CN113679348B - Blood glucose prediction method, blood glucose prediction device, apparatus, and storage medium - Google Patents
Blood glucose prediction method, blood glucose prediction device, apparatus, and storage medium Download PDFInfo
- Publication number
- CN113679348B CN113679348B CN202110991121.7A CN202110991121A CN113679348B CN 113679348 B CN113679348 B CN 113679348B CN 202110991121 A CN202110991121 A CN 202110991121A CN 113679348 B CN113679348 B CN 113679348B
- Authority
- CN
- China
- Prior art keywords
- data
- blood glucose
- medical record
- record information
- time difference
- 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 312
- 210000004369 blood Anatomy 0.000 title claims abstract description 312
- 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 213
- 239000008103 glucose Substances 0.000 title claims abstract description 213
- 238000000034 method Methods 0.000 title claims abstract description 40
- 201000010099 disease Diseases 0.000 claims abstract description 88
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 88
- 239000003814 drug Substances 0.000 claims abstract description 77
- 229940079593 drug Drugs 0.000 claims abstract description 75
- 230000002159 abnormal effect Effects 0.000 claims abstract description 39
- 238000012216 screening Methods 0.000 claims abstract description 19
- 230000002554 disease preventive effect Effects 0.000 claims abstract description 11
- 230000011218 segmentation Effects 0.000 claims description 47
- 230000015654 memory Effects 0.000 claims description 27
- 235000005911 diet Nutrition 0.000 claims description 20
- 238000005259 measurement Methods 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 16
- 230000037213 diet Effects 0.000 claims description 15
- 230000006806 disease prevention Effects 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 12
- 230000007787 long-term memory Effects 0.000 claims description 11
- 230000006403 short-term memory Effects 0.000 claims description 10
- 235000015097 nutrients Nutrition 0.000 claims description 9
- 230000003449 preventive effect Effects 0.000 claims description 8
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 230000000378 dietary effect Effects 0.000 claims description 5
- 238000012512 characterization method Methods 0.000 claims 3
- 238000013473 artificial intelligence Methods 0.000 abstract description 8
- 206010012601 diabetes mellitus Diseases 0.000 description 18
- 238000005516 engineering process Methods 0.000 description 11
- 208000024891 symptom Diseases 0.000 description 11
- 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 8
- 238000012549 training Methods 0.000 description 8
- 208000019901 Anxiety disease Diseases 0.000 description 7
- 206010044565 Tremor Diseases 0.000 description 7
- 230000036506 anxiety Effects 0.000 description 7
- 238000004140 cleaning Methods 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 201000001421 hyperglycemia Diseases 0.000 description 6
- 230000007246 mechanism Effects 0.000 description 6
- 238000001514 detection method Methods 0.000 description 5
- 235000003642 hunger Nutrition 0.000 description 5
- 235000012054 meals Nutrition 0.000 description 5
- 230000027939 micturition Effects 0.000 description 5
- 230000000291 postprandial effect Effects 0.000 description 5
- 102000004877 Insulin Human genes 0.000 description 4
- 108090001061 Insulin Proteins 0.000 description 4
- 230000005856 abnormality Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 229940125396 insulin Drugs 0.000 description 4
- 230000037351 starvation Effects 0.000 description 4
- 201000004384 Alopecia Diseases 0.000 description 3
- 206010033425 Pain in extremity Diseases 0.000 description 3
- 231100000360 alopecia Toxicity 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000002641 glycemic effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 210000001519 tissue Anatomy 0.000 description 3
- 238000011282 treatment Methods 0.000 description 3
- 208000024172 Cardiovascular disease Diseases 0.000 description 2
- 206010010071 Coma Diseases 0.000 description 2
- 208000013016 Hypoglycemia Diseases 0.000 description 2
- 240000007594 Oryza sativa Species 0.000 description 2
- 235000007164 Oryza sativa Nutrition 0.000 description 2
- 239000003472 antidiabetic agent Substances 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 208000026106 cerebrovascular disease Diseases 0.000 description 2
- 230000001684 chronic effect Effects 0.000 description 2
- 238000013145 classification model Methods 0.000 description 2
- 230000002526 effect on cardiovascular system Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 235000013305 food Nutrition 0.000 description 2
- 230000002218 hypoglycaemic effect Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 210000000056 organ Anatomy 0.000 description 2
- 102000004169 proteins and genes Human genes 0.000 description 2
- 108090000623 proteins and genes Proteins 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 235000009566 rice Nutrition 0.000 description 2
- 208000011580 syndromic disease Diseases 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 240000007087 Apium graveolens Species 0.000 description 1
- 235000015849 Apium graveolens Dulce Group Nutrition 0.000 description 1
- 235000010591 Appio Nutrition 0.000 description 1
- 238000012935 Averaging Methods 0.000 description 1
- 206010008132 Cerebral thrombosis Diseases 0.000 description 1
- 240000008067 Cucumis sativus Species 0.000 description 1
- 235000010799 Cucumis sativus var sativus Nutrition 0.000 description 1
- 208000007342 Diabetic Nephropathies Diseases 0.000 description 1
- 208000032131 Diabetic Neuropathies Diseases 0.000 description 1
- 206010012689 Diabetic retinopathy Diseases 0.000 description 1
- 241000233866 Fungi Species 0.000 description 1
- 201000001429 Intracranial Thrombosis Diseases 0.000 description 1
- 235000009811 Momordica charantia Nutrition 0.000 description 1
- 208000022329 Protein metabolism disease Diseases 0.000 description 1
- 206010037660 Pyrexia Diseases 0.000 description 1
- 244000300264 Spinacia oleracea Species 0.000 description 1
- 235000009337 Spinacia oleracea Nutrition 0.000 description 1
- 244000078912 Trichosanthes cucumerina Species 0.000 description 1
- 235000008322 Trichosanthes cucumerina Nutrition 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 235000021152 breakfast Nutrition 0.000 description 1
- 150000001720 carbohydrates Chemical class 0.000 description 1
- 235000014633 carbohydrates Nutrition 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 208000033679 diabetic kidney disease Diseases 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 208000002173 dizziness Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 210000001508 eye Anatomy 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 210000002216 heart Anatomy 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 230000003914 insulin secretion Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 210000003734 kidney Anatomy 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 208000027905 limb weakness Diseases 0.000 description 1
- 231100000861 limb weakness Toxicity 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 208000030159 metabolic disease Diseases 0.000 description 1
- 235000013336 milk Nutrition 0.000 description 1
- 239000008267 milk Substances 0.000 description 1
- 210000004080 milk Anatomy 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000011369 optimal treatment Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000002040 relaxant effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 235000013311 vegetables Nutrition 0.000 description 1
- 230000000007 visual effect Effects 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
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
-
- 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
- A61B5/72—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
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- 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
- A61B5/7271—Specific aspects of physiological measurement analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- General Health & Medical Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Molecular Biology (AREA)
- Medical Informatics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Signal Processing (AREA)
- Psychiatry (AREA)
- Physiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Emergency Medicine (AREA)
- Optics & Photonics (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The present application relates to the field of artificial intelligence, and in particular, to a method, apparatus, device, and storage medium for predicting blood glucose, where the method includes: acquiring medical record information, blood sugar data and medication data of a user; screening the medical record information to obtain medical record information corresponding to abnormal blood sugar values, and taking the medical record information corresponding to the abnormal blood sugar values as target medical record information; inputting the target medical record information, the blood glucose data and the medication data into a pre-trained blood glucose prediction model to obtain blood glucose trend data; acquiring state characteristic data of a user, and predicting diseases according to the blood sugar trend data and the state characteristic data to obtain expected diseases; and determining corresponding disease preventive measures according to the state characteristic data and the expected diseases, and pushing the disease preventive measures to a target terminal. Therefore, the user can be reminded to prevent possible diseases, and the user experience is improved.
Description
Technical Field
The present application relates to the field of artificial intelligence, and in particular, to a blood glucose prediction method, a blood glucose prediction device, a computer device, and a storage medium.
Background
Diabetes is a high-incidence disease at present, particularly for the aged, people are always threatening the health and life of people, and although the diabetes treatment technology is rapidly developed, many people cannot detect blood sugar regularly, so that many people cannot detect diabetes, and meanwhile, the diabetes is not treated timely, so that the condition is easy to worsen, and the situation of coma and death is caused.
Today, users usually go to hospitals actively to review or spontaneously to measure blood sugar, and when the measured blood sugar value is high, the users often find that the users may suffer from diabetes. The existing blood sugar prediction model only predicts through real-time blood sugar data of a user, and does not consider the influence of factors such as insulin injection, blood sugar reducing medicine administration or past morbidity record on future blood sugar trend, so that the blood sugar prediction result is inaccurate, diseases such as diabetes cannot be prevented in advance, and the user misses the optimal treatment opportunity.
Disclosure of Invention
The application provides a blood sugar prediction method, a blood sugar prediction device, computer equipment and a storage medium, which aim to solve the problems that a user cannot perceive own illness, treatment is not timely carried out, illness state is easy to deteriorate, and therefore coma is caused and death is even caused.
To achieve the above object, the present application provides a blood glucose prediction method, the method comprising:
acquiring medical record information, blood sugar data and medication data of a user;
screening the medical record information to obtain medical record information corresponding to abnormal blood sugar values, and taking the medical record information corresponding to the abnormal blood sugar values as target medical record information;
inputting the target medical record information, the blood glucose data and the medication data into a pre-trained blood glucose prediction model to obtain blood glucose trend data;
acquiring state characteristic data of a user, and predicting diseases according to the blood sugar trend data and the state characteristic data to obtain expected diseases;
and determining corresponding disease preventive measures according to the state characteristic data and the expected diseases, and pushing the disease preventive measures to a target terminal.
To achieve the above object, the present application also provides a blood glucose prediction device, including:
the data acquisition module is used for acquiring medical record information, blood sugar data and medication data of a user;
the medical record information screening module is used for screening the medical record information to obtain medical record information corresponding to the abnormal blood glucose value, and taking the medical record information corresponding to the abnormal blood glucose value as target medical record information;
The blood sugar prediction module is used for inputting the target medical record information, the blood sugar data and the medication data into a pre-trained blood sugar prediction model to obtain blood sugar trend data;
the disease prediction acquisition module is used for acquiring state characteristic data of a user, and carrying out disease prediction according to the blood glucose trend data and the state characteristic data to obtain expected diseases;
and the preventive measure generation module is used for determining corresponding preventive measures of diseases according to the state characteristic data and the expected diseases and pushing the preventive measures of the diseases to a target terminal.
In addition, to achieve the above object, the present application further provides a computer apparatus including a memory and a processor; the memory is used for storing a computer program; the processor is configured to execute the computer program and implement any one of the blood glucose prediction methods provided in the embodiments of the present application when the computer program is executed.
In addition, to achieve the above object, the present application further provides a computer readable storage medium storing a computer program, where the computer program when executed by a processor causes the processor to implement the blood glucose prediction method provided in any one of the embodiments of the present application.
According to the blood glucose prediction method, the blood glucose prediction device, the equipment and the storage medium disclosed by the embodiment of the application, the blood glucose value of the user is monitored according to the medical record data of the user and the real-time blood glucose value data, and the future blood glucose trend is predicted, so that the user is reminded to prevent diseases possibly happening, and the user experience is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic view of a scenario of a blood glucose prediction method provided in an embodiment of the present application;
fig. 2 is a schematic flow chart of a blood glucose prediction method according to an embodiment of the present application;
FIG. 3 is a schematic block diagram of a blood glucose prediction device according to an embodiment of the present application;
fig. 4 is a schematic block diagram of a computer device 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 evident that the embodiments described are some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
The flow diagrams depicted in the figures are merely illustrative and not necessarily all of the elements and operations/steps are included or performed in the order described. For example, some operations/steps may be further divided, combined, or partially combined, so that the order of actual execution may be changed according to actual situations. In addition, although the division of the functional modules is performed in the apparatus schematic, in some cases, the division of the modules may be different from that in the apparatus schematic.
The term "and/or" as used in this specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
The sugar in blood is called blood glucose, which is the most common. The energy required for the activity of tissue cells in the body is mostly derived from glucose, so that the blood glucose must be maintained at a certain level to maintain the needs of the organs and tissues in the body. The concentration of fasting blood sugar of normal people is 3.9-6.1 mmol/L. Fasting blood glucose concentrations exceeding 7.0mmol/L are referred to as hyperglycemia.
Diabetes also often occurs once hyperglycemia is detected. Diabetes is a group of metabolic diseases characterized by chronic hyperglycemia due to multiple causes, due to defects in insulin secretion and/or utilization. Long-term carbohydrate, fat, protein metabolism disorders can cause multisystem damage, resulting in chronic progressive lesions, hypofunction and failure of tissues and organs such as eyes, kidneys, nerves, heart, blood vessels, etc. Monitoring and prediction of blood glucose levels is therefore particularly important for the prevention of diabetes.
In general, hyperglycemia is liable to cause diabetes mellitus, and also causes complications such as diabetic nephropathy, diabetic peripheral neuropathy, diabetic retinopathy, etc.; the blood sugar is too low to cause cardiovascular and cerebrovascular diseases, so that the blood sugar of a user is controlled in a good range, and complications of the user are not easy to occur.
In order to solve the problems, the application provides a blood glucose prediction method which is applied to a server, so that the blood glucose of a user can be monitored according to medical record data of the user and real-time blood glucose data and future blood glucose trend is predicted, thereby reminding the user to prevent diseases which can occur and improving the use experience of the user.
The terminal device may include a fixed terminal such as a cell phone, tablet computer, personal digital assistant (Personal Digital Assistant, PDA), etc. The servers may be, for example, individual servers or a cluster of servers. However, for the sake of easy understanding, the following embodiments will be described in detail with respect to a blood glucose prediction method applied to a server.
Some embodiments of the present application are described in detail below with reference to the accompanying drawings. The following embodiments and features of the embodiments may be combined with each other without conflict.
As shown in fig. 1, the blood glucose prediction method provided in the embodiment of the present application may be applied to an application environment as shown in fig. 1. The application environment includes a terminal device 110 and a server 120, where the terminal device 110 may communicate with the server 120 through a network. Specifically, the server 120 acquires medical record information, blood glucose data and medication data at the current moment and status feature data sent by the terminal device 110, and the server 120 determines future blood glucose trends, expected diseases and disease preventive measures according to the acquired data and sends the future blood glucose trends, expected diseases and disease preventive measures to the terminal device 110 so as to remind a user of disease prevention. The server 120 may be an independent server, or may be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms. The terminal device 110 may be, but is not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, etc. The terminal and the server may be directly or indirectly connected through wired or wireless communication, which is not limited herein.
Referring to fig. 2, fig. 2 is a schematic flowchart of a blood glucose prediction method according to an embodiment of the present application. The blood glucose prediction method can be applied to a server, so that the blood glucose of a user can be monitored according to medical record data of the user and real-time blood glucose data, future blood glucose trend is predicted, the user is reminded to prevent diseases possibly occurring, and user experience is improved.
As shown in fig. 2, the blood glucose prediction method includes steps S101 to S105.
S101, acquiring medical record information, blood sugar data and medication data of a user.
The medical record information of the user is illness state information recorded when the user performs the doctor seeing, and specifically can be electronic medical record information, including various diseases of the user and corresponding doctor seeing records. The server can be in communication connection with the hospital system, so that medical record information of the user can be obtained from the hospital system, or the user can input the medical record information from the terminal device. The blood glucose data is a blood glucose value measured by the user last time, and may specifically include a fasting blood glucose value of the user, which refers to a blood glucose value measured in a state in which no caloric food is consumed for at least 8 hours or more, a blood glucose value of two hours after meal, a random blood glucose value, and a historical blood glucose value. The random blood glucose level is a blood glucose level at any time regardless of the meal time, and can be measured at any time before or after a meal. The historical blood glucose values are fasting blood glucose values of the user for the last few days (e.g., the last 3 days), two hours postprandial blood glucose values, and random blood glucose values. The medication data is hypoglycemic drugs or insulin and the like which are taken by the user last time.
The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Among these, artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use knowledge to obtain optimal results.
Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.
In some embodiments, after blood glucose data is obtained, cleaning the blood glucose data of the user to obtain cleaned blood glucose data; and carrying out smooth denoising treatment on the cleaned data to obtain treated blood sugar data, and taking the treated blood sugar data as blood sugar data input into the pre-trained blood sugar prediction model. Standard blood glucose data can thus be obtained and used as input to a pre-trained blood glucose prediction model.
Specifically, when data cleaning is performed on blood glucose data of a user, abnormal detection is performed on the blood glucose data of the user, so that corresponding abnormal blood glucose data is obtained, cleaning processing is performed on the abnormal blood glucose data, and cleaned blood glucose data is obtained. The cleaned blood glucose data is subjected to smooth denoising by using Kalman filtering, and the cleaned data set is subjected to smooth denoising by using Kalman filtering, so that smaller time delay can be ensured, and meanwhile, the high-frequency noise is obviously reduced.
In some embodiments, the cleaning processing of the blood glucose data of the user is performed by performing abnormality detection on each blood glucose data, and if there is an abnormal blood glucose value in the blood glucose data, performing cleaning processing on the abnormal blood glucose value to obtain cleaned blood glucose data.
Specifically, abnormality detection is performed on each of the blood glucose data; if the blood glucose data has abnormal blood glucose values, cleaning the abnormal blood glucose values to obtain cleaned blood glucose data; and if the blood glucose data does not have abnormal blood glucose values, the blood glucose data is considered to be blood glucose data after cleaning.
The abnormal detection of each blood sugar data is determined by detecting whether the blood sugar difference value of any two data points in the same group of blood sugar data exceeds a preset blood sugar difference value threshold value; if the blood glucose difference value of any two data points in the same group of blood glucose data exceeds a preset blood glucose difference value threshold value, the group of blood glucose data is considered to be abnormal; if the blood glucose difference value of any two data points in the same group of blood glucose data does not exceed the preset blood glucose difference value threshold value, the group of blood glucose data is considered to have no abnormality. Wherein, the same group of blood glucose data can be fasting blood glucose values, two hours postprandial blood glucose values and random blood glucose values collected on the same day. The preset blood glucose difference threshold may be any value.
By way of example, if within the same set of glycemic data, fasting blood glucose values are 3.4mmol/L, postprandial two hours blood glucose values are 11.5mmol/L, and normoglycemia is normalized by fasting 3.9-6.1mmol/L, postprandial 2 hours blood glucose < 7.8mmol/L, the set of glycemic data is abnormal if the preset glycemic difference threshold is 4 mmol/L. And at this time, the fasting blood glucose level showed hypoglycemia, and the blood glucose level of two hours after meal showed hyperglycemia, which did not conform to the conventional, and therefore, it was confirmed that the blood glucose data of the group had abnormality.
S102, screening the medical record information to obtain medical record information corresponding to the abnormal blood glucose level, and taking the medical record information corresponding to the abnormal blood glucose level as target medical record information.
The target medical record information is related medical record information corresponding to abnormal blood sugar levels, for example, if symptoms such as frequent urination, starvation, anxiety and tremors occur, the medical record information corresponding to the symptoms is the target medical record information when the measured blood sugar levels are abnormal, or the medical record information corresponding to the abnormal blood sugar levels is directly obtained from the medical record information, and the medical record information corresponding to the abnormal blood sugar levels is taken as the target medical record information.
Because the medical record information of the user comprises a lot of information which is irrelevant to complications caused by hyperglycemia or hypoglycemia, for example, the user burns to a hospital for a certain time to see the doctor, symptoms such as dizziness, limb weakness and the like are recorded in the medical record, but obviously, the symptoms caused by fever are not related with the target medical record information required by the blood glucose prediction method of the application, or the medical record information comprises a lot of symptoms, but the blood glucose value is not abnormal, so that the information needs to be screened.
Meanwhile, the medical record information of the user may further comprise blood sugar data in the illness state and medication data in the illness state, wherein the blood sugar data in the illness state are fasting blood sugar values, postprandial blood sugar values and random blood sugar values in the illness state of the user, and the medication data in the illness state are hypoglycemic drugs or insulin and the like taken in the illness state of the user. The data can be obtained by screening medical record information of the user, and the data is input into a pre-trained blood glucose prediction model to conduct blood glucose prediction.
In some embodiments, based on a word segmentation algorithm, word segmentation processing is performed on the medical record information to obtain a word segmentation result corresponding to the medical record information; and screening the word segmentation result to obtain target medical record information. The word segmentation algorithm can be a word segmentation algorithm based on a hidden Markov model, a word segmentation algorithm based on a conditional random field and the like. Therefore, the related medical record information causing the rise or the fall of blood sugar can be obtained by firstly performing word segmentation processing on the medical record information and then screening the word segmentation result and is used as input of a blood sugar prediction model.
Specifically, based on a preset medical knowledge base, word segmentation is carried out on the medical record information to obtain a plurality of word segmentation results; wherein the preset medical knowledge base comprises standard names, hyponyms, association syndrome words and the like.
For example, the medical record information may be segmented based on a knowledge base (preset medical knowledge base) corresponding to the blood sugar complications, where the knowledge base includes standard names, hyponyms, association-syndrome words, and the like corresponding to the blood sugar complications, so as to obtain a segmentation result.
For example, word segmentation is performed on one medical record information to obtain corresponding related symptom words including frequent urination, hunger, anxiety, tremor, foot pain, alopecia and the like, and then word segmentation results are screened, so that it is obvious that the two symptoms of foot pain and alopecia have a small relation with complications corresponding to abnormal blood sugar, and therefore the two symptoms are screened out to obtain target medical record information.
In some embodiments, the word segmentation result is screened to obtain the target medical record information, and word sense prediction is performed on each word segment in the word segmentation result based on a medical word sense prediction model to obtain a word sense prediction result corresponding to each word segment, and the word segmentation result is screened based on the word sense prediction result to obtain the target medical record information. Thus, medical record information related to complications caused by abnormal blood sugar can be rapidly screened and obtained.
The medical word sense prediction model is used for predicting the similarity degree of word segmentation results and standard medical word segmentation, the medical word sense prediction model is obtained by training a semantic matching model and a standard medical word segmentation database, the semantic prediction model can comprise an LSTM matching model, an MV-DSSM model, an ESIM model and the like, the standard medical word segmentation database is used for storing a database of standard medical word segmentation, the standard medical word segmentation is medical word segmentation related to complications caused by abnormal blood sugar, and the word sense prediction results are the similarity degree of each word segmentation and the standard medical word segmentation in the standard medical word segmentation database.
Specifically, medical word segmentation matching can be performed on a standard medical word segmentation database corresponding to the blood sugar complications through a medical word sense prediction model, the similarity of each word segmentation and the standard medical word segmentation in the standard medical word segmentation database is calculated, and each word sense prediction result is ranked according to the similarity to obtain a ranking result; and screening the word segmentation result based on the sequencing result to obtain standard medical data.
For example, if the word segmentation result includes frequent urination, starvation, anxiety, tremor, foot pain, alopecia, etc., the frequent urination, starvation, anxiety, tremor are common symptoms of abnormal blood sugar, so that the frequent urination, starvation, anxiety, tremor have high similarity, and the anxiety and tremor do not belong to common symptoms of abnormal blood sugar, so that the anxiety and tremor have low similarity, so that the word segmentation with low similarity may be screened in subsequent screening, so as to obtain the word segmentation corresponding to the complications of blood sugar, and medical record information including the word segmentation is used as target medical record information.
S103, inputting the target medical record information, the blood glucose data and the medication data into a pre-trained blood glucose prediction model to obtain blood glucose trend data.
The pre-trained blood sugar prediction model is used for predicting future blood sugar trend of a user and is used for laying a cushion for subsequent disease diagnosis.
In some embodiments, determining a time of illness in the target medical record information, a time of measurement of the blood glucose data, and a time of administration of the administration data, respectively; determining weight proportions corresponding to the target medical record information, the blood glucose data and the medication data according to the illness time, the measurement time and the medication time; and inputting the target medical record information, the blood glucose data, the medication data and the weight proportion corresponding to each to a pre-trained blood glucose prediction model to obtain blood glucose trend data. The disease time is the occurrence time corresponding to symptoms in the target medical record information, the measurement time is the acquisition time corresponding to each blood glucose value, and the administration time is the time corresponding to taking medicine or injecting insulin. Therefore, higher weight proportion can be allocated to the blood glucose data with adjacent date, and accuracy of blood glucose prediction is improved.
By way of example, if it is determined that the patient time in the target medical record information is 7 months 21 days-7 months 27 days, the measurement time of the blood glucose data is 8 months 3 days, and the medication time of the medication data is 8 months 1 day; the longer the time interval is, the lower the credibility of the data is, so the lower the weight proportion is allocated to the data with the longer time interval, and the higher the weight proportion is allocated to the data with the shorter time interval. Therefore, the weight ratios corresponding to the target medical record information, the blood glucose data and the medication data can be determined to be 20%, 50% and 30% respectively.
In some embodiments, differences of the disease time, the measurement time, the medication time and the data acquisition time are respectively determined to obtain corresponding first time difference, second time difference and third time difference; determining a weight proportion corresponding to the target medical record information according to the first time difference; determining a weight proportion corresponding to the blood glucose data according to the second time difference; and determining the weight proportion corresponding to the medication data according to the third time difference. Therefore, the time interval of each data can be determined through the time difference, the credibility of the data is determined, a higher weight proportion is allocated to the blood sugar data with adjacent date, and the accuracy of blood sugar prediction is improved.
The data acquisition time is the time corresponding to the medical record information of the user, the blood sugar data of the user and the medication data acquired by the server. The first time difference is the difference between the illness time and the time corresponding to the medical record information of the user acquired by the server, the second time difference is the difference between the measurement time and the time corresponding to the blood sugar data of the user acquired by the server, and the third time difference is the difference between the medication time and the time corresponding to the medication data of the user acquired by the server.
For example, if the first time difference is 24h, the second time difference is 14h, and the third time difference is 18h, it can be seen that the second time difference is the shortest, that is, the reliability of the blood glucose data is the highest, and the reliability of the target medical record information is the lowest, so that the weight ratios corresponding to the target medical record information, the blood glucose data, and the medication data can be determined to be 20%, 50%, and 30%, respectively.
In some embodiments, differences of the disease time, the measurement time, the medication time and the data acquisition time are respectively determined to obtain corresponding first time difference, second time difference and third time difference; calculating to obtain a total time difference according to the first time difference, the second time difference and the third time difference; determining the ratio of the first time difference to the total time difference, and determining the weight proportion corresponding to the target medical record information according to the ratio of the first time difference to the total time difference; determining the ratio of the second time difference to the total time difference, and determining the weight proportion corresponding to the blood glucose data according to the ratio of the second time difference to the total time difference; and determining the ratio of the third time difference to the total time difference, and determining the weight proportion corresponding to the medication data according to the ratio of the third time difference to the total time difference.
For example, if the first time difference is 1h, the second time difference is 4h, and the third time difference is 5h, the total time difference is calculated to be 10h, the ratio of the first time difference to the total time difference is determined to be 10%, the ratio of the second time difference to the total time difference is determined to be 40%, and the ratio of the first time difference to the total time difference is determined to be 50%, so that the weight ratios corresponding to the target medical record information, the blood glucose data and the medication data can be determined to be 10%, 40% and 50%, respectively.
In some embodiments, medical record information, blood glucose data and medication data of a target user are obtained, and the medical record information, the blood glucose data and the medication data of the target user are used as a training set; and training a preset blood sugar prediction model through the training set to obtain a pre-trained blood sugar prediction model. The preset blood sugar prediction model is an untrained blood sugar prediction model, and the target user is a user with diabetes or a user corresponding to complications caused by too high and too low blood sugar. Specifically, the preset blood glucose prediction model comprises a recurrent neural network based on a long-short-term memory model, the neural network can learn long-term dependency, and meanwhile, a storage unit in the long-short-term memory model can help the neural network to combine stored memory and data, so that the prediction accuracy is improved. And because a two-way long-short-term memory model is used, the neural network can refer to outdated data and infer future data, so that training time is accelerated.
Specifically, medical record information, blood sugar data and medication data of a target user are obtained, the data are divided into a training set, a verification set and a test set, a preset blood sugar prediction model is trained through the training set, so that converted medical record information, blood sugar data and medication data are learned, and parameters of the blood sugar prediction model are adjusted through the verification set, such as selecting the number of hidden units in a neural network and determining parameters of network structure or control model complexity. Finally, the resolution capability of the trained model is tested through the test set, such as the recognition rate and the like, so that the performance of the test is achieved, and the optimal model is selected as a target detection model. The neural network can be trained through the open source data set, so that a corresponding and trained blood glucose prediction model is obtained.
In some embodiments, the target medical record information, the blood glucose data and the medication data are input into a first long-short-term memory network in a pre-trained blood glucose prediction model to be operated, so as to obtain a hidden state vector sequence in the first long-short-term memory network, wherein the blood glucose prediction model comprises a first long-short-term memory network for encoding and a second long-short-term memory network for decoding; and inputting the hidden state vector sequence into the second long-term and short-term memory network for operation to obtain blood sugar trend data.
Wherein the second long-short term memory model is a recurrent neural network model based on an attention mechanism. The first long-short-term memory model is different from the second long-short-term memory model in that a concentration mechanism is added into the second long-short-term memory model. Attention mechanisms refer to the human visual attention mechanisms, which are brain signal processing mechanisms specific to human vision.
Specifically, determining an influence factor corresponding to a future trend of blood sugar according to the target medical record information, the blood sugar data and the medication data; inputting the influence factors corresponding to the future trend of the blood sugar into a first long-term and short-term memory network in a blood sugar prediction model for operation to obtain a hidden state vector sequence in the first long-term and short-term memory network. Inputting the hidden state vector sequence into the second long-short-term memory network for operation, so as to obtain a high-dimensional vector sequence output by the second long-short-term memory network; and according to the corresponding relation between the preset component vector and the meaning of the prediction result, the high-dimensional vector sequence is interpreted, so that blood glucose trend data in different time periods in the future are obtained.
And the target medical record information, the blood sugar data and the influence factors corresponding to the future trend of the blood sugar are used as input of a long-term and short-term memory model. The hidden state vector sequence can be used as a decoding basis of a second long-short-term memory network, and the second long-term memory network outputs a high-dimensional vector sequence which represents the prediction results of different time periods, wherein the component of the high-dimensional vector represents blood glucose trend data. And obtaining blood glucose trend data in different time periods in the future according to the corresponding relation between the preset component vector and the meaning of the prediction result.
S104, acquiring state characteristic data of a user, and carrying out disease prediction according to the blood glucose trend data and the state characteristic data to obtain an expected disease.
The state characteristic data comprise physiological data, dietary data and exercise data, the physiological data can comprise height information, weight information and the like of a user, the dietary data can be lunch and dinner foods which are the latest of the user, and the exercise data is the latest body building or running time of the user. The expected disease is a disease which may occur in the future of the user.
In some embodiments, based on a pre-trained disease classification prediction model, and based on the blood glucose trend data, the physiological data, the dietary data, and the exercise data, performing disease prediction, determining an expected illness and a corresponding probability; wherein the disease classification predictive model includes a fully connected layer and a softmax layer.
Specifically, the blood glucose trend data, the physiological data, the diet data and the exercise data are input into a pre-trained disease classification model, and finally the disease classification model can output expected illness and corresponding probability of a user.
The full connection layer multiplies the weight matrix by the input vector and adds bias to map n disease types into k real numbers, and meanwhile, the softmax layer maps k real numbers into k real numbers (probabilities) in (0, 1) and ensures that the sum of the k real numbers is 1. Illustratively, the probability of diabetes is 70%, the probability of cardiovascular and cerebrovascular diseases is 20% and the probability of cerebral thrombosis is 10%, respectively, can be obtained last.
The specific probability calculation formula is as follows:
γ＝softmax(z)＝softmax(W T x+b)
wherein x is input of a full-connection layer, and blood glucose trend data, physiological data, diet data and exercise data can be specifically used as input of the full-connection layer, W is weight, and W T x is the inner product of the weight and the input of the fully connected layer, b is the bias term, and gamma is the probability of Softmax output, so that each expected illness and the corresponding probability can be calculated. Through the full-connection layer, K disease types (- ≡infinity) are obtained, + -infinity) range of scores, to obtain the probability corresponding to each disease type, the score is obtained byMapping to (0, + -infinity), and normalizing to (0, 1) to obtain the corresponding probability, the specific softmax calculation formula is:
wherein z is j Can be expressed as a corresponding score for the jth diseased class, in particular, z j ＝w j *x+b j Wherein b j Bias term corresponding to jth disease category, w j And (3) obtaining the score of each category by weighting and summing the characteristics for the data weight corresponding to the jth diseased category, namely the importance degree of each data and the influence degree of each data on the final score, and mapping the score into probability through a Softmax function.
In some embodiments, feature extraction is performed on the blood glucose trend data, the physiological data, the diet data and the exercise data respectively to obtain feature information corresponding to each data; and performing feature matching in a preset disease library according to the feature information to obtain expected illness and corresponding probability. The preset disease library comprises a series of diseases caused by too low blood sugar or too high blood sugar, and each disease corresponds to characteristic information of various data.
Specifically, feature extraction can be performed on blood glucose trend data, physiological data, diet data and exercise data through a pre-trained decoder such as a ResNet decoder, a Glove decoder and the like, so that feature information corresponding to each data can be obtained. And (3) comparing the extracted characteristic information with characteristic information corresponding to each disease in a preset disease library item by item to obtain comprehensive similarity, wherein the comprehensive similarity can be considered as probability corresponding to each disease, and the disease with the highest comprehensive similarity can be considered as expected disease.
For example, if the blood glucose characteristic information is matched with diabetes in a disease library, the blood glucose characteristic information is compared with blood glucose characteristic information corresponding to diabetes to obtain the similarity of 80%, and then physiological characteristic information, diet characteristic information and exercise characteristic information are sequentially compared to obtain the similarity of 70%, 60% and 90% respectively. The overall similarity can be found to be 75% by averaging the similarities, and the expected disease can be considered to be diabetes and the corresponding probability to be 75%.
In some embodiments, after determining the expected illness and corresponding probability, a hint of completion of the prediction may also be output.
The prompting information mode can specifically include application program (APP) or means such as Email, short message, chat tool, e.g. WeChat, qq and the like.
Illustratively, after determining the type of illness and the corresponding probability, the Application (APP) may transmit a pop-up window to alert the user that the disease prediction has been completed, and the user may also view the predicted expected illness and the corresponding probability on the Application (APP).
S105, corresponding disease prevention measures are determined according to the state characteristic data and the expected diseases, and the disease prevention measures are pushed to a target terminal.
Wherein the disease prevention measures are measures for preventing expected diseases, and specifically may include a diet plan and a exercise plan, wherein the target terminal is a terminal device of a user.
In some embodiments, determining a required amount of a nutrient element and a target amount of exercise from the physiological data, the dietary data, the exercise data, and the expected disease; corresponding disease prevention measures are determined according to the required amount of the nutrient elements and the target movement amount, and the disease prevention measures comprise a diet plan and a movement plan.
Specifically, the required amount of the nutrient element is determined according to the physiological data, the diet data and the expected disease, and the corresponding diet plan is determined according to the required amount of the nutrient element. And determining the required quantity of the nutrient elements according to the physiological data, the exercise data and the expected diseases, and determining a corresponding disease preventive measure exercise plan according to the target exercise quantity.
Illustratively, if the user is fat in size, analyzing the diet data results in a recent lack of intake of sufficient protein, analyzing the exercise data results in an exercise duration of 20 minutes per week, and the expected disease probability is at most diabetes.
Since the highest expected disease probability is diabetes and no sufficient protein has been ingested recently, the corresponding diet program can be confirmed as follows: breakfast can eat an egg, 150 ml milk and a steamed stuffed bun. The Chinese meal can be cooked rice, and the fried dish can be cooked with edible tree fungus, fried bitter gourd, cucumber, etc. The dinner can be used for eating a small amount of rice, and the vegetables can be used for eating celery and stir-fried spinach. Since the expected disease probability is at the highest diabetes, the user is fat and has insufficient movement, the corresponding movement plan can be confirmed as follows: running for 15 min each day, relaxing muscle for 5 min, and training strength for 10 min.
Referring to fig. 3, fig. 3 is a schematic block diagram of a blood glucose prediction device according to an embodiment of the present application, where the blood glucose prediction device may be configured in a server for executing the blood glucose prediction method described above.
As shown in fig. 3, the blood glucose prediction device 200 includes: a data acquisition module 201, a medical record information screening module 202, a blood glucose prediction module 203, a disease prediction acquisition module 204, and a preventive measure generation module 205.
A data acquisition module 201, configured to acquire medical record information, blood glucose data, and medication data of a user;
the medical record information screening module 202 is configured to screen the medical record information to obtain medical record information corresponding to an abnormal blood glucose level, and take the medical record information corresponding to the abnormal blood glucose level as target medical record information;
The blood glucose prediction module 203 is configured to input the target medical record information, the blood glucose data, and the medication data into a pre-trained blood glucose prediction model, so as to obtain blood glucose trend data;
a disease prediction obtaining module 204, configured to obtain status feature data of a user, and perform disease prediction according to the blood glucose trend data and the status feature data, so as to obtain an expected disease;
and a preventive measure generating module 205, configured to determine a corresponding preventive measure for a disease according to the status characteristic data and the expected disease, and push the preventive measure for the disease to a target terminal.
It should be noted that, for convenience and brevity of description, specific working processes of the above-described apparatus and each module, unit may refer to corresponding processes in the foregoing method embodiments, which are not repeated herein.
The methods and apparatus of the present application are operational with numerous general purpose or special purpose computing system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.
By way of example, the methods, apparatus described above may be implemented in the form of a computer program that is executable on a computer device as shown in fig. 4.
Referring to fig. 4, fig. 4 is a schematic diagram of a computer device according to an embodiment of the present application. The computer device may be a server.
As shown in fig. 4, the computer device includes a processor, a memory, and a network interface connected by a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.
The non-volatile storage medium may store an operating system and a computer program. The computer program comprises program instructions that, when executed, cause the processor to perform any of a number of methods for predicting blood glucose.
The processor is used to provide computing and control capabilities to support the operation of the entire computer device.
The internal memory provides an environment for the execution of a computer program in a non-volatile storage medium that, when executed by a processor, causes the processor to perform any of a number of blood glucose prediction methods.
The network interface is used for network communication such as transmitting assigned tasks and the like. It will be appreciated by those skilled in the art that the structure of the computer device is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or less components than those shown, or may combine some of the components, or have a different arrangement of components.
It should be appreciated that the processor may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field-programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. Wherein the general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Wherein in some embodiments the processor is configured to run a computer program stored in the memory to implement the steps of: acquiring medical record information, blood sugar data and medication data of a user; screening the medical record information to obtain medical record information corresponding to abnormal blood sugar values, and taking the medical record information corresponding to the abnormal blood sugar values as target medical record information; inputting the target medical record information, the blood glucose data and the medication data into a pre-trained blood glucose prediction model to obtain blood glucose trend data; acquiring state characteristic data of a user, and predicting diseases according to the blood sugar trend data and the state characteristic data to obtain expected diseases; and determining corresponding disease preventive measures according to the state characteristic data and the expected diseases, and pushing the disease preventive measures to a target terminal.
In some embodiments, the processor is further configured to: based on a word segmentation algorithm, performing word segmentation processing on the medical record information to obtain a word segmentation result corresponding to the medical record information; performing word sense prediction on each word in the word segmentation result based on a medical word sense prediction model to obtain a word sense prediction result corresponding to each word; and screening the word segmentation result based on the word sense prediction result to obtain target medical record information.
In some embodiments, the processor is further configured to: respectively determining the illness time, the measurement time of the blood sugar data and the medication time of the medication data in the target medical record information; determining weight proportions corresponding to the target medical record information, the blood glucose data and the medication data according to the illness time, the measurement time and the medication time; and inputting the target medical record information, the blood glucose data, the medication data and the weight proportion corresponding to each data into a pre-trained blood glucose prediction model to obtain blood glucose trend data.
In some embodiments, the processor is further configured to: respectively determining differences among the illness time, the measurement time, the medication time and the data acquisition time to obtain corresponding first time difference, second time difference and third time difference; calculating to obtain a total time difference according to the first time difference, the second time difference and the third time difference; determining the ratio of the first time difference to the total time difference, and determining the weight proportion corresponding to the target medical record information according to the ratio of the first time difference to the total time difference; determining the ratio of the second time difference to the total time difference, and determining the weight proportion corresponding to the blood glucose data according to the ratio of the second time difference to the total time difference; and determining the ratio of the third time difference to the total time difference, and determining the weight proportion corresponding to the medication data according to the ratio of the third time difference to the total time difference.
In some embodiments, the processor is further configured to: inputting the target medical record information, the blood glucose data, the medication data and the weight proportion corresponding to the blood glucose data and the medication data into a first long-term and short-term memory network in a pre-trained blood glucose prediction model for operation to obtain a hidden state vector sequence in the first long-term and short-term memory network; and inputting the hidden state vector sequence into the second long-term and short-term memory network for operation to obtain blood sugar trend data.
In some embodiments, the processor is further configured to: based on a pre-trained disease classification prediction model, and according to the blood glucose trend data, the physiological data, the diet data and the exercise data, disease prediction is carried out, and expected illness and corresponding probability are obtained; wherein the disease classification predictive model includes a fully connected layer and a softmax layer.
In some embodiments, the processor is further configured to: determining a required amount of a nutrient element and a target movement amount according to the state characteristic data and the expected disease; corresponding disease prevention measures are determined according to the required amount of the nutrient elements and the target movement amount, wherein the disease prevention measures comprise a diet plan and a movement plan.
The embodiment of the application also provides a computer readable storage medium, and a computer program is stored on the computer readable storage medium, wherein the computer program comprises program instructions, and when the program instructions are executed, any blood glucose prediction method provided by the embodiment of the application is realized.
The computer readable storage medium may be an internal storage unit of the computer device according to the foregoing embodiment, for example, a hard disk or a memory of the computer device. The computer readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), or the like, which are provided on the computer device.
Further, the computer-readable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created from the use of blockchain nodes, and the like.
The invention refers to a novel application mode of computer technologies such as storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like of a blockchain language model. The Blockchain (Blockchain), which is essentially a decentralised database, is a string of data blocks that are generated by cryptographic means in association, each data block containing a batch of information of network transactions for verifying the validity of the information (anti-counterfeiting) and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.
While the invention has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions of equivalents may be made and equivalents will be apparent to those skilled in the art without departing from the scope of the invention. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (8)
1. A method of predicting blood glucose, the method comprising:
acquiring medical record information, blood sugar data and medication data of a user;
screening the medical record information to obtain medical record information corresponding to abnormal blood sugar values, and taking the medical record information corresponding to the abnormal blood sugar values as target medical record information;
inputting the target medical record information, the blood glucose data and the medication data into a pre-trained blood glucose prediction model to obtain blood glucose trend data;
acquiring state characteristic data of a user, and predicting diseases according to the blood sugar trend data and the state characteristic data to obtain expected diseases;
determining corresponding disease prevention measures according to the state characteristic data and the expected diseases, and pushing the disease prevention measures to a target terminal;
The step of inputting the target medical record information, the blood glucose data and the medication data into a pre-trained blood glucose prediction model to obtain blood glucose trend data comprises the following steps:
respectively determining the illness time, the measurement time of the blood sugar data and the medication time of the medication data in the target medical record information;
acquiring data acquisition time corresponding to the disease time, the measurement time and the medication time respectively;
determining weight proportions corresponding to the target medical record information, the blood glucose data and the medication data according to the illness time, the measurement time, the medication time and the respective corresponding data acquisition time;
inputting the target medical record information, the blood glucose data, the medication data and the weight proportion corresponding to each data into a pre-trained blood glucose prediction model to obtain blood glucose trend data;
the determining the weight proportion corresponding to the target medical record information, the blood sugar data and the medication data according to the illness time, the measurement time and the medication time comprises the following steps:
respectively determining differences of the illness time, the measurement time and the medication time and the corresponding data acquisition time to obtain corresponding first time difference, second time difference and third time difference;
Calculating to obtain a total time difference according to the first time difference, the second time difference and the third time difference;
determining the ratio of the first time difference to the total time difference, and determining the weight proportion corresponding to the target medical record information according to the ratio of the first time difference to the total time difference;
determining the ratio of the second time difference to the total time difference, and determining the weight proportion corresponding to the blood glucose data according to the ratio of the second time difference to the total time difference;
and determining the ratio of the third time difference to the total time difference, and determining the weight proportion corresponding to the medication data according to the ratio of the third time difference to the total time difference.
2. The method according to claim 1, wherein the step of screening the medical record information to obtain medical record information corresponding to an abnormal blood glucose level, and taking the medical record information corresponding to the abnormal blood glucose level as target medical record information, comprises:
based on a word segmentation algorithm, performing word segmentation processing on the medical record information to obtain a word segmentation result corresponding to the medical record information;
performing word sense prediction on each word in the word segmentation result based on a medical word sense prediction model to obtain a word sense prediction result corresponding to each word;
And screening the word segmentation result based on the word sense prediction result to obtain target medical record information.
3. The method of claim 1, wherein the blood glucose prediction model comprises a first long-short-term memory network and a second long-short-term memory network, the inputting the target medical record information, the blood glucose data and the medication data and the corresponding weight ratios into the pre-trained blood glucose prediction model to obtain blood glucose trend data, comprising:
inputting the target medical record information, the blood glucose data, the medication data and the weight proportion corresponding to the blood glucose data and the medication data into a first long-term and short-term memory network in a pre-trained blood glucose prediction model for operation to obtain a hidden state vector sequence in the first long-term and short-term memory network;
and inputting the hidden state vector sequence into the second long-term and short-term memory network for operation to obtain blood sugar trend data.
4. The method of claim 1, wherein the status characterization data includes physiological data, dietary data, and exercise data, wherein the predicting a disease based on the status characterization data results in an expected disease and a corresponding probability, comprising:
Based on a pre-trained disease classification prediction model, and according to the blood glucose trend data, the physiological data, the diet data and the exercise data, disease prediction is carried out, and expected illness and corresponding probability are obtained; wherein the disease classification predictive model includes a fully connected layer and a softmax layer.
5. The method of claim 1, wherein determining a corresponding disease prevention measure based on the status characterization data and the expected disease and pushing the disease prevention measure to a target terminal comprises:
determining a required amount of a nutrient element and a target movement amount according to the state characteristic data and the expected disease;
corresponding disease prevention measures are determined according to the required amount of the nutrient elements and the target movement amount, wherein the disease prevention measures comprise a diet plan and a movement plan.
6. A blood glucose prediction device, comprising:
the data acquisition module is used for acquiring medical record information, blood sugar data and medication data of a user;
the medical record information screening module is used for screening the medical record information to obtain medical record information corresponding to the abnormal blood glucose value, and taking the medical record information corresponding to the abnormal blood glucose value as target medical record information;
The blood sugar prediction module is used for inputting the target medical record information, the blood sugar data and the medication data into a pre-trained blood sugar prediction model to obtain blood sugar trend data;
the disease prediction acquisition module is used for acquiring state characteristic data of a user, and carrying out disease prediction according to the blood glucose trend data and the state characteristic data to obtain expected diseases;
a preventive measure generating module, configured to determine a corresponding disease preventive measure according to the status feature data and the expected disease, and push the disease preventive measure to a target terminal;
the step of inputting the target medical record information, the blood glucose data and the medication data into a pre-trained blood glucose prediction model to obtain blood glucose trend data comprises the following steps:
respectively determining the illness time, the measurement time of the blood sugar data and the medication time of the medication data in the target medical record information;
acquiring data acquisition time corresponding to the disease time, the measurement time and the medication time respectively;
determining weight proportions corresponding to the target medical record information, the blood glucose data and the medication data according to the illness time, the measurement time, the medication time and the respective corresponding data acquisition time;
Inputting the target medical record information, the blood glucose data, the medication data and the weight proportion corresponding to each data into a pre-trained blood glucose prediction model to obtain blood glucose trend data;
the determining the weight proportion corresponding to the target medical record information, the blood sugar data and the medication data according to the illness time, the measurement time and the medication time comprises the following steps:
respectively determining differences of the illness time, the measurement time and the medication time and the corresponding data acquisition time to obtain corresponding first time difference, second time difference and third time difference;
calculating to obtain a total time difference according to the first time difference, the second time difference and the third time difference;
determining the ratio of the first time difference to the total time difference, and determining the weight proportion corresponding to the target medical record information according to the ratio of the first time difference to the total time difference;
determining the ratio of the second time difference to the total time difference, and determining the weight proportion corresponding to the blood glucose data according to the ratio of the second time difference to the total time difference;
and determining the ratio of the third time difference to the total time difference, and determining the weight proportion corresponding to the medication data according to the ratio of the third time difference to the total time difference.
7. A computer device, the computer device comprising a memory and a processor;
the memory is used for storing a computer program;
the processor is configured to execute the computer program and implement when executing the computer program:
the method for predicting blood glucose according to any one of claims 1 to 5.
8. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program which, when executed by a processor, causes the processor to implement the blood glucose prediction method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110991121.7A CN113679348B (en) | 2021-08-26 | 2021-08-26 | Blood glucose prediction method, blood glucose prediction device, apparatus, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110991121.7A CN113679348B (en) | 2021-08-26 | 2021-08-26 | Blood glucose prediction method, blood glucose prediction device, apparatus, and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113679348A CN113679348A (en) | 2021-11-23 |
| CN113679348B true CN113679348B (en) | 2024-02-06 |
Family
ID=78583055
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110991121.7A Active CN113679348B (en) | 2021-08-26 | 2021-08-26 | Blood glucose prediction method, blood glucose prediction device, apparatus, and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113679348B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115862795A (en) * | 2022-09-22 | 2023-03-28 | 深圳可孚生物科技有限公司 | Blood glucose prediction method based on pre-training and data decomposition two stages |
| CN115662617B (en) * | 2022-10-24 | 2023-06-27 | 重庆联芯致康生物科技有限公司 | Result illness state prediction system based on CGM and prediction method thereof |
| CN116504355B (en) * | 2023-04-27 | 2024-04-02 | 广东食品药品职业学院 | Closed-loop insulin infusion control method, device and storage medium based on neural network |
| CN116671906A (en) * | 2023-08-01 | 2023-09-01 | 亿慧云智能科技（深圳）股份有限公司 | Noninvasive blood glucose measurement method and noninvasive blood glucose measurement system for smart watch |
| CN116936133B (en) * | 2023-09-18 | 2023-12-08 | 四川互慧软件有限公司 | Nutrient condition monitoring method and system based on nursing morning shift data |
| CN117352182A (en) * | 2023-09-19 | 2024-01-05 | 北京华益精点生物技术有限公司 | Training method, device, equipment and storage medium of blood glucose prediction model |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108852365A (en) * | 2018-05-04 | 2018-11-23 | 中国人民解放军第三〇九医院 | Blood Sugar Monitoring system, intelligent medicine box terminal and blood Sugar Monitoring main system |
| CN109166608A (en) * | 2018-09-17 | 2019-01-08 | 新华三大数据技术有限公司 | Electronic health record information extracting method, device and equipment |
| CN109299239A (en) * | 2018-09-29 | 2019-02-01 | 福建弘扬软件股份有限公司 | ES-based electronic medical record retrieval method |
| CN109692007A (en) * | 2019-01-25 | 2019-04-30 | 北京中器华康科技发展有限公司 | A kind of gestational period blood Sugar Monitoring system based on blood sugar monitoring and the application in disease monitoring system |
| CN109920547A (en) * | 2019-03-05 | 2019-06-21 | 北京工业大学 | A kind of diabetes prediction model construction method based on electronic health record data mining |
| CN109977422A (en) * | 2019-04-18 | 2019-07-05 | 中国石油大学(华东) | A kind of case history key message extraction model based on participle technique |
| CN109978022A (en) * | 2019-03-08 | 2019-07-05 | 腾讯科技（深圳）有限公司 | A kind of medical treatment text message processing method and device, storage medium |
| CN110459324A (en) * | 2019-06-27 | 2019-11-15 | 平安科技（深圳）有限公司 | Disease forecasting method, apparatus and computer equipment based on shot and long term memory models |
| CN111274806A (en) * | 2020-01-20 | 2020-06-12 | 医惠科技有限公司 | Method and device for recognizing word segmentation and part of speech and method and device for analyzing electronic medical record |
| CN111297329A (en) * | 2020-02-24 | 2020-06-19 | 苏州大学 | Method and system for predicting dynamic morbidity risk of cardiovascular complications of diabetic patients |
| CN111329491A (en) * | 2020-02-27 | 2020-06-26 | 京东方科技集团股份有限公司 | Blood glucose prediction method and device, electronic equipment and storage medium |
| CN112951430A (en) * | 2021-03-29 | 2021-06-11 | 阿斯利康投资(中国)有限公司 | Multiple chronic disease joint management apparatus and computer-readable storage medium |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1672629A (en) * | 2005-04-29 | 2005-09-28 | 天津大学 | USB-based multiple physiological parameter detecting system |
| US8030891B2 (en) * | 2008-04-10 | 2011-10-04 | Smiths Medical Asd, Inc. | Ambulatory medical device with electrical isolation from connected peripheral device |
| CN101536900B (en) * | 2009-03-24 | 2011-06-29 | 重庆大学 | Method and system for constructing modular medical instrument based on USB HUB |
| CN206151451U (en) * | 2016-08-15 | 2017-05-10 | 深圳市理邦精密仪器股份有限公司 | Multi -parameter guardianship equipment |
| EP3610397A4 (en) * | 2017-04-13 | 2020-12-23 | Intuity Medical, Inc. | Systems and methods for managing chronic disease using analyte and patient data |
| CN115412106A (en) * | 2022-07-12 | 2022-11-29 | 北京中科睿谱科技有限公司 | Multi-channel high-isolation differential switch matrix circuit |
-
2021
- 2021-08-26 CN CN202110991121.7A patent/CN113679348B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108852365A (en) * | 2018-05-04 | 2018-11-23 | 中国人民解放军第三〇九医院 | Blood Sugar Monitoring system, intelligent medicine box terminal and blood Sugar Monitoring main system |
| CN109166608A (en) * | 2018-09-17 | 2019-01-08 | 新华三大数据技术有限公司 | Electronic health record information extracting method, device and equipment |
| CN109299239A (en) * | 2018-09-29 | 2019-02-01 | 福建弘扬软件股份有限公司 | ES-based electronic medical record retrieval method |
| CN109692007A (en) * | 2019-01-25 | 2019-04-30 | 北京中器华康科技发展有限公司 | A kind of gestational period blood Sugar Monitoring system based on blood sugar monitoring and the application in disease monitoring system |
| CN109920547A (en) * | 2019-03-05 | 2019-06-21 | 北京工业大学 | A kind of diabetes prediction model construction method based on electronic health record data mining |
| CN109978022A (en) * | 2019-03-08 | 2019-07-05 | 腾讯科技（深圳）有限公司 | A kind of medical treatment text message processing method and device, storage medium |
| CN109977422A (en) * | 2019-04-18 | 2019-07-05 | 中国石油大学(华东) | A kind of case history key message extraction model based on participle technique |
| CN110459324A (en) * | 2019-06-27 | 2019-11-15 | 平安科技（深圳）有限公司 | Disease forecasting method, apparatus and computer equipment based on shot and long term memory models |
| CN111274806A (en) * | 2020-01-20 | 2020-06-12 | 医惠科技有限公司 | Method and device for recognizing word segmentation and part of speech and method and device for analyzing electronic medical record |
| CN111297329A (en) * | 2020-02-24 | 2020-06-19 | 苏州大学 | Method and system for predicting dynamic morbidity risk of cardiovascular complications of diabetic patients |
| CN111329491A (en) * | 2020-02-27 | 2020-06-26 | 京东方科技集团股份有限公司 | Blood glucose prediction method and device, electronic equipment and storage medium |
| CN112951430A (en) * | 2021-03-29 | 2021-06-11 | 阿斯利康投资(中国)有限公司 | Multiple chronic disease joint management apparatus and computer-readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113679348A (en) | 2021-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113679348B (en) | Blood glucose prediction method, blood glucose prediction device, apparatus, and storage medium | |
| Dave et al. | Feature-based machine learning model for real-time hypoglycemia prediction | |
| US20200375549A1 (en) | Systems for biomonitoring and blood glucose forecasting, and associated methods | |
| Reddy et al. | Prediction of hypoglycemia during aerobic exercise in adults with type 1 diabetes | |
| Afsaneh et al. | Recent applications of machine learning and deep learning models in the prediction, diagnosis, and management of diabetes: a comprehensive review | |
| US20140316220A1 (en) | Personal Health Monitoring System | |
| Wang et al. | A novel adaptive-weighted-average framework for blood glucose prediction | |
| KR102400740B1 (en) | System for monitoring health condition of user and analysis method thereof | |
| US20210241916A1 (en) | Forecasting and explaining user health metrics | |
| WO2019183510A1 (en) | System and method for assessing heart health and communicating the assessment to a patient | |
| Meneghetti et al. | Detection of insulin pump malfunctioning to improve safety in artificial pancreas using unsupervised algorithms | |
| US20220336109A1 (en) | Application for tracking progression and isolating causes of adverse medical conditions | |
| Kardas et al. | Type 2 diabetes patients benefit from the COMODITY12 mHealth system: results of a randomised trial | |
| Alenizi et al. | Machine learning approach for diabetes prediction | |
| De Falco et al. | A genetic programming-based regression for extrapolating a blood glucose-dynamics model from interstitial glucose measurements and their first derivatives | |
| Meneghetti et al. | Machine learning-based anomaly detection algorithms to alert patients using sensor augmented pump of infusion site failures | |
| CN109065126A (en) | A kind of life style management method, device and terminal | |
| US20230255556A1 (en) | Biosensors and food logger systems for personalized health and self-care | |
| Panyamit et al. | Identification of factors in the survival rate of heart failure patients using machine learning models and principal component analysis | |
| WO2022050992A1 (en) | Virtually monitoring glucose levels in a patient using machine learning and digital twin technology | |
| Zhang | Predicting occurrences of acute hypoglycemia during insulin therapy in the intensive care unit | |
| Zope et al. | Hybrid Ensemble and Features Selection for Heart Disease Prediction Using Machine Learning | |
| ZareMehrjardi et al. | Assessment of the Relationship Between Physical Behaviours, Psychological Factors, Medication Use, Body Composition and Weight Regain Following Bariatric Surgery | |
| US20230420142A1 (en) | Application for tracking progression and isolating causes of adverse medical conditions | |
| Chaki et al. | Computer and Information Sciences |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220922 Address after: Room 2601 (Unit 07), Qianhai Free Trade Building, No. 3048, Xinghai Avenue, Nanshan Street, Qianhai Shenzhen-Hong Kong Cooperation Zone, Shenzhen, Guangdong 518000 Applicant after: Shenzhen Ping An Smart Healthcare Technology Co.,Ltd. Address before: 1-34 / F, Qianhai free trade building, 3048 Xinghai Avenue, Mawan, Qianhai Shenzhen Hong Kong cooperation zone, Shenzhen, Guangdong 518000 Applicant before: Ping An International Smart City Technology Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |