CN112842297B - Signal processing method, apparatus, computer device and storage medium - Google Patents
Signal processing method, apparatus, computer device and storage medium Download PDFInfo
- Publication number
- CN112842297B CN112842297B CN202011488258.2A CN202011488258A CN112842297B CN 112842297 B CN112842297 B CN 112842297B CN 202011488258 A CN202011488258 A CN 202011488258A CN 112842297 B CN112842297 B CN 112842297B
- Authority
- CN
- China
- Prior art keywords
- signal
- early warning
- detection
- moment
- time
- 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
- 238000003672 processing method Methods 0.000 title abstract description 18
- 238000000034 method Methods 0.000 claims abstract description 29
- 230000036772 blood pressure Effects 0.000 claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 14
- 230000005236 sound signal Effects 0.000 claims abstract description 14
- 238000001514 detection method Methods 0.000 claims description 156
- 230000002159 abnormal effect Effects 0.000 claims description 72
- 238000004590 computer program Methods 0.000 claims description 26
- 230000005856 abnormality Effects 0.000 claims description 24
- 239000012634 fragment Substances 0.000 claims description 23
- 238000013135 deep learning Methods 0.000 claims description 7
- 238000004458 analytical method Methods 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 7
- 208000024891 symptom Diseases 0.000 description 7
- 238000012937 correction Methods 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 3
- 230000006793 arrhythmia Effects 0.000 description 2
- 206010003119 arrhythmia Diseases 0.000 description 2
- 230000000747 cardiac effect Effects 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 208000010125 myocardial infarction Diseases 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 206010048623 Atrial hypertrophy Diseases 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 208000031225 myocardial ischemia Diseases 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 230000029058 respiratory gaseous exchange Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000002861 ventricular Effects 0.000 description 1
Abstract
The application relates to a signal processing method, a signal processing device, a computer device and a storage medium. The method comprises the following steps: acquiring a first comprehensive signal corresponding to a first moment, wherein the first comprehensive signal comprises a gyroscope signal, a sound signal, a blood pressure signal and a photoelectric signal; and inputting the first integrated signal into an early warning model to obtain first early warning information, wherein the first early warning information comprises an inferred event corresponding to a second moment and occurrence probability corresponding to the inferred event, and the second moment is after the first moment. Based on the method, the inferred event and the probability of the inferred event occurring after the current moment are prejudged according to the comprehensive signals acquired in real time, so that the illness state of the patient is early warned in advance to take corresponding relief measures, and the situation that the user is suddenly happened and cannot take corresponding measures in time is avoided.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a signal processing method, a signal processing device, a computer device, and a storage medium.
Background
ECG is an electrical cardiac signal and is often used to examine various arrhythmias, ventricular atrial hypertrophy, myocardial infarction, myocardial ischemia, and the like. Conventional 12-lead ECG is commonly used in intensive care units, where there is little movement of the patient and thus signals can be acquired smoothly. Holter electrocardiograph (dynamic electrocardiogram) acquisition devices, which are popular at present, generally do not limit the activities of patients and can monitor the electrocardiogram of the patients for a long time.
Today, these monitors are often used as a screening scheme to identify from the already recorded signals whether the recorded time period signal is a disease signal. However, the detection of some cardiac inference events requires more than an ECG signal, more signals such as respiration, blood oxygen, etc. as an aid. However, when symptoms such as arrhythmia and myocardial infarction occur as sudden temporary events, the existing electrocardiogram acquisition device can only check according to the symptoms which have occurred, and early warning can not be performed before the symptoms occur.
Disclosure of Invention
In order to solve the technical problems, the application provides a signal processing method, a signal processing device, a computer device and a storage medium.
In a first aspect, the present application provides a signal processing method, including:
Acquiring a first comprehensive signal corresponding to a first moment, wherein the first comprehensive signal comprises a gyroscope signal, a sound signal, a blood pressure signal and a photoelectric signal;
And inputting the first integrated signal into an early warning model to obtain first early warning information, wherein the first early warning information comprises an inferred event corresponding to a second moment and occurrence probability corresponding to the inferred event, and the second moment is after the first moment.
Optionally, after the first integrated signal is input to the early warning model to obtain the first early warning information, the method further includes:
And inputting the first integrated signal into a detection model to obtain a first detection result, wherein the first detection result comprises an abnormal state and an abnormal type corresponding to the first moment.
Optionally, the time difference between the second time and the first time is a preset interval, and after the first integrated signal is input into the detection model to obtain a detection result corresponding to the first time, the method further includes:
Acquiring a second comprehensive signal corresponding to the second moment;
inputting the second comprehensive signal into the detection model to obtain a second detection result;
and correcting the early warning model according to the second detection result.
Optionally, the correcting the early warning model according to the second detection result includes:
when the second detection result is not matched with the first early warning information, correcting the early warning model according to the second detection result to obtain a corrected early warning model, wherein the corrected early warning model is used for carrying out early warning analysis on the comprehensive signals corresponding to the third moment.
Optionally, after the first integrated signal is input to the early warning model to obtain the first early warning information, the method further includes:
And when the occurrence probability corresponding to the current inferred event in the first early warning information is larger than a preset threshold value of the corresponding event, generating an early warning report according to the first comprehensive signal, and sending out an early warning signal.
Optionally, before the obtaining the first integrated signal corresponding to the first time, the method further includes:
acquiring a sample comprehensive signal within a preset time period, wherein the sample comprehensive signal comprises a plurality of detection signals;
marking each detection signal according to an abnormality detection algorithm to obtain an abnormality mark corresponding to each detection signal;
dividing each detection signal according to the abnormal mark corresponding to each detection signal to obtain a corresponding early warning signal slice;
And performing deep learning according to the early warning signal slices and the abnormal labels corresponding to the detection signals, and establishing the early warning model.
Optionally, the slicing the detection signals according to the abnormal time corresponding to the detection signals to obtain corresponding early warning signal slices includes:
dividing each detection signal to obtain a plurality of fragment signals, wherein the fragment signals carry corresponding starting time and ending time;
Before the abnormal time corresponding to each detection signal, selecting a fragment signal corresponding to the ending time of a preset interval from the abnormal time as the early warning signal slice.
In a second aspect, the present application provides a signal processing apparatus comprising:
the first signal acquisition module is used for acquiring a first comprehensive signal corresponding to a first moment, wherein the first comprehensive signal comprises a gyroscope signal, a sound signal, a blood pressure signal and a photoelectric signal;
The early warning module is used for inputting the first comprehensive signals into an early warning model to obtain first early warning information, wherein the first early warning information comprises an inferred event corresponding to a second moment and occurrence probability corresponding to the inferred event, and the second moment is larger than the first moment.
A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of:
Acquiring a first comprehensive signal corresponding to a first moment, wherein the first comprehensive signal comprises a gyroscope signal, a sound signal, a blood pressure signal and a photoelectric signal;
And inputting the first integrated signal into an early warning model to obtain first early warning information, wherein the first early warning information comprises an inferred event corresponding to a second moment and occurrence probability corresponding to the inferred event, and the second moment is after the first moment.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:
Acquiring a first comprehensive signal corresponding to a first moment, wherein the first comprehensive signal comprises a gyroscope signal, a sound signal, a blood pressure signal and a photoelectric signal;
And inputting the first integrated signal into an early warning model to obtain first early warning information, wherein the first early warning information comprises an inferred event corresponding to a second moment and occurrence probability corresponding to the inferred event, and the second moment is after the first moment.
The signal processing method, the device, the computer equipment and the storage medium, wherein the method comprises the following steps: acquiring a first comprehensive signal corresponding to a first moment, wherein the first comprehensive signal comprises a gyroscope signal, a sound signal, a blood pressure signal and a photoelectric signal; and inputting the first integrated signal into an early warning model to obtain first early warning information, wherein the first early warning information comprises an inferred event corresponding to a second moment and occurrence probability corresponding to the inferred event, and the second moment is after the first moment. Based on the method, the inferred event and the probability of the inferred event occurring after the current moment are prejudged according to the comprehensive signals acquired in real time, so that the illness state of the patient is early warned in advance to take corresponding relief measures, and the situation that the user is suddenly happened and cannot take corresponding measures in time is avoided.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, and it will be obvious to a person skilled in the art that other drawings can be obtained from these drawings without inventive effort.
FIG. 1 is a diagram of an application environment of a signal processing method in one embodiment;
FIG. 2 is a flow chart of a signal processing method according to an embodiment;
FIG. 3 is a schematic diagram of a signal processing principle in one embodiment;
FIG. 4 is a block diagram of a signal processing apparatus in one embodiment;
fig. 5 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
FIG. 1 is a diagram of an application environment for a signal processing method in one embodiment. Referring to fig. 1, the signal processing method is applied to a signal processing system. The signal processing system includes a signal acquisition terminal 110 and a server 120. The signal acquisition terminal 110 and the server 120 are connected through a network. The signal acquisition terminal 110 includes at least one of an ECG sensor, a high-precision gyroscope sensor, a sound sensor, a photoelectric sensor, a blood pressure sensor, or other biological sensor, etc. The server 120 may be implemented as a stand-alone server or as a server cluster composed of a plurality of servers.
In one embodiment, fig. 2 is a flow chart of a signal processing method in one embodiment, and referring to fig. 2, a signal processing method is provided. The embodiment is mainly exemplified by the method applied to the signal acquisition terminal 110 (or the server 120) in fig. 1, and the signal processing method specifically includes the following steps:
Step S210, a first integrated signal corresponding to a first time is obtained.
Specifically, a first integrated signal acquired by a plurality of biological sensors at a first moment is acquired, wherein the integrated signal comprises physiological parameters such as a gyroscope signal, a sound signal, a blood pressure signal, a photoelectric signal and the like.
Step S220, inputting the first integrated signal into an early warning model to obtain first early warning information, where the first early warning information includes an inferred event corresponding to a second time and an occurrence probability corresponding to the inferred event, and the second time is after the first time.
Specifically, the early warning model is a neural network model trained in advance, the first early warning information is an inferred result obtained by the early warning model according to a first comprehensive signal, the first early warning information comprises at least one inferred event and occurrence probability corresponding to each inferred event, the occurrence probability of each signal corresponding event is estimated according to the early warning model, so that the inferred event and the occurrence probability of the inferred event which will occur at a second moment after the first moment are comprehensively pre-determined, the inferred event is an abnormal event which will occur according to the inference of the comprehensive signal, corresponding rescue measures can be extracted and taken before symptoms of a patient occur according to the inferred event, the condition of a patient is prevented from being determined according to the symptoms when the user occurs, and a large amount of rescue time is wasted by taking corresponding measures according to the condition of the patient.
In one embodiment, the first integrated signal is input into a detection model to obtain a first detection result, where the first detection result includes an abnormal state and an abnormal type corresponding to the first moment.
Specifically, the first integrated signal is input into the detection model and the early warning model at the same time, the early warning model is used for detecting the inferred event corresponding to the second moment and the probability of occurrence of the inferred event, namely, early warning is carried out according to the inferred event according to the probability that the early warning model is used for detecting the occurrence of the subsequent abnormal event. The detection model is used for detecting an abnormal state and an abnormal type corresponding to the first comprehensive signal at the first moment, the first detection result is a diagnosis result obtained according to the first comprehensive signal at the first moment, namely, the detection model is used for detecting whether an abnormal event occurs at the current moment, the abnormal state comprises normal and abnormal, and the abnormal type is a disease type obtained according to the first comprehensive signal.
In one embodiment, the time difference between the second time and the first time is a preset interval, and the first integrated signal is input into a detection model, and after a detection result corresponding to the first time is obtained, a second integrated signal corresponding to the second time is obtained; inputting the second comprehensive signal into the detection model to obtain a second detection result; and correcting the early warning model according to the second detection result.
Specifically, an inferred event and occurrence probability of the inferred event are obtained according to the first integrated signal, a second integrated signal corresponding to the second moment is obtained at the second moment, a second detection result corresponding to the second integrated signal, namely, a diagnosis result obtained after the first moment, can be detected according to the detection model, the diagnosis result corresponding to the second moment can be used for detecting the inferred event and the occurrence probability accuracy of the inferred event obtained according to the early warning model at the first moment, and if the second detection result corresponding to the second moment is matched with the inferred event corresponding to the first moment, the inferred result output by the early warning model at the first moment is verified to be accurate.
In one embodiment, when the second detection result is not matched with the first early warning information, the early warning model is modified according to the second detection result to obtain a modified early warning model, and the modified early warning model is used for carrying out early warning analysis on the integrated signal corresponding to the third moment.
Specifically, if the second detection result corresponding to the second moment is not matched with the inferred event corresponding to the first moment, correcting the early warning model according to the second detection result corresponding to the second moment to obtain a corrected early warning model, performing pre-judgment on an abnormal event on the subsequently acquired integrated signal according to the corrected early warning model, and so on, correcting the inferred result output by the early warning model at the last moment according to the detection result output by the detection model at the current moment, and performing iterative update on the early warning model, thereby improving the accuracy of the output result of the early warning model.
In one embodiment, after the first integrated signal is input into the early warning model to obtain the first early warning information, when the occurrence probability corresponding to the current inferred event in the first early warning information is greater than a preset threshold value of the corresponding event, an early warning report is generated according to the first integrated signal, and an early warning signal is sent out.
Specifically, the current inferred event is any inferred event in the first early warning information corresponding to the first moment, each inferred event has a corresponding preset threshold, the preset threshold is a probability value, if the occurrence probability corresponding to the current inferred event is larger than the preset threshold corresponding to the event, which indicates that the occurrence probability of the current inferred event is higher, an early warning report is generated according to the first comprehensive signal, and the early warning report is used for uploading a server background or is sent to a supervision medical staff of a patient and a mobile terminal held by a guardian, timely notifying relevant staff of the impending symptom of the patient, and simultaneously sending an early warning signal timely notifying personnel nearby or guardianship staff holding a remote monitoring terminal to take corresponding rescue measures, so that the patient is prevented from waiting until the symptom is not later rescued.
In one embodiment, before the first integrated signal corresponding to the first time is obtained, a sample integrated signal is obtained within a preset duration, where the sample integrated signal includes a plurality of detection signals; marking each detection signal according to an abnormality detection algorithm to obtain an abnormality mark corresponding to each detection signal; dividing each detection signal according to the abnormal mark corresponding to each detection signal to obtain a corresponding early warning signal slice; and performing deep learning according to the early warning signal slices and the abnormal labels corresponding to the detection signals, and establishing the early warning model.
Specifically, the sample comprehensive signal is a signal continuously collected within a preset duration, the sample comprehensive signal comprises a plurality of detection signals, the signal length of each detection signal corresponds to the preset duration, the detection signals can be physiological parameters such as gyroscope signals, sound signals, blood pressure signals and photoelectric signals, whether the detection signals are abnormal or not is detected according to an abnormality detection algorithm, abnormal fragments generated in the detection signals are marked to obtain abnormal marks corresponding to the detection signals, the detection signals are segmented according to the abnormal marks on the detection signals, each detection signal is segmented into a plurality of fragment signals, each fragment signal carries corresponding time information, the time length of each fragment signal from the abnormal mark is different, any one fragment signal is selected from the plurality of fragment signals to serve as an early warning signal slice, and the early warning models obtained by training different fragment signals and the abnormal labels are also different.
Referring to fig. 3, the continuously recorded integrated multi-modal signal is any one of the integrated signals, and a plurality of the integrated signals are segmented, wherein the early warning signal is segmentedSignal slice for early warning/>Signal slice for early warning/>For the three fragment signals obtained by segmentation in the detection signal, if before the selection of the abnormal mark/>Corresponding fragment signals are used as early warning signal slices/>I.e. all detection signals are selected to be at a distance/>, from the abnormality marker in each detection signalThe segment signals of the detection signals are used as early warning signal slices of the detection signals, and an early warning model obtained through training according to the early warning signal slices corresponding to the detection signals and the abnormal labels can be used for deducing the distance/>, between the early warning model and the first momentAn abnormal event after a length of time, i.e. the time interval between the first instant and the second instant is/>. Different fragment signals are selected as early warning signal slices, so that the early warning model can determine how far the early warning model can predict the abnormal event.
In one embodiment, each detection signal is segmented to obtain a plurality of segment signals, wherein the segment signals carry corresponding starting time and ending time; before the abnormal time corresponding to each detection signal, selecting a fragment signal corresponding to the ending time of a preset interval from the abnormal time as the early warning signal slice.
Specifically, each segment signal carries corresponding time information, the time information includes a start time and an end time of the segment signal, the abnormal mark in each detection signal also carries a corresponding abnormal time, the abnormal time is the start time of occurrence of the abnormality, the preset interval is a time interval which is away from the abnormal time before the abnormal time, the greater the preset interval is, the farther the early warning signal slice is from the segment signal of the abnormal mark, but the stronger the signal corresponding to the early warning signal slice which is closer to the abnormal mark is, namely the higher the accuracy of predicting the abnormal event is, and the preset interval can be customized according to actual conditions.
FIG. 2 is a flow chart of a signal processing method in one embodiment. It should be understood that, although the steps in the flowchart of fig. 2 are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 2 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of the sub-steps or stages of other steps or other steps.
In one embodiment, as shown in fig. 4, there is provided a signal processing apparatus including:
a first signal obtaining module 310, configured to obtain a first integrated signal corresponding to a first time, where the first integrated signal includes a gyroscope signal, a sound signal, a blood pressure signal, and a photoelectric signal;
The early warning module 320 is configured to input the first integrated signal into an early warning model to obtain first early warning information, where the first early warning information includes an inferred event corresponding to a second time and an occurrence probability corresponding to the inferred event, and the second time is greater than the first time.
In one embodiment, the apparatus further comprises:
the first signal detection module is used for inputting the first integrated signal into a detection model to obtain a first detection result, and the first detection result comprises an abnormal state and an abnormal type corresponding to the first moment.
In one embodiment, the apparatus further comprises:
The second signal acquisition module is used for acquiring a second comprehensive signal corresponding to the second moment;
the second signal detection module is used for inputting the second comprehensive signal into the detection model to obtain a second detection result;
and the correction module is used for correcting the early warning model according to the second detection result.
In one embodiment, the correction module includes:
And the correction unit is used for correcting the early warning model according to the second detection result when the second detection result is not matched with the first early warning information, so as to obtain a corrected early warning model, and the corrected early warning model is used for carrying out early warning analysis on the comprehensive signal corresponding to the third moment.
In one embodiment, the apparatus further comprises:
and the early warning report module is used for generating an early warning report according to the first comprehensive signal and sending out an early warning signal when the occurrence probability corresponding to the current inferred event in the first early warning information is larger than the preset threshold value of the corresponding event.
In one embodiment, the apparatus further comprises:
The sample acquisition module is used for acquiring a sample comprehensive signal within a preset duration, wherein the sample comprehensive signal comprises a plurality of detection signals;
the abnormality marking module is used for marking each detection signal according to an abnormality detection algorithm to obtain an abnormality mark corresponding to each detection signal;
The signal segmentation module is used for segmenting each detection signal according to the corresponding abnormal mark of each detection signal to obtain a corresponding early warning signal slice;
and the model training module is used for performing deep learning according to the early warning signal slices and the abnormal labels corresponding to the detection signals and establishing the early warning model.
In one embodiment, the signal slicing module includes:
The signal segmentation unit is used for segmenting each detection signal to obtain a plurality of segment signals, wherein the segment signals carry corresponding starting time and ending time;
And the segment selection unit is used for selecting a segment signal corresponding to the ending time of a preset interval from the abnormal time before the abnormal time corresponding to each detection signal as the early warning signal slice.
FIG. 5 illustrates an internal block diagram of a computer device in one embodiment. The computer device may be specifically the signal acquisition terminal 110 (or the server 120) in fig. 1. As shown in fig. 5, the computer device includes a processor, a memory, a network interface, an input device, and a display screen connected by a system bus. The memory includes a nonvolatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system, and may also store a computer program which, when executed by a processor, causes the processor to implement a signal processing method. The internal memory may also store a computer program which, when executed by the processor, causes the processor to perform the signal processing method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the structure shown in FIG. 5 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, the signal processing apparatus provided by the present application may be implemented in the form of a computer program that is executable on a computer device as shown in fig. 5. The memory of the computer device may store various program modules constituting the signal processing apparatus, such as the first signal acquisition module 310 and the early warning module 320 shown in fig. 4. The computer program constituted by the respective program modules causes the processor to execute the steps in the signal processing method of the respective embodiments of the present application described in the present specification.
The computer device shown in fig. 5 may perform acquisition of a first integrated signal corresponding to a first time by the first signal acquisition module 310 in the signal processing apparatus shown in fig. 4, where the first integrated signal includes a gyroscope signal, a sound signal, a blood pressure signal, and a photoelectric signal. The computer device may perform inputting the first integrated signal into the early warning model through the early warning module 320 to obtain first early warning information, where the first early warning information includes an inferred event corresponding to a second time and an occurrence probability corresponding to the inferred event, and the second time is greater than the first time.
In one embodiment, a computer device is provided comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps of when executing the computer program: acquiring a first comprehensive signal corresponding to a first moment, wherein the first comprehensive signal comprises a gyroscope signal, a sound signal, a blood pressure signal and a photoelectric signal; and inputting the first integrated signal into an early warning model to obtain first early warning information, wherein the first early warning information comprises an inferred event corresponding to a second moment and occurrence probability corresponding to the inferred event, and the second moment is after the first moment.
In one embodiment, the processor when executing the computer program further performs the steps of: and inputting the first integrated signal into a detection model to obtain a first detection result, wherein the first detection result comprises an abnormal state and an abnormal type corresponding to the first moment.
In one embodiment, the processor when executing the computer program further performs the steps of: acquiring a second comprehensive signal corresponding to the second moment; inputting the second comprehensive signal into the detection model to obtain a second detection result; and correcting the early warning model according to the second detection result.
In one embodiment, the processor when executing the computer program further performs the steps of: when the second detection result is not matched with the first early warning information, correcting the early warning model according to the second detection result to obtain a corrected early warning model, wherein the corrected early warning model is used for carrying out early warning analysis on the comprehensive signals corresponding to the third moment.
In one embodiment, the processor when executing the computer program further performs the steps of: and when the occurrence probability corresponding to the current inferred event in the first early warning information is larger than a preset threshold value of the corresponding event, generating an early warning report according to the first comprehensive signal, and sending out an early warning signal.
In one embodiment, the processor when executing the computer program further performs the steps of: acquiring a sample comprehensive signal within a preset time period, wherein the sample comprehensive signal comprises a plurality of detection signals; marking each detection signal according to an abnormality detection algorithm to obtain an abnormality mark corresponding to each detection signal; dividing each detection signal according to the abnormal mark corresponding to each detection signal to obtain a corresponding early warning signal slice; and performing deep learning according to the early warning signal slices and the abnormal labels corresponding to the detection signals, and establishing the early warning model.
In one embodiment, the processor when executing the computer program further performs the steps of: dividing each detection signal to obtain a plurality of fragment signals, wherein the fragment signals carry corresponding starting time and ending time; before the abnormal time corresponding to each detection signal, selecting a fragment signal corresponding to the ending time of a preset interval from the abnormal time as the early warning signal slice.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of: acquiring a first comprehensive signal corresponding to a first moment, wherein the first comprehensive signal comprises a gyroscope signal, a sound signal, a blood pressure signal and a photoelectric signal; and inputting the first integrated signal into an early warning model to obtain first early warning information, wherein the first early warning information comprises an inferred event corresponding to a second moment and occurrence probability corresponding to the inferred event, and the second moment is after the first moment.
In one embodiment, the computer program when executed by the processor further performs the steps of: and inputting the first integrated signal into a detection model to obtain a first detection result, wherein the first detection result comprises an abnormal state and an abnormal type corresponding to the first moment.
In one embodiment, the computer program when executed by the processor further performs the steps of: acquiring a second comprehensive signal corresponding to the second moment; inputting the second comprehensive signal into the detection model to obtain a second detection result; and correcting the early warning model according to the second detection result.
In one embodiment, the computer program when executed by the processor further performs the steps of: when the second detection result is not matched with the first early warning information, correcting the early warning model according to the second detection result to obtain a corrected early warning model, wherein the corrected early warning model is used for carrying out early warning analysis on the comprehensive signals corresponding to the third moment.
In one embodiment, the computer program when executed by the processor further performs the steps of: and when the occurrence probability corresponding to the current inferred event in the first early warning information is larger than a preset threshold value of the corresponding event, generating an early warning report according to the first comprehensive signal, and sending out an early warning signal.
In one embodiment, the computer program when executed by the processor further performs the steps of: acquiring a sample comprehensive signal within a preset time period, wherein the sample comprehensive signal comprises a plurality of detection signals; marking each detection signal according to an abnormality detection algorithm to obtain an abnormality mark corresponding to each detection signal; dividing each detection signal according to the abnormal mark corresponding to each detection signal to obtain a corresponding early warning signal slice; and performing deep learning according to the early warning signal slices and the abnormal labels corresponding to the detection signals, and establishing the early warning model.
In one embodiment, the computer program when executed by the processor further performs the steps of: dividing each detection signal to obtain a plurality of fragment signals, wherein the fragment signals carry corresponding starting time and ending time; before the abnormal time corresponding to each detection signal, selecting a fragment signal corresponding to the ending time of a preset interval from the abnormal time as the early warning signal slice.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program, which may be stored on a non-transitory computer readable storage medium, and which, when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SYNCHLINK) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
It should be noted that in this document, relational terms such as "first" and "second" and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. A method of signal processing, the method comprising:
Acquiring a first comprehensive signal corresponding to a first moment, wherein the first comprehensive signal comprises a gyroscope signal, a sound signal, a blood pressure signal and a photoelectric signal;
Inputting the first comprehensive signal into an early warning model to obtain first early warning information, wherein the first early warning information comprises an inferred event corresponding to a second moment and occurrence probability corresponding to the inferred event, and the second moment is after the first moment;
before the first integrated signal corresponding to the first moment is obtained, the method further includes:
acquiring a sample comprehensive signal within a preset time period, wherein the sample comprehensive signal comprises a plurality of detection signals;
marking each detection signal according to an abnormality detection algorithm to obtain an abnormality mark corresponding to each detection signal;
dividing each detection signal according to the abnormal mark corresponding to each detection signal to obtain a corresponding early warning signal slice;
deep learning is carried out according to the early warning signal slices and the abnormal marks corresponding to the detection signals, and the early warning model is established;
The step of slicing each detection signal according to the corresponding abnormal mark of each detection signal to obtain a corresponding early warning signal slice comprises the following steps:
Dividing each detection signal to obtain a plurality of fragment signals, wherein the fragment signals carry corresponding starting time and ending time, and the abnormal marks in each detection signal also carry corresponding abnormal time, and the abnormal time is the starting time of occurrence of the abnormality;
Before the abnormal time corresponding to each detection signal, selecting a fragment signal corresponding to the ending time of a preset interval from the abnormal time as the early warning signal slice.
2. The method of claim 1, wherein after inputting the first integrated signal into the early warning model to obtain the first early warning information, the method further comprises:
And inputting the first integrated signal into a detection model to obtain a first detection result, wherein the first detection result comprises an abnormal state and an abnormal type corresponding to the first moment.
3. The method according to claim 2, wherein the time difference between the second time and the first time is a preset interval, and the method further comprises, after the first integrated signal is input into the detection model to obtain the detection result corresponding to the first time:
Acquiring a second comprehensive signal corresponding to the second moment;
inputting the second comprehensive signal into the detection model to obtain a second detection result;
and correcting the early warning model according to the second detection result.
4. A method according to claim 3, wherein said correcting said pre-warning model based on said second detection result comprises:
when the second detection result is not matched with the first early warning information, correcting the early warning model according to the second detection result to obtain a corrected early warning model, wherein the corrected early warning model is used for carrying out early warning analysis on the comprehensive signals corresponding to the third moment.
5. The method of claim 1, wherein after inputting the first integrated signal into the early warning model to obtain the first early warning information, the method further comprises:
And when the occurrence probability corresponding to the current inferred event in the first early warning information is larger than a preset threshold value of the corresponding event, generating an early warning report according to the first comprehensive signal, and sending out an early warning signal.
6. A signal processing apparatus, the apparatus comprising:
the first signal acquisition module is used for acquiring a first comprehensive signal corresponding to a first moment, wherein the first comprehensive signal comprises a gyroscope signal, a sound signal, a blood pressure signal and a photoelectric signal;
The early warning module is used for inputting the first comprehensive signals into an early warning model to obtain first early warning information, wherein the first early warning information comprises an inferred event corresponding to a second moment and occurrence probability corresponding to the inferred event, and the second moment is larger than the first moment;
the apparatus further comprises:
The sample acquisition module is used for acquiring a sample comprehensive signal within a preset duration, wherein the sample comprehensive signal comprises a plurality of detection signals;
the abnormality marking module is used for marking each detection signal according to an abnormality detection algorithm to obtain an abnormality mark corresponding to each detection signal;
The signal segmentation module is used for segmenting each detection signal according to the corresponding abnormal mark of each detection signal to obtain a corresponding early warning signal slice;
the model training module is used for performing deep learning according to the early warning signal slices and the abnormal marks corresponding to the detection signals and establishing the early warning model;
The signal segmentation module comprises:
The signal segmentation unit is used for segmenting each detection signal to obtain a plurality of segment signals, wherein the segment signals carry corresponding starting time and ending time, the abnormality marks in each detection signal also carry corresponding abnormality time, and the abnormality time is the starting time of abnormality occurrence;
And the segment selection unit is used for selecting a segment signal corresponding to the ending time of a preset interval from the abnormal time before the abnormal time corresponding to each detection signal as the early warning signal slice.
7. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1 to 5 when the computer program is executed by the processor.
8. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011488258.2A CN112842297B (en) | 2020-12-16 | Signal processing method, apparatus, computer device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011488258.2A CN112842297B (en) | 2020-12-16 | Signal processing method, apparatus, computer device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112842297A CN112842297A (en) | 2021-05-28 |
| CN112842297B true CN112842297B (en) | 2024-05-14 |
Family
ID=
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104473648A (en) * | 2014-09-24 | 2015-04-01 | 上海大学 | Physiological parameter monitoring-combined human body tumble warning and detecting method |
| CN108143407A (en) * | 2017-12-25 | 2018-06-12 | 四川大学 | A kind of heart sound segmentation method for automatically extracting heart sound envelope characteristic |
| CN109924965A (en) * | 2018-10-12 | 2019-06-25 | 王平 | A kind of cardiovascular and cerebrovascular rehabilitation training system |
| CN110200619A (en) * | 2019-06-06 | 2019-09-06 | 中山大学孙逸仙纪念医院 | It is a kind of to establish major cardiovascular disease risks Early-warning Model method and device |
| CN110537909A (en) * | 2019-09-03 | 2019-12-06 | 深圳旭宏医疗科技有限公司 | First-aid method and device based on electrocardiogram monitoring equipment and computer equipment |
| CN110720903A (en) * | 2019-10-22 | 2020-01-24 | 深圳旭宏医疗科技有限公司 | Electrocardiosignal processing method and device |
| CN111000551A (en) * | 2019-12-05 | 2020-04-14 | 常州市第一人民医院 | Heart disease risk diagnosis method based on deep convolutional neural network model |
| CN112017784A (en) * | 2020-10-22 | 2020-12-01 | 平安科技（深圳）有限公司 | Coronary heart disease risk prediction method based on multi-modal data and related equipment |
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104473648A (en) * | 2014-09-24 | 2015-04-01 | 上海大学 | Physiological parameter monitoring-combined human body tumble warning and detecting method |
| CN108143407A (en) * | 2017-12-25 | 2018-06-12 | 四川大学 | A kind of heart sound segmentation method for automatically extracting heart sound envelope characteristic |
| CN109924965A (en) * | 2018-10-12 | 2019-06-25 | 王平 | A kind of cardiovascular and cerebrovascular rehabilitation training system |
| CN110200619A (en) * | 2019-06-06 | 2019-09-06 | 中山大学孙逸仙纪念医院 | It is a kind of to establish major cardiovascular disease risks Early-warning Model method and device |
| CN110537909A (en) * | 2019-09-03 | 2019-12-06 | 深圳旭宏医疗科技有限公司 | First-aid method and device based on electrocardiogram monitoring equipment and computer equipment |
| CN110720903A (en) * | 2019-10-22 | 2020-01-24 | 深圳旭宏医疗科技有限公司 | Electrocardiosignal processing method and device |
| CN111000551A (en) * | 2019-12-05 | 2020-04-14 | 常州市第一人民医院 | Heart disease risk diagnosis method based on deep convolutional neural network model |
| CN112017784A (en) * | 2020-10-22 | 2020-12-01 | 平安科技（深圳）有限公司 | Coronary heart disease risk prediction method based on multi-modal data and related equipment |
Non-Patent Citations (1)
| Title |
|---|
| 高上凯.《脑-计算机交互研究前沿》.上海交通大学出版社,2019,第295-296页. * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11147463B2 (en) | Method and apparatus for high accuracy photoplethysmogram based atrial fibrillation detection using wearable device | |
| EP3367897B1 (en) | An automatic method to delineate or categorize an electrocardiogram | |
| US10779744B2 (en) | Automatic method to delineate or categorize an electrocardiogram | |
| JP7108267B2 (en) | Biological information processing system, biological information processing method, and computer program | |
| CN109564586B (en) | System monitor and system monitoring method | |
| US11185269B1 (en) | Context-aware post traumatic stress disorder monitoring and intervention | |
| US10456087B2 (en) | Method for score confidence interval estimation when vital sign sampling frequency is limited | |
| JP2018513727A (en) | Cardiovascular deterioration warning score | |
| KR20210064029A (en) | Method, system and non-transitory computer-readable recording medium for estimating arrhythmia by using artificial neural network | |
| EP3219253B1 (en) | System for detecting arrhythmia using photoplethysmogram signal | |
| WO2022034045A1 (en) | Systems and methods for generating health risk assessments | |
| US20220218289A1 (en) | Predicting transient ischemic events using ecg data | |
| JP2021533938A (en) | Advanced cardiac waveform analysis | |
| CN112842297B (en) | Signal processing method, apparatus, computer device and storage medium | |
| JP2022042304A (en) | Stress estimation device, method, and program | |
| JP2022502201A (en) | Continuous monitoring of user health using mobile devices | |
| Madona et al. | Classification of ECG signals using the Naïve Bayes classification method and its implementation in android-based smart health care | |
| JP7002168B1 (en) | ECG analyzer, ECG analysis method and program | |
| US20210298625A1 (en) | System and method for detecting and predicting an occurrence of cardiac events from electrocardiograms | |
| CN112842297A (en) | Signal processing method, signal processing device, computer equipment and storage medium | |
| Thakur et al. | Patient Health Monitoring and Inferencing Arrhythmia Using ECG_Data | |
| KR20210157444A (en) | Machine Learning-Based Diagnosis Method Of Schizophrenia And System there-of | |
| Dey et al. | Patient Health Observation and Analysis with Machine Learning and IoT Based in Realtime Environment | |
| US20230346290A1 (en) | Overriding longest rr intervals | |
| Kuzmin et al. | Intelligent data processing scheme for mobile heart monitoring system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |