CN107595243B - Disease evaluation method and terminal equipment - Google Patents
Disease evaluation method and terminal equipment Download PDFInfo
- Publication number
- CN107595243B CN107595243B CN201710632865.3A CN201710632865A CN107595243B CN 107595243 B CN107595243 B CN 107595243B CN 201710632865 A CN201710632865 A CN 201710632865A CN 107595243 B CN107595243 B CN 107595243B
- Authority
- CN
- China
- Prior art keywords
- data
- physiological data
- physiological
- dynamic
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 61
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 41
- 201000010099 disease Diseases 0.000 title claims abstract description 40
- 238000012545 processing Methods 0.000 claims abstract description 52
- 238000000034 method Methods 0.000 claims abstract description 39
- 238000013136 deep learning model Methods 0.000 claims abstract description 37
- 201000002859 sleep apnea Diseases 0.000 claims abstract description 23
- 238000012549 training Methods 0.000 claims abstract description 10
- 230000003068 static effect Effects 0.000 claims description 38
- 230000015654 memory Effects 0.000 claims description 24
- 238000007781 pre-processing Methods 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 10
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 230000036760 body temperature Effects 0.000 claims description 7
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 5
- 239000008280 blood Substances 0.000 claims description 5
- 210000004369 blood Anatomy 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 5
- 229910052760 oxygen Inorganic materials 0.000 claims description 5
- 239000001301 oxygen Substances 0.000 claims description 5
- 230000008859 change Effects 0.000 claims description 3
- 238000012417 linear regression Methods 0.000 claims description 2
- 238000013528 artificial neural network Methods 0.000 description 11
- 238000004891 communication Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 8
- 230000000306 recurrent effect Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000013135 deep learning Methods 0.000 description 6
- 208000024891 symptom Diseases 0.000 description 6
- 208000001797 obstructive sleep apnea Diseases 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 208000008784 apnea Diseases 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000013139 quantization Methods 0.000 description 2
- 230000029058 respiratory gaseous exchange Effects 0.000 description 2
- 208000007590 Disorders of Excessive Somnolence Diseases 0.000 description 1
- 206010020591 Hypercapnia Diseases 0.000 description 1
- 206010021143 Hypoxia Diseases 0.000 description 1
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 206010041235 Snoring Diseases 0.000 description 1
- 206010041349 Somnolence Diseases 0.000 description 1
- 206010042434 Sudden death Diseases 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 208000026106 cerebrovascular disease Diseases 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 208000029078 coronary artery disease Diseases 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 206010012601 diabetes mellitus Diseases 0.000 description 1
- 208000035475 disorder Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007954 hypoxia Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000006386 memory function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008506 pathogenesis Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000002560 therapeutic procedure Methods 0.000 description 1
Images
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The embodiment of the invention provides a disease evaluation method and terminal equipment, wherein the method comprises the following steps: acquiring original physiological data of a user, wherein the original physiological data is used for evaluating sleep apnea syndrome; processing the original physiological data to obtain target physiological data; and taking the target physiological data as the input of a deep learning model, and calculating to obtain a disease evaluation result, wherein the deep learning model is obtained by training according to the historical physiological data of the user. By adopting the invention, the disease evaluation result of the user can be calculated in real time or periodically by utilizing the multi-dimensional user physiological data so as to help a doctor to provide effective reference information and help treatment in time.
Description
Technical Field
The invention relates to the field of medicine and information intellectualization, in particular to a disease evaluation method and terminal equipment.
Background
The sleep apnea hypopnea syndrome (sign) is a symptom with unclear etiology and pathogenesis at present, and the clinical manifestations mainly comprise: the night sleep snoring is accompanied by symptoms such as apnea and daytime sleepiness. The apnea can cause repeated hypercapnia and night hypoxia, so that complications such as coronary heart disease, diabetes, cerebrovascular disease and the like can be caused, and even sudden death at night can be caused in severe cases. How to accurately diagnose sleep apnea hypopnea syndrome is an important ring of nighttime medicine.
Therefore, a reasonable and accurate evaluation scheme needs to be provided.
Disclosure of Invention
The technical problem to be solved by the embodiments of the present invention is to provide a method and a terminal device for evaluating a disease condition, which can utilize multi-dimensional user physiological data to calculate a disease condition evaluation result of a user in real time or periodically, so as to help a doctor to cure the user timely and reliably, thereby improving the practicability.
In a first aspect, embodiments of the present invention provide a method for evaluating a condition, the method including:
acquiring original physiological data of a user, wherein the original physiological data is used for evaluating sleep apnea syndrome;
processing the original physiological data to obtain target physiological data;
and taking the target physiological data as the input of a deep learning model, and calculating to obtain a disease evaluation result, wherein the deep learning model is obtained by training according to the historical physiological data of the user.
In some possible embodiments, the processing the raw physiological data to obtain the target physiological data includes:
classifying the original physiological data to obtain the dynamic physiological data, the static physiological data and the ECG data;
respectively preprocessing the dynamic physiological data and the static physiological data to obtain dynamic intermediate data and static intermediate data, wherein the preprocessing comprises at least one of the following items: data deduplication processing, abnormal data processing and data missing filling processing;
performing feature extraction on the ECG data to obtain ECG feature data;
fusing the dynamic intermediate data, the static intermediate data and the ECG characteristic data to obtain intermediate physiological data;
and converting the intermediate physiological data into target physiological data with a preset format.
In some possible embodiments, the ECG characteristic data comprises at least one of: time domain feature data, frequency domain feature data, nonlinear domain feature data.
In some possible embodiments, the method further comprises:
and when the disease evaluation result is a target evaluation result, prompting the target evaluation result, and sending the target evaluation result to a prestored contact person, wherein the target evaluation result is used for indicating that the user suffers from sleep apnea syndrome.
In some possible embodiments, the deep learning model is an N-layer deep learning model, N being a positive integer greater than 0, the deep learning mode including any one of: a long and short time memory network LSTM model, a gate control repeat unit network GRU model, a recurrent neural network RNN model and a recurrent neural network RNNs model.
In a second aspect, an embodiment of the present invention provides a terminal device, where the terminal device includes a functional unit configured to execute the method of the first aspect.
In a third aspect, an embodiment of the present invention provides a terminal device, including: a processor, a memory, a communication interface, and a bus; the processor, the memory and the communication interface are connected through the bus and complete mutual communication; the memory stores executable program code; the processor executes a program corresponding to the executable program code by reading the executable program code stored in the memory for performing a disorder evaluation method; wherein the method is the method of any one of the first aspect.
In a fourth aspect, the invention provides a computer-readable storage medium storing program code for execution by a computing device. The program code comprises instructions for performing the method of any of the first aspects.
In a fifth aspect, the present invention provides a computer program product comprising instructions which, when run on a computer, cause the computer to perform the method of any of the first aspects described above.
In the embodiment of the invention, terminal equipment can acquire original physiological data of a user, wherein the original physiological data is used for evaluating sleep apnea syndrome, then the original physiological data is processed to obtain target physiological data, and finally the target physiological data is used as the input of a deep learning model which is obtained by training according to historical physiological data of the user; therefore, the deep learning model can be used for calculating the multi-dimensional user physiological data to obtain a disease evaluation result, and assisting doctors in treatment in time, so that the practicability of disease evaluation is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for evaluating a condition according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of an LSTM model according to an embodiment of the present invention;
FIG. 3 is a flow chart of a method for evaluating a condition according to another embodiment of the present invention;
fig. 4 is a schematic structural diagram of a terminal device according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a terminal device according to another embodiment of the present invention;
fig. 6 is a schematic structural diagram of a terminal device according to another embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terms "first," "second," and "third" (if any) in the description and claims of the invention and the above-described drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "comprises" and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.
Referring to fig. 1, a flow chart of a method for evaluating a disease state according to an embodiment of the present invention is shown, and the method according to the embodiment of the present invention may be applied to terminal devices with a communication network function, such as a smart phone, a tablet computer, and a smart wearable device, and may be specifically implemented by processors of the terminal devices. The method of embodiments of the present invention further includes the following steps.
Step S102, the terminal device obtains original physiological data of a user, and the original physiological data is used for evaluating sleep apnea syndrome.
In the present application, the raw physiological data is data for assessing or determining whether a user suffers from sleep apnea syndrome, the raw physiological data being associated with the sleep apnea syndrome. Typically, the raw physiological data includes, but is not limited to, Electrocardiographic (ECG) data, heart rate data, respiration data, body temperature data, pulse data, height data, weight data, and the like, and the invention is not limited thereto.
And step S104, the terminal equipment processes the original physiological data to obtain target physiological data.
The terminal device may process the raw physiological data, such as format conversion processing, to obtain target physiological data, which is described in detail below. Generally, the target physiological data is data having a preset format, and the preset format includes, but is not limited to, an integer format, a decimal format, an eighteen-digit format, and the like, which is not limited in the present invention.
And S106, the terminal equipment takes the target physiological data as the input of a deep learning model, and calculates to obtain a disease evaluation result, wherein the deep learning model is obtained by training according to historical physiological data of a user.
And the terminal equipment takes the target physiological data as the input of a deep learning model, and calculates the corresponding disease evaluation result by using the deep learning model related to the time series. The condition assessment result is used to indicate whether the user has sleep apnea syndrome (also referred to as sleep apnea syndrome).
The following describes specific embodiments of the present invention.
First, there are three specific embodiments in step S104 as follows:
in a first implementation manner, the terminal device quantizes the original physiological data at the same data magnitude, and converts the original physiological data into target physiological data with a uniform preset format.
The raw physiological data may be detected by the terminal device using a sensor, or may be obtained by the terminal device from a server side through a network, which is not limited in the present invention.
The preset format includes, but is not limited to, decimal, eighteen decimal, and the like. Preferably, the present application converts the raw physiological data into an integer type of physiological data of the user. For example, the target physiological data after quantization of the same order of magnitude is controlled to be between 0 and 10 by adopting a quantization rule of ten orders of magnitude.
In a second embodiment, the terminal device processes the original physiological data, for example, performs a deduplication process, to obtain intermediate physiological data. Further, the terminal device quantizes the intermediate physiological data in the same data magnitude, and converts the intermediate physiological data into target physiological data with a uniform preset format.
For the quantification of the intermediate physiological data, reference may be made to the related description of the foregoing first embodiment, and details are not repeated here.
In a third embodiment, the terminal device obtains raw physiological data of a user. Wherein the raw physiological data may include dynamic physiological data, static physiological data, and ECG data. The terminal device processes the original physiological data to obtain intermediate physiological data, and the intermediate physiological data comprises: the terminal equipment classifies the original physiological data to obtain the dynamic physiological data, the static physiological data and the ECG data; the terminal equipment respectively preprocesses the dynamic physiological data and the static physiological data to obtain dynamic intermediate data and static intermediate data; the terminal equipment performs feature extraction on the ECG data to obtain ECG feature data; the terminal device fuses the dynamic intermediate data, the static intermediate data and the ECG characteristic data to obtain the intermediate physiological data, and the preprocessing comprises at least one of the following steps: data deduplication processing, abnormal data processing and data missing filling processing.
The dynamic physiological data refers to other data, besides ECG data, used for characterizing the physiological index of the user, which may change dynamically with time, such as heart rate data, pulse data, respiration data, etc. of the user. The static physiological data is used for representing user physiological index data which does not change along with time or within a fixed time (such as 1 month), such as height, weight and the like of the user.
Illustratively, the terminal device may utilize the data analysis module to divide the raw physiological data into three categories, i.e. dynamic physiological data, static physiological data and ECG data, according to the identification (e.g. name) corresponding to each item of data in the raw physiological data. Then, the terminal device can utilize the dynamic data processing module to perform data deduplication, abnormal data removal, missing data filling and other processing on the dynamic physiological data to obtain intermediate dynamic data. Correspondingly, the terminal device may utilize a static data processing module to perform learning, training, and other processing on the static physiological data, for example, utilize one-hot coding to encode the static data, so as to obtain static intermediate data. Accordingly, the terminal device can utilize an ECG data processing module to perform feature extraction on the ECG data, so as to obtain ECG feature data. Further, the terminal device can utilize a data splicing module to splice and fuse the dynamic intermediate data, the static intermediate data and the ECG characteristic data, so as to obtain complete and accurate intermediate physiological data. And finally, the terminal equipment can convert the intermediate physiological data into target physiological data with a preset format by using a characteristic engineering module.
The following describes the data preprocessing in detail by taking the filling of missing data as an example. Since the physiological data of the user collected by the hospital is irregular, for example, the heart rate data may be collected every minute, but the collection frequency of the body temperature data is not fixed. In the dynamic physiological data processing flow, the physiological data (including the body temperature data, of course) of the user needs to be uniformly processed into a format with the same length. Therefore, missing values need to be filled, and because the body temperature and the time may have a linear relationship, a linear regression method can be considered to be used for filling the missing values.
As another example, in one possible scenario, blood oxygen data may not be acquired every minute. If the data format is to be kept uniform, the missing value can be filled by adopting a near filling method, so that the purpose of processing is to prevent the blood oxygen data from generating large fluctuation, thereby not influencing data distortion or abnormity and not influencing the learning of the model.
In an optional embodiment, after the terminal device performs feature extraction on the ECG data by using a feature extraction module, any one or more of the following three feature data are obtained. The three characteristic data are respectively as follows: time domain signature data, frequency domain signature data, and nonlinear domain signature data. That is, the ECG characteristic data includes any one or more of time domain characteristic data, frequency domain characteristic data, and nonlinear domain characteristic data, which is not limited in the embodiments of the present invention. Preferably, the ECG characteristic data includes the above three characteristic data.
The time domain feature data is typically a relationship directly calculated and analyzed for time series between their connected heartbeats using a continuously measured electrocardiogram waveform (i.e., ECG signal or ECG data), such as: the time domain feature data may be any one of: SDNN, SDANN, NN50 count, NN50 count, and the like. Among them, SDNN is the Standard Deviation of Normal heart beat spacing, and is called Standard development of Normal to Normal in English. SDANN is the average of the Standard deviations of normal heart beat spacing over five minutes, and is generally referred to in English as Standard definition of the averages of NNs in all 5-minute segments of the entry recording. NN50 count is the Number of normal beat interval differences exceeding 50ms for each pair, which is called Number of calls of ad jacent NNintermediates differential by more than 50ms in the entry recording. NN50 count is the ratio of the number of NN50 to the total number of all normal heartbeat intervals, which is called NN50 count divided by the total number of all NN intervals.
The frequency domain feature data is generally obtained by a method that a first step terminal device finds an identification point of each cycle of the ECG data (i.e. the ECG signal), such as a zero crossing point, a maximum minimum extreme point and some points which can be easily detected, and generally uses the time between RR points in the ECG signal (i.e. the ECG data) as a corresponding cycle, so that the first step is to detect the R point of the ECG signal; the second step is to calculate the value of each period, i.e. the time between RR points. Thirdly, interpolation; and finally, carrying out Fourier transform to obtain frequency domain characteristic data. The embodiments of the present invention are not described in detail and limited with respect to the acquisition of frequency domain feature data.
The nonlinear domain characteristic data can be obtained by a method that the terminal device explores the ECG signal (i.e. the ECG data) through a nonlinear system theory and method, and obtains the nonlinear domain data by processing a poincare scattergram. The embodiments of the present invention do not describe or limit the details of the obtaining of the nonlinear domain characteristic data.
Before step S106, the terminal device may acquire one or more sets of historical physiological data and historical condition results of the user. The historical physiological data and the historical condition outcome may be actual data collected over a time horizon. Optionally, the historical physiological data and the historical disease result may be pre-stored in the terminal device, or may be obtained from other terminal devices or a server through a network, which is not limited in the present invention. The terminal device can then train and learn the historical physiological data and the historical disease results by using a mathematical model related to a time series, so as to obtain a deep learning model.
Accordingly, in step S106, the terminal device may use the target physiological data in step S104 as an input of the deep learning model, and calculate the target physiological data through the deep learning model, so as to obtain a disease evaluation result. The condition evaluation result is used for indicating whether the user suffers from sleep apnea syndrome.
In an alternative embodiment, the deep learning model may be a deep learning model of n layers, n being a positive integer greater than 0. The deep learning mode includes any one of: a Long Short-Term Memory Network (LSTM) model, a gated repeat unit Network GRU model, a Recurrent Neural Network (RNN) model, a Recurrent Neural Network (RNNs) model, a BP Neural Network model, or other data models associated with time series, and the like.
For example, take the LSTM model of three-layer neural network as an example. As shown in fig. 2, the LSTM model includes 2 layers of LSTM cells (shown as lstmcells) and one layer of Fully Connected (FC) neural network cells (shown as FCs). Wherein the number of LSTMcell and FC per layer is not limited. The model input end can input target physiological data (specifically dynamic intermediate data in the target physiological data) of the user which is acquired at each moment by the terminal equipment and processed, and corresponding disease evaluation results are output from the model output end through calculation of the three-layer LSTM model.
Taking the time t4 as an example, the model input end inputs the target physiological data of the user at the time, and the LTSMcell of the first layer and the second layer performs feature extraction and learning on the input target physiological data, for example, learning a linear or nonlinear relationship between data, and the like. Since the LSTMcell allows selective transmission of information, physiological data of the user (i.e., historical physiological data) before time t4 is selectively transmitted together with target physiological data at time t4 through the FC layer to calculate a disease assessment result of the user, and the calculated disease assessment result is output to the output end of the model through the FC layer.
Specifically, on one hand, dynamic intermediate data in the target physiological data is input at the input end of the model, processed by two layers of LSTMcell units to obtain processed dynamic intermediate data, and then input to the FC layer; on the other hand, the processed ECG characteristic data and static intermediate data in the target physiological data are directly input to the FC layer. Accordingly, after receiving the three data, the FC layer can classify the data to diagnose the sleep apnea hypopnea syndrome event, thereby outputting a visualized symptom evaluation result from the output end of the model.
It should be noted that the deep learning model can output the disease evaluation result at any time (i.e., in real time). Specifically, as shown in fig. 2, the output time of the LSTM model is arbitrarily adjustable, and the parameter n in fig. 2 may be set to any real number greater than 0. The benefit of this adjustable parameter is that the outcome of the condition assessment can be real-time, which enables real-time intervention and intervention of sleep apnea hypopnea syndrome to assist the physician in treatment.
In an alternative embodiment, the terminal device may prompt the condition assessment result. The prompting method includes, but is not limited to, a prompting method such as sound, subtitles, pictures, floating windows, vibration, and the like, and the embodiment of the present invention is not limited thereto.
In an optional embodiment, when the disease condition evaluation result is a target evaluation result, prompting the target evaluation result and sending the target evaluation result to a prestored contact person, wherein the target evaluation result is used for indicating that the user suffers from sleep apnea syndrome.
Specifically, when the disease evaluation result is used for indicating that the user has sleep apnea syndrome, an alarm can be sent to inform relevant doctors to timely cure the disease, and the like. Optionally, the terminal device may further send the disease evaluation result to a pre-stored contact, for example, a hospital, or a cloud server, an expert therapy system, or the like, so as to assist a doctor in completing treatment.
In an alternative embodiment, the deep learning model can be an LSTM neural network model in deep learning, the biggest difference between LSTM and traditional neural network algorithms is that information before can be traced back, and LSTM has a memory function due to unique gate structure effect, and has unique advantages for very long-spaced events in time sequence.
The deep learning framework adopted by the deep learning model is a TensorFlow platform under Google, and the platform supports the implementation modes of online learning, large-scale distributed operation, Spark cluster, GPU calculation and hundreds of millions of magnitude data processing. Therefore, the deep learning model can analyze the physiological data of the user related to the time series so as to reliably complete the related evaluation of the symptoms of the sleep apnea by combining the correlation before and after the symptoms of the sleep apnea, thereby improving the accuracy of the evaluation of the symptoms of the sleep apnea.
The terminal device may include a User Equipment (UE), a smart phone (such as an Android phone, an IOS phone, etc.), a personal computer, a tablet computer, a palmtop computer, a Mobile Internet device (MID, Mobile Internet Devices), a wearable smart device, and other Internet Devices, which is not limited in the embodiments of the present invention.
In the embodiment of the invention, terminal equipment can acquire original physiological data of a user, wherein the original physiological data is used for evaluating sleep apnea syndrome, then the original physiological data is processed to obtain target physiological data, and finally the target physiological data is used as the input of a deep learning model which is obtained by training according to historical physiological data of the user; therefore, the deep learning model can be used for calculating the multi-dimensional user physiological data to obtain a disease evaluation result, and assisting doctors in treatment in time, so that the practicability of disease evaluation is improved.
Please refer to fig. 3, which illustrates another method for evaluating a disease state according to an embodiment of the present invention. The method as shown in fig. 3, comprising the following implementation steps:
step S302, the terminal device acquires original physiological data of the user.
Step S304, the terminal device classifies the original physiological data to obtain the dynamic physiological data, the static physiological data and the ECG data;
step S306, the terminal device respectively preprocesses the dynamic physiological data and the static physiological data to obtain dynamic intermediate data and static intermediate data, wherein the preprocessing comprises at least one of the following items: data deduplication processing, abnormal data processing and data missing filling processing.
And S308, the terminal equipment performs feature extraction on the ECG data to obtain ECG feature data.
The execution sequence of steps S306 and S308 is not limited, for example, step S308 may be executed first and then step S306 is executed, and the embodiment of the present invention is not limited.
And S310, the terminal equipment fuses the dynamic intermediate data, the static intermediate data and the ECG characteristic data to obtain the intermediate physiological data.
Step S312, the terminal device converts the intermediate physiological data into target physiological data with a preset format.
And S314, the terminal equipment takes the target physiological data as the input of a deep learning model, and calculates to obtain a disease evaluation result, wherein the deep learning model is obtained by training according to historical physiological data and historical disease results of the user.
In an optional embodiment, the deep learning model is an N-layer deep learning model, N is a positive integer greater than 0, and the deep learning mode includes any one of: a long and short time memory network LSTM model, a gate control repeat unit network GRU model, a recurrent neural network RNN model and a recurrent neural network RNNs model.
And step S316, when the disease evaluation result is a target evaluation result, prompting the target evaluation result, and sending the target evaluation result to a prestored contact person, wherein the target evaluation result is used for indicating that the user suffers from sleep apnea syndrome.
For parts which are not shown and not described in the embodiment of the present invention, reference may be made to the description related to the embodiment described in fig. 1, which is not described herein again.
Referring to fig. 4, a schematic structural diagram of a terminal device according to an embodiment of the present invention is shown, where the terminal device 400 according to the embodiment of the present invention includes: an acquisition unit 402, a processing unit 404, and a calculation unit 406; wherein:
the acquiring unit 402 acquires original physiological data of a user, wherein the original physiological data is used for evaluating sleep apnea syndrome;
the processing unit 404 is configured to process the original physiological data to obtain target physiological data;
the prompting unit 406 is configured to calculate a disease evaluation result by using the target physiological data as an input of a deep learning model, where the deep learning model is obtained by training according to historical physiological data of a user.
Please refer to fig. 5, which is a schematic structural diagram of another terminal device according to an embodiment of the present invention, where the terminal device 400 according to the embodiment of the present invention includes the obtaining unit 402, the processing unit 404, and the calculating unit 406; wherein: the raw physiological data comprises dynamic physiological data, static physiological data and electrocardio ECG data,
the processing unit 404 is configured to classify the raw physiological data to obtain the dynamic physiological data, the static physiological data, and the ECG data;
the processing unit 404 is further configured to perform preprocessing on the dynamic physiological data and the static physiological data respectively to obtain dynamic intermediate data and static intermediate data, where the preprocessing includes at least one of the following: data deduplication processing, abnormal data processing and data missing filling processing;
the processing unit 404 is further configured to perform feature extraction on the ECG data to obtain ECG feature data;
the processing unit 404 is further configured to fuse the dynamic intermediate data, the static intermediate data, and the ECG characteristic data to obtain intermediate physiological data;
the processing unit 404 is further configured to convert the intermediate physiological data into target physiological data having a preset format.
In some possible embodiments, the ECG characteristic data comprises at least one of: time domain feature data, frequency domain feature data, nonlinear domain feature data.
In some possible embodiments, the terminal device further comprises a prompting unit 410,
the prompting unit 410 is configured to prompt the target evaluation result when the disease evaluation result is a target evaluation result, and send the target evaluation result to a prestored contact person, where the target evaluation result is used to indicate that the user has sleep apnea syndrome.
For specific implementation of each unit related in the embodiments of the present invention, reference may be made to descriptions of related functional units or implementation steps in the embodiments corresponding to fig. 1 to fig. 3, which are not described herein again.
Referring to fig. 6, fig. 6 is a schematic structural diagram of a terminal device according to an embodiment of the present invention. The terminal device 400 of the present embodiment includes: at least one processor 601, a communication interface 602, a user interface 603 and a memory 604, wherein the processor 601, the communication interface 602, the user interface 603 and the memory 604 can be connected by a bus or other means, and the embodiment of the present invention is exemplified by being connected by the bus 605. Wherein,
The communication interface 602 may be a wired interface (e.g., an ethernet interface) or a wireless interface (e.g., a cellular network interface or using a wireless local area network interface) for communicating with other terminals or websites. In the embodiment of the present invention, the communication interface 602 is specifically configured to obtain the physiological data of the user.
The user interface 603 may specifically be a touch panel, including a touch screen and a touch screen, for detecting an operation instruction on the touch panel, and the user interface 603 may also be a physical button or a mouse. The user interface 603 may also be a display screen for outputting, displaying images or data.
acquiring original physiological data of a user, wherein the original physiological data is used for evaluating sleep apnea syndrome;
processing the original physiological data to obtain target physiological data;
and taking the target physiological data as the input of a deep learning model, and calculating to obtain a disease evaluation result, wherein the deep learning model is obtained by training according to the historical physiological data of the user.
In some possible embodiments, the raw physiological data includes dynamic physiological data, static physiological data, and electrocardiographic ECG data, and the processor 601 is configured to:
classifying the original physiological data to obtain the dynamic physiological data, the static physiological data and the ECG data;
respectively preprocessing the dynamic physiological data and the static physiological data to obtain dynamic intermediate data and static intermediate data, wherein the preprocessing comprises at least one of the following items: data deduplication processing, abnormal data processing and data missing filling processing;
performing feature extraction on the ECG data to obtain ECG feature data;
fusing the dynamic intermediate data, the static intermediate data and the ECG characteristic data to obtain intermediate physiological data;
and converting the intermediate physiological data into target physiological data with a preset format.
In some possible embodiments, the ECG characteristic data comprises at least one of: time domain feature data, frequency domain feature data, nonlinear domain feature data.
In some possible embodiments, the processor 601 is further configured to:
and when the disease evaluation result is a target evaluation result, prompting the target evaluation result, and sending the target evaluation result to a prestored contact person, wherein the target evaluation result is used for indicating that the user suffers from sleep apnea syndrome.
In some possible embodiments, the deep learning model is an N-layer deep learning model, N being a positive integer greater than 0, the deep learning mode including any one of: a long and short time memory network LSTM model, a gate control repeat unit network GRU model, a recurrent neural network RNN model and a recurrent neural network RNNs model.
An embodiment of the present invention further provides a computer storage medium, where the computer storage medium may store a program, and the program includes, when executed, some or all of the implementation steps in the method embodiments described in the above method embodiments.
It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the invention. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required by the invention.
In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus may be implemented in other manners. For example, the above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one type of division of logical functions, and there may be other divisions when actually implementing, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of some interfaces, devices or units, and may be an electric or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (3)
1. A terminal device, comprising:
an acquisition unit for acquiring original physiological data of a user, the original physiological data being used for assessing sleep apnea syndrome;
the processing unit is used for processing the original physiological data to obtain target physiological data;
the calculation unit is used for taking the target physiological data as the input of a deep learning model, and calculating to obtain a disease evaluation result, wherein the deep learning model is obtained by training according to historical physiological data of a user;
the original physiological data comprise dynamic physiological data, static physiological data and Electrocardiogram (ECG) data, wherein the dynamic physiological data are other physiological index data of the user, except the ECG data, which are used for representing dynamic change along with time;
the processing unit is further configured to classify the raw physiological data to obtain the dynamic physiological data, the static physiological data, and the ECG data;
the processing unit is further configured to perform preprocessing on the dynamic physiological data to obtain dynamic intermediate data, and encode the static physiological data by using a one-hot code to obtain static intermediate data, where the preprocessing includes: data deduplication processing, abnormal data processing and data missing filling processing; the dynamic physiological data comprises body temperature data, and the data missing filling processing of the body temperature data comprises filling missing values of the body temperature data by using a linear regression method; the dynamic physiological data comprises blood oxygen data, and the data missing filling processing of the blood oxygen data comprises the filling of missing values of the blood oxygen data according to a near filling method;
the processing unit is further used for performing feature extraction on the ECG data to obtain ECG feature data;
the processing unit is further configured to fuse the dynamic intermediate data, the static intermediate data and the ECG characteristic data to obtain intermediate physiological data;
the processing unit is further used for converting the intermediate physiological data into target physiological data with a preset format;
the deep learning model is a long-time memory network LSTM model, the LSTM model comprises 2 layers of LSTM units and 1 layer of full-connection layer, the target physiological data is used as the input of the deep learning model, and the disease evaluation result is obtained through calculation, and the method comprises the following steps:
inputting dynamic intermediate data in the target physiological data at an input end of the LSTM model, processing by a 2-layer LSTM unit to obtain processed dynamic intermediate data, and inputting the processed dynamic intermediate data to a full connection layer; and directly inputting ECG characteristic data and static intermediate data in the target physiological data into a full connection layer, classifying the ECG characteristic data and the static intermediate data by the full connection layer, and outputting a visualized disease evaluation result from the output end of the LSTM model.
2. Terminal device according to claim 1, characterized in that the ECG characteristic data comprises at least one of the following: time domain feature data, frequency domain feature data, nonlinear domain feature data.
3. The terminal device according to any of claims 1-2, wherein the terminal device further comprises a prompting unit,
the prompting unit is used for prompting the target evaluation result when the disease evaluation result is the target evaluation result, and sending the target evaluation result to a pre-stored contact person, wherein the target evaluation result is used for indicating that the user suffers from sleep apnea syndrome.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710632865.3A CN107595243B (en) | 2017-07-28 | 2017-07-28 | Disease evaluation method and terminal equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710632865.3A CN107595243B (en) | 2017-07-28 | 2017-07-28 | Disease evaluation method and terminal equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107595243A CN107595243A (en) | 2018-01-19 |
| CN107595243B true CN107595243B (en) | 2021-08-17 |
Family
ID=61059652
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710632865.3A Active CN107595243B (en) | 2017-07-28 | 2017-07-28 | Disease evaluation method and terminal equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107595243B (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108282533A (en) * | 2018-01-29 | 2018-07-13 | 百度在线网络技术（北京）有限公司 | intelligent monitoring method, device, wearable device and server |
| CN108652615B (en) * | 2018-03-24 | 2021-07-13 | 上海图灵医疗科技有限公司 | Electrocardio data pathological feature quantitative analysis method and device |
| KR20190114694A (en) * | 2018-03-30 | 2019-10-10 | 삼성에스디에스 주식회사 | Method for learning and analyzing time series data by using artificial intelligence |
| CN109102896A (en) * | 2018-06-29 | 2018-12-28 | 东软集团股份有限公司 | A kind of method of generating classification model, data classification method and device |
| CN109087706B (en) * | 2018-07-13 | 2022-12-02 | 浙江清华长三角研究院 | Human health assessment method and system based on sleep big data |
| CN109411062A (en) * | 2018-09-30 | 2019-03-01 | 天津科技大学 | A method of diabetes and its complication are judged automatically using physiological parameter based on depth learning technology |
| CN109602410A (en) * | 2018-11-16 | 2019-04-12 | 青岛真时科技有限公司 | A kind of wearable device and its monitoring of pulse method |
| CN109637670A (en) * | 2018-12-04 | 2019-04-16 | 泰康保险集团股份有限公司 | Data processing method, device, medium and electronic equipment |
| US11564633B2 (en) * | 2018-12-21 | 2023-01-31 | Industrial Technology Research Institute | State assessment system, diagnosis and treatment system, and method for operating the diagnosis and treatment system |
| CN109833031B (en) * | 2019-03-12 | 2020-08-14 | 西安交通大学 | Automatic sleep staging method based on LSTM and utilizing multiple physiological signals |
| CN110223280B (en) * | 2019-06-03 | 2021-04-13 | Oppo广东移动通信有限公司 | Venous thrombosis detection method and venous thrombosis detection device |
| CN110348500A (en) * | 2019-06-30 | 2019-10-18 | 浙江大学 | Sleep disturbance aided diagnosis method based on deep learning and infrared thermal imagery |
| CN111446002A (en) * | 2020-03-07 | 2020-07-24 | 华中科技大学 | Novel coronavirus patient state of illness classification system based on artificial intelligence |
| CN111671426B (en) * | 2020-05-13 | 2022-07-12 | 北京航空航天大学 | Human body respiration state monitoring system and method based on flexible sensing and deep learning |
| CN111603138A (en) * | 2020-05-19 | 2020-09-01 | 杭州电子科技大学 | Sleep apnea monitoring system based on millimeter wave radar |
| CN111862068B (en) * | 2020-07-28 | 2022-09-13 | 福州大学 | Three-model comprehensive decision emotion prediction method fusing data missing data and images |
| CN113729709B (en) * | 2021-09-23 | 2023-08-11 | 中科效隆(深圳)科技有限公司 | Nerve feedback device, nerve feedback method, and computer-readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106328122A (en) * | 2016-08-19 | 2017-01-11 | 深圳市唯特视科技有限公司 | Voice identification method using long-short term memory model recurrent neural network |
| CN106682220A (en) * | 2017-01-04 | 2017-05-17 | 华南理工大学 | Online traditional Chinese medicine text named entity identifying method based on deep learning |
| CN106691378A (en) * | 2016-12-16 | 2017-05-24 | 深圳市唯特视科技有限公司 | Deep learning vision classifying method based on electroencephalogram data |
| CN106901723A (en) * | 2017-04-20 | 2017-06-30 | 济南浪潮高新科技投资发展有限公司 | A kind of electrocardiographic abnormality automatic diagnosis method |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104545818A (en) * | 2015-01-29 | 2015-04-29 | 吉林大学 | Sleep apnea syndrome detection method based on pulse and blood oxygen signals |
| CN204971271U (en) * | 2015-08-24 | 2016-01-20 | 华南理工大学 | Old person health service system based on health service robot |
| CN106361277A (en) * | 2016-08-26 | 2017-02-01 | 中山大学 | Sleep apnea syndrome assessment method based on electrocardiogram signals |
| CN106473700B (en) * | 2016-08-26 | 2019-11-12 | 复旦大学 | The detection system of obstructive sleep apnea syndrome based on HRV frequency-domain analysis |
-
2017
- 2017-07-28 CN CN201710632865.3A patent/CN107595243B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106328122A (en) * | 2016-08-19 | 2017-01-11 | 深圳市唯特视科技有限公司 | Voice identification method using long-short term memory model recurrent neural network |
| CN106691378A (en) * | 2016-12-16 | 2017-05-24 | 深圳市唯特视科技有限公司 | Deep learning vision classifying method based on electroencephalogram data |
| CN106682220A (en) * | 2017-01-04 | 2017-05-17 | 华南理工大学 | Online traditional Chinese medicine text named entity identifying method based on deep learning |
| CN106901723A (en) * | 2017-04-20 | 2017-06-30 | 济南浪潮高新科技投资发展有限公司 | A kind of electrocardiographic abnormality automatic diagnosis method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107595243A (en) | 2018-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107595243B (en) | Disease evaluation method and terminal equipment | |
| CN107491638A (en) | A kind of ICU user's prognosis method and terminal device based on deep learning model | |
| EP3383271B1 (en) | System for screening delirium patients for the presence of encephalopathy | |
| WO2019182510A1 (en) | Systems and methods for personalized medication therapy management | |
| US20190328251A1 (en) | Arrhythmia detection method, arrhythmia detection device and arrhythmia detection system | |
| CA2883218A1 (en) | Methods and systems for calculating and using statistical models to predict medical events | |
| CN108091391A (en) | Illness appraisal procedure, terminal device and computer-readable medium | |
| KR102494310B1 (en) | Method, apparatus and computer program for determining the possibility of shock occurrence through the analysis of visualized vital sign data | |
| US20180310867A1 (en) | System and method for stress level management | |
| CN109674474B (en) | Sleep apnea recognition method, device and computer readable medium | |
| Ronca et al. | A video-based technique for heart rate and eye blinks rate estimation: a potential solution for telemonitoring and remote healthcare | |
| JP7002168B1 (en) | ECG analyzer, ECG analysis method and program | |
| JP2024016180A (en) | Device and method for stress evaluation using physiological data | |
| US8805484B2 (en) | System, apparatus and method for diagnosing seizures | |
| CN108305688A (en) | Illness appraisal procedure, terminal device and computer-readable medium | |
| CN108198617A (en) | Illness appraisal procedure, terminal device and computer-readable medium | |
| CN108182974B (en) | Disease evaluation method, terminal device and computer readable medium | |
| CN112914506A (en) | Sleep quality detection method, device and computer readable storage medium | |
| D'Costa et al. | AI and the cardiologist: when mind, heart and machine unite | |
| Lee et al. | Experimental verification of the possibility of reducing photoplethysmography measurement time for stress index calculation | |
| EP4124287A1 (en) | Regularized multiple-input pain assessment and trend | |
| US20210169415A1 (en) | Machine classification of significant psychophysiological response | |
| US20230284948A1 (en) | Test protocol for detecting significant psychophysiological response | |
| CN109394197B (en) | Heart rate variability measuring method, device and equipment based on time-frequency analysis | |
| CN113499037A (en) | Method and device for determining sleep quality of user, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |