CN108937954A - Artificial intelligence deep learning method corrects the monitoring method for continuing blood glucose - Google Patents
Artificial intelligence deep learning method corrects the monitoring method for continuing blood glucose Download PDFInfo
- Publication number
- CN108937954A CN108937954A CN201710370218.XA CN201710370218A CN108937954A CN 108937954 A CN108937954 A CN 108937954A CN 201710370218 A CN201710370218 A CN 201710370218A CN 108937954 A CN108937954 A CN 108937954A
- Authority
- CN
- China
- Prior art keywords
- blood glucose
- blood
- artificial intelligence
- monitoring
- data
- 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.)
- Pending
Links
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
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Surgery (AREA)
- Public Health (AREA)
- Pathology (AREA)
- Veterinary Medicine (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Emergency Medicine (AREA)
- Optics & Photonics (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention belongs to glucose monitoring techniques fields, specifically disclose artificial intelligent depth learning method and correct the monitoring method for continuing blood glucose, comprise the concrete steps that: (1) collecting the accurate blood glucose value that single patient persistently changes in for a period of time；(2) accurate blood glucose value is measured by carrying out multiple finger tip blood collection method；(3) blood glucose value measured by lasting blood sugar monitoring instrument is corrected with accurate blood glucose value；(4) correlation data carries out the artificial learning correction of depth；(5) data measured using the sustainable blood sugar monitoring instrument of artificial intelligence autocoder correction of typist's errors；(6) and by the numerical value diversity mode that this artificial intelligence approach is identified blood glucose correction mode is summarized；(7) after artificial intelligence training finishes, the sustainable blood sugar monitoring, and the blood glucose value recorded is corrected using the obtained data calibration method of artificial intelligence training.This method can provide accurately and reliably blood glucose real time data for diabetic, preferably to manage own bodies situation.
Description
Technical field
The invention belongs to glucose monitoring techniques field, in particular to a kind of artificial intelligence deep learning method, which corrects, continues blood
The monitoring method of sugar.
Background technique
Diabetes, which have become, seriously threatens one of chronic disease of human health, according to the statistics of the World Health Organization, glycosuria
Patient's cumulative year after year, the diabetic of countries in the world has reached as many as 200,000,000, and China is even more to become the high-incidence of diabetes
Area, for China diabetic more than 100,000,000, diabetes morbidity is up to 10%, increases by 5,500,000 every year on average, increases daily
1.5 ten thousand, increases by 600 per hour, increase by 10 per minute, worse behind increasingly increased diabetic,
Prediabetes are in there are also 2.64 hundred million people, blood glucose is also unsatisfactory for diabetes diagnostic criterion not within normal value, from glycosuria
Disease is just poor one step away, if not being careful in one's diet and taking exercise, may making them, sb.'s illness took a turn for the worse and becomes diabetes patient.
Diabetes be by inherent cause, immunologic function disorder, microorganism infection and its toxin, free radical toxin, spirit because
The various virulence factors of element etc. act on sugar, the protein, rouge that body leads to hypoinsulinism, insulin resistance etc. and causes
A series of metabolic disorder syndromes such as fat, water and electrolyte, clinically using hyperglycemia as main feature, typical case may occur in which more
The performance, i.e. " three-many-one-little " symptom such as urine, more drinks, more food, syntexis.If diabetic is without properly treating or treating mistake
When being easiest to cause many complication, such as fundus hemorrhage, eye-blurred, have a weak heart, blood pressure injustice is hidden, hyperlipemia, trick
Numb or picotement, shank take out crick contraction, the refractory conjunction of oedema, skin disease, wound, gum swelling bleeding, fatty liver etc..If dragged
Prolong with the passing of time, fundus hemorrhage and peripheral nerve symptom can be easy to cause.Severe diabetes mellitus is easy to appear ketoacidosis, stupor, kidney
The illnesss such as dirty syndrome, albuminuria, headstroke.
Method without eradicating diabetes at this stage, can only be adjusted by frequently monitoring blood glucose value the feed of patient with
Medication.Blood sugar test can find the blood glucose value abnormal conditions (blood glucose value is high or low) of patient in time, some hyperglycemic patients are not
It is abnormal that itself can be discovered, can only be found in time by blood sugar monitoring, to prevent ketosis acid, ketosis poisoning etc..Hypoglycemia can also be made
It will lead to brain tissue blood supply insufficiency at great harm, such as hypoglycemia, it is possible to cause death.Patient needs dense according to blood glucose
Degree is to adjust diet and drug.
The method of traditional blood sugar test includes following several method: the detection that phlebotomizes, quick blood sugar radiomete detection,
Blood sugar test paper colorimetric detection etc..
The first, venous blood samples blood sugar test is to obtain blood with centrifuge separation blood after phlebotomizing using biochemical instruments
Slurry, blood plasma monitor system measurement by generating hydrogen peroxide after reacting oxidizing glucose with glucose oxidase, with another
The number of hydrogen peroxide, and obtain blood-sugar content, the usual detection time of this method is slow, and is difficult to accomplish at once after vein haemospasia
It examines, and the longer blood glucose of blood samples storage time will be reduced due to glycolysis.
Second, quick blood sugar radiomete detection has many advantages, such as small in size, portable, easy to operate, quick detection, fastly
What fast blood glucose meter measured is the blood glucose value of peripheral blood, and peripheral blood is arteriovenous mixing blood, and theoretically its blood glucose level compares venous blood
Sugar is slightly higher, and especially postprandial blood sugar differs greatly.The advantages of fast blood glucose meter is simple and efficient, but its result is inevitably slightly missed
Difference.
The third, blood sugar test paper colorimetric method blood sugar test: this method, which does not need detector, can measure blood sugar concentration, but its
It is a kind of semidefinite weight testing method, test paper can only be used at 18-35 DEG C.The blood glucose meter of early stage glucose oxidase colorimetric method,
Change color after test paper and reaction of blood, to the time after erase drop of blood and place into blood glucose meter blood glucose value is obtained by measurement chromatography.
The above method requires needle needle-holding hand and refers to or venous blood, is detected with blood.Although some methods are being widely used, some is
Among research, but whether be a kind of that method they will extract the blood of patient, be to belong to invasive detection, patient can be caused
Do not accommodate increase infection chance, it is also necessary to expensive consumables, so not being ideal detection method.
In order to solve inconvenience brought by invasive inspection and pain, noninvasive dynamics monitoring has become competing both at home and abroad in recent years
The mutually hot spot of research and development, the detection of Woundless blood sugar include near infrared spectroscopy, mid-infrared light spectrometry, metabolism thermal method etc., these methods
All is mostly that blood sugar concentration is determined using spectrum analysis, some can also the parameters such as combination temperature, humidity analyzed, but due to
Human body environment is complicated, and everyone body difference, and existing blood glucose meter precision is not generally high, therefore clinical and market is all
It is not widely popularized.The general big rise and fall of the blood glucose value of diabetes patient, invasive monitoring and Woundless blood sugar values all at present
Blood glucose value can only be read at a small number of time points, the blood glucose meter that can continuously monitor becomes the demand of numerous diabetics, also takes
Certain achievement was obtained, such as someone has developed the blood glucose meter of sustainable monitoring, this is a kind of minimally invasive blood glucose meter, and probe is implanted into skin
Under, the blood glucose fluctuation in subcutaneous space interstitial fluid can be continuously recorded, can be completed over time to blood glucose trend
Tracking, while inflammation can be also generated, so that monitoring signals is affected, but continous way blood sugar monitoring is the hair appointed be so from now on
Direction is opened up, research and development temperature is high, and technology is left to be desired.
Existing glucose monitoring techniques are mostly finger tip blood drawing monitoring, due to finger tip blood drawing skin can be generated certain wound and
Pain, patient are often unwilling often to draw blood, and can generate conflict psychology, this is just likely to the best period of delay treatment,
Cause serious consequence, and this method can only spot measurement, cannot reflect the trend of change of blood sugar, be unfavorable for the prison of the state of an illness
Control and treatment, some non-invasive blood sugar instruments occurred currently on the market equally have this limitation, and the disadvantage of non-invasive blood sugar instrument maximum
Be to measure it is inaccurate, it is most of by infrared this is because non-invasive blood sugar instrument is not direct detection human blood glucose concentration
The variation of tissue fluid is detected, blood glucose value then is calculated using algorithm, but since human internal environment is complicated, and person to person
Between body it is also variant, while being influenced again by external environment, therefore although noninvasive dynamics monitoring avoids make patient
Body is by wound, however but there is very big errors, also just can not really react glucose situation.Sustainable blood glucose meter energy
The blood glucose fluctuation of enough preferable reflection diabetics, it is to be implanted by probe subcutaneously, records the blood glucose in subcutaneous space interstitial fluid
Fluctuation, in the blood glucose value that a few days ago can accurately measure patient of implantation, but with the growth of probe Implantation Time, meeting
Inflammatory reaction is induced, the exact value of measurement is influenced.
With machine learning, the rise of artificial intelligence, artificial intelligence technology is used for diabetes glucose value by many scholars
The combination of detection, especially Woundless blood sugar monitoring and machine learning.Since machine learning needs biggish data collection capacity ability
Preferable learning effect is obtained, and data collected there must be certain accuracy, but often due to non-invasive blood sugar instrument
It can only just be capable of measuring at some time point, be unable to continuous acquisition data, this just brings biggish workload to the acquisition of data, number
It is less not accurate enough plus data according to measuring, therefore there is no good learning effects.In addition, in addition, such method can not
Achieve the effect that personalization, it, can be from many diabetics when acquiring data in order to make up the inadequate defect of above-mentioned data volume
Trained data are obtained with it, are then gone to detect other patients again with this result, it will be clear that since interpersonal body is poor
The effect of different reason, detection is not very good, moreover, such method obviously considerably increases workload, therefore cannot be obtained
To extensive use.
Therefore, research and develop it is a kind of for diabetic can provide accurately and reliably blood glucose real time data, preferably to manage
The monitoring method for managing the lasting blood glucose of artificial intelligence deep learning method correction of own bodies situation is extremely urgent.
Summary of the invention
The purpose of the present invention is overcome the deficiencies in the prior art, and it is lasting to disclose a kind of artificial intelligence deep learning method correction
The monitoring method of blood glucose, it is inaccurate that this method preferably solves sustainable blood glucose meter late detection as brought by biocompatibility
The problem of the sustainable blood glucose meter worn of patient from starting to wear up to the blood glucose meter end-of-life, be from start to finish all patient
Accurate blood glucose fluctuating change trend at any time and specific blood glucose value are provided, so that patient preferably manages itself blood
Sugar, in addition, it can be directed to different patients, the study that this method can be adaptive with different patients passes through conclusion
Study obtains blood glucose calibration model.
In order to reach above-mentioned technical purpose, the present invention is realized by following technical scheme:
Artificial intelligence deep learning method of the present invention corrects the monitoring method for continuing blood glucose, comprises the concrete steps that:
(1) it collects the accurate blood glucose value that single patient persistently changes in for a period of time: being supervised first with sustainable blood glucose
Survey the blood glucose value that instrument glucose monitoring techniques collect consecutive variations inaccuracy in a period of time from single patient；
(2) accurate blood glucose value is measured by carrying out the blood sugar monitoring methods of multiple finger tip blood sampling；
(3) blood glucose value measured by lasting Russia's blood sugar monitoring instrument is corrected with resulting accurate blood glucose value；
(4) by carrying out blood glucose level data acquisition with a large amount of patients, the lasting blood glucose of collected a large amount of patients is supervised
Result before the Data correction of survey and after correction compares, and using comparing result as sample, these data are carried out depth people
Work learning correction；
(5) artificial intelligence autocoder technology is utilized by the collected data using sustainable detection blood glucose meter,
Deep learning, multilayer neural network method are trained, and the result obtained after training is used further to correct sustainable blood sugar monitoring
The data that instrument is measured；
(6) it is made to identify the difference of result of the data for continuing blood sugar monitoring instrument before correction with patient and after correction
Rule and mode obtain the degree rule of blood glucose value acquired in sustainable glucose monitoring techniques and practical blood glucose value difference, and
The numerical value diversity mode that this artificial intelligence approach is identified is summarized into blood glucose correction mode.
(7) after artificial intelligence training finishes, the sustainable blood sugar monitoring, and it is obtained using artificial intelligence training
Data calibration method is corrected the blood glucose value recorded.
As the further improvement of above-mentioned technology, above-mentioned steps read continuous data in (1) from single patient,
800-1000 blood glucose level data of acquisition daily continues 0.5-10 days.
As the further improvement of above-mentioned technology, number of taking a blood sample daily in above-mentioned steps (2) is 2-10 times.
As the further improvement of above-mentioned technology, the numberical range of a large amount of patients is 100- in above-mentioned steps (4)
100000.
Compared with prior art, the beneficial effects of the present invention are:
(1) artificial intelligence deep learning method of the present invention corrects the monitoring method for continuing blood glucose, will be traditional
Sustainable blood glucose meter and technology artificial intelligence emerging at present have carried out good combination, carry out blood glucose inspection using the method for the present invention
It surveys, on the one hand alleviates intrusive blood glucose method of surveying to pain brought by diabetic and inconvenience, be on the other hand directed to again
The current sustainable biggish defect of blood sugar monitoring methods later period measurement error is optimized.
(2) present invention improves accuracy rate by artificial intelligence depth learning technology, in the blood glucose measurement side for solving script
But also with personalized and adaptive advantage on the basis of method, blood glucose shape can be preferably monitored for different diabetics
Condition, and the sustainable blood glucose meter worn of patient from starting to wear up to the blood glucose meter end-of-life, be from start to finish all patient
Accurate blood glucose fluctuating change trend at any time and specific blood glucose value are provided, so that patient preferably manages itself blood
Sugar.
(3) for different patients, the study that this method can be adaptive with different patients passes through the present invention
Inductive learning obtains the calibration model for being specially suitble to oneself.Error for artificial intelligence technology resulting corrected value and true value
Within 5%, the accuracy of blood glucose value is substantially increased
Detailed description of the invention
The present invention is described in detail in the following with reference to the drawings and specific embodiments:
Fig. 1 is that artificial intelligence deep learning method of the present invention corrects the monitoring method schematic diagram for continuing blood glucose.
Specific embodiment
As shown in Figure 1, artificial intelligence deep learning method of the present invention corrects the monitoring method for continuing blood glucose, tool
Body step is:
(1) it collects the accurate blood glucose value that single patient persistently changes in for a period of time: being supervised first with sustainable blood glucose
The blood glucose value that instrument glucose monitoring techniques collect consecutive variations inaccuracy in a period of time from single patient is surveyed, is acquired daily
800-1000 blood glucose level data continues 0.5-10 days view patient's concrete conditions and determines；
(2) accurate blood glucose value is measured by carrying out the blood sugar monitoring methods of multiple finger tip blood sampling, take a blood sample 2-10 daily
It is secondary,；
(3) blood glucose value measured by lasting Russia's blood sugar monitoring instrument is corrected with resulting accurate blood glucose value；
(4) by carrying out blood glucose level data acquisition with a large amount of patients (100-100000), by collected a large amount of trouble
Result before the Data correction of the lasting blood sugar monitoring of person and after correction compares, using comparing result as sample, by these
Data carry out the artificial learning correction of depth；By carrying out blood glucose level data acquisition with a large amount of patients (100-100000)
(5) artificial intelligence autocoder technology is utilized by the collected data using sustainable detection blood glucose meter,
Deep learning, multilayer neural network method are trained, and the result obtained after training is used further to correct sustainable blood sugar monitoring
The data that instrument is measured；
(6) it is made to identify the difference of result of the data for continuing blood sugar monitoring instrument before correction with patient and after correction
Rule and mode obtain the degree rule of blood glucose value acquired in sustainable glucose monitoring techniques and practical blood glucose value difference, and
The numerical value diversity mode that this artificial intelligence approach is identified is summarized into blood glucose correction mode, while can be by patient by crowd point
Class, such as age-based stratum, gender, the factors such as diabetes type are classified, and the accuracy of artificial intelligence identification is improved.
(7) after artificial intelligence training finishes, the sustainable blood sugar monitoring, and it is obtained using artificial intelligence training
Data calibration method is corrected after artificial intelligence training finishes the blood glucose value recorded, the sustainable blood sugar monitoring,
And the blood glucose value recorded is corrected using the obtained data calibration method of artificial intelligence training.
The inspection of method very good solution of the present invention sustainable blood glucose meter later period as brought by biocompatibility
Indeterminable problem, traditional solution are corrected again using finger tip blood drawing, this can bring secondary insult to patient, and this
Method also avoids the problem well, and the sustainable blood glucose meter that patient is worn is from wearing is started up to blood glucose meter service life knot
Beam from start to finish all provides accurate blood glucose fluctuating change trend at any time and specific blood glucose value for patient, to suffer from
Person preferably manages own blood glucose.For different patients, the study that this method can be adaptive with different patients,
Blood glucose calibration model is obtained by inductive learning.
Below by way of specific example, the present invention will be described:
(1) firstly, the blood glucose meter of sustainable monitoring is worn on the body of healthy aglycosuria patient, certain particular time such as
Blood glucose value is detected after meal, for examining this blood glucose meter for the correctness of change of blood sugar reflection trend, if variation tendency and normal
Desired value is not obviously inconsistent, then the meaning that this blood glucose meter does not correct, if variation tendency is consistent with desired value, the detection of this blood glucose meter
Data can be used as reference data in subsequent artefacts' intelligence timing.
(2) then, after the sustainable blood glucose meter for obtaining correction meaning, this blood glucose meter is worn on diabetic
3-4 days, and periodically read this and glucose value daily from this blood glucose meter.
(3) then, training of enough exact values for artificial intelligence is read in 3-4 days, because with the increasing of time
It is long, measure data can increasingly be not allowed, therefore manually intelligence corrects sustainable blood glucose meter and examines data measured.
(4) for artificial intelligence using autocoder as calibration model, variation autocoder conduct is unsupervised in recent years
The new hot spot of deep learning, main be characterized by are introduced into the probability distribution that probability interpretation is come in learning data.For
The problem of blood glucose level data corrects, since the data of Different Individual have a certain difference, it is assumed that it meets certain probability
Distribution is rationally, such as normal distribution.Therefore fitting is corrected to blood glucose level data using variation autocoder, compared to tradition
Autocoder model, output result can be advanced optimized.
(5) finally, the 6th day or so time point, the same of glucose value is measured with the blood glucose meter of this sustainable monitoring
When, it is drawn blood using finger tip and measures the blood glucose value of the patient, by the volume blood glucose of the blood glucose value and manually intelligent correction of finger tip blood drawing
Value compares, and for verifying the accuracy of artificial intelligence technology, if error is larger, then plans or improve again calibration model.
It is all that the present invention is not departed to various changes or modifications of the invention the invention is not limited to above embodiment
Spirit and scope, if these modification and variations belong within the scope of claim and equivalent technologies of the invention, then this hair
It is bright to also imply that comprising these modification and variations.
Claims (4)
1. artificial intelligence deep learning method corrects the monitoring method for continuing blood glucose, comprise the concrete steps that:
(1) the accurate blood glucose value that single patient persistently changes in for a period of time is collected: first with sustainable blood sugar monitoring instrument
Glucose monitoring techniques collect the blood glucose value of consecutive variations inaccuracy in a period of time from single patient；
(2) accurate blood glucose value is measured by carrying out the blood sugar monitoring methods of multiple finger tip blood sampling；
(3) blood glucose value measured by lasting Russia's blood sugar monitoring instrument is corrected with resulting accurate blood glucose value；
(4) by carrying out blood glucose level data acquisition with a large amount of patients, by the lasting blood sugar monitoring of collected a large amount of patients
Result before Data correction and after correction compares, and using comparing result as sample, these data are carried out depth ergonomics
Practise correction；
(5) artificial intelligence autocoder technology, depth are utilized by the collected data using sustainable detection blood glucose meter
Study, multilayer neural network method are trained, and the result obtained after training is used further to correct sustainable blood sugar monitoring instrument survey
Data out；
(6) it is made to identify the rule of the difference of result of the data for continuing blood sugar monitoring instrument before correction with patient and after correction
With mode, the degree for obtaining blood glucose value acquired in sustainable glucose monitoring techniques and practical blood glucose value difference is regular, and by this
The numerical value diversity mode that artificial intelligence approach is identified summarizes blood glucose correction mode；
(7) after artificial intelligence training finishes, the sustainable blood sugar monitoring, and utilize the obtained numerical value of artificial intelligence training
Bearing calibration is corrected the blood glucose value recorded.
2. artificial intelligence deep learning method according to claim 1 corrects the monitoring method for continuing blood glucose, feature exists
In: above-mentioned steps read continuous data in (1) from single patient, acquire 800-1000 blood glucose level data daily, hold
It is 0.5-10 days continuous.
3. artificial intelligence deep learning method according to claim 1 corrects the monitoring method for continuing blood glucose, feature exists
In: number of taking a blood sample daily in above-mentioned steps (2) is 2-10 times.
4. artificial intelligence deep learning method according to claim 1 corrects the monitoring method for continuing blood glucose, feature exists
In: the numberical range of a large amount of patients is 100-100000 in above-mentioned steps (4).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710370218.XA CN108937954A (en) | 2017-05-23 | 2017-05-23 | Artificial intelligence deep learning method corrects the monitoring method for continuing blood glucose |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710370218.XA CN108937954A (en) | 2017-05-23 | 2017-05-23 | Artificial intelligence deep learning method corrects the monitoring method for continuing blood glucose |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108937954A true CN108937954A (en) | 2018-12-07 |
Family
ID=64494176
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710370218.XA Pending CN108937954A (en) | 2017-05-23 | 2017-05-23 | Artificial intelligence deep learning method corrects the monitoring method for continuing blood glucose |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108937954A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111401378A (en) * | 2020-03-23 | 2020-07-10 | 北京糖护科技有限公司 | Glucometer display screen information identification method and system based on OCR |
| CN111588384A (en) * | 2020-05-27 | 2020-08-28 | 京东方科技集团股份有限公司 | Method, device and equipment for obtaining blood sugar detection result |
| CN112438704A (en) * | 2019-08-31 | 2021-03-05 | 深圳硅基传感科技有限公司 | Calibration system of physiological parameter monitor |
| CN117373586A (en) * | 2023-08-28 | 2024-01-09 | 北京华益精点生物技术有限公司 | Blood glucose data comparison method and related equipment |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101915744A (en) * | 2010-07-05 | 2010-12-15 | 北京航空航天大学 | Near infrared spectrum nondestructive testing method and device for material component content |
-
2017
- 2017-05-23 CN CN201710370218.XA patent/CN108937954A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101915744A (en) * | 2010-07-05 | 2010-12-15 | 北京航空航天大学 | Near infrared spectrum nondestructive testing method and device for material component content |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112438704A (en) * | 2019-08-31 | 2021-03-05 | 深圳硅基传感科技有限公司 | Calibration system of physiological parameter monitor |
| CN112438704B (en) * | 2019-08-31 | 2024-02-23 | 深圳硅基传感科技有限公司 | Calibration system of physiological parameter monitor |
| CN111401378A (en) * | 2020-03-23 | 2020-07-10 | 北京糖护科技有限公司 | Glucometer display screen information identification method and system based on OCR |
| CN111588384A (en) * | 2020-05-27 | 2020-08-28 | 京东方科技集团股份有限公司 | Method, device and equipment for obtaining blood sugar detection result |
| CN111588384B (en) * | 2020-05-27 | 2023-08-22 | 京东方科技集团股份有限公司 | Method, device and equipment for obtaining blood glucose detection result |
| CN117373586A (en) * | 2023-08-28 | 2024-01-09 | 北京华益精点生物技术有限公司 | Blood glucose data comparison method and related equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bolinder et al. | Self-monitoring of blood glucose in type I diabetic patients: comparison with continuous microdialysis measurements of glucose in subcutaneous adipose tissue during ordinary life conditions | |
| CN108937955A (en) | The adaptive wearable blood glucose bearing calibration of personalization and its means for correcting based on artificial intelligence | |
| Maran et al. | Continuous subcutaneous glucose monitoring in diabetic patients: a multicenter analysis | |
| US6998247B2 (en) | Method and apparatus using alternative site glucose determinations to calibrate and maintain noninvasive and implantable analyzers | |
| CN108937954A (en) | Artificial intelligence deep learning method corrects the monitoring method for continuing blood glucose | |
| Abd Salam et al. | The evolution of non-invasive blood glucose monitoring system for personal application | |
| EP1309271B1 (en) | Device for prediction of hypoglycemic events | |
| WO2018107915A1 (en) | General noninvasive blood glucose prediction method based on timing analysis | |
| RU2477078C2 (en) | Method, system and software product for estimating changeability of glucose content in blood in case of diabetes by self-control data | |
| Li et al. | Non-invasive monitoring of three glucose ranges based on ECG by using DBSCAN-CNN | |
| US20100324398A1 (en) | Non-invasive characterization of a physiological parameter | |
| CA2622986A1 (en) | Non-invasive glucose monitoring | |
| Tronstad et al. | Non-invasive prediction of blood glucose trends during hypoglycemia | |
| US6949070B2 (en) | Non-invasive blood glucose monitoring system | |
| Chen et al. | Non-invasive blood glucose measurement of 95% certainty by pressure regulated Mid-IR | |
| CN101138500A (en) | Non-invasive blood-sugar detecting instrument based on conservation of energy | |
| Javier et al. | Portable non-invasive glucometer using near-infrared sensor and raspberry Pi | |
| CN112568902A (en) | Noninvasive blood glucose calibration method based on blood glucose value | |
| Albalat et al. | Non-invasive blood glucose sensor: A feasibility study | |
| Wei et al. | Noninvasive glucose evaluation by human skin oxygen saturation level | |
| CN116035571A (en) | Intelligent noninvasive glucometer | |
| CN108852367A (en) | Blood extracorporeal circulation formula high-precision blood glucose continuous monitor device based on light microfluid | |
| CN204318765U (en) | A kind of infrared spectrum technology non-invasive blood-sugar detecting instrument | |
| WO2020029898A1 (en) | Method for non-invasive blood glucose monitoring | |
| Amaral et al. | Effects of glucose in blood and skin impedance spectroscopy |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181207 |