CN116313083B - Intelligent diabetes interconnection remote real-time monitoring management system based on algorithm and big data - Google Patents

Intelligent diabetes interconnection remote real-time monitoring management system based on algorithm and big data Download PDF

Info

Publication number
CN116313083B
CN116313083B CN202310207448.XA CN202310207448A CN116313083B CN 116313083 B CN116313083 B CN 116313083B CN 202310207448 A CN202310207448 A CN 202310207448A CN 116313083 B CN116313083 B CN 116313083B
Authority
CN
China
Prior art keywords
data
patient
feature
point
image
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
Application number
CN202310207448.XA
Other languages
Chinese (zh)
Other versions
CN116313083A (en
Inventor
曲斌斌
许亚慧
宋超
李伯彬
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Qingdao Central Hospital
Original Assignee
Qingdao Central Hospital
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Qingdao Central Hospital filed Critical Qingdao Central Hospital
Priority to CN202310207448.XA priority Critical patent/CN116313083B/en
Publication of CN116313083A publication Critical patent/CN116313083A/en
Application granted granted Critical
Publication of CN116313083B publication Critical patent/CN116313083B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
    • G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
    • G06V10/757—Matching configurations of points or features
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
    • G16H40/67—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for remote operation
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H80/00—ICT specially adapted for facilitating communication between medical practitioners or patients, e.g. for collaborative diagnosis, therapy or health monitoring
    • 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

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Pathology (AREA)
  • Multimedia (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Business, Economics & Management (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • Evolutionary Computation (AREA)
  • Software Systems (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

The invention provides an intelligent interconnected remote real-time monitoring management system for diabetes based on an algorithm and big data, which comprises a data acquisition module, a data analysis module, a disease early warning module and a remote medical consultation module, wherein the data acquisition module is used for uniformly acquiring diabetes disease data of a patient, the data analysis module is used for carrying out real-time analysis on the data of the patient and generating corresponding monitoring reports, providing reference bases for doctors and the patient, the disease early warning module is used for carrying out prediction and analysis based on historical data and real-time data of the patient and timely early warning and prompting the patient and the doctor, and the remote medical consultation module is used for realizing online communication between the doctor and the patient, so that the doctor can grasp the disease condition and the data of the patient in real time and timely give diagnosis and treatment comments and suggestions.

Description

Intelligent diabetes interconnection remote real-time monitoring management system based on algorithm and big data
Technical Field
The present disclosure relates to the field of smart medical and data processing, and more particularly, to an algorithm and big data based diabetes smart interconnection remote real-time monitoring management system.
Background
Diabetes is a metabolic disease, and the number of diabetics worldwide has exceeded 4.60 billion, accounting for 9.3% of the world population. Moreover, this number is also increasing. Diabetes has become a epidemic disease, particularly in developing countries. Diabetes can present many health problems to patients, including cardiovascular disease, kidney disease, eye disease, neuropathy, etc., which can cause significant disturbance to the patient's life. In addition, diabetes is a chronic disease that requires long-term treatment and monitoring. At present, the traditional diabetes monitoring method mainly depends on a glucometer and a blood glucose test paper for monitoring, but the method has a plurality of defects. First, the blood glucose level needs to be measured manually, which is inconvenient and error-prone to operate. Secondly, only short-term blood glucose data can be provided, and the disease condition of the patient cannot be comprehensively known. In addition, frequent blood glucose monitoring is required for patients, which can place a significant burden and discomfort on the patient. For diabetics, monitoring and managing own disease conditions is very important, because the disease conditions can be found and controlled in time, the severity of the disease conditions can be reduced, and complications can be prevented. The conventional diabetes monitoring method is to let the patient regularly go to the hospital for blood glucose monitoring, but this method has some drawbacks such as requiring a lot of time and money, and inconveniencing the work and life of the patient. In addition, the data acquisition and analysis of the method are slow, and real-time monitoring and early warning are difficult to achieve. With the development of intelligent medical technology, an intelligent diabetes interconnection remote real-time monitoring management system based on algorithms and big data is generated. The system can monitor physiological parameters such as blood sugar and blood pressure of a patient in real time through the intelligent sensor and the mobile equipment, and perform data analysis and processing through technologies such as cloud computing and artificial intelligence, so that a personalized medical scheme and early warning of the patient are provided, all-weather and omnibearing health management of the diabetic patient is realized, the illness state is effectively controlled, and complications are prevented.
Disclosure of Invention
Aiming at the problems, the invention aims to provide an intelligent interconnected remote real-time monitoring management system for diabetes based on an algorithm and big data.
The aim of the invention is realized by the following technical scheme:
The diabetes intelligent interconnection remote real-time monitoring management system based on the algorithm and the big data comprises a data acquisition module, a data analysis module, a disease early warning module and a remote medical consultation module, wherein the data acquisition module comprises two parts of hardware and software, diabetes disease data of a patient are uniformly acquired, the data analysis module has the functions of data cleaning, feature extraction, data modeling and monitoring report generation, real-time analysis is carried out on the data of the patient, corresponding monitoring reports are generated, reference bases are provided for doctors and the patient, the disease early warning module extracts image features on the basis of historical data and real-time data of the patient, and utilizes an improved scale invariant feature transformation algorithm for extracting image features of a disease feature map of the diabetes patient, and timely early warning and prompting the patient and the doctor, and the remote medical consultation module realizes online communication between the doctor and the patient, so that the doctor can master the disease condition and the data of the patient in real time and timely give diagnosis and suggestion.
Further, the data acquisition module comprises two parts, namely hardware and software: in terms of hardware, existing diabetes monitoring equipment such as a blood glucose meter, a blood pressure meter and a weight meter is used for data transmission with a system through Bluetooth, wiFi and other technologies; in terms of software, the data acquisition module communicates with equipment of different models through various interfaces and API modes to acquire and analyze data transmitted by the equipment, and meanwhile, the data acquisition module also supports data input of various formats, such as Excel and CSV formats; in the process of data acquisition, the data acquisition module can perform real-time verification, verification and filtration on the data, so that the accuracy and the integrity of the data are ensured, the acquired data can be optimized through data compression, encryption and other technologies, and the transmission efficiency and the safety of the data are improved; in addition, the data acquisition module not only can acquire the diabetes monitoring data of the patient, but also can acquire other related information, such as basic information and treatment records of the patient, and the acquisition of the data provides references when a doctor diagnoses the patient, so that the doctor can better know the illness state of the patient and formulate a more effective treatment scheme.
Further, the data analysis module has the following functions:
(1) Data cleaning: the collected data is subjected to denoising, duplication removing, complement and other treatments, so that the quality and the integrity of the data are ensured;
(2) Feature extraction: extracting key features in the patient diabetes condition data, such as blood glucose level, insulin dosage, exercise amount and the like, so as to further analyze and predict;
(3) Modeling data: based on machine learning and data mining algorithms, constructing a model to predict the trend and risk of the diabetes mellitus of the patient, and evaluating the current health state of the patient;
(4) Monitoring report generation: generating a monitoring report from the data analysis result, wherein the monitoring report comprises contents such as the current condition, the disease change trend, the potential risk and the like of a patient, and the contents are used for reference and analysis of doctors and the patient;
Through real-time analysis and monitoring of the data analysis module, doctors and patients can more comprehensively understand the illness state and the health state of the patients, and timely take corresponding treatment measures and adjustment schemes, so that the treatment effect and the life quality are improved.
Further, the disease early warning module is used for predicting and analyzing based on historical data and real-time data of a patient, early warning and prompting the patient and doctors in time, and the main functions of the disease early warning module comprise data preprocessing, disease prediction, early warning prompting and data visualization, wherein the data preprocessing function is used for processing acquired patient data, removing abnormal values and noise in the data and guaranteeing the accuracy and reliability of the data; the disease prediction function predicts the disease of the patient based on historical data and real-time data by adopting an improved scale invariant feature transformation algorithm, analyzes the disease change trend of the patient, and predicts the possible disease change of the patient through a disease feature map of the patient, wherein the method comprises the following steps:
For a disease feature map of a diabetic patient, extracting image features by using an improved scale invariant feature transformation algorithm, firstly constructing a scale space, converting an image into the scale space, carrying out convolution calculation on an image I (x, y) by using a Gaussian function G (x, y, sigma), converting a two-dimensional image on a plane into the scale space image, and calculating as follows: l (x, y, σ) =g (x, y, σ) ×i (x, y), where L (x, y, σ) is an image in scale space, x is an abscissa of the image, y is an ordinate of the image, σ is a spatial scale factor, and G (x, y, σ) is a gaussian kernel function, satisfying: I (x, y) is a two-dimensional image in a plane, and in order to obtain an extreme point D (x, y, σ) corresponding to a scale space, it is calculated as follows:
D(x，y，σ)＝(G(x，y，kσ)-G(x，y，σ))*I(x，y)＝L(x，y，kσ)-L(x，y，σ)
Wherein k is a multiplier factor for changing the scale size of the adjacent scale;
Then, through two steps of image Gaussian blur and image downsampling, a Gaussian differential pyramid is generated, a specific gradual change image is obtained through image downsampling, then the gradual change image is arranged according to a sequence, an original image is at the lowest end, each layer of the Gaussian differential pyramid is obtained through two adjacent layers of result differential operation of the Gaussian pyramid, then extreme points of the image are detected, each point in the image is firstly screened for one time, whether each point has an extreme value or not is analyzed, whether the extreme value point is judged, whether the extreme point is the maximum value or the minimum value is judged by comparing the point with 26 points of adjacent layers, if the extreme point is the maximum value, the point is remained, the positioning of characteristic points is obtained through function fitting, namely, the secondary Taylor expansion operation is carried out on Gaussian differential operators D (x, y and sigma), and then the sampling point is calculated:
deriving the above equation, and letting D (X) =0, there are: Wherein/> When the offset is greater than 0.5 in one of the three dimensions x, y and sigma, the characteristic point is adjusted to an adjacent position, and then the offset corresponding to the characteristic point is estimated until the characteristic point is converged in all dimensions, and when the position of the characteristic point exceeds the boundary range, the characteristic point is discarded; clearing out the response points with poor reliability through the sea plug matrix, specifically: the principal curvature is analyzed, the reliability of the characteristic points is judged, the principal curvature is obtained through the sea plug matrix of the characteristic points, and the calculation formula is as follows:
Wherein, D xx represents the second order bias derivative of x after solving the bias derivative of x for the gaussian difference operator, D xy represents the second order bias derivative of y after solving the bias derivative of x for the gaussian difference operator, D yy represents the second order bias derivative of y after solving the bias derivative of y for the gaussian difference operator, and the analysis matrix H feature values a and b sequentially refer to gradients corresponding to feature points in the two-axis direction, which can be known:
tr(H)＝Dxx+Dyy＝a+b
det(H)＝DxxDyy-(Dyy)2＝ab
Where tr (·) is the trace operation of the matrix, det (·) is the determinant operation of the matrix, and c is the ratio of two eigenvalues of a and b, with: a=cb, satisfying: in the deployment test, if If true, the point is indicated to be a stable point, otherwise, the point needs to be removed.
Further, the distribution of the directions of the feature points is performed, and in the process of scale-invariant feature transformation, because each feature point is distributed with a relative direction according to statistics, the rotation of the image is not realized, the gradient value and the gradient direction of the adjacent pixels of the feature points are calculated, and the direction parameters of the feature points are obtained, and are calculated as follows:
θ(x，y)＝tan-1((L(x，y+1)-L(x，y-1))/L(x+1，y)-L(x-1，y))
Wherein L (x, y) refers to a feature value corresponding to a pixel point (x, y), m (x, y) refers to a gradient modulus value corresponding to the pixel point (x, y), θ (x, y) mainly refers to a gradient direction value corresponding to the pixel point (x, y), after the feature point neighborhood gradient information is subjected to the expansion operation, a Len dimension feature vector is required to be generated, detailed statistics must be expanded through a gradient histogram, the gradient histogram divides the direction into 36 columns, each column refers to a specific gradient direction, the direction of the column where the extremum is selected to serve as a main direction thereof, the direction of the column where the second limit is selected to serve as a corresponding auxiliary direction, 45 ° is divided into 8 directions, namely, east, south, west, north, northeast, southeast, northwest and southwest, and then the feature point is generated, in order to ensure that rotation invariance is well ensured, and standardized conversion is expanded to the main direction:
Wherein x 'and y' are respectively the horizontal and vertical coordinates of the image pixel point after rotation transformation,
Wherein,Is a feature descriptor,/>Normalized feature descriptors define a matching image feature descriptor as x= (X 1,x2,...,xLen), a feature descriptor in the image to be registered is mainly y= (Y 1,y2,...,yLen), and if their euclidean distance ratio does not exceed a specific threshold, then this feature point is considered to achieve effective matching, and the euclidean distance is expressed as follows:
Further, the improvement on the scale-invariant feature transform algorithm comprises the following steps: reducing the number of sub-pixel areas by re-dividing the rectangular pixel areas so as to reduce the dimension of the feature vector of the feature point; in the matching stage, the matching condition is not only the ratio of Euclidean distance, but also the vector correlation coefficient is integrated to ensure the matching accuracy of the feature points, the dimension of the lightweight original feature vector is updated to be Len *, and the matching condition is obtained after normalization processing:
Wherein D is a descriptor, The normalized descriptor is then fused into a vector correlation coefficient for a scale invariant and feature transformation algorithm, the length similarity of the vector X, Y is defined as a (X, Y), the length similarity is a standard for measuring the vector size, and b is: Wherein c (X ', Y) =a (X', Y) ·b (X ', Y'), wherein c (X, Y) ∈ [ -1,1], when θ=ζ/2, the two vectors are orthogonal, where the similarity coefficient c (X, Y) =0, when θ=0, and the norms are the same, the similarity coefficient c (X, Y) =1; when θ=ζ, and the norms are the same, the similarity coefficient c (X, Y) = -1 of the two vectors;
in addition, when the illness state of the patient is abnormal or possibly abnormal, the illness state early-warning module can send out early-warning prompt to prompt the patient to seek medical attention or adjust the treatment scheme in time; the data visualization function intuitively displays analysis results in the form of charts and the like, so that doctors and patients clearly know the change trend of the illness state, doctors are helped to make more reasonable diagnosis and treatment schemes, and the treatment effect of the patients is improved.
Further, the remote medical consultation module is mainly used for realizing online communication between a doctor and a patient, enabling the doctor to master the illness state and data of the patient in real time and timely giving diagnosis and treatment comments and suggestions, realizing remote communication between the doctor and the patient in a mode of Internet, mobile application programs and the like, uploading the illness state and data of the patient by the patient, checking the data and illness state report uploaded by the patient by a doctor through a login system, diagnosing and analyzing the illness state of the patient, giving treatment comments and guidance, simultaneously enabling the doctor to conduct online communication with the patient through an instant communication function of the system, solving the doubt of the patient, and providing professional medical consultation service.
The intelligent diabetes interconnection remote real-time monitoring management system based on the algorithm and the big data has the beneficial effects in many aspects: the method has the advantages that the distinguishing performance is good, the rapid and accurate distinguishing information can be matched in a massive image database, a large number of feature vectors can be generated even if only a single object exists, the realization speed is high, the feature vector matching can be rapidly performed, the feature vectors can be combined and expanded with other forms of feature vectors, the innovation point is that the correlation coefficient is used for further classifying the number of image bits so as to reduce the complexity of calculation, the correlation coefficient can be used for measuring the relation of variables through numbers and has directivity, 1 represents positive correlation, 1 represents negative correlation, the intensity of the variable relation can be measured, and the closer to 0, the weaker the correlation is 1; in addition, the invention has the beneficial effects of real-time monitoring, intelligent early warning, personalized treatment, improving the life quality of patients and relieving the medical care burden, and is specifically described as follows: 1. and (3) real-time monitoring: the system can monitor the physiological index of a diabetic patient in real time, and provides more accurate data for doctors. Under the supervision of doctors, higher or lower data can be found in time, so that illness state is better controlled, and hospitalization rate is reduced. 2, intelligent early warning: the system analyzes physiological data of a patient based on an intelligent algorithm, identifies possible diseases and gives early warning in advance, so that doctors can take measures in time, and the deterioration of the illness state is avoided. 3. Personalized treatment: the system can tailor the treatment regimen for each patient according to the personal characteristics and physiological index of the diabetic patient. By means of the accurate treatment scheme, the diabetes condition can be better controlled, and complications are reduced. 4, improving the life quality of the patient: the system provides more convenient and quick medical service for patients. Through real-time monitoring and intelligent early warning, the patient can control the illness state better, reduce the times of hospitalization and visit, improve quality of life. 5, medical care burden is reduced: the system can automatically collect and analyze physiological data of diabetics, lighten the workload of medical staff, and enable doctors to monitor the diabetics more efficiently, thereby improving the working efficiency.
Drawings
The invention will be further described with reference to the accompanying drawings, in which embodiments do not constitute any limitation on the invention, and other drawings can be obtained by one of ordinary skill in the art without undue effort from the following drawings.
Fig. 1 is a schematic diagram of the structure of the present invention.
Detailed Description
The diabetes intelligent interconnection remote real-time monitoring management system based on the algorithm and the big data comprises a data acquisition module, a data analysis module, a disease early warning module and a remote medical consultation module, wherein the data acquisition module comprises two parts of hardware and software, diabetes disease data of a patient are uniformly acquired, the data analysis module has the functions of data cleaning, feature extraction, data modeling and monitoring report generation, real-time analysis is carried out on the data of the patient, corresponding monitoring reports are generated, reference bases are provided for doctors and the patient, the disease early warning module extracts image features on the basis of historical data and real-time data of the patient, and utilizes an improved scale invariant feature transformation algorithm for extracting image features of a disease feature map of the diabetes patient, and timely early warning and prompting the patient and the doctor, and the remote medical consultation module realizes online communication between the doctor and the patient, so that the doctor can master the disease condition and the data of the patient in real time and timely give diagnosis and suggestion.
Specifically, the data acquisition module comprises two parts, namely hardware and software: in terms of hardware, existing diabetes monitoring equipment such as a blood glucose meter, a blood pressure meter and a weight meter is used for data transmission with a system through Bluetooth, wiFi and other technologies; in terms of software, the data acquisition module communicates with equipment of different models through various interfaces and API modes to acquire and analyze data transmitted by the equipment, and meanwhile, the data acquisition module also supports data input of various formats, such as Excel and CSV formats; in the process of data acquisition, the data acquisition module can perform real-time verification, verification and filtration on the data, so that the accuracy and the integrity of the data are ensured, the acquired data can be optimized through data compression, encryption and other technologies, and the transmission efficiency and the safety of the data are improved; in addition, the data acquisition module not only can acquire the diabetes monitoring data of the patient, but also can acquire other related information, such as basic information and treatment records of the patient, and the acquisition of the data provides references when a doctor diagnoses the patient, so that the doctor can better know the illness state of the patient and formulate a more effective treatment scheme.
Specifically, the data analysis module has the following functions:
(1) Data cleaning: the collected data is subjected to denoising, duplication removing, complement and other treatments, so that the quality and the integrity of the data are ensured;
(2) Feature extraction: extracting key features in the patient diabetes condition data, such as blood glucose level, insulin dosage, exercise amount and the like, so as to further analyze and predict;
(3) Modeling data: based on machine learning and data mining algorithms, constructing a model to predict the trend and risk of the diabetes mellitus of the patient, and evaluating the current health state of the patient;
(4) Monitoring report generation: generating a monitoring report from the data analysis result, wherein the monitoring report comprises contents such as the current condition, the disease change trend, the potential risk and the like of a patient, and the contents are used for reference and analysis of doctors and the patient;
Through real-time analysis and monitoring of the data analysis module, doctors and patients can more comprehensively understand the illness state and the health state of the patients, and timely take corresponding treatment measures and adjustment schemes, so that the treatment effect and the life quality are improved.
Preferably, the disease early warning module is used for predicting and analyzing based on historical data and real-time data of a patient, early warning and prompting the patient and doctors in time, and the main functions of the disease early warning module comprise data preprocessing, disease prediction, early warning prompting and data visualization, wherein the data preprocessing function is used for processing acquired patient data, removing abnormal values and noise in the data and ensuring the accuracy and reliability of the data; the disease prediction function predicts the disease of the patient based on historical data and real-time data by adopting an improved scale invariant feature transformation algorithm, analyzes the disease change trend of the patient, and predicts the possible disease change of the patient through a disease feature map of the patient, wherein the method comprises the following steps:
For a disease feature map of a diabetic patient, extracting image features by using an improved scale invariant feature transformation algorithm, firstly constructing a scale space, converting an image into the scale space, carrying out convolution calculation on an image I (x, y) by using a Gaussian function G (x, y, sigma), converting a two-dimensional image on a plane into the scale space image, and calculating as follows: l (x, y, σ) =g (x, y, σ) ×i (x, y), where L (x, y, σ) is an image in scale space, x is an abscissa of the image, y is an ordinate of the image, σ is a spatial scale factor, and G (x, y, σ) is a gaussian kernel function, satisfying: I (x, y) is a two-dimensional image in a plane, and in order to obtain an extreme point D (x, y, σ) corresponding to a scale space, it is calculated as follows:
D(x，y，σ)＝(G(x，y，kσ)-G(x，y，σ))*I(x，y)＝L(x，y，kσ)-L(x，y，σ)
Wherein k is a multiplier factor for changing the scale size of the adjacent scale;
Then, through two steps of image Gaussian blur and image downsampling, a Gaussian differential pyramid is generated, a specific gradual change image is obtained through image downsampling, then the gradual change image is arranged according to a sequence, an original image is at the lowest end, each layer of the Gaussian differential pyramid is obtained through two adjacent layers of result differential operation of the Gaussian pyramid, then extreme points of the image are detected, each point in the image is firstly screened for one time, whether each point has an extreme value or not is analyzed, whether the extreme value point is judged, whether the extreme point is the maximum value or the minimum value is judged by comparing the point with 26 points of adjacent layers, if the extreme point is the maximum value, the point is remained, the positioning of characteristic points is obtained through function fitting, namely, the secondary Taylor expansion operation is carried out on Gaussian differential operators D (x, y and sigma), and then the sampling point is calculated:
deriving the above equation, and letting D (X) =0, there are: Wherein/> When the offset is greater than 0.5 in one of the three dimensions x, y and sigma, the characteristic point is adjusted to an adjacent position, and then the offset corresponding to the characteristic point is estimated until the characteristic point is converged in all dimensions, and when the position of the characteristic point exceeds the boundary range, the characteristic point is discarded; clearing out the response points with poor reliability through the sea plug matrix, specifically: the principal curvature is analyzed, the reliability of the characteristic points is judged, the principal curvature is obtained through the sea plug matrix of the characteristic points, and the calculation formula is as follows:
Wherein D xx represents the second order bias derivative of x after solving the bias derivative of x for the gaussian difference operator, dx y represents the second order bias derivative of y after solving the bias derivative of x for the gaussian difference operator, D yy represents the second order bias derivative of y after solving the bias derivative of y for the gaussian difference operator, and the analysis matrix H feature values a and b sequentially refer to gradients corresponding to feature points in the two-axis direction, which can be known:
tr(H)＝Dxx+Dyy＝a+b
det(H)＝DxxDyy-(Dyy)2＝ab
Where tr (·) is the trace operation of the matrix, det (·) is the determinant operation of the matrix, and c is the ratio of two eigenvalues of a and b, with: a=cb, satisfying: in the deployment test, if If true, the point is indicated to be a stable point, otherwise, the point needs to be removed.
Preferably, the distribution of the directions of the feature points is performed, and in the scale-invariant feature transformation process, because each feature point is distributed with a relative direction according to statistics, the rotation of the image is not realized, the gradient value and the gradient direction of the adjacent pixels of the feature point are calculated, and the direction parameters of the feature points are obtained, and are calculated as follows:
θ(x，y)＝tan-1((L(x，y+1)-L(x，y-1))/L(x+1，y)-L(x-1，y))
Wherein L (x, y) refers to a feature value corresponding to a pixel point (x, y), m (x, y) refers to a gradient modulus value corresponding to the pixel point (x, y), θ (x, y) mainly refers to a gradient direction value corresponding to the pixel point (x, y), after the feature point neighborhood gradient information is subjected to the expansion operation, a Len dimension feature vector is required to be generated, detailed statistics must be expanded through a gradient histogram, the gradient histogram divides the direction into 36 columns, each column refers to a specific gradient direction, the direction of the column where the extremum is selected to serve as a main direction thereof, the direction of the column where the second limit is selected to serve as a corresponding auxiliary direction, 45 ° is divided into 8 directions, namely, east, south, west, north, northeast, southeast, northwest and southwest, and then the feature point is generated, in order to ensure that rotation invariance is well ensured, and standardized conversion is expanded to the main direction:
where x 'and y' are the abscissa and ordinate of the rotated and transformed image pixel respectively,
Wherein,Is a feature descriptor,/>Normalized feature descriptors define a matching image feature descriptor as x= (X 1,x2,...,xLen), a feature descriptor in the image to be registered is mainly y= (Y 1,y2,...,yLen), and if their euclidean distance ratio does not exceed a specific threshold, then this feature point is considered to achieve effective matching, and the euclidean distance is expressed as follows:
Specifically, the scale-invariant feature transform algorithm is further improved, which comprises the following steps: reducing the number of sub-pixel areas by re-dividing the rectangular pixel areas so as to reduce the dimension of the feature vector of the feature point; in the matching stage, the matching condition is not only the ratio of Euclidean distance, but also the vector correlation coefficient is integrated to ensure the matching accuracy of the feature points, the dimension of the lightweight original feature vector is updated to be Len *, and the matching condition is obtained after normalization processing:
Wherein D is a descriptor, The normalized descriptor is then fused into a vector correlation coefficient for a scale invariant and feature transformation algorithm, the length similarity of the vector X, Y is defined as a (X, Y), the length similarity is a standard for measuring the vector size, and b is: Wherein c (X ', Y) =a (X', Y) ·b (X ', Y'), wherein c (X, Y) ∈ [ -1,1], when θ=ζ/2, the two vectors are orthogonal, wherein the similarity coefficient c (X, Y) =0, when θ=0, and the norms are the same, the similarity coefficient c (X, Y) =1; when θ=ζ, and the norms are the same, the similarity coefficient c (X, Y) = -1 of the two vectors;
in addition, when the illness state of the patient is abnormal or possibly abnormal, the illness state early-warning module can send out early-warning prompt to prompt the patient to seek medical attention or adjust the treatment scheme in time; the data visualization function intuitively displays analysis results in the form of charts and the like, so that doctors and patients clearly know the change trend of the illness state, doctors are helped to make more reasonable diagnosis and treatment schemes, and the treatment effect of the patients is improved.
Specifically, the remote medical consultation module is mainly used for realizing online communication between a doctor and a patient, enabling the doctor to master the illness state and data of the patient in real time and timely giving diagnosis and treatment comments and suggestions, realizing remote communication between the doctor and the patient in a mode of Internet, mobile application programs and the like, uploading the illness state and the data of the patient by the patient, checking the data and illness state reports uploaded by the patient by a doctor through a login system, diagnosing and analyzing the illness state of the patient, giving treatment comments and guidance, simultaneously, enabling the doctor to conduct online communication with the patient through an instant communication function of the system, solving the doubt of the patient, and providing professional medical consultation service.
The intelligent diabetes interconnection remote real-time monitoring management system based on the algorithm and the big data has the beneficial effects in many aspects: the method has the advantages that the distinguishing performance is good, the rapid and accurate distinguishing information can be matched in a massive image database, a large number of feature vectors can be generated even if only a single object exists, the realization speed is high, the feature vector matching can be rapidly performed, the feature vectors can be combined and expanded with other forms of feature vectors, the innovation point is that the correlation coefficient is used for further classifying the number of image bits so as to reduce the complexity of calculation, the correlation coefficient can be used for measuring the relation of variables through numbers and has directivity, 1 represents positive correlation, 1 represents negative correlation, the intensity of the variable relation can be measured, and the closer to 0, the weaker the correlation is 1; in addition, the invention has the beneficial effects of real-time monitoring, intelligent early warning, personalized treatment, improving the life quality of patients and relieving the medical care burden, and is specifically described as follows: 1. and (3) real-time monitoring: the system can monitor the physiological index of a diabetic patient in real time, and provides more accurate data for doctors. Under the supervision of doctors, higher or lower data can be found in time, so that illness state is better controlled, and hospitalization rate is reduced. 2, intelligent early warning: the system analyzes physiological data of a patient based on an intelligent algorithm, identifies possible diseases and gives early warning in advance, so that doctors can take measures in time, and the deterioration of the illness state is avoided. 3. Personalized treatment: the system can tailor the treatment regimen for each patient according to the personal characteristics and physiological index of the diabetic patient. By means of the accurate treatment scheme, the diabetes condition can be better controlled, and complications are reduced. 4, improving the life quality of the patient: the system provides more convenient and quick medical service for patients. Through real-time monitoring and intelligent early warning, the patient can control the illness state better, reduce the times of hospitalization and visit, improve quality of life. 5, medical care burden is reduced: the system can automatically collect and analyze physiological data of diabetics, lighten the workload of medical staff, and enable doctors to monitor the diabetics more efficiently, thereby improving the working efficiency.
Finally, it should be noted that the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the scope of the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.

Claims (4)

1. The intelligent interconnected remote real-time monitoring management system for the diabetes is characterized by comprising a data acquisition module, a data analysis module, a disease condition early warning module and a remote medical consultation module, wherein the data acquisition module comprises two parts of hardware and software, the diabetes condition data of a patient are uniformly acquired, the data analysis module has the functions of data cleaning, feature extraction, data modeling and monitoring report generation, the data of the patient are analyzed in real time, corresponding monitoring reports are generated, reference bases are provided for doctors and the patient, the disease condition early warning module extracts image features based on historical data and real-time data of the patient, and an improved scale-invariant feature transformation algorithm is utilized for the disease condition feature map of the diabetes patient, prediction and analysis are carried out, the patient and doctor are timely warned and prompted, and the remote medical consultation module realizes online communication between the doctor and the patient, so that the doctor can grasp the disease condition and the data of the patient in real time, and diagnosis and suggestion are timely given;
The disease early warning module is used for predicting and analyzing based on historical data and real-time data of a patient, early warning and prompting the patient and doctors in time, and has the main functions of data preprocessing, disease prediction, early warning prompting and data visualization, wherein the data preprocessing function is used for processing acquired patient data, removing abnormal values and noise in the data and ensuring the accuracy and reliability of the data; the disease prediction function predicts the disease of the patient based on historical data and real-time data by adopting an improved scale invariant feature transformation algorithm, analyzes the disease change trend of the patient, and predicts the possible disease change of the patient through a disease feature map of the patient, wherein the method comprises the following steps:
For a disease feature map of a diabetic patient, extracting image features by using an improved scale invariant feature transformation algorithm, firstly constructing a scale space, converting an image into the scale space, carrying out convolution calculation on an image I (x, y) by using a Gaussian function G (x, y, sigma), converting a two-dimensional image on a plane into the scale space image, and calculating as follows: l (x, y, σ) =g (x, y, σ) ×i (x, y), where L (x, y, σ) is an image in scale space, x is an abscissa of the image, y is an ordinate of the image, σ is a spatial scale factor, and G (x, y, σ) is a gaussian kernel function, satisfying: I (x, y) is a two-dimensional image in a plane, and in order to obtain an extreme point D (x, y, σ) corresponding to a scale space, it is calculated as follows:
D(x，y，σ)＝(G(x，y，kσ)-G(x，y，σ))*I(x，y)＝L(x，y，kσ)-L(x，y，σ)
Wherein k is a multiplier factor for changing the scale size of the adjacent scale;
Then, through two steps of image Gaussian blur and image downsampling, a Gaussian differential pyramid is generated, a specific gradual change image is obtained through image downsampling, then the gradual change image is arranged according to a sequence, an original image is at the lowest end, each layer of the Gaussian differential pyramid is obtained through two adjacent layers of result differential operation of the Gaussian pyramid, then extreme points of the image are detected, each point in the image is firstly screened for one time, whether each point has an extreme value or not is analyzed, whether the extreme value point is judged, whether the extreme point is the maximum value or the minimum value is judged by comparing the point with 26 points of adjacent layers, if the extreme point is the maximum value, the point is remained, the positioning of characteristic points is obtained through function fitting, namely, the secondary Taylor expansion operation is carried out on Gaussian differential operators D (x, y and sigma), and then the sampling point is calculated:
deriving the above equation, and letting D (X) =0, there are: Wherein/> When the offset is greater than 0.5 in one of the three dimensions x, y and sigma, the characteristic point is adjusted to an adjacent position, and then the offset corresponding to the characteristic point is estimated until the characteristic point is converged in all dimensions, and when the position of the characteristic point exceeds the boundary range, the characteristic point is discarded; clearing out the response points with poor reliability through the sea plug matrix, specifically: the principal curvature is analyzed, the reliability of the characteristic points is judged, the principal curvature is obtained through the sea plug matrix of the characteristic points, and the calculation formula is as follows:
Wherein, D xx represents the second order bias derivative of x after solving the bias derivative of x for the gaussian difference operator, D xy represents the second order bias derivative of y after solving the bias derivative of x for the gaussian difference operator, D yy represents the second order bias derivative of y after solving the bias derivative of y for the gaussian difference operator, and the analysis matrix H feature values a and b sequentially refer to gradients corresponding to feature points in the two-axis direction, which can be known:
tr(H)＝Dxx+Dyy＝a+b
det(H)＝DxxDyy-(Dyy)2＝ab
where tr (·) is the trace operation of the matrix, det (·) is the determinant operation of the matrix, and is recorded For/>And/>The ratio of the two eigenvalues of (2) is: /(I)The method meets the following conditions: /(I)In the deployment test, ifIf true, indicating that the point is a stable point, otherwise, eliminating the point;
And (3) distributing the directions of the feature points, wherein in the scale-invariant feature transformation process, each feature point is distributed with a relative direction according to statistics, so that the rotation of the image is not realized, the gradient value and gradient direction of adjacent pixels of the feature points are calculated, and the feature point direction parameters are obtained, and are calculated as follows:
θ(x，y)＝tan-1((L(x，y+1)-L(x，y-1))/L(x+1，y)-L(x-1，y))
Wherein L (x, y) refers to a feature value corresponding to a pixel point (x, y), m (x, y) refers to a gradient modulus value corresponding to the pixel point (x, y), θ (x, y) mainly refers to a gradient direction value corresponding to the pixel point (x, y), after the feature point neighborhood gradient information is subjected to the expansion operation, a Len dimension feature vector is required to be generated, detailed statistics must be expanded through a gradient histogram, the gradient histogram divides the direction into 36 columns, each column refers to a specific gradient direction, the direction of the column where the extremum is selected to serve as a main direction thereof, the direction of the column where the second limit is selected to serve as a corresponding auxiliary direction, 45 ° is divided into 8 directions, namely, east, south, west, north, northeast, southeast, northwest and southwest, and then the feature point is generated, in order to ensure that rotation invariance is well ensured, and standardized conversion is expanded to the main direction:
Wherein x 'and y' are respectively the horizontal and vertical coordinates of the image pixel point after rotation transformation,
Defining the feature descriptors of the matched images as x= (X 1,x2,...,xLen), wherein a certain feature descriptor in the images to be registered is mainly y= (Y 1,y2,...,yLen), and if the euclidean distance ratio of the feature descriptors does not exceed a specific threshold value, the feature point is considered to realize effective matching, and the euclidean distance d (X, Y) is expressed as follows:
Further improvements to scale invariant feature transform algorithms, including: reducing the number of sub-pixel areas by re-dividing the rectangular pixel areas so as to reduce the dimension of the feature vector of the feature point; in the matching stage, the matching condition is not only the ratio of Euclidean distance, but also the vector correlation coefficient is integrated to ensure the matching accuracy of the feature points, the dimension of the lightweight original feature vector is updated to be Len *, and the matching condition is obtained after normalization processing:
Wherein Dis is a descriptor of the type, The normalized descriptor is then fused into a vector correlation coefficient for a scale invariant and feature transformation algorithm, the length similarity of the vector X, Y is defined as a (X, Y), the length similarity is a standard for measuring the vector size, and b is: /(I)Wherein c (X ', Y) =a (X', Y) ·b (X ', Y'), wherein c (X, Y) ∈ [ -1,1], when θ=pi/2, the two vectors are orthogonal, where the similarity coefficient c (X, Y) =0, when θ=0, and the norms are the same, the similarity coefficient c (X, Y) =1; when θ=pi, and the norms are the same, the similarity coefficient c (X, Y) = -1 of the two vectors;
In addition, when the illness state of the patient is abnormal or possibly abnormal, the illness state early-warning module can send out early-warning prompt to prompt the patient to seek medical attention or adjust the treatment scheme in time; the data visualization function intuitively displays the analysis result in a chart form, so that doctors and patients clearly know the change trend of the illness state, and the doctors are helped to make a more reasonable diagnosis and treatment scheme, and the treatment effect of the patients is improved.
2. The intelligent interconnected remote real-time monitoring management system for diabetes based on algorithm and big data according to claim 1, wherein the data acquisition module comprises two parts of hardware and software: in terms of hardware, the existing diabetes monitoring equipment, including a blood glucose meter, a blood pressure meter and a weight meter, is used for data transmission with a system through Bluetooth and WiFi technologies; in terms of software, the data acquisition module communicates with equipment of different models through various interfaces and API modes to acquire and analyze data transmitted by the equipment, and meanwhile, the data acquisition module also supports data input of various formats, including Excel and CSV formats; in the process of data acquisition, the data acquisition module can perform real-time verification, verification and filtration on the data, so that the accuracy and the integrity of the data are ensured, the acquired data can be optimized through data compression and encryption technology, and the transmission efficiency and the safety of the data are improved; in addition, the data acquisition module not only can acquire the diabetes monitoring data of the patient, but also can acquire other related information including basic information and treatment records of the patient, and the acquisition of the data provides references when a doctor diagnoses the patient, so that the doctor can better know the illness state of the patient and formulate a more effective treatment scheme.
3. The intelligent interconnected remote real-time monitoring management system for diabetes based on the algorithm and the big data according to claim 1, wherein the data analysis module has the following functions:
(1) Data cleaning: denoising, deduplication and complement processing are carried out on the acquired data, so that the quality and the integrity of the data are ensured;
(2) Feature extraction: extracting key features in the diabetes condition data of the patient, including blood glucose level, insulin dosage and exercise amount, so as to further analyze and predict;
(3) Modeling data: based on machine learning and data mining algorithms, constructing a model to predict the trend and risk of the diabetes mellitus of the patient, and evaluating the current health state of the patient;
(4) Monitoring report generation: generating a monitoring report from the data analysis result, wherein the monitoring report comprises the current condition, the disease change trend and the potential risk content of the patient, and the monitoring report is used for reference and analysis of doctors and the patient;
Through real-time analysis and monitoring of the data analysis module, doctors and patients can more comprehensively understand the illness state and the health state of the patients, and timely take corresponding treatment measures and adjustment schemes, so that the treatment effect and the life quality are improved.
4. The intelligent interconnected remote real-time monitoring management system for diabetes based on the algorithm and the big data according to claim 1, wherein the remote medical consultation module is mainly used for realizing online communication between doctors and patients, enabling the doctors to master the conditions and data of the patients in real time and timely giving diagnosis and treatment comments and suggestions, the module is used for realizing remote communication between the doctors and the patients in an Internet and mobile application program mode, the patients upload the conditions and data of the patients through the system, the doctors check the data and the condition reports uploaded by the patients through a login system, diagnose and analyze the conditions of the patients, give treatment comments and guidance, and simultaneously, enable the doctors to communicate with the patients online through the instant communication function of the system, answer the questions of the patients and provide professional medical consultation services.
CN202310207448.XA 2023-03-06 2023-03-06 Intelligent diabetes interconnection remote real-time monitoring management system based on algorithm and big data Active CN116313083B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310207448.XA CN116313083B (en) 2023-03-06 2023-03-06 Intelligent diabetes interconnection remote real-time monitoring management system based on algorithm and big data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310207448.XA CN116313083B (en) 2023-03-06 2023-03-06 Intelligent diabetes interconnection remote real-time monitoring management system based on algorithm and big data

Publications (2)

Publication Number Publication Date
CN116313083A CN116313083A (en) 2023-06-23
CN116313083B true CN116313083B (en) 2024-06-04

Family

ID=86812567

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310207448.XA Active CN116313083B (en) 2023-03-06 2023-03-06 Intelligent diabetes interconnection remote real-time monitoring management system based on algorithm and big data

Country Status (1)

Country Link
CN (1) CN116313083B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116936102A (en) * 2023-08-24 2023-10-24 南方医科大学南方医院 Diagnosis and treatment early warning system and method for middle-aged and elderly patients based on big data
CN117079807A (en) * 2023-08-24 2023-11-17 南方医科大学南方医院 Intelligent diagnosis support system and method for common diseases of middle-aged and elderly people
CN116825336A (en) * 2023-08-30 2023-09-29 山东志诚普惠健康科技有限公司 AI-based medical information intelligent management method and system
CN116844733B (en) * 2023-08-31 2023-11-07 吉林大学第一医院 Medical data integrity analysis method based on artificial intelligence
CN117219247B (en) * 2023-11-08 2024-02-23 厦门培邦信息科技有限公司 Intelligent management system for patient treatment
CN117275699B (en) * 2023-11-23 2024-02-13 四川省医学科学院·四川省人民医院 Wisdom ward system

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20040101721A (en) * 2003-05-26 2004-12-03 주식회사 케어앤케어 System and method for remote taking care of diabetic
CN105740378A (en) * 2016-01-27 2016-07-06 北京航空航天大学 Digital pathology whole slice image retrieval method
CN111145908A (en) * 2019-12-26 2020-05-12 糖医生健康管理南京有限公司 Health assessment comprehensive information service platform

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102009038364A1 (en) * 2009-08-23 2011-02-24 Friedrich-Alexander-Universität Erlangen-Nürnberg Method and system for automatic object recognition and subsequent object tracking according to the object shape

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20040101721A (en) * 2003-05-26 2004-12-03 주식회사 케어앤케어 System and method for remote taking care of diabetic
CN105740378A (en) * 2016-01-27 2016-07-06 北京航空航天大学 Digital pathology whole slice image retrieval method
CN111145908A (en) * 2019-12-26 2020-05-12 糖医生健康管理南京有限公司 Health assessment comprehensive information service platform

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于SIFT算子融合最大相异系数的自适应图像匹配算法;陈虹;肖越;肖成龙;宋好;;计算机应用;20171226(第05期);全文 *

Also Published As

Publication number Publication date
CN116313083A (en) 2023-06-23

Similar Documents

Publication Publication Date Title
CN116313083B (en) Intelligent diabetes interconnection remote real-time monitoring management system based on algorithm and big data
WO2018120942A1 (en) System and method for automatically detecting lesions in medical image by means of multi-model fusion
WO2021093448A1 (en) Image processing method and apparatus, server, medical image processing device and storage medium
CN104424386A (en) Multi-parameter magnetic resonance image based prostate cancer computer auxiliary identification system
CN112465905A (en) Characteristic brain region positioning method of magnetic resonance imaging data based on deep learning
CN112966792B (en) Blood vessel image classification processing method, device, equipment and storage medium
CN117009924B (en) Multi-mode self-adaptive multi-center data fusion method and system guided by electronic medical records
CN112750531A (en) Automatic inspection system, method, equipment and medium for traditional Chinese medicine
CN117711578B (en) Medical image data analysis management system
Elayaraja et al. An efficient approach for detection and classification of cancer regions in cervical images using optimization based CNN classification approach
CN112767374A (en) Alzheimer disease focus region semantic segmentation algorithm based on MRI
WO2023110477A1 (en) A computer implemented method and a system
CN116433660A (en) Medical image data processing device, electronic apparatus, and computer-readable storage medium
CN117854749A (en) Endocrine metabolism disease remote monitoring and early warning system based on artificial intelligence
CN115331071A (en) Tuberculous meningoencephalitis prediction method and system based on multi-scale feature map
CN117557840B (en) Fundus lesion grading method based on small sample learning
CN113724185B (en) Model processing method, device and storage medium for image classification
CN113284613A (en) Face diagnosis system based on deep learning
CN113592769A (en) Abnormal image detection method, abnormal image model training method, abnormal image detection device, abnormal image model training device and abnormal image model training medium
CN113610746A (en) Image processing method and device, computer equipment and storage medium
CN116091446A (en) Method, system, medium and equipment for detecting abnormality of esophageal endoscope image
CN117496323B (en) Multi-scale second-order pathological image classification method and system based on transducer
Hilal et al. Design of Intelligent Alzheimer Disease Diagnosis Model on CIoT Environment
CN116738352B (en) Method and device for classifying abnormal rod cells of retinal vascular occlusion disease
CN115831356B (en) Auxiliary prediction diagnosis method based on artificial intelligence algorithm

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
TA01 Transfer of patent application right

Effective date of registration: 20240102

Address after: Room 501, Room 502, room 503, room 504, room 505, floor 5, building 10, Industrialization Center, No.19 Alishan Road, Songshanhu high tech Industrial Development Zone, Dongguan City, Guangdong Province 523000

Applicant after: Guangdong deao smart Medical Technology Co.,Ltd.

Address before: People's Hospital, No. 504 Qingnian Road, Xishan District, Kunming City, Yunnan Province 650000

Applicant before: Song Chao

CB03 Change of inventor or designer information
CB03 Change of inventor or designer information

Inventor after: Qu Binbin

Inventor after: Xu Yahui

Inventor after: Song Chao

Inventor after: Li Bobin

Inventor before: Song Chao

Inventor before: Li Bobin

TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20240513

Address after: 266042 No. four, No. 127 South Road, Shandong, Qingdao

Applicant after: QINGDAO CENTRAL Hospital

Country or region after: China

Address before: Room 501, Room 502, room 503, room 504, room 505, floor 5, building 10, Industrialization Center, No.19 Alishan Road, Songshanhu high tech Industrial Development Zone, Dongguan City, Guangdong Province 523000

Applicant before: Guangdong deao smart Medical Technology Co.,Ltd.

Country or region before: China

GR01 Patent grant
GR01 Patent grant