WO2017131276A1

WO2017131276A1 - Infant safety monitoring system using smartphone-based wearable device

Info

Publication number: WO2017131276A1
Application number: PCT/KR2016/001656
Authority: WO
Inventors: 서지윤; 노윤홍; 박준모; 정도운
Original assignee: 동서대학교 산학협력단
Priority date: 2016-01-29
Filing date: 2016-02-18
Publication date: 2017-08-03
Also published as: KR101755594B1

Abstract

The present invention relates to an infant safety monitoring system using a smartphone-based wearable device, the infant safety monitoring system comprising: an SpO₂ sensor, which is an integrated type of IR/Red LED for pulse wave measurement and a photo diode, and in which a light-emitting element (IR/Red LED) and a light-receiving sensor (photo diode) are packaged for reflective measurement; a constant-current LED switching unit for driving the LED of the SpO₂ sensor; a Bluetooth module for Bluetooth communication with a smartphone; an MCU for processing the detection of SpO₂ and a pulse rate, the detection of a posture, and communication from a measurement signal obtained by the SpO₂ sensor; and a three-axial acceleration sensor obtaining a rotation angle by using a gravitational acceleration direction reflected in each axis, and then detecting only a supine posture and a prone posture as the postures, wherein firmware of the MCU extracts AC/DC components of a signal by applying a digital LPF in order to cancel noise in an SpO₂ and pulse rate detection process and a QV algorithm in order to detect a baseline, detects a rotation angle and motion of the acceleration sensor with respect to the postures, and transmits the SpO₂, the pulse rate, and posture information to the smartphone through the Bluetooth module.

Description

Infant safety monitoring system using smartphone-based wearable device

The present invention relates to an infant safety monitoring system using a smartphone-based wearable device, and more specifically, to prevent sudden death of infants and infants through the monitoring application (band) of the wearable device in the form of a band and a smartphone. The present invention relates to an infant safety monitoring system using a smartphone-based wearable device for providing posture monitoring.

Various researches for the safety management of infants and young children are conducted at home or in infant care facilities.

In the prior art, an infant safety management system for attaching or paying a tag storing personal information and guardian information to infants, and periodically notifying the guardian of whether the infant is present in a protected area through a reader unit has been studied and widely used. have.

However, the infant safety management system using the conventional tag is a method in which the guardian simply transmits only whether the infant is located in the protected area. Therefore, the guardian cannot accurately recognize the current state of the infant and the infant's infant. There is a limit to not being aware of an accident and not being able to take immediate action against it.

The present invention is to solve the above problems, infant safety monitoring using a smart phone-based wearable device to provide oxygen saturation and posture monitoring to prevent sudden death of infants through the band-type wearable device and monitoring application It is to provide a system.

In addition, the present invention using a band-type wearable device to wear on the thighs (thighs) of infants in real time to detect oxygen saturation and posture, smart phone for automatically transmitting wireless detection information to the smartphone via Bluetooth To provide an infant safety monitoring system using a wearable device based on the.

In addition, the present invention utilizes a smart phone monitoring application to display the oxygen saturation and posture received from the wearable device in real time, infant safety using a smart phone-based wearable device to provide an alarm when determining the emergency situation It is to provide a monitoring system.

However, the objects of the present invention are not limited to the above-mentioned objects, and other objects not mentioned will be clearly understood by those skilled in the art from the following description.

Infant safety monitoring system using a smart phone-based wearable device according to an embodiment of the present invention, the band-type wearable device 100 and the smart phone 200, including the smart phone 200 to achieve the above object Infant safety monitoring system using wearable device, IR / Red LED for pulse wave measurement, photo diode integrated with IR / Red LED and photo-diode sensor packaged for reflective measurement SpO2 sensor 110; A constant current LED switching unit 120 for driving the LED of the SpO2 sensor 110; Bluetooth module 140 for Bluetooth communication with the smart phone 200; And a MCU 160 which processes SpO2 and pulse rate detection, posture detection, and communication performance from the measurement signal by the SpO2 sensor 110. And a three-axis acceleration sensor 170 which detects only a supine and a prone posture immediately lying in a posture after obtaining a rotation angle using the gravitational acceleration direction projected on each axis. The firmware of the MCU 160 extracts AC / DC components of a signal by applying a digital LPF for noise reduction and a QV algorithm for baseline detection in spO2 and pulse rate detection, and 3-axis acceleration for posture. It detects the rotation angle and the movement of the sensor 170, and transmits the SpO2, pulse rate, posture information to the smart phone 20 via the Bluetooth module 140.

Infant safety monitoring system using a smart phone-based wearable device according to an embodiment of the present invention, through the band-type wearable device and monitoring applications to provide real-time oxygen saturation and posture monitoring to prevent sudden death of infants It works.

In addition, the infant safety monitoring system using a smart phone-based wearable device according to another embodiment of the present invention, by wearing a band-type wearable device on the thigh (thigh) of the infant, it detects the oxygen saturation and posture in real time In addition, by automatically transmitting wirelessly detected information to a smartphone via Bluetooth, it minimizes inconvenience for infants and provides the effect that the guardian can easily check the infant's status information.

1 is a schematic diagram showing an infant safety monitoring system using a smart phone-based wearable device according to an embodiment of the present invention.

FIG. 2 is a block diagram illustrating hardware for the band-type wearable device of FIG. 1. FIG.

Figure 3 is a schematic diagram showing the configuration of the SpO2 sensor of the band-type wearable device of FIG.

4 is a diagram illustrating a pulse wave measuring circuit (except for a communication and a power supply) of the band-type wearable device of FIG. 1;

5 is a diagram illustrating a circuit of a posture measuring unit in the band-type wearable device of FIG. 1.

FIG. 6 is a circuit diagram illustrating an MCU and a peripheral circuit of the band-type wearable device of FIG. 1. FIG.

FIG. 7 is a block diagram illustrating a Bluetooth module of the band-type wearable device of FIG. 1. FIG.

FIG. 8 is a circuit diagram illustrating a Bluetooth module of the wearable device in the band form of FIG. 1. FIG.

9 is a circuit diagram illustrating a charging circuit in a power supply unit of the band-type wearable device of FIG. 1.

FIG. 10 is a diagram illustrating a SpO2 and pulse rate calculation process flow chart provided in firmware of an MCU of the band-type wearable device of FIG. 1. FIG.

FIG. 11 is a flowchart illustrating a method of determining whether a calculated SpO2 value and a pulse rate value provided by firmware of an MCU among the band-type wearable devices of FIG. 1 are valid; FIG.

FIG. 12 is a diagram illustrating a process of performing a double weighted QV algorithm provided by firmware of an MCU among wearable devices having a band form of FIG. 1.

FIG. 13 is a flowchart illustrating a posture calculation process provided by firmware of an MCU of the band-type wearable device of FIG. 1. FIG.

FIG. 14 illustrates the X, Y, and Z axes of the Cartesian coordinate system in the case where the 3-axis acceleration sensor rotates among the band-type wearable devices of FIG. 1 when the acceleration of gravity of 1 g changes φ and θ by 0 to 2π. A 3D representation of the magnitude (SMA, | X | + | Y | + | Z |, excluding integrals) of the magnitude projected to the absolute value plus the absolute value.

15 is a view showing a case of the wearable device in the form of a band of FIG.

FIG. 16 is a diagram illustrating a user interface screen for acquiring data through Bluetooth communication with a band-type wearable device of FIG. 1 and observing a calculation result of firmware in real time. FIG.

FIG. 17 is a diagram illustrating a communication format of a wearable device in band form of FIG. 1. FIG.

18 to 21 are views for explaining a measurement experiment for the purpose of determining the skin portion that can be easily attached and measured to the band-type wearable device of FIG.

FIG. 22 is a reference diagram showing a user interface screen implemented by an application on the smartphone of FIG. 1; FIG.

Hereinafter, the detailed description of the preferred embodiment of the present invention will be described with reference to the accompanying drawings. In the following description of the present invention, detailed descriptions of well-known functions or configurations will be omitted when it is deemed that they may unnecessarily obscure the subject matter of the present invention.

In the present specification, when one component 'transmits' data or a signal to another component, the component may directly transmit the data or signal to another component, and through at least one other component. This means that data or signals can be transmitted to other components.

1 is a schematic diagram showing a baby safety monitoring system using a smart phone-based wearable device according to an embodiment of the present invention. Referring to FIG. 1, an infant safety monitoring system using a smartphone-based wearable device includes a wearable device 100 and a smartphone 200 in a band form.

FIG. 2 is a block diagram illustrating hardware of the wearable device 100 in the band form of FIG. 1. Referring to FIG. 2, the band-type wearable device 100 includes an IR / Red LED, a photodiode integrated SpO2 sensor 110, and a constant current LED switching unit for driving LEDs of the SpO2 sensor 110 for pulse wave measurement. 120, the I / V converter 130 for converting the sensor output current into a voltage, the Bluetooth module 140 for Bluetooth communication, the power supply unit 150, and the MCU 160 are configured as main components.

In addition, the band-type wearable device 100 uses the 3-axis acceleration sensor 170 to measure the attitude, the user interface 180 is a power on / off switch 181, the connection state of the Bluetooth communication and battery charging It consists of

LEDs

182 and 183 indicating a state. The power source is driven by the battery 151, and the battery can be charged through the micro USB connector 152.

Meanwhile, the pulse wave measuring unit of the band-type wearable device 100 according to the present invention will be described. As the pulse wave measuring unit, an SpO2 sensor (IR / Red LED, photo diode integrated type) 110 is used. The pulse wave sensor uses an integrated sensor in which light-emitting (IR / Red LED) elements and photo-diode sensors are packaged for reflective measurements.

3 is a diagram showing that the wavelengths IR 905 nm and Red 660 nm are used as the SpO2 sensor 110. That is, as shown in FIG. 2, the constant current LED switching unit 120 supplies about 20 mA of current through the constant current control to the light emitting device, and controls switching in three stages (IR LED On → Red LED On → All LED Off). do. The duty ratio of each stage is 1/3 and the switching frequency is about 300 Hz. The light receiving sensor of the SpO2 sensor 110 generates a current signal in proportion to the amount of incident light. The current output signal of the SpO2 sensor 110 is converted into a voltage signal through an I / V converter 130 and then transferred to the 16-bit ADC 161 of the MCU 160. The ADC 161 is driven at a sampling rate of 50 kSPS.

4 is a diagram illustrating a pulse wave measuring unit circuit (except the communication and power supply unit) 100a of the band-type wearable device 100 of FIG. 1. Referring to FIG. 4, a circuit diagram related to the switching 120, the constant current LED switching unit 120, and the current-voltage converter 130 of the SpO2 sensor 110 is performed. The circuit operation is performed through U1 and Q1 is performed. It controls the current of Red / IR LED to be constant current. The amount of current is designed to be set by the value of MCU's DAC (DAC_CH2 signal line), and it is set to flow about 20mA. After performing I / V conversion using OP Amp (U3B), the output is transferred to the 16-bit ADC 161 of the MCU 160. The amplification factor of the I / V conversion is 75 kV / I. It is designed to reduce the offset through R5 to increase the degree of amplification.

FIG. 5 is a diagram illustrating a circuit of the posture measuring unit 100b of the wearable device 100 in the band form of FIG. 1. Referring to FIG. 5, the posture in the posture measuring unit 110b is calculated after obtaining a rotation angle by using the gravity acceleration direction projected on each axis of the 3-axis acceleration sensor 170. Posture detects only supine and prone positions. The three-axis acceleration sensor 170 is expressed at a resolution of 256 levels per 1g. The communication method uses I2C.

FIG. 6 is a circuit diagram illustrating the MCU 160 and the peripheral circuit of the band-type wearable device 100 of FIG. 1. Referring to FIG. 6, the MCU 160 includes a 16-bit ADC 161 for miniaturization. The MCU 160 performs not only communication control but also calculation functions of posture, SpO2, and HR. The calculation of SpO2 value should detect all AC / DC values of IR / Red signal, DC component is very large compared to AC component and IR signal is larger than Red signal. In the device using the 16-bit ADC 161, the AC component of the red signal appears differently depending on the person, but has a value of about 200 levels when measured by the finger. If a 12-bit quantization rate ADC is applied, the resolution is reduced to about 13 levels, which causes a problem in SpO2 operation. Therefore, the 16-bit ADC 161 is adopted.

FIG. 7 is a module block diagram of the Bluetooth module 140 of the wearable device 100 in the band form of FIG. 1, and FIG. 8 illustrates the Bluetooth module 140 of the wearable device 100 in the band form of FIG. 1. A circuit diagram is shown. 7 and 8, the Bluetooth module 140 is adopted for communication with the smartphone 200, and is a small module incorporating an FW, an RF unit, an antenna, etc., and has a size of 10 mm × 17 mm × 2 mm. . The Bluetooth module 140 communicates with the MCU 160 at UART 961200 bps and has a built-in V2.0 stack for Bluetooth communication.

FIG. 9 is a circuit diagram illustrating the charging circuit 150a of the power supply unit 150 of the band-type wearable device 100 of FIG. 1. Referring to FIG. 9, the power supply unit 150 includes a charging unit and a constant voltage power supply for charging a lithium polymer battery. The constant voltage for driving the Bluetooth module 140, the MCU 160, etc. in the power supply unit 150 is implemented using the TCR2EF30, and the external power supply uses the micro USB connector 152 in consideration of ease of use and miniaturization. Adopt. The charging current of the power supply unit 150 is set to about 200 mA, and the battery 151 is designed to be used for 4 hours or more with a capacity of 430 mAh.

Meanwhile, the user interface 180 includes a power on / off switch 181 and two

LEDs

182 and 183 for transmitting information to the user. Power On / Off switch (181) is turned on when pressed for more than 1 second, and operates to turn off when pressed for more than 1 second. The two

LEDs

182 and 183 include a charge LED 183 indicating a charging state when the battery 151 is charged and a connect LED 182 indicating a Bluetooth connection state. The charging LED 183 repeats 0.2 seconds on and 1 second off when charging, and continues to turn on when charging is completed, and 0.1 seconds on and 0.1 seconds off when a charging error occurs. The connection LED 182 repeats 0.2 seconds on and 1 second off in the connection standby state, and continues to be on when connected.

In addition, referring to the firmware of the MCU 160, the firmware programming and debugging of the MCU 160 is performed by SWO communication through a JLink emulator, and the software is developed through IAR Embedded Workbench for ARM V6.6.

The firmware of the MCU 160 basically processes SpO2 and pulse rate detection, posture detection, communication, and user interface functions. The firmware of the MCU 160 extracts AC / DC components of a signal by applying a digital LPF and a QV algorithm for baseline detection to remove noise during SpO2 and pulse rate detection.

Here, the pulse rate is detected by using the maximum detection method of the window method, the posture is inferred by detecting the rotation angle and the movement of the three-axis acceleration sensor 170. Since the Bluetooth module 140 is controlled through the UART, communication with the PC is implemented to control and perform the UART port through DMA.

On the other hand, SpO2 calculation relations are defined as shown in Equation 1 below. The AC and DC components must be extracted for the signal value when the Red / IR LED is lit. In Equation 1, A and B are constants, and the values suggested in TI's Application Note, that is, A is 110 and B is 25 are applied.

Equation 1

10 is a diagram illustrating a flowchart of SpO2 and pulse rate calculation processing provided by the firmware of the MCU 160. Referring to FIG. 10, the SpO2 sensor 100 switches the LEDs in the order of IR LED on / Red LED on / all LED off, thereby outputting the output signal of the generated SpO2 sensor at a sampling rate of 50kSPS through the built-in ADC 161. Acquire data (S11). In order to remove noise, a moving average is performed at a size of 22 samples, and down sampling is performed at 4.5kSPS (S12).

Since the LEDs of the SpO2 sensor 110 are switched to about 300 Hz, 300SPS for three states: IR light is on, Red LED is on, all LEDs are off, in synchronization with the switching cycle. The signal waveform of is extracted (S13). LPF is passed through each signal to further remove high frequency noise (S14).

The pulse rate is extracted after extracting the change amount component (AC component) of 300SPS data and performing peak detection, and calculating the time (sample number) between the maximum values. When the time between maximum values is ΔT and the pulse rate is HR, the pulse rate is calculated as shown in Equation 2 below. At this time, if only the pulse wave waveform is to be observed, a sampling rate of 50 SPS is also suitable. However, if the sampling rate is low, the resolution of HR decreases due to a decrease in the resolution of ΔT.

Equation 2

The AC component for pulse rate calculation is extracted by removing the baseline change component from the 300SPS signal (16c). Baseline is detected by applying the QV algorithm. The QV algorithm, which processes 300SPS and 5 seconds, is implemented through a square matrix, and requires a large amount of computation to obtain a result value, thus exceeding the processing capability of the MCU 160 applied to the present invention. In order to apply the QV algorithm, after downsampling 300SPS data to 30SPS data (S15), baseline is detected through the QV algorithm, and linear interpolation is performed again at 300SPS (S16a). Since a time difference occurs between the baseline (S16b) obtained through the QV algorithm and the linear interpolation and the signal passed through the LPF, the time delay component is compensated for through the time delay circular queue (S16e). Since the baseline result by the QV algorithm has a problem that a discontinuity point occurs in each calculation section, a weighted dual QV algorithm is proposed to solve the problem.

On the other hand, SpO2 value is calculated through IR and Red signals of 30SPS. In Equation 1, which calculates the SpO2 value, the baseline to which the QV algorithm is applied is calculated as the DC component, and the value obtained by subtracting the baseline value from the signal value is the AC component (S17a and S17b).

11 is a flowchart illustrating a method of determining whether a calculated SpO2 value and a pulse rate value provided by the firmware of the MCU 160 are valid. Referring to FIG. 11, SpO2 is normally measured only when the SpO2 sensor 110 is close to the measurement target. If the distance between the SpO2 sensor 110 and the measurement object is far, it is meaningless to calculate SpO2 and the pulse rate. Therefore, a method of determining whether the calculated SpO2 value and the pulse rate value are valid is proposed. Immediately after the power is turned on, the sensor waits for a certain time until the QV square matrix becomes stable and then switches to the Sensor Off state.

When the DC signal of the IR is greater than or equal to the threshold, it is determined that the SpO2 sensor 110 is normally connected to the sensor On state. Two thresholds (Thres_IR1 and Thres_IR2) are used, and hysteresis is applied to prevent rapid state changes even with slight DC value changes.

After entering the Sensor On state, if the pulse rate is detected more than 7 times, it returns to the normal measurement state (Normal). If the AC value of the IR is too small, it is assumed that the measurement is not performed normally and the state is changed to the Error_No Signal state. Do not display SpO2 or pulse rate in any condition other than normal.

12 is a diagram for describing a procedure of performing a double weighted QV algorithm provided by the firmware of the MCU 160. Referring to FIG. 12, a double weighted QV algorithm is proposed to solve a problem in which a discontinuity point appears at the end of a section in the QV algorithm. The QV algorithm divides and processes data intervals by square matrix intervals. In FIG. 12, a section is processed for each 150 samples, and a portion indicated by A or B is one section. Section B is a section in which section A is processed by delaying 75 samples, and the size of each section is 150. The result of processing the signal a with the baseline variation as the A and dividing the interval as B and B as the result of applying the QV algorithm is shown in c. In b and c, the straight part represents the center line and the value is zero. Therefore, the QV algorithm shows that the baseline fluctuation component included in the signal a is removed very well, and thus the AC component is well represented. For reference, the line passing near the vertical center of the a signal is the baseline calculated by subtracting the b or c signal by applying the QV algorithm by dividing the A signal into the A and B sections. However, the QV algorithm has the potential to represent discontinuities at the end of each interval. Circle 1 of the end points calculated by dividing by section A in the figure (here '1' refers to '①', the same below). Looking at the intersections of the signals, we can see that there are discontinuities. For reference, discontinuities do not appear well in a section where signal baseline fluctuations are small.

The double weighted QV is described as follows. Calculate the QV output in each section where one section is delayed by one-half section, and then multiply the QV output with the weight of a triangular wave of 0 ~ 1 magnitude repeated in each section. The triangular wave weights have a form of multiplying the end point of the interval by 0 and multiplying the center by 1, and the result of multiplying d and e by the shape of the weight and the b and c signals, which are the QV outputs, and the weight. The results of adding the results of d and e to each other are shown in the f row. Since the weight is zero at the end of the interval, the effect of discontinuities at the end of the interval is eliminated. The signal which added the signal of d and e multiplied by the weight is shown to f. Looking at f, it can be seen that it is not affected by the signal value at the discontinuous point of the section. It can be seen that there is no discontinuity at the intersection of the circular arc 1 and the vertical arc 2 and the f signal where the discontinuity point occurred.

When the double weighted QV is processed, a sudden increase of the value does not occur even when the derivative is performed, compared to when there is a discontinuity point, so it can be used for the detection of the maximum value through the derivative.

13 is a flowchart illustrating a posture calculation process provided by the firmware of the MCU 160. Referring to FIG. 13, the posture is implemented to express four states (movement, leg up / down, straight down, upside down) every 1 second using the 3-axis acceleration sensor 170. The device attachment portion is above the thigh, and when the device is attached, the Y axis of the 3-axis acceleration sensor 170 indicates the long axis direction of the thigh. If the legs are placed horizontally on the floor, the Y-axis is 0, the X-axis is 0, and the Z-axis is 1g. Therefore, if the legs are rotated in the long axis direction (the axis perpendicular to the direction of gravity) while the legs are horizontal to the floor, the rotation angle can be known according to the values of the X and Z axes. There is no big problem in calculating the rotation angle even if the legs are raised or lowered horizontally on the floor. If the leg is raised or lowered horizontally on the floor, the value of the Y-axis increases from 0 to + or decreases to-, and the threshold is detected to detect the entry into the uncalculated state.

Referring to the flowchart of FIG. 13, after calculating the maximum, minimum, and average values of each axis every second, the maximum acceleration change amount (maximum value-minimum value) of the 3-axis acceleration sensor 170 is calculated, and then the maximum acceleration change amount is a threshold value (Thres_Move). ) Is determined to be in a moving state. If there is no movement and the leg is not raised, it is determined whether the person is lying directly or lying down through the rotation angle. On the block diagram, it is determined based on 180 degrees, but the actual code has a hysteresis of 10 degrees based on the current posture so that the display of the posture is stable even when moving slightly around 0 and 180 degrees.

For reference, the reason why the SMA is not selected for detecting the motion state is that the SMA value is approximately 1.73 times different depending on the direction of the three-axis acceleration sensor 170 even without the motion. When the three-axis acceleration sensor 170 rotates, the magnitude of the gravity acceleration projected in the X, Y, and Z axis directions of the Cartesian coordinate system changes.

14 is an absolute value of the magnitude projected on the X, Y, and Z axes of the Cartesian coordinate system when the gravitational acceleration of 1 g changes φ and θ by 0 to 2π in the spherical coordinate system assuming that the 3-axis acceleration sensor 170 rotates. 3D shows the result (SMA, | X | + | Y | + | Z |, excluding the integral) after taking the? That is, FIG. 14 shows a change in the SMA value when φ and θ are changed by 0 to 2π in the spherical coordinate system.

The minimum value is 1 and the maximum value is about 1.73. If the threshold value is set to 1.75 for motion detection, the motion in the direction of maximum value should be moved by 0.02 and 0.75 in the direction of minimum value. Due to the large difference in sensitivity that detects motion depending on the direction in which the sensor is placed, SMA is not adopted for motion detection.

15 is a view showing a case 1 of the wearable device 100 in the form of a band. Referring to FIG. 15, the case 1 includes a PCB and a battery, and is configured to be easily fixed to the thigh. It is designed to be attached to the thigh through the band at the thigh. The SpO2 sensor 110 is in contact with the skin area. The size is 54 mm x 67 mm x 14 mm.

On the other hand, all operations related to the algorithm are performed in the firmware of the wearable device 100 in the form of a band. However, since the real-time observation of the internal calculation results during the firmware development is difficult, the PC SW is developed to check the communication between the band-type wearable device 100 and the PC and to easily observe the measurement signal. The development tool uses Visual C ++. The software screen is shown in FIG. 16, and acquires data through Bluetooth communication with the band-type wearable device 100 to observe the calculation result of the firmware in real time.

16 indicates an SpO2 measurement state, a SpO2 measurement value and a heart rate, and posture information to be displayed on the smartphone 200. The measurement state represents five states of "Pulse Rate / SpO2 Computable State Transition State Diagram of FIG. 11", and the SpO2 measurement value and heart rate are expressed as an integer. In the case of posture, four states of “posture calculation block diagram of FIG. 13” are displayed.

The process of extracting the signal from the ADC data of 4500 SPS with the IR / Red signal value of 300 SPS is shown in

reference numerals

10 and 11 of FIG. 16. 16 indicates ADC data of 4500 SPS which has been averaged by moving average, and reference numeral 11 of FIG. 16 indicates a position at which IR, Red, and all LEDs of 300 SPS are extracted from the reference numeral 10 of FIG. Display. This is where the low triangle extracts the IR signal, the middle height triangle is red, and the high triangle is off.

The obtained 300 SPS IR / Red SPS signals are shown in

circles

2 and 4 of FIG. 16, and the LPF signals are shown in

circles

3 and 5 of FIG. 16. After down sampling the LPF signal to 30 SPS, the result of applying the double weighted QV algorithm can be observed in

symbols

6 and 7 of FIG. 16. SpO2 measurement state calculated by the algorithm is shown in circle 8 of FIG. 16, SpO2 measurement value is circle 15 of FIG. 16, and information on posture is circle 17 of FIG. 16. Information on the rotation angle used in the posture calculation is shown in circle 16 of FIG. The calculated heart rate can be observed in symbol 9 of FIG. 16.

The output values for the X, Y, and Z axes of the 3-axis acceleration sensor 170 used as an input when calculating the attitude are shown in the

symbols

12, 13, and 14 of FIG. 16, and the 3-axis acceleration sensor when the movement occurs or rotates. (170) Make it easy to observe the change in output.

17 shows a diagram illustrating a communication format of the wearable device 100 in the form of a band. Referring to FIG. 17, packets communicating with the Bluetooth module 140 of the band-type wearable device 100 and the PC or the smartphone 200 are arranged. The sampling rate column is the sampling rate described in FIG. The PC SW column indicates the number displayed on the “PC SW screen of FIG. 16” and means information to be displayed on the graph or edit window in the PC SW. Data communication is transmitted in packet units, and packets are transmitted 30 times per second. For example, in the table IR300 [10], 10 samples are sent 30 times per second, so 300 samples are sent in 1 second.

On the other hand, the measurement experiment was carried out for the purpose of determining the skin portion that can be easily attached to the band-type wearable device 100 and can be measured. The measurement was performed only in one 2 year old child because it was not easy to secure the subject. Due to the resolution of the laptop, only part of the screen was acquired. 18-21 show some waveforms of the IR / Red signal in the double-weighted QV graph and the stable SpO2 values. That is, FIG. 18 shows the measurement at the heel, FIG. 19 shows the measurement inside the calf, and FIG. 20 shows the measurement above the thigh.

As a result of these experiments, it is judged that the heel is the most appropriate if the posture detection is not considered. If the posture detection is considered, it is appropriate to measure at the front of the thigh which is highly related to the rotation of the body, which is a criterion for determining the posture.

21 and 22 are reference diagrams illustrating a user interface screen implemented by an application of the smartphone 200.

The invention can also be embodied as computer readable code on a computer readable recording medium. Computer-readable recording media include all kinds of recording devices that store data that can be read by a computer system.

Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disks, optical data storage devices, and the like, which are also implemented in the form of carrier waves (eg, transmission over the Internet). It also includes.

The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion. And functional programs, codes and code segments for implementing the present invention can be easily inferred by programmers in the art to which the present invention belongs.

As described above, the present specification and drawings have been described with respect to preferred embodiments of the present invention, although specific terms are used, it is only used in a general sense to easily explain the technical contents of the present invention and to help the understanding of the present invention. It is not intended to limit the scope of the present invention. It will be apparent to those skilled in the art that other modifications based on the technical idea of the present invention can be carried out in addition to the embodiments disclosed herein.

Infant safety monitoring system using a smart phone-based wearable device according to an embodiment of the present invention includes a band-type wearable device 100 and the smartphone 200, the band-type wearable device 100 IR / Red LED, SpO2 sensor 110 integrated with photo diode, constant current LED switching unit 120 for driving LED of SpO2 sensor 110 for pulse wave measurement, and I / V for changing sensor output current to voltage The converter 130, the Bluetooth module 140 for Bluetooth communication, the power supply unit 150, and the MCU 160 are configured as main components. In addition, the band-type wearable device 100 uses the 3-axis acceleration sensor 170 to measure the attitude, the user interface 180 is a power on / off switch 181, the connection state of the Bluetooth communication and battery charging It consists of

LEDs

At this time, the firmware of the MCU 160, for SpO2 operation, after extracting the AC component and DC component for the signal value when the Red / IR LED is turned on,

It is preferable to apply A to 110 and B to 25 in the calculation.

In addition, it is preferable that the SpO2 sensor 110 uses wavelength IR905 nm and red 660 nm.

In addition, the constant current LED switching unit 120 supplies about 20 mA of current to the light emitting (IR / Red LED) device through constant current control, and performs three steps (IR LED On → Red LED On → All LED Off) in one cycle. Switching is performed, and the duty ratio of each step is preferably controlled to 1/3 and the switching frequency to 300 Hz.

In addition, the photo-diode sensor of the SpO2 sensor 110 generates a current signal in proportion to the amount of incident light, and the current output signal of the SpO2 sensor 110 is a current of the band-type wearable device 100. It is converted into a voltage signal through the voltage converter (I / V converter) 130 and then transferred to the 16-bit ADC 161 of the MCU 160, the ADC 161 is preferably driven at a sampling rate of 50kSPS Do.

Infant safety monitoring system using a smart phone-based wearable device according to an embodiment of the present invention, by using the monitoring application of the smart phone display the oxygen saturation and posture received from the wearable device in real time, when judging emergency By providing an alarm, the guardian can conveniently use it even if he or she leaves the baby for a while.

Claims

In the infant safety monitoring system using a smartphone-based wearable device, including a band-type wearable device 100 and a smartphone 200,

An IR / Red LED for pulse wave measurement, a SpO2 sensor 110 in which a light-emitting (IR / Red LED) element and a photo-diode sensor are integrally packaged for reflection measurement; A constant current LED switching unit 120 for driving the LED of the SpO2 sensor 110; Bluetooth module 140 for Bluetooth communication with the smart phone 200; And a MCU 160 which processes SpO2 and pulse rate detection, posture detection, and communication performance from the measurement signal by the SpO2 sensor 110. And a three-axis acceleration sensor 170 which detects only a supine and a prone posture immediately lying in a posture after obtaining a rotation angle using the gravitational acceleration direction projected on each axis. Including;

The firmware of the MCU 160 extracts AC / DC components of a signal by applying a digital LPF and a QV algorithm for baseline detection to remove noise during SpO2 and pulse rate detection, and a 3-axis acceleration sensor 170 for posture. Infant safety monitoring system using a smart phone-based wearable device, characterized in that for detecting the rotation angle and movement, and transmits the SpO2, pulse rate, posture information to the smart phone 20 through the Bluetooth module 140.
The method of claim 1, wherein the firmware of the MCU 160, for SpO2 operation,

After extracting the AC and DC components for the signal values when the Red / IR LED is on,

Calculated by, A is 110, B is a baby safety monitoring system using a wearable device based on a smartphone, characterized in that 25 applied.
The method according to claim 1, SpO2 sensor 110,

Infant safety monitoring system using a smartphone-based wearable device, characterized in that the use of wavelengths IR 905 nm, Red 660 nm.
The method according to claim 1, the constant current LED switching unit 120,

The light emitting (IR / Red LED) device flows about 20mA of current through constant current control, and switching is performed in three stages (IR LED On → Red LED On → All LED Off) in one cycle. Infant safety monitoring system using a smartphone-based wearable device, characterized in that the switching frequency is controlled to 300Hz.
The method according to claim 4,

The photo-diode sensor of the SpO2 sensor 110 generates a current signal in proportion to the amount of incident light, and the current output signal of the SpO2 sensor 110 is a current-voltage of the band-type wearable device 100. After converting into a voltage signal through the converter (I / V Converter 130) to be transmitted to the 16-bit ADC (161) of the MCU (160),

ADC (161), the infant safety monitoring system using a smart phone-based wearable device, characterized in that driven at a sampling rate of 50kSPS.