CN108903930A - A kind of fetal heart rate curve categorizing system, method and device - Google Patents
A kind of fetal heart rate curve categorizing system, method and device Download PDFInfo
- Publication number
- CN108903930A CN108903930A CN201810382930.6A CN201810382930A CN108903930A CN 108903930 A CN108903930 A CN 108903930A CN 201810382930 A CN201810382930 A CN 201810382930A CN 108903930 A CN108903930 A CN 108903930A
- Authority
- CN
- China
- Prior art keywords
- heart rate
- fetal heart
- rate curve
- classification
- curve
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
- A61B5/02411—Detecting, measuring or recording pulse rate or heart rate of foetuses
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Animal Behavior & Ethology (AREA)
- Cardiology (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Veterinary Medicine (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Physiology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Gynecology & Obstetrics (AREA)
- Pregnancy & Childbirth (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The present invention is applicable in field of artificial intelligence,Provide a kind of fetal heart rate curve categorizing system,Method and device,The system includes medical treatment acquisition equipment,User equipment and Cloud Server,Medical treatment acquisition equipment,Cloud Server is connect with user equipment respectively,Medical treatment acquisition equipment is used to acquire the fetal heart rate curve of fetus in user's abdomen,Fetal heart rate curve is sent to user equipment,User equipment is used to fetal heart rate curve being sent to Cloud Server,And receive Cloud Server return,The healthy classification of fetal heart rate curve simultaneously exports,Cloud Server is used to fetal heart rate curve being divided into fetal heart rate curve section,The healthy classification of each fetal heart rate curve section is identified by trained convolutional neural networks,The healthy classification of fetal heart rate curve is determined according to the healthy classification of fetal heart rate curve section,And the healthy classification of fetal heart rate curve is sent to user equipment,To effectively improve the accuracy of fetal heart rate curve classification,Efficiency and intelligence degree.
Description
Technical field
The invention belongs to field of artificial intelligence more particularly to a kind of fetal heart rate curve categorizing systems, method and device.
Background technique
Efficiently, intelligentized electronic fetal monitoring system is the developing direction of fetal monitoring system, to guarantee as much as possible
The health of pregnant woman and fetus.In electronic fetal monitoring (Electronic Fetal Monitoring, EFM), Fetal Heart Rate is identified
Whether (Fetal Heart Rate, FHR) curve is that extremely critical, domestic fetal heart rate curve identification is differentiated with doctor extremely
Based on, supplemented by computer statistics method, it is clear that the identification of fetal heart rate curve not up to automates and precision.Currently, most of
Fetal heart rate curve identifying system is all based on the feature extraction of basic statistical method, and this method needs great feature extraction work
Journey, and accuracy is also not very high.
Summary of the invention
The purpose of the present invention is to provide a kind of fetal heart rate curve categorizing systems, method and device, it is intended to solve existing skill
It is fetal heart rate curve classification accuracy in art, inefficient, and the problem of intelligence degree deficiency.
On the one hand, the present invention provides a kind of fetal heart rate curve categorizing system, the system comprises：
The fetal heart rate curve is sent to by medical treatment acquisition equipment for acquiring the fetal heart rate curve of fetus in user's abdomen
The user equipment of connection；
The user equipment being connect with Cloud Server and the medical treatment acquisition equipment, for sending out the fetal heart rate curve
The Cloud Server is given, and receives the healthy classification of the fetal heart rate curve that the Cloud Server returns, described and exports；And
The Cloud Server being connect with the user equipment, for the fetal heart rate curve to be divided into preset number
Fetal heart rate curve section identifies the healthy classification of each fetal heart rate curve section by trained convolutional neural networks, according to
The healthy classification of each fetal heart rate curve section, determines the healthy classification of the fetal heart rate curve, by the fetal heart rate curve
Healthy classification be sent to the user equipment.
On the other hand, the present invention provides a kind of fetal heart rate curve classification method, the method includes the following steps：
When receiving the acquisition instructions of user, the Fetal Heart Rate song that equipment acquires fetus in user's abdomen is acquired by medical treatment
Line, and the fetal heart rate curve is sent to user equipment；
The fetal heart rate curve is sent to Cloud Server by the user equipment, it on the cloud server will be described
Fetal heart rate curve is divided into preset number fetal heart rate curve section, and passes through trained convolutional Neural net on the Cloud Server
Network identifies the healthy classification of each fetal heart rate curve section；
According to the healthy classification of each fetal heart rate curve section, the healthy classification of the fetal heart rate curve is determined, and lead to
It crosses the Cloud Server and the healthy classification of the fetal heart rate curve is sent to the user equipment；
It is exported in the user equipment and shows the healthy classification of the fetal heart rate curve.
On the other hand, the present invention provides a kind of fetal heart rate curve sorter, described device includes：
Fetal Heart Rate acquisition unit, for acquiring equipment by medical treatment and acquiring user when receiving the acquisition instructions of user
The fetal heart rate curve of fetus in abdomen, and the fetal heart rate curve is sent to user equipment；
Curved section classification recognition unit, for the fetal heart rate curve to be sent to cloud service by the user equipment
The fetal heart rate curve is divided into preset number fetal heart rate curve section on the cloud server, and passes through the cloud by device
Trained convolutional neural networks identify the healthy classification of each fetal heart rate curve section on server；
Curve category determination unit determines the fetal rhythm for the healthy classification according to each fetal heart rate curve section
The healthy classification of rate curve, and the healthy classification of the fetal heart rate curve is sent to by the user by the Cloud Server and is set
It is standby；And
Curve category output unit, for being exported in the user equipment and showing the healthy class of the fetal heart rate curve
Not.
Medical treatment acquisition equipment acquires the fetal heart rate curve of fetus in user's abdomen in the present invention, and fetal heart rate curve is sent to use
Family equipment, fetal heart rate curve is sent to Cloud Server by the user equipment connecting with medical treatment acquisition equipment, Cloud Server, and is received
Healthy classification that Cloud Server returns, fetal heart rate curve simultaneously exports, and fetal heart rate curve is divided into Fetal Heart Rate song by Cloud Server
Line segment identifies the healthy classification of each fetal heart rate curve section by trained convolutional neural networks, according to these healthy classifications
The healthy classification of fetal heart rate curve is sent to user equipment, to improve fetal rhythm by the healthy classification for determining fetal heart rate curve
The degree of automation of curve classification effectively improves accuracy, efficiency and the intelligence degree of fetal heart rate curve classification.
Detailed description of the invention
Fig. 1 is a kind of structural schematic diagram for fetal heart rate curve categorizing system that the embodiment of the present invention one provides；
Fig. 2 is the healthy classification example for the corresponding fetal heart rate curve of fetal heart rate curve section that the embodiment of the present invention one provides
Figure；
Fig. 3 is a kind of preferred structure schematic diagram for fetal heart rate curve categorizing system that the embodiment of the present invention one provides；
Fig. 4 is a kind of implementation flow chart of fetal heart rate curve classification method provided by Embodiment 2 of the present invention；
Fig. 5 is a kind of structural schematic diagram of the sorter for fetal heart rate curve that the embodiment of the present invention three provides；And
Fig. 6 is a kind of structural schematic diagram of the sorter for fetal heart rate curve that the embodiment of the present invention four provides.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
Specific implementation of the invention is described in detail below in conjunction with specific embodiment：
Embodiment one：
Fig. 1 shows a kind of structure of fetal heart rate curve categorizing system of the offer of the embodiment of the present invention one, for the ease of saying
Bright, only parts related to embodiments of the present invention are shown.
In embodiments of the present invention, fetal heart rate curve categorizing system includes medical treatment acquisition equipment 11 and medical treatment acquisition equipment
The user equipment 12 of 11 connections and the Cloud Server 13 being connect with user equipment 12, wherein：
Fetal heart rate curve is sent to company for acquiring the fetal heart rate curve of fetus in user's abdomen by medical treatment acquisition equipment 11
The user equipment 12 connect.
In embodiments of the present invention, user is pregnant and lying-in women, and medical treatment acquisition equipment 11 is setting of being guarded to fetus
It is standby, such as ultrasonic Doppler fetal monitor, user equipment 12 can be that equipment, the users such as mobile phone, smartwatch, computer pass through
The fetal heart rate curve of fetus in medical treatment acquisition 11 multi collect abdomen of equipment is led to these fetal heart rate curves by medical treatment acquisition equipment 11
It crosses wireless (such as bluetooth, Wi-Fi) or wired connection mode is sent to user equipment 12.
In embodiments of the present invention, doctor would generally analyze 20 minutes fetal heart rate datas in clinical practice, therefore
In the specific implementation process, medical treatment acquisition equipment 11 is after 20 minutes Fetal Heart Rates of continuous collecting, so that it may generate a Fetal Heart Rate
Curve, acquisition can generate a plurality of fetal heart rate curve for a long time.Wherein, fetal heart rate curve recorded data is a high dimension
According to, such as ultrasonic Doppler fetal monitor is when acquiring fetal heart rate curve, it is per second to can record 4 heart rate datas, then 20 minutes
Fetal heart rate curve on record 4*60*20=4800 heart rate data.
User equipment 12 for fetal heart rate curve to be sent to Cloud Server 13, and receives the return of Cloud Server 13, tire
The healthy classification of heart rate curve simultaneously exports.
In embodiments of the present invention, user equipment 12 can be connected with Cloud Server by wireless network or data network
It connects, when user equipment 12 is even connected to the Net, fetal heart rate curve can be sent to Cloud Server 13, by Cloud Server 13 to these
Fetal heart rate curve is classified, and the healthy classification of these fetal heart rate curves is obtained, and receives Cloud Server 13 in user equipment 12
After the healthy classification of these fetal heart rate curves returned, the healthy classification of these fetal heart rate curves is exported, so that user checks, from
And improve the degree of automation of fetal heart rate curve classification.Preferably, the healthy classification of fetal heart rate curve include fetal heart rate curve just
Often, fetal heart rate curve is suspicious and fetal heart rate curve is abnormal, clearly to reflect whether fetus is healthy to user.
Cloud Server 13, for fetal heart rate curve to be divided into preset number fetal heart rate curve section, by trained
Convolutional neural networks identify the healthy classification of each fetal heart rate curve section, according to the healthy classification of each fetal heart rate curve section, really
The healthy classification of fetal heart rate curve is sent to user equipment 12 by the healthy classification for determining fetal heart rate curve.
In embodiments of the present invention, every fetal heart rate curve is divided into preset number Fetal Heart Rate song on Cloud Server 13
Line segment to increase the curve quantity of convolutional neural networks processing, and then improves the accuracy of fetal heart rate curve classification.Pass through training
Good convolutional neural networks identify the healthy classification of each fetal heart rate curve section, further according to the healthy class of these fetal heart rate curve sections
Not, the healthy classification of these corresponding fetal heart rate curves of fetal heart rate curve section is obtained.
Preferably, Cloud Server 13 obtains these fetal heart rate curve sections point in the healthy classification according to fetal heart rate curve section
When the healthy classification of not corresponding fetal heart rate curve, it can be obtained according to the healthy classification of fetal heart rate curve section by ballot mode
The classification frequency of these corresponding fetal heart rate curves of fetal heart rate curve section counts different health classifications in the fetal heart rate curve
The frequency of occurrences sets the highest healthy classification of classification frequency to the healthy classification of fetal heart rate curve, to effectively improve
The accuracy of fetal heart rate curve classification.For example, passing through as shown in Fig. 2, fetal heart rate curve is divided into 10 sections of fetal heart rate curve sections
The healthy classification of each fetal heart rate curve section can be obtained in convolutional neural networks, and it is final to obtain fetal heart rate curve by ballot mode
Healthy classification, it is abnormal, normal and suspicious that P, N, S respectively indicate fetal heart rate curve.
Preferably, as shown in figure 3, fetal heart rate curve categorizing system further includes connecting with user equipment 12, Cloud Server 13
Doctor's custodial care facility 34, wherein：
User equipment 12 is sent to doctor's monitoring and sets while the fetal heart rate curve of acquisition is sent to Cloud Server 13
Standby 34, doctor can check these fetal heart rate curves on doctor's custodial care facility 34, and by carrying out healthy class to fetal heart rate curve
The mode not marked judges the foetus health situation that these fetal heart rate curves reflect, when convolutional neural networks are in
When the training stage, doctor's custodial care facility 34 will be set as Fetal Heart Rate training curve by the fetal heart rate curve of category label, by tire
The category label of heart rate training curve is sent to Cloud Server 13, and Cloud Server 13 is according to the category label of Fetal Heart Rate training curve
With preset classification indicators, Training is carried out to convolutional neural networks, to effectively improve fetal heart rate curve classification
Accuracy.Wherein, classification indicators can be sensitivity (TPR), specificity (SPC), false positive rate (FPR), positive prediction of classification
Value (PPV), negative predictive value (NPV) and accuracy (ACC) etc..
In embodiments of the present invention, medical treatment acquisition equipment 11 acquires the fetal heart rate curve of fetus in user's abdomen, by Fetal Heart Rate
Curve is sent to user equipment 12, and the user equipment 12 connecting with medical treatment acquisition equipment 11, Cloud Server 13 is by fetal heart rate curve
It is sent to Cloud Server 13, and receives the return of Cloud Server 13, fetal heart rate curve healthy classification and exports, Cloud Server 13
Fetal heart rate curve is divided into fetal heart rate curve section, identifies each fetal heart rate curve section by trained convolutional neural networks
Healthy classification determines the healthy classification of fetal heart rate curve according to these health classifications, the healthy classification of fetal heart rate curve is sent
To user equipment 12, to improve the degree of automation of fetal rhythm curve classification, fetal heart rate curve classification is effectively improved
Accuracy, efficiency and intelligence degree.
Embodiment two：
Fig. 4 shows a kind of implementation process of fetal heart rate curve classification method provided by Embodiment 2 of the present invention, in order to just
In explanation, only parts related to embodiments of the present invention are shown, and details are as follows：
In step S401, when receiving the acquisition instructions of user, equipment is acquired by medical treatment and acquires user's abdomen inner tube of a tyre
The fetal heart rate curve of youngster, and fetal heart rate curve is sent to user equipment.
Fetal heart rate curve categorizing system of the embodiment of the present invention suitable for embodiment one, user acquire equipment by medical treatment
The fetal heart rate curve of fetus in multi collect abdomen, by medical treatment acquisition equipment by these fetal heart rate curves by wirelessly or non-wirelessly connecting
Mode is sent to user equipment, so that user checks.Doctor would generally analyze 20 minutes fetal heart frequency numbers in clinical practice
According to, therefore in the specific implementation process, medical treatment acquisition equipment is after 20 minutes Fetal Heart Rates of continuous collecting, so that it may generate a tire
Heart rate curve, acquisition can generate a plurality of fetal heart rate curve for a long time.
In step S402, fetal heart rate curve is sent to by Cloud Server by user equipment, by tire on Cloud Server
Heart rate curve is divided into preset number fetal heart rate curve section, and passes through convolutional neural networks identification trained on Cloud Server
The healthy classification of each fetal heart rate curve section.
In embodiments of the present invention, when user equipment is even connected to the Net, fetal heart rate curve can be sent out by user equipment
Cloud Server is given, is classified by Cloud Server to these fetal heart rate curves, obtains the healthy classification of these fetal heart rate curves.
Preferably, the healthy classification of fetal heart rate curve includes that fetal heart rate curve is normal, fetal heart rate curve is suspicious and fetal heart rate curve is abnormal,
Clearly to reflect whether fetus is healthy to user.
In embodiments of the present invention, every fetal heart rate curve is divided into preset number Fetal Heart Rate song on Cloud Server
Line segment to increase the curve quantity that convolutional neural networks are handled on Cloud Server, and then improves the accurate of fetal heart rate curve classification
Degree.It can recognize the healthy classification of each fetal heart rate curve section by trained convolutional neural networks.
In step S403, according to the healthy classification of each fetal heart rate curve section, the healthy classification of fetal heart rate curve is determined,
And the healthy classification of fetal heart rate curve is sent to by user equipment by Cloud Server.
In embodiments of the present invention, on Cloud Server, according to the healthy classification of these fetal heart rate curve sections, these are obtained
The healthy classification of the corresponding fetal heart rate curve of fetal heart rate curve section.It preferably, can be according to the healthy class of fetal heart rate curve section
Not, the classification frequency of these corresponding fetal heart rate curves of fetal heart rate curve section is obtained by ballot mode, that is, counts the Fetal Heart Rate
The frequency of occurrences of different health classifications in curve sets the highest healthy classification of classification frequency to the healthy class of fetal heart rate curve
Not, to effectively improve the accuracy of fetal heart rate curve classification.
In step s 404, it is exported in user equipment and shows the healthy classification of fetal heart rate curve.
In embodiments of the present invention, the healthy class of these fetal heart rate curves of Cloud Server return is received in user equipment
After not, the healthy classification of these fetal heart rate curves is exported by user equipment, so that user checks, to improve fetal heart rate curve
The degree of automation of classification.
Preferably, while the fetal heart rate curve of acquisition is sent to Cloud Server by user equipment, pass through user
These fetal heart rate curves are sent to doctor's custodial care facility by equipment, and doctor can check these Fetal Heart Rates song on doctor's custodial care facility
Line, and by way of carrying out healthy category label to fetal heart rate curve, the foetus health that these fetal heart rate curves are reflected
Situation judges, will be by the tire of category label by doctor's custodial care facility when convolutional neural networks are in the training stage
Heart rate curve is set as Fetal Heart Rate training curve, the category label of Fetal Heart Rate training curve is sent to Cloud Server, then in cloud
According to the category label of Fetal Heart Rate training curve and preset classification indicators on server, supervision has been carried out to convolutional neural networks
Training, to effectively improve the accuracy of fetal heart rate curve classification.
In embodiments of the present invention, the fetal heart rate curve of fetus in user's abdomen of medical treatment acquisition equipment acquisition is sent to use
Fetal heart rate curve is sent to Cloud Server by user equipment by family equipment, by being divided into fetal heart rate curve on Cloud Server
Multiple fetal heart rate curve sections identify the healthy classification of fetal heart rate curve section, by trained convolutional neural networks to determine tire
The healthy classification of heart rate curve effectively improves fetal heart rate curve to improve the degree of automation of fetal rhythm curve classification
Accuracy, efficiency and the intelligence degree of classification.
Embodiment three：
Fig. 5 shows a kind of structure of fetal heart rate curve sorter of the offer of the embodiment of the present invention three, for the ease of saying
Bright, only parts related to embodiments of the present invention are shown, including：
Fetal Heart Rate acquisition unit 51, for acquiring equipment acquisition by medical treatment and using when receiving the acquisition instructions of user
The fetal heart rate curve of fetus in the abdomen of family, and fetal heart rate curve is sent to user equipment.
Fetal heart rate curve categorizing system of the embodiment of the present invention suitable for embodiment one, user acquire equipment by medical treatment
The fetal heart rate curve of fetus in multi collect abdomen, by medical treatment acquisition equipment by these fetal heart rate curves by wirelessly or non-wirelessly connecting
Mode is sent to user equipment, so that user checks.
Curved section classification recognition unit 52, for fetal heart rate curve to be sent to Cloud Server by user equipment, in cloud
Fetal heart rate curve is divided into preset number fetal heart rate curve section on server, and passes through trained convolution on Cloud Server
The healthy classification of each fetal heart rate curve section of neural network recognization.
In embodiments of the present invention, when user equipment is even connected to the Net, fetal heart rate curve can be sent out by user equipment
Cloud Server is given, is classified by Cloud Server to these fetal heart rate curves, obtains the healthy classification of these fetal heart rate curves.
Preferably, the healthy classification of fetal heart rate curve includes that fetal heart rate curve is normal, fetal heart rate curve is suspicious and fetal heart rate curve is abnormal,
Clearly to reflect whether fetus is healthy to user.
In embodiments of the present invention, every fetal heart rate curve is divided into preset number Fetal Heart Rate song on Cloud Server
Line segment to increase the curve quantity that convolutional neural networks are handled on Cloud Server, and then improves the accurate of fetal heart rate curve classification
Degree.It can recognize the healthy classification of each fetal heart rate curve section by trained convolutional neural networks.
Curve category determination unit 53 determines fetal heart rate curve for the healthy classification according to each fetal heart rate curve section
Healthy classification, and the healthy classification of fetal heart rate curve is sent to by user equipment by Cloud Server.
In embodiments of the present invention, on Cloud Server, according to the healthy classification of these fetal heart rate curve sections, these are obtained
The healthy classification of the corresponding fetal heart rate curve of fetal heart rate curve section.It preferably, can be according to the healthy class of fetal heart rate curve section
Not, the classification frequency of these corresponding fetal heart rate curves of fetal heart rate curve section is obtained by ballot mode, that is, counts the Fetal Heart Rate
The frequency of occurrences of different health classifications in curve sets the highest healthy classification of classification frequency to the healthy class of fetal heart rate curve
Not, to effectively improve the accuracy of fetal heart rate curve classification.
Curve category output unit 54, for being exported in user equipment and showing the healthy classification of fetal heart rate curve.
In embodiments of the present invention, the healthy class of these fetal heart rate curves of Cloud Server return is received in user equipment
After not, the healthy classification of these fetal heart rate curves is exported by user equipment, so that user checks, to improve fetal heart rate curve
The degree of automation of classification.
In embodiments of the present invention, the fetal heart rate curve of fetus in user's abdomen of medical treatment acquisition equipment acquisition is sent to use
Fetal heart rate curve is sent to Cloud Server by user equipment by family equipment, by being divided into fetal heart rate curve on Cloud Server
Multiple fetal heart rate curve sections identify the healthy classification of fetal heart rate curve section, by trained convolutional neural networks to determine tire
The healthy classification of heart rate curve effectively improves fetal heart rate curve to improve the degree of automation of fetal rhythm curve classification
Accuracy, efficiency and the intelligence degree of classification.
In embodiments of the present invention, a kind of each unit of fetal heart rate curve sorter can be by corresponding hardware or software list
Member realizes that each unit can be independent soft and hardware unit in fetal heart rate curve categorizing system, also can integrate for one it is soft,
Hardware cell, herein not to limit the present invention.
Example IV：
Fig. 6 shows a kind of structure of fetal heart rate curve sorter of the offer of the embodiment of the present invention four, for the ease of saying
Bright, only parts related to embodiments of the present invention are shown, including：
Fetal Heart Rate acquisition unit 61, for acquiring equipment acquisition by medical treatment and using when receiving the acquisition instructions of user
The fetal heart rate curve of fetus in the abdomen of family, and fetal heart rate curve is sent to user equipment.
Curve output unit 62 passes through doctor for fetal heart rate curve to be sent to doctor's custodial care facility by user equipment
Raw custodial care facility exports fetal heart rate curve.
Category label unit 63, for receiving doctor to the category label of fetal heart rate curve by doctor's custodial care facility, and
Fetal Heart Rate training curve is set by the fetal heart rate curve for having carried out category label.
Transmission unit 64 is marked, for the category label of Fetal Heart Rate training curve to be sent to use by doctor's custodial care facility
Family equipment and Cloud Server.
In embodiments of the present invention, the fetal heart rate curve of acquisition is being sent to by the same of Cloud Server by user equipment
When, these fetal heart rate curves are sent to by doctor's custodial care facility by user equipment, doctor can check on doctor's custodial care facility
These fetal heart rate curves, and by way of carrying out healthy category label to fetal heart rate curve, these fetal heart rate curves are reflected
Foetus health situation out judges, and when convolutional neural networks are in the training stage, will be passed through by doctor's custodial care facility
The fetal heart rate curve of category label is set as Fetal Heart Rate training curve, and the category label of Fetal Heart Rate training curve is sent to cloud clothes
Business device, then according to the category label of Fetal Heart Rate training curve and preset classification indicators on Cloud Server, to convolutional Neural net
Network carries out Training, to effectively improve the accuracy of fetal heart rate curve classification.
Curved section classification recognition unit 65, for fetal heart rate curve to be sent to Cloud Server by user equipment, in cloud
Fetal heart rate curve is divided into preset number fetal heart rate curve section on server, and passes through trained convolution on Cloud Server
The healthy classification of each fetal heart rate curve section of neural network recognization.
In embodiments of the present invention, the fetal heart rate curve of trained convolutional neural networks lock identification is without doctor's class
The fetal heart rate curve not marked.
Curve category determination unit 66 determines fetal heart rate curve for the healthy classification according to each fetal heart rate curve section
Healthy classification, and the healthy classification of fetal heart rate curve is sent to by user equipment by Cloud Server.
Curve category output unit 67, for being exported in user equipment and showing the healthy classification of fetal heart rate curve.
In embodiments of the present invention, the fetal heart rate curve of fetus in user's abdomen of medical treatment acquisition equipment acquisition is sent to use
Fetal heart rate curve is sent to Cloud Server by user equipment by family equipment, by being divided into fetal heart rate curve on Cloud Server
Multiple fetal heart rate curve sections identify the healthy classification of fetal heart rate curve section, by trained convolutional neural networks to determine tire
The healthy classification of heart rate curve effectively improves fetal heart rate curve to improve the degree of automation of fetal rhythm curve classification
Accuracy, efficiency and the intelligence degree of classification.
In embodiments of the present invention, a kind of each unit of the sorter of fetal heart rate curve can be by corresponding hardware or software
Unit realizes that each unit can be independent soft and hardware unit in fetal heart rate curve categorizing system, and also can integrate is one
Soft and hardware unit, herein not to limit the present invention.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
1. a kind of fetal heart rate curve categorizing system, which is characterized in that the system comprises：
The fetal heart rate curve is sent to connection for acquiring the fetal heart rate curve of fetus in user's abdomen by medical treatment acquisition equipment
User equipment；
The user equipment being connect with Cloud Server and the medical treatment acquisition equipment, for the fetal heart rate curve to be sent to
The Cloud Server, and receive the healthy classification of the fetal heart rate curve that the Cloud Server returns, described and export；And
The Cloud Server being connect with the user equipment, for the fetal heart rate curve to be divided into preset number fetal rhythm
Rate curved section identifies the healthy classification of each fetal heart rate curve section by trained convolutional neural networks, according to described
The healthy classification of each fetal heart rate curve section determines the healthy classification of the fetal heart rate curve, by the strong of the fetal heart rate curve
Health classification is sent to the user equipment.
2. the system as claimed in claim 1, which is characterized in that the system also includes：
The doctor's custodial care facility being connect with the user equipment, the Cloud Server, for receiving the user equipment transmission
Fetal heart rate curve, and doctor is received to the category label of the fetal heart rate curve, the tire of the category label will have been carried out
Heart rate curve is set as Fetal Heart Rate training curve, and the category label of the Fetal Heart Rate training curve is sent to the user and is set
The standby and described Cloud Server.
3. system as claimed in claim 2, which is characterized in that the Cloud Server is according to the class of the Fetal Heart Rate training curve
Training Biao Ji not be carried out to the convolutional neural networks with preset classification indicators.
4. the system as claimed in claim 1, which is characterized in that the Cloud Server is according to each fetal heart rate curve section
Healthy classification calculates the classification frequency of the fetal heart rate curve, according to the classification frequency of the fetal heart rate curve, determines the tire
The healthy classification of heart rate curve.
5. a kind of fetal heart rate curve classification method based on any fetal heart rate curve categorizing system of claim 1-4, special
Sign is that the method includes the following steps：
When receiving the acquisition instructions of user, the fetal heart rate curve that equipment acquires fetus in user's abdomen is acquired by medical treatment, and
The fetal heart rate curve is sent to user equipment；
The fetal heart rate curve is sent to Cloud Server by the user equipment, on the cloud server by the fetal rhythm
Rate curve is divided into preset number fetal heart rate curve section, and is known by convolutional neural networks trained on the Cloud Server
The healthy classification of not described each fetal heart rate curve section；
According to the healthy classification of each fetal heart rate curve section, the healthy classification of the fetal heart rate curve is determined, and pass through institute
It states Cloud Server and the healthy classification of the fetal heart rate curve is sent to the user equipment；
It is exported in the user equipment and shows the healthy classification of the fetal heart rate curve.
6. method as claimed in claim 5, which is characterized in that the step of fetal heart rate curve is sent to user equipment it
Afterwards, before the step of fetal heart rate curve being sent to Cloud Server by the user equipment, the method also includes：
The fetal heart rate curve is sent to doctor's custodial care facility by the user equipment, is set by doctor monitoring
It is standby to export the fetal heart rate curve；
Doctor is received to the category label of the fetal heart rate curve by doctor's custodial care facility, and will carry out the classification
The fetal heart rate curve of label is set as Fetal Heart Rate training curve；
The category label of the Fetal Heart Rate training curve is sent to the user equipment and institute by doctor's custodial care facility
State Cloud Server.
7. method as claimed in claim 6, which is characterized in that by doctor's custodial care facility that Fetal Heart Rate training is bent
The category label of line was sent to after the step of user equipment and Cloud Server, the method also includes：
On the cloud server, according to the category label of the Fetal Heart Rate training curve and preset classification indicators, to described
Convolutional neural networks carry out Training.
8. method as claimed in claim 5, which is characterized in that according to the healthy classification of each fetal heart rate curve section, really
The step of healthy classification of the fixed fetal heart rate curve, including：
According to the healthy classification of each fetal heart rate curve section, the classification frequency of the fetal heart rate curve is calculated；
According to the classification frequency of the fetal heart rate curve, the healthy classification of the fetal heart rate curve is determined.
9. a kind of fetal heart rate curve sorter based on any fetal heart rate curve categorizing system of claim 1-4, special
Sign is that described device includes：
Fetal Heart Rate acquisition unit, for acquiring equipment by medical treatment and acquiring in user's abdomen when receiving the acquisition instructions of user
The fetal heart rate curve of fetus, and the fetal heart rate curve is sent to user equipment；
Curved section classification recognition unit, for the fetal heart rate curve to be sent to Cloud Server by the user equipment,
The fetal heart rate curve is divided into preset number fetal heart rate curve section on the Cloud Server, and passes through the Cloud Server
Upper trained convolutional neural networks identify the healthy classification of each fetal heart rate curve section；
Curve category determination unit determines that the Fetal Heart Rate is bent for the healthy classification according to each fetal heart rate curve section
The healthy classification of line, and the healthy classification of the fetal heart rate curve is sent to by the user equipment by the Cloud Server；
And
Curve category output unit, for being exported in the user equipment and showing the healthy classification of the fetal heart rate curve.
10. device as claimed in claim 9, which is characterized in that described device further includes：
Curve output unit, for the fetal heart rate curve to be sent to doctor's custodial care facility by the user equipment,
The fetal heart rate curve is exported by doctor's custodial care facility；
Category label unit, for receiving doctor to the category label of the fetal heart rate curve by doctor's custodial care facility,
And Fetal Heart Rate training curve is set by the fetal heart rate curve for having carried out the category label；And
Transmission unit is marked, for being sent to the category label of the Fetal Heart Rate training curve by doctor's custodial care facility
The user equipment and the Cloud Server.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810382930.6A CN108903930A (en) | 2018-04-26 | 2018-04-26 | A kind of fetal heart rate curve categorizing system, method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810382930.6A CN108903930A (en) | 2018-04-26 | 2018-04-26 | A kind of fetal heart rate curve categorizing system, method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108903930A true CN108903930A (en) | 2018-11-30 |
Family
ID=64403405
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810382930.6A Pending CN108903930A (en) | 2018-04-26 | 2018-04-26 | A kind of fetal heart rate curve categorizing system, method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108903930A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102302363A (en) * | 2011-06-29 | 2012-01-04 | 深圳市理邦精密仪器股份有限公司 | Device and method for automatically identifying fetal heart rate baseline |
| CN102319064A (en) * | 2011-10-13 | 2012-01-18 | 深圳市理邦精密仪器股份有限公司 | Device and method for improving accuracy of recognizing deceleration of fetal heart rate data |
| CN105105740A (en) * | 2015-09-06 | 2015-12-02 | 深圳市理邦精密仪器股份有限公司 | Fetus monitoring method and apparatus |
| CN105118503A (en) * | 2015-07-13 | 2015-12-02 | 中山大学 | Ripped audio detection method |
| CN105615855A (en) * | 2016-04-08 | 2016-06-01 | 江苏省人民医院 | Pregnant and lying-in woman health management system and method based on wireless transmission technology and mobile internet |
| CN106955097A (en) * | 2017-03-31 | 2017-07-18 | 福州大学 | A kind of fetal heart frequency state classification method |
-
2018
- 2018-04-26 CN CN201810382930.6A patent/CN108903930A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102302363A (en) * | 2011-06-29 | 2012-01-04 | 深圳市理邦精密仪器股份有限公司 | Device and method for automatically identifying fetal heart rate baseline |
| CN102319064A (en) * | 2011-10-13 | 2012-01-18 | 深圳市理邦精密仪器股份有限公司 | Device and method for improving accuracy of recognizing deceleration of fetal heart rate data |
| CN105118503A (en) * | 2015-07-13 | 2015-12-02 | 中山大学 | Ripped audio detection method |
| CN105105740A (en) * | 2015-09-06 | 2015-12-02 | 深圳市理邦精密仪器股份有限公司 | Fetus monitoring method and apparatus |
| CN105615855A (en) * | 2016-04-08 | 2016-06-01 | 江苏省人民医院 | Pregnant and lying-in woman health management system and method based on wireless transmission technology and mobile internet |
| CN106955097A (en) * | 2017-03-31 | 2017-07-18 | 福州大学 | A kind of fetal heart frequency state classification method |
Non-Patent Citations (1)
| Title |
|---|
| MIROSLAV BURSA ET AL: "The use of convolutional neural networks in biomedical processing", 《INFORMATION TECHNOLOGY IN BIO-AND MEDICAL INFORMATIONS》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110706786B (en) | Non-contact intelligent psychological parameter analysis and evaluation system | |
| CN102319064B (en) | Device and method for improving accuracy of recognizing deceleration of fetal heart rate data | |
| CN102509010B (en) | Remote medical detection system | |
| CN111613335A (en) | Health early warning system and method | |
| CN111968746A (en) | Cerebral apoplexy risk prediction method and device based on hybrid deep transfer learning | |
| CN111062021B (en) | Method and device for identity authentication based on wearable equipment | |
| CN103440421A (en) | Medical data processing method and system | |
| CN114334169B (en) | Medical object category decision method and device, electronic equipment and storage medium | |
| CN117334331B (en) | Medical diagnosis system for health condition based on artificial intelligence | |
| CN111370124A (en) | Health analysis system and method based on facial recognition and big data | |
| CN113889252A (en) | Remote internet big data intelligent medical system based on vital sign big data clustering core algorithm and block chain | |
| CN104055506B (en) | A kind of fetal monitoring data processing method and device | |
| CN113810792B (en) | Edge data acquisition and analysis system based on cloud computing | |
| CN114707608A (en) | Medical quality control data processing method, apparatus, device, medium, and program product | |
| CN114628033A (en) | Disease risk prediction method, device, equipment and storage medium | |
| CN109933680A (en) | Image data screening technique, device, computer equipment and its storage medium | |
| CN111261298B (en) | Medical data quality prejudging method and device, readable medium and electronic equipment | |
| CN108903930A (en) | A kind of fetal heart rate curve categorizing system, method and device | |
| CN112599250A (en) | Postoperative data analysis method and device based on deep neural network | |
| Beaulah et al. | Lung Melanoma Recognition through Platelet Count Composition Using Supervised ML Technique | |
| CN110974220A (en) | Brain wave bone marrow cell classification counting system and method | |
| CN114167837A (en) | Intelligent fault diagnosis method and system for railway signal system | |
| CN107194918A (en) | Data analysing method and device | |
| CN114088605A (en) | Blood cell sorting method and related apparatus | |
| CN113035351A (en) | System and method for analyzing data of re-diagnosis treatment of liver cancer patient |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181130 |