CN116195998A - Blood oxygen detection method and device, computer equipment and storage medium - Google Patents
Blood oxygen detection method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN116195998A CN116195998A CN202310465046.XA CN202310465046A CN116195998A CN 116195998 A CN116195998 A CN 116195998A CN 202310465046 A CN202310465046 A CN 202310465046A CN 116195998 A CN116195998 A CN 116195998A
- Authority
- CN
- China
- Prior art keywords
- channel image
- determining
- amplitude
- blood oxygen
- light data
- 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.)
- Granted
Links
- 239000008280 blood Substances 0.000 title claims abstract description 94
- 210000004369 blood Anatomy 0.000 title claims abstract description 94
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 title claims abstract description 92
- 229910052760 oxygen Inorganic materials 0.000 title claims abstract description 92
- 239000001301 oxygen Substances 0.000 title claims abstract description 92
- 238000001514 detection method Methods 0.000 title claims abstract description 55
- 238000000034 method Methods 0.000 claims abstract description 19
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 9
- 238000004364 calculation method Methods 0.000 claims description 29
- 238000012549 training Methods 0.000 claims description 23
- 230000006870 function Effects 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 7
- 238000002496 oximetry Methods 0.000 claims 1
- 238000012545 processing Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000011897 real-time detection Methods 0.000 description 3
- 206010052428 Wound Diseases 0.000 description 2
- 208000027418 Wounds and injury Diseases 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 206010021143 Hypoxia Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000010241 blood sampling Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 210000000624 ear auricle Anatomy 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 210000001061 forehead Anatomy 0.000 description 1
- 230000007954 hypoxia Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 210000000707 wrist Anatomy 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/1455—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue using optical sensors, e.g. spectral photometrical oximeters
- A61B5/14551—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue using optical sensors, e.g. spectral photometrical oximeters for measuring blood gases
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention discloses a blood oxygen detection method, a blood oxygen detection device, computer equipment and a storage medium, wherein the method comprises the following steps: acquiring first light data of the first wavelength infrared light reflected by the skin, and acquiring second light data of the second wavelength infrared light reflected by the skin; generating a first single-channel image corresponding to the first light data and a second single-channel image corresponding to the second light data according to the first light data and the second light data; and detecting the first single-channel image and the second single-channel image according to a trained network model, determining the blood oxygen concentration corresponding to the skin, wherein the network model is used for classifying and detecting the first single-channel image and the second single-channel image, and the network model is a convolutional neural network. Through the first single-channel image and the second single-channel image, the blood oxygen concentration corresponding to the skin can be obtained, and the accuracy and the efficiency of blood oxygen concentration detection are improved.
Description
Technical Field
The present invention relates to the field of blood oxygen detection technology, and in particular, to a blood oxygen detection method, a blood oxygen detection device, a computer device, and a storage medium.
Background
The existing blood oxygen content detection is generally divided into invasive blood oxygen detection, transmission blood oxygen detection and reflection blood oxygen detection. The invasive blood oxygen detection is usually completed by adopting a blood oxygen analyzer, the blood oxygen analyzer detects the oxygen saturation in blood, and the detection of the blood oxygen content is obtained by collecting human blood; the transmission type blood oxygen detection and the reflection type blood oxygen detection avoid wounds and realize real-time detection, the transmission type blood oxygen detection and the reflection type blood oxygen detection mainly use light sources to emit light to human skin, then use a light receiving device to receive the light after transmitting the skin and the light after reflecting the skin, finally analyze the received light, and perform real-time noninvasive calculation to obtain the blood oxygen content.
The invasive blood oxygen detection needs to be carried out by adopting a human body blood sampling mode, blood needs to be drawn every time of sampling, pain is brought to a patient, the blood collection and preservation conditions are strict, the analysis time is long, and the blood oxygen content cannot be detected in real time. Although the transmission type blood oxygen detection avoids wounds and realizes real-time detection, the transmission type blood oxygen detection is required to be clung to the skin to prevent light from interfering with detection results, and meanwhile, strict requirements are required on measurement parts, such as fingers and earlobes, the measurement range is smaller, and the blood oxygen content of larger or thicker local tissues cannot be measured. The reflective blood oxygen detection also realizes noninvasive and real-time detection, but has requirements on detection parts, and the detection area is required to be flat, such as forehead and wrist, and the blood oxygen content detection error is larger under the condition of hypoxia. Meanwhile, the current reflective blood oxygen detection technology is easy to be interfered by external illumination, power frequency noise and the like. And the formula generated by using artificial experience is needed, so that the blood oxygen content of various conditions cannot be well estimated.
At present, the blood oxygen content detection technology has the defects of invasiveness, non-real time, contact, non-real time, poor anti-interference capability, limited use area and the like. Therefore, how to provide a blood oxygen detection method to accurately and efficiently detect the blood oxygen concentration becomes a technical problem to be solved at present.
Disclosure of Invention
The embodiment of the invention provides a blood oxygen detection method, a blood oxygen detection device, computer equipment and a storage medium, which can improve the accuracy and efficiency of blood oxygen concentration detection.
In order to solve the above technical problems, an embodiment of the present application provides a blood oxygen detection method, including the following steps:
acquiring first light data of the first wavelength infrared light reflected by the skin, and acquiring second light data of the second wavelength infrared light reflected by the skin;
generating a first single-channel image corresponding to the first light data and a second single-channel image corresponding to the second light data according to the first light data and the second light data;
and detecting the first single-channel image and the second single-channel image according to a trained network model, determining the blood oxygen concentration corresponding to the skin, wherein the network model is used for classifying and detecting the first single-channel image and the second single-channel image, and the network model is a convolutional neural network.
Further, the determining, according to the first light ray data, a corresponding first single-channel image, and determining, according to the second light ray data, a corresponding second single-channel image includes:
determining a first direct current component and a first alternating current component corresponding to the first light ray data, and determining a second direct current component and a second alternating current component corresponding to the second light ray data;
the first single-channel image is determined according to each first direct current component and each first alternating current component, and the second single-channel image is determined according to each second direct current component and each second alternating current component.
Further, the determining the first dc component and the first ac component corresponding to the first light data, and determining the second dc component and the second ac component corresponding to the second light data includes:
calculating standard deviation of peak-to-valley heights in every five pulse periods according to the first light ray data, taking a result obtained by each calculation as one first alternating current component, calculating an average value of peak heights in every five pulse periods, and taking the result obtained by each calculation as one first direct current component;
and calculating standard deviation of peak-to-valley heights in every five pulse periods according to the second light ray data, taking a result obtained by each calculation as one second alternating current component, calculating an average value of peak heights in every five pulse periods, and taking a result obtained by each calculation as one second direct current component.
Further, the determining the first single-channel image according to the first direct current component and the first alternating current component, and the determining the second single-channel image according to the second direct current component and the second alternating current component includes:
determining a first amplitude of the first dc component and a second amplitude of the first ac component, and determining a third amplitude of the second dc component and a fourth amplitude of the second ac component;
and generating the first single-channel image according to the first amplitude and the second amplitude, and generating the second single-channel image according to the third amplitude and the fourth amplitude.
Further, the generating the first single-channel image according to the first amplitude and the second amplitude, and the generating the second single-channel image according to the third amplitude and the fourth amplitude includes:
determining each first sequence according to the first amplitude and the second amplitude, and determining each second sequence according to the third amplitude and the fourth amplitude;
and generating the first single-channel image according to each first sequence, and generating the second single-channel image according to each second sequence.
Further, before the step of obtaining the first light data of the first wavelength infrared light reflected by the skin and the second light data of the second wavelength infrared light reflected by the skin, the step of obtaining includes:
acquiring third light data of the first wavelength infrared light reflected by the skin and fourth light data of the second wavelength infrared light reflected by the skin, acquiring actual blood oxygen concentration, and taking the actual blood oxygen concentration as a training label;
determining a corresponding third single-channel image according to the third light ray data, and determining a corresponding fourth single-channel image according to the fourth light ray data;
and taking the third single-channel image, the fourth single-channel image and the training label as training samples, and training an initial network model by adopting the training samples to obtain a network model.
Further, the initial network model includes a first convolution layer, a spatial attention layer, a second convolution layer, a third convolution layer, a full connection layer, and an output layer that are sequentially connected, and training the initial network model using the training sample to obtain the network model includes:
inputting the third single-channel image and the fourth single-channel image into a first convolution layer to obtain a first feature, and inputting the first feature into a spatial attention layer to obtain a second feature;
inputting the second feature into the second convolution layer to obtain a third feature, inputting the third feature into the third convolution layer to obtain a fourth feature, inputting the fourth feature into a full connection layer, and determining a fifth feature;
inputting the fifth characteristic to an output layer to obtain various probability results, and determining a probability result with the maximum probability value as a detected blood oxygen concentration in the probability results;
and determining a classification loss function according to the error of the detected blood oxygen concentration and the actual blood oxygen concentration, and adjusting parameters and weights in the initial network model through the classification loss function to obtain the trained network model.
In order to solve the above technical problem, an embodiment of the present application further provides a blood oxygen detection device, including:
the acquisition module is used for acquiring first light data of the first wavelength infrared light reflected by the skin and acquiring second light data of the second wavelength infrared light reflected by the skin;
the generation module is used for generating a first single-channel image corresponding to the first light data and a second single-channel image corresponding to the second light data according to the first light data and the second light data;
the determining module is used for detecting the first single-channel image and the second single-channel image according to a trained network model, determining the blood oxygen concentration corresponding to the skin, and the network model is used for classifying and detecting the first single-channel image and the second single-channel image, wherein the network model is a convolutional neural network.
To solve the above technical problem, embodiments of the present application further provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the above method when executing the computer program.
To solve the above technical problem, embodiments of the present application further provide a computer readable storage medium storing a computer program, which when executed by a processor, implements the steps of the above method.
According to the blood oxygen detection method, the blood oxygen detection device, the computer equipment and the storage medium, first light data of the first wavelength infrared light reflected by the skin are obtained, and second light data of the second wavelength infrared light reflected by the skin are obtained; generating a first single-channel image corresponding to the first light data and a second single-channel image corresponding to the second light data according to the first light data and the second light data; and detecting the first single-channel image and the second single-channel image according to a trained network model, determining the blood oxygen concentration corresponding to the skin, wherein the network model is used for classifying and detecting the first single-channel image and the second single-channel image, and the network model is a convolutional neural network. Through the first single-channel image and the second single-channel image, the blood oxygen concentration corresponding to the skin can be obtained, and the accuracy and the efficiency of blood oxygen concentration detection are improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments of the present invention will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is an exemplary system architecture diagram in which the present application may be applied;
FIG. 2 is a flow chart of one embodiment of a blood oxygen detection method of the present application;
FIG. 3 is a schematic illustration of a single channel image of the blood oxygen detection method of the present application;
FIG. 4 is a block diagram of an initial network model of the blood oxygen detection method of the present application;
FIG. 5 is a schematic diagram of the structure of one embodiment of an oxygen detection device according to the present application;
FIG. 6 is a schematic structural diagram of one embodiment of a computer device according to the present application.
Detailed Description
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used in the description of the applications herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having" and any variations thereof in the description and claims of the present application and in the description of the figures above are intended to cover non-exclusive inclusions. The terms first, second and the like in the description and in the claims or in the above-described figures, are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the present application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The blood oxygen detection method provided by the embodiment of the invention can be applied to an application environment shown in fig. 1, wherein the acquisition equipment is communicated with the computing equipment. The method comprises the steps that an acquisition device acquires first light data of first wavelength infrared light after skin reflection and second light data of second wavelength infrared light after skin reflection, then a first single-channel image corresponding to the first light data and a second single-channel image corresponding to the second light data are generated according to the first light data and the second light data, a trained network model is arranged in a computing device, the computing device acquires the first single-channel image and the second single-channel image, the first single-channel image and the second single-channel image are detected through the network model, and therefore blood oxygen concentration corresponding to skin is determined. Wherein the acquisition device may be an infrared acquisition device. The computing device may be an image data processing capable device, and the computing device may be implemented as a stand-alone server or as a server cluster of multiple servers.
It should be noted that, the blood oxygen detection method provided in the embodiment of the present application is executed by a server, and accordingly, the blood oxygen detection device is disposed in the server.
It should be understood that the number of terminal devices, networks and servers in fig. 1 is merely illustrative. Any number of terminal devices, networks and servers may be provided according to implementation requirements, and the terminal device 110 in the embodiment of the present application may specifically correspond to an application system in actual production.
Referring to fig. 2, fig. 2 shows a blood oxygen detection method according to an embodiment of the present invention, and the method is applied to the server in fig. 1 for illustration, as follows.
S201, acquiring first light data of the first wavelength infrared light reflected by the skin and acquiring second light data of the second wavelength infrared light reflected by the skin;
wherein the first wavelength infrared light may be 940nm wavelength infrared light and the second wavelength infrared light may be 660nm wavelength infrared light.
Specifically, the first light ray data and the second light ray data may be collected by a collecting device, where the collecting device is provided with an infrared light emitting end with a wavelength of 940nm, an infrared light emitting end with a wavelength of 660nm, and a light receiving end.
As an example, the 940nm wavelength infrared light emitting end emits 940nm wavelength infrared light, and the light receiving end receives the light L of the 940nm wavelength infrared light after being reflected by the skin A 。
As yet another example, the 660nm wavelength infrared light emitting end emits 660nm wavelength red light, and the light receiving end receives light L after skin reflection of the 660nm wavelength red light B 。
S202, generating a first single-channel image corresponding to the first light data and a second single-channel image corresponding to the second light data according to the first light data and the second light data;
in this embodiment, a first single-channel image corresponding to the first light data may be generated by the first light data, and a second single-channel image corresponding to the second light data may be generated by the second light data.
Further, in an embodiment, S202 further includes:
s2021, determining a first direct current component and a first alternating current component corresponding to the first light ray data, and determining a second direct current component and a second alternating current component corresponding to the second light ray data;
s2022, determining the first single-channel image according to each of the first dc components and each of the first ac components, and determining the second single-channel image according to each of the second dc components and each of the second ac components.
Specifically, first, a first direct current component and a first alternating current component corresponding to first light data are calculated through the first light data, and a plurality of second direct current components and a plurality of second alternating current components corresponding to second light data are calculated through the second light data.
Then, a first single-channel image is generated by each first direct current component and each first alternating current component, and a second single-channel image is generated by each second direct current component and each second alternating current component.
Further, in an embodiment, S2021 further includes:
s20211, calculating standard deviation of peak-to-valley heights in every five pulse periods according to the first light ray data, taking a result obtained by each calculation as one first alternating current component, calculating an average value of peak heights in every five pulse periods, and taking the result obtained by each calculation as one first direct current component;
s20212, calculating standard deviation of peak-to-valley heights in every five pulse periods according to the second light ray data, taking the result obtained by each calculation as one second alternating current component, calculating the average value of peak heights in every five pulse periods, and taking the result obtained by each calculation as one second direct current component.
In this embodiment, for the first light ray data, calculating the standard deviation of the peak-to-valley heights in each five pulse periods, taking the result obtained by each calculation as a first alternating current component, calculating the average value of the peak heights in each five pulse periods, and taking the result obtained by each calculation as a first direct current component; for the second light ray data, calculating standard deviation of peak-to-valley heights in every five pulse periods, taking the result obtained by each calculation as a second alternating current component, calculating the average value of the peak heights in every five pulse periods, and taking the result obtained by each calculation as a second direct current component.
In another implementation manner, for the first light ray data, calculating a standard deviation of peak-to-valley heights in every ten pulse periods, taking a result obtained by each calculation as a first alternating current component, calculating a mean value of peak heights in every ten pulse periods, and taking the result obtained by each calculation as a first direct current component; for the second light ray data, calculating standard deviation of peak-to-valley heights in every ten pulse periods, taking the result obtained by each calculation as a second alternating current component, calculating the average value of the peak heights in every ten pulse periods, and taking the result obtained by each calculation as a second direct current component.
Further, in an embodiment, S2022 further includes:
s20221, determining a first amplitude of the first dc component and a second amplitude of the first ac component, and determining a third amplitude of the second dc component and a fourth amplitude of the second ac component;
s20222, generating the first single-channel image according to the first amplitude and the second amplitude, and generating the second single-channel image according to the third amplitude and the fourth amplitude.
The amplitude is the maximum value that the physical quantity of vibration can reach, and represents the physical quantity of the vibration range and intensity.
In the present embodiment, the first single-channel image may be generated by the first amplitude of the first direct current component and the second amplitude of the first alternating current component, and the second single-channel image may be generated by the third amplitude of the second direct current component and the fourth amplitude of the second alternating current component.
As an example, the first amplitude of the first DC component may be represented by I DC940 A representation; the second amplitude of the first AC component may be represented by I AC940 A representation; the third amplitude of the second DC component may be represented by I DC660 A representation; the fourth amplitude of the second AC component may be represented by I AC660 And (3) representing.
Further, in one embodiment, S20222 includes:
A. determining each first sequence according to the first amplitude and the second amplitude, and determining each second sequence according to the third amplitude and the fourth amplitude;
B. and generating the first single-channel image according to each first sequence, and generating the second single-channel image according to each second sequence.
In this embodiment, each first sequence is calculated by the first amplitude and the second amplitude, and each second sequence is calculated according to the third amplitude and the fourth amplitude.
As an example, the second amplitude is divided by the first amplitude to obtain a calculation result, and the calculation result is the first sequence. Dividing the fourth amplitude by the third amplitude to obtain a calculation result, i.e. the second sequence, e.g. lambda 940 For the first sequence and lambda 660 As a second sequence lambda 940 =I AC940 /I DC940 ，λ 660 =I AC660 /I DC660 。
As yet another example, the first amplitude is divided by the second amplitude to obtain a calculation result, and the calculation result is the first sequence. Dividing the third amplitude by the fourth amplitude to obtain a calculation result, i.e. the second sequence, e.g. lambda 940 For the first sequence and lambda 660 As a second sequence lambda 940 =I DC940 /I AC940 ，λ 660 =I DC660 /I AC660 。
Next, a first single-channel image is generated by each first sequence, and a second single-channel image is generated by each second sequence. For example, lambda is obtained by acquisition and calculation 940 And lambda (lambda) 660 100 values each, lambda 940 And lambda (lambda) 660 Each of the 100 values of (2) generates a single pass single image having a width and height of 100 x 100.
As an example, 100 lambda values are presented in a plane rectangular coordinate system, in which the X-axis is the number of values, for example, the 0 th, 1 st, and 2 nd, and the Y-axis is the nth lambda value, so as to obtain a single-channel image, as shown in fig. 3, and fig. 3 is a single-channel image composed of lambda values.
S203, detecting the first single-channel image and the second single-channel image according to a trained network model, determining the blood oxygen concentration corresponding to the skin, wherein the network model is used for classifying and detecting the first single-channel image and the second single-channel image, and the network model is a convolutional neural network.
The network model can be used for classifying and detecting the first single-channel image and the second single-channel image.
Specifically, the first single-channel image and the second single-channel image are input into a trained network model for detection, and a detection result is output, wherein the detection result is the blood oxygen concentration corresponding to the skin.
Further, in an embodiment, S203 includes:
s2031, acquiring third light data of the first wavelength infrared light reflected by the skin and fourth light data of the second wavelength infrared light reflected by the skin, acquiring actual blood oxygen concentration corresponding to the skin, and taking the actual blood oxygen concentration as a training label;
s2032, determining a corresponding third single-channel image according to the third light ray data, and determining a corresponding fourth single-channel image according to the fourth light ray data;
and S2033, taking the third single-channel image, the fourth single-channel image and the training label as training samples, and training an initial network model by adopting the training samples to obtain a network model.
In this embodiment, the corresponding third single-channel image is determined according to the third light ray data, and the corresponding processing procedure of the fourth single-channel image is determined according to the fourth light ray data, which is similar to the processing procedure of determining the corresponding first single-channel image according to the first light ray data and determining the corresponding second single-channel image according to the second light ray data in the above embodiment, and is not repeated herein.
And finally, taking the third single-channel image, the fourth single-channel image and the training label as training samples, and training the initial network model to obtain the network model.
Further, in an embodiment, the initial network model includes a first convolution layer, a spatial attention layer, a second convolution layer, a third convolution layer, a full connection layer, and an output layer connected in sequence, and S2033 includes:
s20331, inputting the third single-channel image and the fourth single-channel image into a first convolution layer to obtain a first feature, and inputting the first feature into a spatial attention layer to obtain a second feature;
s20332, inputting the second feature into the second convolution layer to obtain a third feature, inputting the third feature into the third convolution layer to obtain a fourth feature, inputting the fourth feature into a full connection layer, and determining a fifth feature;
s20333, inputting the fifth feature to an output layer to obtain various probability results, and determining a probability result with the maximum probability value as a detected blood oxygen concentration in the probability results;
s20334, determining a classification loss function according to the error of the detected blood oxygen concentration and the actual blood oxygen concentration, and adjusting parameters and weights in the initial network model through the classification loss function to obtain the trained network model.
Specifically, as shown in fig. 4, the data is sent to an initial network model for training, the INPUT size of the initial network model is 100×100×2, that is, the third single-channel image and the fourth single-channel image may be 100×100, the first feature is obtained through conv3×3 first convolution layer processing, the first feature is sent to a spatial attention layer Spatial Attention, the spatial attention layer outputs a second feature, and the spatial attention layer is used to remove environmental interference factors, so that the algorithm has an anti-noise function. Then, the second feature is processed by the second convolution layer to obtain a third feature of 25×25×64, the third feature is processed by the third convolution layer to generate a fourth feature of 12×12×128, the fourth feature is sent to the full connection layer to output a fifth feature of 1×1×256, and finally 100 probability values are output, wherein the label represented by the maximum probability value is the detected blood oxygen concentration detected by the initial network model, for example, the 90 th probability value is the largest, and the detected blood oxygen concentration is 90%.
And finally, determining a classification loss function according to the error of the detected blood oxygen concentration and the actual blood oxygen concentration, and adjusting parameters and weights in the initial network model through the classification loss function to obtain a trained network model.
According to the blood oxygen detection method, first light data of the first wavelength infrared light reflected by the skin are obtained, and second light data of the second wavelength infrared light reflected by the skin are obtained; generating a first single-channel image corresponding to the first light data and a second single-channel image corresponding to the second light data according to the first light data and the second light data; and detecting the first single-channel image and the second single-channel image according to a trained network model, determining the blood oxygen concentration corresponding to the skin, wherein the network model is used for classifying and detecting the first single-channel image and the second single-channel image, and the network model is a convolutional neural network. Through the first single-channel image and the second single-channel image, the blood oxygen concentration corresponding to the skin can be obtained, and the accuracy and the efficiency of blood oxygen concentration detection are improved.
Fig. 5 shows a schematic block diagram of the blood oxygen detecting device 300 according to the above embodiment. As shown in fig. 5, the apparatus 300 includes an acquisition module 310, a generation module 320, and a determination module 330. The functional modules are described in detail below.
An acquisition module 310, configured to acquire first light data of the first wavelength infrared light reflected by the skin, and acquire second light data of the second wavelength infrared light reflected by the skin;
the generating module 320 is configured to generate a first single-channel image corresponding to the first light data and a second single-channel image corresponding to the second light data according to the first light data and the second light data;
the determining module 330 is configured to detect the first single-channel image and the second single-channel image according to a trained network model, determine an oxygen concentration of blood corresponding to the skin, and the network model is configured to perform classification detection on the first single-channel image and the second single-channel image, where the network model is a convolutional neural network.
For specific limitations of the blood oxygen detection device, reference may be made to the above limitations of the blood oxygen detection method, and no further description is given here. The respective modules in the blood oxygen detecting apparatus described above may be realized in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In order to solve the technical problems, the embodiment of the application also provides computer equipment. Referring specifically to fig. 5, fig. 5 is a basic structural block diagram of a computer device according to the present embodiment.
The computer device 400 includes a memory 410, a processor 420, and a network interface 430 communicatively coupled to each other via a system bus. It should be noted that only computer device 400 having component connection memory 410, processor 420, and network interface 430 is shown in the figures, but it should be understood that not all of the illustrated components need be implemented and that more or fewer components may be implemented instead. It will be appreciated by those skilled in the art that the computer device herein is a device capable of automatically performing numerical calculations and/or blood oxygen measurements in accordance with predetermined or stored instructions, the hardware of which includes, but is not limited to, microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.
The computer equipment can be a desktop computer, a notebook computer, a palm computer, a cloud server and other computing equipment. The computer equipment can perform man-machine interaction with a user through a keyboard, a mouse, a remote controller, a touch pad or voice control equipment and the like.
The memory 410 includes at least one type of readable storage medium including flash memory, hard disk, multimedia card, card memory (e.g., SD or D interface display memory, etc.), random Access Memory (RAM), static Random Access Memory (SRAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), programmable Read Only Memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 410 may be an internal storage unit of the computer device 400, such as a hard disk or a memory of the computer device 400. In other embodiments, the memory 410 may also be an external storage device of the computer device 400, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card), etc. that are provided on the computer device 400. Of course, the memory 410 may also include both internal storage units and external storage devices of the computer device 400. In this embodiment, the memory 410 is typically used to store an operating system and various application software installed on the computer device 400, such as program codes for controlling electronic files. In addition, the memory 410 may also be used to temporarily store various types of data that have been output or are to be output.
The processor 420 may be a central processing unit (Central Processing Unit, CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. The processor 420 is generally used to control the overall operation of the computer device 400. In this embodiment, the processor 420 is configured to execute the program code stored in the memory 410 or process data, such as program code for executing control of an electronic file.
The network interface 430 may include a wireless network interface or a wired network interface, the network interface 430 typically being used to establish a communication connection between the computer device 400 and other electronic devices.
The present application also provides another embodiment, namely, a computer-readable storage medium storing an interface display program executable by at least one processor to cause the at least one processor to perform the steps of the method as described above.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk), comprising several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method described in the embodiments of the present application.
It is apparent that the embodiments described above are only some embodiments of the present application, but not all embodiments, the preferred embodiments of the present application are given in the drawings, but not limiting the patent scope of the present application. This application may be embodied in many different forms, but rather, embodiments are provided in order to provide a more thorough understanding of the present disclosure. Although the present application has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described in the foregoing, or equivalents may be substituted for elements thereof. All equivalent structures made by the specification and the drawings of the application are directly or indirectly applied to other related technical fields, and are also within the protection scope of the application.
Claims (10)
1. A blood oxygen detection method, characterized in that the blood oxygen detection method comprises the steps of:
acquiring first light data of the first wavelength infrared light reflected by the skin, and acquiring second light data of the second wavelength infrared light reflected by the skin;
generating a first single-channel image corresponding to the first light data and a second single-channel image corresponding to the second light data according to the first light data and the second light data;
and detecting the first single-channel image and the second single-channel image according to a trained network model, determining the blood oxygen concentration corresponding to the skin, wherein the network model is used for classifying and detecting the first single-channel image and the second single-channel image, and the network model is a convolutional neural network.
2. The method of claim 1, wherein generating a first single-channel image corresponding to first light data and a second single-channel image corresponding to second light data according to the first light data and the second light data comprises:
determining a first direct current component and a first alternating current component corresponding to the first light ray data, and determining a second direct current component and a second alternating current component corresponding to the second light ray data;
the first single-channel image is determined according to each first direct current component and each first alternating current component, and the second single-channel image is determined according to each second direct current component and each second alternating current component.
3. The method of claim 2, wherein determining the first dc component and the first ac component corresponding to the first light data and determining the second dc component and the second ac component corresponding to the second light data comprises:
calculating standard deviation of peak-to-valley heights in every five pulse periods according to the first light ray data, taking a result obtained by each calculation as one first alternating current component, calculating an average value of peak heights in every five pulse periods, and taking the result obtained by each calculation as one first direct current component;
and calculating standard deviation of peak-to-valley heights in every five pulse periods according to the second light ray data, taking a result obtained by each calculation as one second alternating current component, calculating an average value of peak heights in every five pulse periods, and taking a result obtained by each calculation as one second direct current component.
4. The method of claim 2, wherein determining the first single channel image from the first dc component and the first ac component and determining the second single channel image from the second dc component and the second ac component comprises:
determining a first amplitude of the first dc component and a second amplitude of the first ac component, and determining a third amplitude of the second dc component and a fourth amplitude of the second ac component;
and generating the first single-channel image according to the first amplitude and the second amplitude, and generating the second single-channel image according to the third amplitude and the fourth amplitude.
5. The method of claim 4, wherein generating the first single channel image based on the first amplitude and the second amplitude and generating the second single channel image based on the third amplitude and the fourth amplitude comprises:
determining each first sequence according to the first amplitude and the second amplitude, and determining each second sequence according to the third amplitude and the fourth amplitude;
and generating the first single-channel image according to each first sequence, and generating the second single-channel image according to each second sequence.
6. The method of claim 1, wherein the step of obtaining the first light data of the first wavelength infrared light reflected by the skin and the second light data of the second wavelength infrared light reflected by the skin comprises:
acquiring third light data of the first wavelength infrared light reflected by the skin and fourth light data of the second wavelength infrared light reflected by the skin, acquiring actual blood oxygen concentration, and taking the actual blood oxygen concentration as a training label;
determining a corresponding third single-channel image according to the third light ray data, and determining a corresponding fourth single-channel image according to the fourth light ray data;
and taking the third single-channel image, the fourth single-channel image and the training label as training samples, and training an initial network model by adopting the training samples to obtain a network model.
7. The method of claim 6, wherein the initial network model includes a first convolution layer, a spatial attention layer, a second convolution layer, a third convolution layer, a full connection layer, and an output layer connected in sequence, and wherein training the initial network model using the training sample to obtain the network model includes:
inputting the third single-channel image and the fourth single-channel image into a first convolution layer to obtain a first feature, and inputting the first feature into a spatial attention layer to obtain a second feature;
inputting the second feature into the second convolution layer to obtain a third feature, inputting the third feature into the third convolution layer to obtain a fourth feature, inputting the fourth feature into a full connection layer, and determining a fifth feature;
inputting the fifth characteristic to an output layer to obtain various probability results, and determining a probability result with the maximum probability value as a detected blood oxygen concentration in the probability results;
and determining a classification loss function according to the error of the detected blood oxygen concentration and the actual blood oxygen concentration, and adjusting parameters and weights in the initial network model through the classification loss function to obtain the trained network model.
8. An oximetry device, the device comprising:
the acquisition module is used for acquiring first light data of the first wavelength infrared light reflected by the skin and acquiring second light data of the second wavelength infrared light reflected by the skin;
the generation module is used for generating a first single-channel image corresponding to the first light data and a second single-channel image corresponding to the second light data according to the first light data and the second light data;
the determining module is used for detecting the first single-channel image and the second single-channel image according to a trained network model, determining the blood oxygen concentration corresponding to the skin, and the network model is used for classifying and detecting the first single-channel image and the second single-channel image, wherein the network model is a convolutional neural network.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1 to 7 when executing the computer program.
10. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310465046.XA CN116195998B (en) | 2023-04-27 | 2023-04-27 | Blood oxygen detection method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310465046.XA CN116195998B (en) | 2023-04-27 | 2023-04-27 | Blood oxygen detection method and device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116195998A true CN116195998A (en) | 2023-06-02 |
| CN116195998B CN116195998B (en) | 2023-08-04 |
Family
ID=86508017
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310465046.XA Active CN116195998B (en) | 2023-04-27 | 2023-04-27 | Blood oxygen detection method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116195998B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040073142A1 (en) * | 2002-07-30 | 2004-04-15 | Hamamatsu Photonics K.K. | Mastication monitoring device |
| JP2013220312A (en) * | 2012-04-19 | 2013-10-28 | Advanced Medical Kk | Pulse oximeter |
| CN104224197A (en) * | 2014-09-24 | 2014-12-24 | 天津大学 | Calculation method for arterial oxygen saturation by using characteristic quantity of dual-wavelength photoelectric volume pulse wave |
| CN105030217A (en) * | 2015-06-09 | 2015-11-11 | 吉林大学 | Portable motion monitoring and detecting equipment and motion monitoring and detecting method |
| US20180214088A1 (en) * | 2016-09-24 | 2018-08-02 | Sanmina Corporation | System and method for obtaining health data using a neural network |
| CN108471989A (en) * | 2016-01-15 | 2018-08-31 | 皇家飞利浦有限公司 | The equipment, system and method for the photo-plethysmographic image of vital sign information for generating carrying object |
| CN112819093A (en) * | 2021-02-24 | 2021-05-18 | 浙江工业大学 | Man-machine asynchronous recognition method based on small data set and convolutional neural network |
| CN113288128A (en) * | 2021-04-23 | 2021-08-24 | 深圳阜时科技有限公司 | Blood oxygen detection device and electronic equipment |
| CN114847942A (en) * | 2022-04-22 | 2022-08-05 | 深圳市爱都科技有限公司 | Blood oxygen data processing method, related device and medium |
-
2023
- 2023-04-27 CN CN202310465046.XA patent/CN116195998B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040073142A1 (en) * | 2002-07-30 | 2004-04-15 | Hamamatsu Photonics K.K. | Mastication monitoring device |
| JP2013220312A (en) * | 2012-04-19 | 2013-10-28 | Advanced Medical Kk | Pulse oximeter |
| CN104224197A (en) * | 2014-09-24 | 2014-12-24 | 天津大学 | Calculation method for arterial oxygen saturation by using characteristic quantity of dual-wavelength photoelectric volume pulse wave |
| CN105030217A (en) * | 2015-06-09 | 2015-11-11 | 吉林大学 | Portable motion monitoring and detecting equipment and motion monitoring and detecting method |
| CN108471989A (en) * | 2016-01-15 | 2018-08-31 | 皇家飞利浦有限公司 | The equipment, system and method for the photo-plethysmographic image of vital sign information for generating carrying object |
| US20180214088A1 (en) * | 2016-09-24 | 2018-08-02 | Sanmina Corporation | System and method for obtaining health data using a neural network |
| CN112819093A (en) * | 2021-02-24 | 2021-05-18 | 浙江工业大学 | Man-machine asynchronous recognition method based on small data set and convolutional neural network |
| CN113288128A (en) * | 2021-04-23 | 2021-08-24 | 深圳阜时科技有限公司 | Blood oxygen detection device and electronic equipment |
| CN114847942A (en) * | 2022-04-22 | 2022-08-05 | 深圳市爱都科技有限公司 | Blood oxygen data processing method, related device and medium |
Non-Patent Citations (1)
| Title |
|---|
| WAIZ AHMED: "Development of a Photoplethysmography Signal Processing Method for Oxygen Measurement Concentration", 5TH INTERNATIONAL CONFERENCE ON COMPUTER & COMMUNICATION ENGINEERING, pages 40 - 43 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116195998B (en) | 2023-08-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sopic et al. | Real-time event-driven classification technique for early detection and prevention of myocardial infarction on wearable systems | |
| JP5565876B2 (en) | System and method for identifying matching portions of a signal | |
| Djelouat et al. | System-on-chip solution for patients biometric: A compressive sensing-based approach | |
| US20210106241A1 (en) | Heart Rate Prediction from a Photoplethysmogram | |
| Reddy et al. | On-device integrated PPG quality assessment and sensor disconnection/saturation detection system for IoT health monitoring | |
| Al Disi et al. | ECG signal reconstruction on the IoT-gateway and efficacy of compressive sensing under real-time constraints | |
| CN105212919B (en) | physiological signal analysis system | |
| CN108573227B (en) | Electrocardiogram data quality evaluation method and device | |
| KR20160094265A (en) | Computing device and method for analyzing biophysiological rate features thereof | |
| CN102551699A (en) | Non-invasive cardiac output determination system | |
| CN107374646A (en) | Blood sugar detection apparatus and method, electronic installation based near infrared technology | |
| Shang | Multiscale Tsallis permutation entropy analysis for complex physiological time series | |
| Oletic et al. | System-level power consumption analysis of the wearable asthmatic wheeze quantification | |
| CN110363090A (en) | Intelligent heart disease detection method, device and computer readable storage medium | |
| CN116311539A (en) | Sleep motion capturing method, device, equipment and storage medium based on millimeter waves | |
| CN114403904B (en) | Device for determining muscle state based on electromyographic signals and muscle blood oxygen saturation | |
| CN114052693B (en) | Heart rate analysis method, device and equipment | |
| CN116195998B (en) | Blood oxygen detection method and device, computer equipment and storage medium | |
| Mortensen et al. | Multi-class stress detection through heart rate variability: A deep neural network based study | |
| US11534112B2 (en) | AI based method for determining oxygen saturation levels | |
| CN116327133A (en) | Multi-physiological index detection method, device and related equipment | |
| CN112598033B (en) | Physiological signal processing method, device, equipment and storage medium | |
| Saini et al. | Edge of Medical Things Implementation for Deep Learning-Based Cognitive Task Recognition | |
| Sivaranjini et al. | Fast quality-aware amdf based pulse rate estimation from compressed ppg measurements for wearable vital signs monitor | |
| JP6706996B2 (en) | Biological signal processing device, abnormality determination method and program |
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 | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: No. 205, Building B1, Huigu Science and Technology Industrial Park, No. 336 Bachelor Road, Bachelor Street, Yuelu District, Changsha City, Hunan Province, 410000 Patentee after: Wisdom Eye Technology Co.,Ltd. Country or region after: China Address before: 410000 building 14, phase I, Changsha Zhongdian Software Park, No. 39, Jianshan Road, high tech Development Zone, Changsha City, Hunan Province Patentee before: Wisdom Eye Technology Co.,Ltd. Country or region before: China |