 | 2012-12-14 | B G Informatica S R L | METHOD FOR THE DEFINITION AND INTERACTIVE MANAGEMENT OF A TREATMENT FOR THE CONTROL OF GLYCEMIC RATE IN THE DIABETIC PATIENT AND INTEGRATED DEVICE THAT IMPLEMENTS THIS METHOD |
| JP5511033B1 (en) * | 2012-12-04 | 2014-06-04 | Ｎｅｃシステムテクノロジー株式会社 | Blood glucose level prediction device, measurement device, blood glucose level prediction method, and program |
| RU2629796C1 (en) * | 2016-05-23 | 2017-09-04 | Общество с ограниченной ответственностью "Лаборатория межклеточных технологий "Интерсел Рэнд" (ООО "Интерсел Рэнд") | Method and multisensor device for noninvasive monitoring of glucose level in blood |
| US11204847B2 (en) * | 2018-12-21 | 2021-12-21 | Microsoft Technology Licensing, Llc | Machine learning model monitoring |
| US20220087577A1 (en) * | 2020-09-18 | 2022-03-24 | Iplemind.inc | Method, apparatus and program for measuring electrocardiogram-based blood glucose using artificial intelligence |
| US20230097161A1 (en) * | 2021-09-15 | 2023-03-30 | University Of Virginia Patent Foundation | Using an Online Disturbance Rejection and Anticipation System to Reduce Hyperglycemia |
-
2024
- 2024-01-18 CN CN202410072794.6A patent/CN117582222B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106980746A (en) * | 2016-12-16 | 2017-07-25 | 清华大学 | A kind of general Woundless blood sugar Forecasting Methodology based on Time-Series analysis |
| CN109171754A (en) * | 2018-07-27 | 2019-01-11 | 清华-伯克利深圳学院筹备办公室 | Training method, device, terminal and the storage medium of blood glucose prediction model |
| CN114207737A (en) * | 2019-05-31 | 2022-03-18 | 明智数据系统公司(d/b/a一滴公司) | System for biological monitoring and blood glucose prediction and associated methods |
| CN113080949A (en) * | 2021-03-30 | 2021-07-09 | 北京京东拓先科技有限公司 | Hypoglycemia early warning method and device and computer-readable storage medium |
| CN114239718A (en) * | 2021-12-15 | 2022-03-25 | 杭州电子科技大学 | High-precision long-term time sequence prediction method based on multivariate time sequence data analysis |
| CN114548158A (en) * | 2022-01-28 | 2022-05-27 | 广东工业大学 | Data processing method for blood sugar prediction |
| CN115062729A (en) * | 2022-07-15 | 2022-09-16 | 展讯通信（上海）有限公司 | Classification model and training method, classification method, device and medium thereof |
| CN116911419A (en) * | 2023-01-28 | 2023-10-20 | 杭州电子科技大学 | Long time sequence prediction method based on trend correlation feature learning |
| CN115827988A (en) * | 2023-02-10 | 2023-03-21 | 成都桉尼维尔信息科技有限公司 | Self-media content popularity prediction method |
Non-Patent Citations (2)
| Title |
|---|
| PCR-RBF-SVM 预测模型在财政数据中的应用;王喆，王有力，孙雯雯，吕巍;《吉林大学学报( 理学版)》;20120131;第50卷(第1期);全文 * |
| Relationship Between Time-Dependent Proteinuria and Risk of Stroke in Population With Different Glucose Tolerance Status;Anxin Wang et al;《Journal of the American Heart Association》;20201231;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117582222A (en) | 2024-02-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117582222B (en) | Informationized blood glucose monitoring system and informationized blood glucose monitoring method | |
| US7621871B2 (en) | Systems and methods for monitoring and evaluating individual performance | |
| Zaitcev et al. | A deep neural network application for improved prediction of $\text {HbA} _ {\text {1c}} $ in type 1 diabetes | |
| EP3558113A1 (en) | Determining wellness using activity data | |
| CN103440421B (en) | medical data processing method and system | |
| WO2017064609A1 (en) | System and method for predicting heart failure decompensation | |
| CN117598700B (en) | Intelligent blood oxygen saturation detection system and method | |
| CN116798631A (en) | Digital healthy living platform based on big data | |
| CN110403611B (en) | Method and apparatus for predicting glycated hemoglobin component value in blood, computer device, and storage medium | |
| JP6053166B2 (en) | Numerical data analysis apparatus and program | |
| WO2021087608A1 (en) | System and method for evaluating glucose homeostasis | |
| US20170011184A1 (en) | Method of Adaptively Predicting Blood-Glucose Level by Collecting Biometric and Activity Data with A User Portable Device | |
| CN116019429A (en) | Health monitoring method, device, equipment and storage medium based on physiological index | |
| Kapur et al. | GlucoBreath: Non-Invasive Glucometer to Detect Diabetes using Breath | |
| Hua et al. | ECG signal classification in wearable devices based on compressed domain | |
| US20140031658A1 (en) | System and method for monitoring glucose or other compositions in an individual | |
| Jain et al. | Predictive Modeling for Early Hyperglycemia Detection in Type 2 Diabetes | |
| EP3327598A1 (en) | A system and a method for automatically analyzing continuous glucose monitoring data indicative of glucose level in a bodily fluid | |
| EP4248842A1 (en) | Personalized patient hydration state measuring device | |
| US20230000446A1 (en) | Apparatus and method for estimating lipid concentration | |
| CN117617921A (en) | Intelligent blood pressure monitoring system and method based on Internet of things | |
| CN115444367A (en) | Monitoring method and system of respiratory function rehabilitation training instrument based on artificial intelligence | |
| Ricaurte et al. | Characterizing Parkinson´ s disease from gait dynamics using chaos descriptors | |
| Haberfehlner et al. | A Novel Video-Based Methodology for Automated Classification of Dystonia and Choreoathetosis in Dyskinetic Cerebral Palsy During a Lower Extremity Task | |
| Gómez et al. | Check for updates Electronic Biosignal Monitoring System for the Prevention of Respiratory Diseases by Applying Artificial Intelligence |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |