WO2023173386A1 - Newborn/non-contact physiological sign monitoring method and system - Google Patents

Abstract

A newborn/non-contact physiological sign monitoring method and system. Fluctuation data of a target part of a target object is acquired by a radar sensor. Such a non-contact mode improves detection accuracy. Moreover, an adjusting device is further provided. The measurement position and/or the viewing angle of the radar sensor are adjusted by the adjusting device, facilitating the radar sensor to be aligned with the target part of the target object, thereby further improving the detection accuracy, and also improving the monitoring accuracy of the target object.

Description

A newborn/non-contact physiological sign monitoring method and system Technical field

The invention relates to the medical field, and in particular to a newborn/non-contact physiological sign monitoring method and system.

Background technique

The neonatology department has clear regulations that require observation and recording of respiratory parameters of neonatal patients every 2 hours. At present, respiratory testing in neonatal departments is mainly based on impedance breathing or manual counting, and some ventilators are also equipped with air bag breathing.

Newborns are placed in incubators or incubators, surrounded on all sides and above. Using the traditional impedance method to measure respiration requires bonding respiratory leads, which is inconvenient to operate. Moreover, the impedance method measures respiration when the patient is crying or moving, etc. It is easy to be disturbed in the scene and the measurement is inaccurate.

The air bag method measures respiration by bonding an air bag to the patient's abdomen or chest. The volume of the air bag is compressed when the patient breathes, and the patient's respiratory rate is calculated by measuring the pressure waveform in the air bag. In actual clinical practice, using the air bag method to measure respiration will encounter a series of problems: (1) For newborn children, the fetal fat on the body has not been fully absorbed, so it is difficult to fix the air bag with tape; (2) The child's abdomen is bulging, and the air bag is very tight. It is easy to tilt up, resulting in weak respiratory wave amplitude and missed detection, and low RR; (3) The air bag catheter is pressed under the child's body, which will affect the pressure amplitude of the air bag detection, resulting in missed respiratory wave detection and low RR; (4) Air bag pipeline During nursing, the child is caught by the incubator door, causing the respiratory wave to straighten, falsely reporting suffocation; (5) The air bag is attached with tape, and the air bag will tilt up after the child moves and cries, causing the amplitude of the monitored respiratory wave to weaken and leakage. Check, RR is low. If the breath detection is inaccurate, the corresponding alarm prompt will also be inaccurate.

On the one hand, manual counting will increase the workload of nurses, and on the other hand, it is impossible to monitor the patient's respiration in real time.

Therefore, the existing monitoring of neonatal breathing needs to be improved and improved.

technical problem

The present invention mainly provides a newborn/non-contact physiological signs monitoring method and system, aiming to improve the accuracy of respiratory monitoring.

Technical solutions

One embodiment provides a neonatal physiological signs monitoring system, including:

Box, used to accommodate newborns;

A radar sensor is arranged on the box and is used to collect fluctuation data of a target part of a target object, and the target object is the newborn;

An adjustment device, connected to the radar sensor, used to adjust the measurement position and/or viewing angle of the radar sensor;

A processor, configured to obtain respiratory data and/or heartbeat data of the target object based on the fluctuation data.

One embodiment provides a non-contact physiological sign monitoring system, including:

radar sensor;

Camera, used to capture one or more frames of images of the target object;

Processor for:

The radar sensor collects the fluctuation data of the target part of the target object, and obtains the respiratory amplitude and/or respiratory frequency of the target part caused by breathing according to the fluctuation data; and obtains the target through the one or more frames of images. information about the object and/or environmental information,

And determine an alarm strategy corresponding to the breathing amplitude and/or breathing frequency based on the target object's information and/or environmental information.

One embodiment provides a non-contact physiological sign monitoring method for newborns, including:

The body position state of the target object is obtained by taking one or more frames of images of the target object located in the box through the camera. The body position state is divided into three types: supine state, prone state and side lying state. Each body position state is pre-set Different suffocation thresholds are associated, and the target object is a newborn; the suffocation threshold includes a suffocation amplitude threshold and/or a suffocation frequency threshold;

Collect the fluctuation data of the target part of the target object through a radar sensor, and obtain the respiratory amplitude and/or respiratory frequency of the fluctuation of the target part caused by breathing based on the fluctuation data;

Determine whether the respiratory amplitude and/or respiratory frequency are lower than the apnea threshold corresponding to the posture state, and if so, output an apnea alarm message, or determine whether the respiratory amplitude and/or respiratory frequency are lower than the apnea threshold corresponding to the posture state, and the The duration of the breathing amplitude and/or breathing frequency lower than the apnea threshold corresponding to the posture state is greater than the preset time threshold, and if so, an apnea alarm message is output.

One embodiment provides a non-contact physiological sign monitoring method for newborns, including:

Use a camera to capture one or more frames of images of a target object located in the box, and the target object is a newborn;

Collect the fluctuation data of the target part of the target object through a radar sensor arranged on the box, and obtain the respiratory amplitude and/or respiratory frequency of the fluctuation of the target part caused by breathing based on the fluctuation data;

The information of the target object and/or the environment information is obtained through the one or more frames of images, and the alarm strategy corresponding to the breathing amplitude and/or the breathing frequency is determined based on the information of the target object and/or the environment information.

One embodiment provides a non-contact physiological sign monitoring method, including:

Take one or more frames of images of the target object through the camera;

Collect the fluctuation data of the target part of the target object through a radar sensor, and obtain the respiratory amplitude and/or respiratory frequency of the fluctuation of the target part caused by breathing based on the fluctuation data;

The information of the target object and/or the environment information is obtained through the one or more frames of images, and the alarm strategy corresponding to the breathing amplitude and/or the breathing frequency is determined based on the information of the target object and/or the environment information.

One embodiment provides a computer-readable storage medium having a program stored on the medium, and the program can be executed by a processor to implement the method as described above.

beneficial effects

According to the newborn/non-contact physiological sign monitoring method and system of the above embodiment, the fluctuation data of the target part of the target object is collected through the radar sensor. This non-contact method improves the accuracy of detection, and is also provided with The adjustment device adjusts the measurement position and/or viewing angle of the radar sensor through the adjustment device, so that the radar sensor can be aligned with the target part of the target object, thereby further improving the accuracy of detection and thus improving the accuracy of monitoring the target object.

Description of the drawings

Figure 1 is a structural block diagram of an embodiment of a newborn/non-contact physiological sign monitoring system provided by the present invention;

Figure 2 is a schematic structural diagram of an embodiment of a neonatal incubator provided by the present invention;

Figure 3 is a schematic diagram of a radar sensor installed at the bottom of the box in the neonatal incubator provided by the present invention;

Figure 4 is a flow chart of an embodiment of a newborn physiological signs monitoring method provided by the present invention;

Figure 5 is a flow chart of an embodiment of the non-contact physiological sign monitoring method provided by the present invention;

Figure 6 is a structural block diagram of another embodiment of the newborn/non-contact physiological sign monitoring system provided by the present invention.

Embodiments of the invention

The present invention will be further described in detail below through specific embodiments in conjunction with the accompanying drawings. Similar elements in different embodiments use associated similar element numbers. In the following embodiments, many details are described in order to make the present application better understood. However, those skilled in the art can readily recognize that some of the features may be omitted in different situations, or may be replaced by other elements, materials, and methods. In some cases, some operations related to the present application are not shown or described in the specification. This is to avoid the core part of the present application being overwhelmed by excessive descriptions. For those skilled in the art, it is difficult to describe these in detail. The relevant operations are not necessary, and they can fully understand the relevant operations based on the descriptions in the instructions and general technical knowledge in the field.

Additionally, the features, operations, or characteristics described in the specification may be combined in any suitable manner to form various embodiments. At the same time, each step or action in the method description can also be sequentially exchanged or adjusted in a manner that is obvious to those skilled in the art. Therefore, the various sequences in the description and drawings are only for clearly describing a certain embodiment, and do not imply a necessary sequence, unless otherwise stated that a certain sequence must be followed.

The serial numbers assigned to components in this article, such as "first", "second", etc., are only used to distinguish the described objects and do not have any sequential or technical meaning. The terms "connection" and "connection" mentioned in this application include direct and indirect connections (connections) unless otherwise specified.

As shown in Figure 1, the non-contact physiological sign monitoring system provided by the present invention includes a processor 10, a camera 20, a radar sensor 30 and an adjustment device 40.

The camera 20 is used to capture one or more frames of images of the target object. This embodiment takes the shooting of multi-frame images as an example for explanation. The multi-frame images may be multi-frame photos taken individually, or may be multiple frames in a video captured. The target audience is patients requiring respiratory monitoring.

The radar sensor 30 is used to collect fluctuation data of a target part of the target object. For example, the radar sensor 30 emits electromagnetic waves to a target part of the target object and receives echo signals of the electromagnetic waves, thereby obtaining fluctuation data of the target part. The fluctuation data can be subsequently processed to obtain the breathing data and/or heartbeat data of the target object. In this embodiment, the radar sensor 30 adopts a millimeter wave radar sensor.

The adjustment device 40 is connected to the radar sensor 30 and is used to adjust the measurement position and/or viewing angle of the radar sensor 30 .

The processor 10 is used to monitor the respiration and/or heart rate of the target subject by controlling the monitoring system. The following is a detailed description through some examples.

The monitoring system provided by the present invention can be used in various situations to monitor the breathing of patients, such as being installed in a ward, a patient's bedroom, etc. This embodiment takes the installation in a neonatal incubator as an example. Correspondingly, the target object Take a newborn as an example. As shown in Figure 2, the monitoring system also includes a box 70 and a thermostat 60.

The box 70 is used to accommodate the newborn.

The constant temperature device 60 is used to provide a constant temperature environment for the newborn in the box 70 .

The camera 20 can be installed on the top or side of the box 70 , and its field of view covers the bed board area of the box 70 . The radar sensor 30 is installed on the adjusting device 40. The adjusting device 40 and the radar sensor 30 can be installed on the top of the box 70 (such as the box cover), as shown in Figure 2; the adjusting device 40 and the radar sensor 30 can also be installed on the box. The bottom of the body 70, as shown in Figure 3. In some embodiments, the adjustment device 40 and the radar sensor 30 can also be installed on the side of the box 70 .

The processor 10 controls the monitoring system to perform non-contact physiological sign monitoring of the newborn. The process is shown in Figure 4 and includes the following steps:

Step 1. The processor 10 obtains the posture state of the target object, which may be obtained from a device outside the system or from one or more frames of images captured by the camera 20. This embodiment takes the latter as an example for explanation. The processor 10 captures one or more frames of images of the target object located in the box through the camera to obtain the posture state of the target object. The posture states are divided into three types: supine state, prone state and side lying state.

Step 2: The processor 10 adjusts the measurement position and/or viewing angle of the radar sensor 30 through the adjustment device 40 according to the posture state of the target object, so that the radar sensor 30 is aligned with the target part of the target object. In this embodiment, the target parts can correspond to the posture states, and there are three types: the chest and abdomen, the back corresponding to the chest and abdomen, and the sides of the chest and abdomen. Specifically, in the supine state, the adjustment device 40 can be used to align the center line of the radar sensor 30 with the chest and abdomen of the newborn; in the prone state, the adjustment device 40 can be used to align the center line of the radar sensor 30 with the baby's chest and abdomen. The back corresponds to the newborn's chest and abdomen; in the side-lying state, the center line of the viewing angle of the radar sensor 30 can be aligned with the side of the newborn's chest and abdomen through the adjustment device 40 .

The adjustment device 40 may have an X-axis moving mechanism and a Y-axis moving mechanism. The X-axis moving mechanism can drive the radar sensor 30 to move along the X-axis direction. The Y-axis moving mechanism can drive the X-axis moving mechanism to move along the Y-axis direction, thereby allowing the radar sensor to move. 30 can be moved to any position on the top of the box, or to any position on the bottom of the box. The adjustment device 40 may also include a rotation mechanism, which includes rotation in two dimensions, so that the center line of the viewing angle of the radar sensor 30 can scan the entire top or bottom surface of the box 70 .

In this embodiment, because the bed board of the box 70 is tilted, the newborn may slide along the bed board from the middle position to one end of the bed board inside the box 70 . Therefore, the newborn is monitored in real time, that is, the camera 20 captures the image of the newborn in real time. The processor 10 obtains the real-time body position and the real-time position of the newborn based on the real-time image of the newborn. The processor 10 obtains the current perspective of the radar sensor 30 (ie, the direction of the radar sensor) through the adjustment device 40, and obtains the real-time position of the newborn based on the real-time image of the newborn. Based on the body position status, real-time position and the current viewing angle of the radar sensor 30, it is determined whether the current viewing angle center line of the radar sensor 30 is located at the target part of the newborn. If it is, there is no need to adjust the sensor viewing angle. If it is not located, it will be based on the real-time position and real-time body position. The viewing angle offset is calculated based on the state and the current viewing angle and sent to the adjustment device 40 . The adjustment device 40 rotates according to the viewing angle offset, so that the radar sensor 30 rotates following the sliding of the newborn, so that the center line of the viewing angle of the radar sensor 30 is maintained at the target part of the newborn. Among them, the processor 10 obtains the real-time position of the newborn based on the real-time image of the newborn. The following method can be used: the processor 10 identifies the newborn's head, feet, hands and other key parts from the real-time image, and then identifies the newborn's real-time position. area of interest (target part). If only the newborn's head is recognized (the newborn is covered with a quilt or swaddle), the position about 10cm below the newborn's head can be used as the newborn's chest or back position.

In some embodiments, the camera 20 can be installed on the adjustment device 40 together with the radar sensor 30 , and the center line of the field of view of the camera 20 coincides with the center line of the field of view of the radar sensor 30 , that is, the camera 20 is aligned with the target part of the target object. This means that the radar sensor 30 is aligned with the target location. The processor 10 can adjust the field of view of the camera 20 through the adjustment device 40 so that the target part of the target object is located in the middle of the image captured by the camera 20 .

Step 3: The processor 10 collects the fluctuation data of the target part of the target object through the radar sensor 30, for example, obtains the fluctuation data of the target part according to the echo signal received by the radar sensor 30. The fluctuation data of the target part of the target object is caused by breathing and/or the heartbeat. Generally, the fluctuation data includes two types: one is the first fluctuation data of the target part caused by breathing, and the other is the third fluctuation data of the target part caused by heartbeat. 2. Fluctuation data. The principle of radar measuring respiration and heartbeat is to measure the rise and fall of the target object's chest and abdomen through radar signals, because both respiration and heartbeat will cause the rise and fall of the target object's chest and abdomen, but the rise and fall of the chest and abdomen caused by breathing are relatively large, while the rise and fall of the chest and abdomen caused by heartbeat are relatively large. The fluctuations are relatively small. The respiratory frequency is relatively low, while the heartbeat frequency is relatively high. The two data can be separated by band-pass filters in different frequency bands. Furthermore, the processor 10 can obtain the respiratory data of the target object based on the first fluctuation data caused by breathing, where the respiratory data includes respiratory frequency and/or respiratory amplitude (ie, the fluctuation amplitude of the target part). The processor 10 further determines whether the respiratory frequency and/or respiratory amplitude are abnormal, and if so, generates corresponding alarm information. For example, a normal interval for the newborn's respiratory frequency (such as 40-60 rpm) can be preset, and the maximum value of the normal interval is rapid. Threshold, the minimum value is the slow threshold. The processor determines whether the respiratory rate is higher than the preset tachypnea threshold. If so, it considers that there is a risk of tachypnea and generates a tachypnea alarm message. The processor determines whether the respiratory frequency is lower than the preset bradypnea threshold. If so, it considers that there is a risk of bradypnea and generates a bradypnea alarm message. The processor can also determine whether the breathing amplitude is lower than the preset suffocation amplitude threshold. If so, the target object is considered to be at risk of suffocation and generates a suffocation alarm message. The processor can also determine whether the breathing frequency is lower than the preset suffocation frequency threshold. If so, the target object is considered to be at risk of suffocation and generates a suffocation alarm message. Of course, the processor may also generate the apnea alarm information only when the breathing amplitude is lower than the preset apnea amplitude threshold and the breathing frequency is lower than the preset apnea frequency threshold.

The processor 10 can obtain the heartbeat data of the target object based on the second fluctuation data caused by the heartbeat. If the heartbeat fluctuates once, it is considered as one heartbeat, thereby obtaining the heart rate. Then determine whether the heartbeat data is abnormal, and if so, generate corresponding alarm information. For example, you can preset a normal interval for newborn heart rate (such as 100-160). The maximum value of this normal interval is the overspeed threshold and the minimum value is the too slow threshold. . The processor determines whether the heart rate is higher than a preset overspeed threshold, and if so, generates an alarm message about a too fast heartbeat; determines whether the heart rate is lower than a preset threshold value when the heartbeat is too slow, and if so, generates an alarm message about a too slow heartbeat.

Each of the above thresholds may be different according to different posture states, that is, each posture state is pre-associated with different threshold values. This embodiment takes as an example that each posture state is pre-associated with different asphyxiation thresholds.

The apnea threshold includes an apnea amplitude threshold and/or apnea frequency threshold. The apnea amplitude threshold and/or the apnea frequency threshold are used to evaluate whether the target subject is apnea. The processor 10 determines whether the breathing amplitude is lower than the apnea amplitude threshold corresponding to the posture state and/or determines whether the breathing frequency is lower than the apnea frequency threshold corresponding to the posture state. If so, the target object is considered to be at risk of suffocation, and then generates a suffocation alarm information. Of course, the processor 10 can also count the duration in which the breathing amplitude is lower than the suffocation amplitude threshold corresponding to the posture state. Only when the duration is greater than the preset time threshold, the target object is considered to be at risk of suffocation, and then generates suffocation alarm information. And/or, the processor 10 counts the duration in which the respiratory frequency is lower than the suffocation frequency threshold corresponding to the posture state, and only considers that the target object is at risk of suffocation when the duration is greater than the preset time threshold, and then generates suffocation alarm information. The time threshold can be set as needed, usually at the second level, such as 10 seconds. Among them, the asphyxiation threshold corresponding to the supine state is greater than the asphyxia threshold corresponding to the prone state, and the asphyxiation threshold corresponding to the prone state is greater than the asphyxiation threshold corresponding to the side lying state. The supine state is the most ideal position, with the largest detected amplitude, followed by the prone state. The side-lying state detects the newborn's chest and abdomen sides, so the detected amplitude is the smallest. Therefore, the three positions correspond to The suffocation thresholds decrease in sequence. In this embodiment, the suffocation threshold corresponding to the prone state may be 1/10 of the suffocation threshold corresponding to the supine state, and the suffocation threshold corresponding to the side lying state may be 1/10 of the asphyxia threshold corresponding to the supine state. /20. By subdividing the suffocation threshold, we can more appropriately and accurately determine whether newborns are at risk of suffocation. Different thresholds are set according to different posture states, and the resulting alarm information is more accurate.

The rise and fall data collected by the radar sensor is susceptible to interference from the outside world and newborns. This embodiment can also remove the interference in the rise and fall data first, and then determine whether the rise and fall data is abnormal. If it is abnormal, alarm information related to the rise and fall data will be generated. The following is an example of how to remove interference.

The processor can obtain the information of the target object and/or the environment information from a device outside the system, or can obtain the information of the target object and/or the environment information through one or more frames of images captured by the camera 20 . This embodiment takes the latter as an example. , obtain target object information and environmental information for explanation. The information of the target object includes at least one of the following: the activity state of the target object, and the skin color of the target object. This embodiment will be described by taking the two as an example. Environmental information includes adult hand information. The processor knows whether the measurement has been disturbed based on information about the target object and/or environmental information.

Considering that when the radar sensor 30 collects the fluctuation data of the target part, the activities of the non-target part of the target object (such as crying, changing diapers, etc.) will cause interference. Therefore, the processor 10 can adjust the target part according to the activity of the target object. The fluctuation data is processed accordingly to reduce interference. Specifically, the processor 10 obtains the activity state of the target object based on the multiple frames of images (for example, it may be multiple consecutive frames). Activity status can be divided into three types according to frequency: high-frequency activity, low-frequency activity and inactivity. Among them, high-frequency activities and low-frequency activities are relative concepts. That is, if the activity frequency of the target object is higher than the preset value, it is considered to be high-frequency activity. A first filter coefficient corresponding to the high-frequency activity can be preset; the activity frequency is greater than A value equal to 0 and less than the preset value is considered to be low-frequency activity, and a second filter coefficient corresponding to the low-frequency activity can be preset. For example, the processor 10 obtains the quantified value of the image difference between the multiple frames of images based on the continuous multi-frame images (such as quantified by similarity), and the quantified value of each image difference forms a waveform curve, thereby obtaining the frequency of the waveform curve, and predetermined Assume high-frequency interval, low-frequency interval and inactive interval. If the frequency of the waveform curve is in the high-frequency interval, the active state is determined to be high-frequency activity. If the frequency of the waveform curve is in the low-frequency interval, the active state is determined to be low-frequency activity. If the frequency of the waveform curve is in the future, it is determined that the active state is high-frequency activity. The activity interval determines the activity status as inactive, etc. The processor 10 can also obtain the current activity state of the target object based on the fluctuation data. Specifically, perform time domain and/or frequency domain analysis on the fluctuation data to obtain the frequency of fluctuations caused by newborn activity, that is, the frequency of the interference signal (if any), and three frequency ranges of the fluctuations of the target part can be preset: The inactive range corresponding to inactivity, the low frequency range corresponding to low-frequency activity, and the high-frequency range corresponding to high-frequency activity. Generally, the frequency of the high-frequency range > the frequency of the low-frequency range > the frequency of the inactive range. The respiratory rate and heart rate of newborns The electrical frequency is generally different from the frequencies in these three ranges. Therefore, the interference caused by newborn activities can be frequency divided into high-frequency range and low-frequency range. Therefore, from the current ups and downs data, if the frequency of ups and downs caused by newborn activity can be extracted, and its frequency is in the high-frequency range, then the current activity state is determined to be high-frequency activity; the extracted frequency of ups and downs caused by newborn activity If it is in the low-frequency range, the current activity state is determined to be low-frequency activity; if the extracted frequency of fluctuations caused by the newborn's activity is in the inactive range, the current activity state is determined to be inactive. If the current activity state is high-frequency activity, the processor 10 performs high-frequency filtering on the fluctuation data to filter out the interference caused by the high-frequency activity, and obtains the target object's breathing data and heartbeat data based on the high-frequency filtered fluctuation data. Then determine whether it is abnormal. The details of determining whether it is abnormal are as mentioned above, and will not be described in detail here. If the current activity state is low-frequency activity, the processor 10 performs low-frequency filtering on the fluctuation data to filter out the interference caused by the low-frequency activity, obtains the target object's breathing data and heartbeat data based on the low-frequency filtered fluctuation data, and then determines whether it is Abnormality. To determine whether it is abnormal or not, see the above content for details and will not be described in detail here. High-frequency filtering and low-frequency filtering are also relative concepts. For example, high-frequency filtering can be filtering using a first filter coefficient, and low-frequency filtering can be filtering using a second filter coefficient. The radar data is filtered according to the newborn's activity, thereby improving the accuracy of breathing and heartbeat detection, and also improving the accuracy of subsequent alarm information.

If the filtering method is not used, the anti-motion algorithm can be used to solve the problem of inaccurate respiration and heartbeat detection caused by target object activity (movement) and thus false alarms. Specifically, the processor 10 obtains the current activity status of the target object based on the multi-frame images. The activity status can also be divided into two types: active and inactive. The above-mentioned high-frequency activities and low-frequency activities are both activities, so the specific activity status is determined. The method can be seen in the previous paragraph. The processor 10 can also obtain the current activity status of the target object based on the fluctuation data. The specific activity status determination method can be found in the previous section. When the activity state is active, the processor 10 uses the preset breathing frequency as the breathing frequency of the target object during the activity period. The preset breathing frequency may be the average breathing frequency before the activity. When the activity state is active, the processor 10 may also use a preset breathing amplitude as the breathing amplitude of the target object during the activity period, and the preset breathing amplitude may be the average breathing amplitude before the activity. Or, when the activity state is active, the processor 10 obtains the breathing frequency during the activity period, then interpolates the breathing frequency before the activity, smoothly interpolates it to the breathing frequency during the activity period, and uses the interpolated breathing frequency as the activity period of the target object. respiratory rate. When the activity state is active, the processor 10 may also interpolate the breathing amplitude before the activity, and use the interpolated breathing amplitude as the breathing amplitude of the target object during the activity period. Alternatively, when the activity state is active, the processor 10 performs smoothing processing on the fluctuation data during the activity period, and obtains the respiratory frequency and/or respiratory amplitude of the target object based on the smoothed fluctuation data. Alternatively, when the activity state is active, the processor 10 uses the breathing frequency before the activity as the breathing frequency of the target object's activity period, for example, uses the breathing frequency for a period of time before the activity as the breathing frequency during the activity period. When the activity state is active, the processor 10 may also use the breathing amplitude before the activity as the breathing amplitude of the target object during the activity period.

If the newborn keeps crying, or moves its hands and feet, or twists its body, etc., its excessive range of activities will cause the respiratory data and heartbeat data to be ineffective, that is, the interference cannot be solved by the above anti-interference method. question. Therefore, the processor 10 can determine whether the breathing amplitude exceeds the preset activity threshold, and if so, output an alarm message indicating that the measurement is interfered with. The activity threshold can be an amplitude that theoretically cannot be reached by breathing and heartbeat. The processor 10 may also obtain a quantified value of the image difference of the multi-frame images based on the continuous multi-frame images, and when the quantified value is within a preset alarm interval, it will output alarm information indicating that the measurement is interfered with.

Medical staff are performing operations on the newborn, such as changing clothes, changing diapers, feeding, etc., which can also cause the newborn to move too much. The processor 10 determines whether there are the above-mentioned operations of the medical staff in the image. For example, it can be implemented using traditional image recognition, or the image can be input into a pre-trained deep learning model, and the deep learning model outputs whether there is an adult's hand. Results, etc.

As shown in Figures 1 and 2, the monitoring system may also include a blood oxygen module 50 and a wireless communication module 80.

The blood oxygen module 50 is used to measure the blood oxygen saturation of the target object. The blood oxygen module 50 is installed on the box 70 , such as on the top or side wall of the box 70 . Its blood oxygen probe can be fixed on the newborn's fingertips and connected to the blood oxygen module body through a blood oxygen probe cable.

The processor 10 measures the blood oxygen saturation of the target object through the blood oxygen module 50 .

The processor 10 can control an alarm device set on or connected to the box to issue an alarm, for example, by displaying alarm information on a display, or by emitting a light signal through an indicator light set or connected on the box, or by setting an alarm on the box. Or the connected built-in speaker can emit an audio signal to prompt an alarm.

In other embodiments, the monitoring system may also include a wireless communication module 80 and a monitoring device. The wireless communication module 80 is used for wireless communication with the monitoring device.

The processor 10 can transmit the images, respiratory data, heartbeat data and blood oxygen saturation data captured by the camera to the monitoring device through the wireless communication module 80 . Monitoring equipment can be a host computer, a monitor, a central station, etc. The external equipment can display respiratory data, heartbeat data, and blood oxygen saturation data. It can also display received alarm information and send out alarm sounds corresponding to the alarm information, etc., to facilitate medical care. Personnel monitor the newborn's breathing, heart rate, blood oxygen and other parameters.

The processor 10 can also obtain the skin color of the target object based on one or more frames of images of the target object, determine whether the skin color is abnormal, and if so, generate and output corresponding alarm information.

The existing technology usually simply generates and outputs corresponding alarm information when a parameter is abnormal. In order to better monitor newborns and improve the accuracy of alarms, the present invention uses target object information and/or environmental information to better monitor newborns. Determine the alarm strategy corresponding to respiratory amplitude and/or respiratory frequency, as shown in Figure 5, including the following steps:

Step 4: The processor 10 obtains the target object information and/or environmental information through one or more frames of images captured by the camera. Environmental information includes information about whether there are adult hands in the environment. For example, the processor 10 determines whether there are adult hands in one or more frames of images. The presence of adult hands in the image means that there are adult hands in the environment. The absence of adult hands in the image means that there are no adult hands in the environment. The judgment result is That is environmental information. For example, the processor 10 obtains the current activity state of the target object according to the multi-frame images. The activity status can be divided into three types according to the amplitude: large activity, small activity and no activity; among them, large activity and small activity are relative concepts. The processor 10 obtains the current activity state of the target object based on the image difference of the target object in consecutive multi-frame images. The greater the image difference, the greater the activity amplitude. The image difference can be quantified (such as using similarity to quantify), and Three preset range intervals: large range, small range and inactive range. If the quantified value of the current image difference is in a large range, the current activity state is determined to be large activity. If the quantified value of the current image difference is in a small range, the current activity state is determined to be small activity. If the quantified value of the current image difference is in an inactive range, the current activity is determined. The status is inactive etc.

The processor 10 can also obtain the current activity state of the target object based on the fluctuation data. Specifically, two ranges of the amplitude of fluctuations of the target part caused by activity can be preset: a small range corresponding to small activities and a large range corresponding to large activities. The amplitudes of the large range and the small range are different from the amplitudes caused by breathing and heartbeat. Generally, the amplitude of the large range > the amplitude of the small range > the amplitude of the non-active range. Therefore, in the current ups and downs data, if the amplitude of ups and downs caused by activity is in a large range, the activity state is determined to be large activity; if the amplitude of ups and downs caused by activity is in a small range, the activity state is determined to be small activity; if the amplitude of ups and downs caused by activity is in the future, the activity state is determined to be large activity. The activity scope determines that the current activity status is inactive.

The processor 10 can also obtain the skin color of the target object based on one or more frames of images captured by the camera; determine whether the skin color belongs to a preset abnormal color; the abnormal color can be in the purple domain, such as RGB[128,0,128] ±30 Color area, purple area can judge the target object is hypoxic.

Step 5: The processor 10 determines an alarm strategy corresponding to the breathing amplitude and/or breathing frequency based on the target object's information and/or environmental information. There are two specific implementation methods, which are introduced one by one below.

In the first method, the respiratory amplitude and/or respiratory frequency are pre-associated with an alarm scheme corresponding to the initial alarm condition. The initial alarm condition is an alarm condition determined solely based on the threshold of respiratory amplitude and/or respiratory frequency. For example, if the respiratory amplitude is lower than the apnea amplitude threshold, or the respiratory frequency is lower than the apnea frequency threshold, or the respiratory frequency is higher than the respiratory-related shortness of breath threshold, etc., it can be said that the initial alarm condition is met.

In this embodiment, the breathing amplitude is pre-associated with an alarm scheme, and the breathing frequency is also pre-associated with an alarm scheme as an example.

The processor 10 meets the initial alarm condition when the breathing amplitude and/or breathing frequency meet the initial alarm condition (that is, the breathing amplitude is lower than the apnea amplitude threshold, and/or the breathing frequency is lower than the apnea frequency threshold), and the target object's information and/or environmental information When the preset conditions are not met, the alarm scheme pre-associated with the respiratory amplitude and/or respiratory frequency is used as the alarm strategy and executed; when the target object's information and/or environmental information meet the preset conditions, the alarm scheme is adjusted to obtain the alarm strategy and executed. . Among them, adjusting the alarm plan includes at least one of the following: not outputting the alarm information corresponding to the alarm plan, adjusting the alarm priority of the alarm information corresponding to the alarm plan before outputting, and delaying the output of the alarm information. The adjustment elements of the alarm plan include the timing and priority of outputting alarm information (alarm priority). In this method, the timing of outputting the alarm information is usually after the alarm information is generated. The priority may be the priority level displayed on the display device, etc., such as the priority display order of alarm information, the degree of conspicuity, etc. The present invention determines whether to adjust the default alarm scheme based on the actual situation of the newborn, thereby obtaining a suitable alarm strategy, optimizing the alarm, and improving the accuracy of the alarm.

The preset conditions for environmental information include the presence of adult hands in the environment. If the processor 10 determines that there are no adult hands in the one or more frames of images, it means that the environmental information does not meet the preset conditions. In this case, the processor 10 determines whether the breathing amplitude is lower than the preset suffocation amplitude threshold. , if so, output the first priority suffocation alarm information. In other words, when the breathing amplitude is lower than the preset suffocation amplitude threshold and there is no adult hand, the processor 10 outputs the first priority suffocation alarm information. Of course, if the breathing amplitude is not lower than the preset suffocation amplitude threshold, it means there is no risk of suffocation and no treatment will be performed. If the processor 10 determines that there are adult hands in the one or more frames of images, it means that the environmental information meets the preset conditions. In this case, the processor 10 determines whether the breathing amplitude is lower than the preset suffocation amplitude threshold. If so, Then the suffocation alarm information is not output, and/or the second priority suffocation alarm information is output, and/or the suffocation alarm information is output after a preset time delay (such as 10s). In other words, when the breathing amplitude is lower than the preset suffocation amplitude threshold and adult hands are present, the processor 10 does not output the suffocation alarm information, and/or outputs the second priority suffocation alarm information, and/or, The suffocation alarm message is output after delaying the preset time. Among them, the first priority is higher than the second priority. The preset condition of the environmental information can also be: the time when the adult's hand is present exceeds the preset time threshold. At this time, the processor 10 also counts the duration during which the breathing amplitude is lower than the suffocation amplitude threshold and the adult's hand is present. During this duration, After the time exceeds the preset time threshold, the second priority suffocation alarm information is output (that is, the alarm information is output after lowering the priority), and/or, the suffocation alarm information is output after a delay of the preset time; if the duration does not If the preset time threshold is exceeded, no alarm information will be output. The time threshold can be a preset time.

The presence of medical care hands indicates that the medical care is in contact with the newborn. At this time, it may be a false alarm caused by the medical care, or there may be an emergency situation where the medical care comes to deal with it. In short, the medical care is handling it at the scene. This situation is less dangerous than just breathing amplitude. The suffocation amplitude threshold should be slight, so the alarm priority of the suffocation alarm information can be reduced (from the first priority to the second priority), or the output can be delayed. It can be seen that the adjusted alarm strategy can better apply to the current scenario. Outputting alarm information is more accurate. When the monitor receives multiple alarm messages, it displays the alarm messages in order from high to low priority.

Similarly, when the breathing frequency is lower than the preset suffocation frequency threshold and there is no adult hand, the processor 10 outputs the first priority suffocation alarm information; when the breathing frequency is lower than the preset suffocation frequency threshold and there is an adult When the hand is turned on, the suffocation alarm information is not output, and/or the second priority suffocation alarm information is output, and/or the suffocation alarm information is output after a preset time delay. Similarly, the processor 10 can count the duration in which the breathing frequency is lower than the apnea frequency threshold and there is an adult's hand, and after the duration exceeds the preset time threshold, output the second priority apnea alarm information, and/or , output the suffocation alarm information after delaying the preset time; if the duration does not exceed the preset time threshold, no alarm information will be output.

In addition to suffocation alarm information, the above solution can also be extended to shortness of breath alarm information and bradypnea alarm information. When the shortness of breath alarm information is generated (the breathing frequency exceeds the preset shortness of breath threshold) and there are no adult hands, the processor 10 outputs the shortness of breath alarm information; when the breathing frequency exceeds the preset shortness of breath threshold and there are adult hands (i.e. When the environmental information meets the preset conditions), the processor 10 does not output shortness of breath alarm information. Similarly, when the bradypnea alarm information is generated (the breathing frequency is greater than the apnea frequency threshold and less than the preset brady threshold) and there is no adult hand, the processor 10 outputs the bradypnea alarm information; when the breathing frequency is greater than the apnea frequency threshold, When the frequency threshold is less than the preset bradypnea threshold and there is an adult hand (that is, the environmental information meets the preset conditions), the processor 10 does not output bradypnea alarm information.

The preset conditions for the target object's information include substantial activity. The processor 10 obtains the activity status of the target object based on the multi-frame images. When the activity status of the target object is small activity or inactivity, the information of the target object does not meet the preset conditions. If the breathing amplitude is lower than the preset apnea amplitude threshold at this time, the processor 10 executes the pre-associated alarm plan, that is, Output the first priority suffocation alarm information. When the activity state of the target object is large-scale activity, the information of the target object meets the preset conditions. If the breathing amplitude is lower than the preset apnea amplitude threshold at this time, the processor 10 adjusts the pre-associated alarm scheme to obtain an alarm strategy, that is, no Output the suffocation alarm information, and/or, output the second priority suffocation alarm information, and/or, output the suffocation alarm information after delaying the preset time. In addition, the preset condition for the target object's information may be that the large-scale activity exceeds a preset time threshold. At this time, it is considered that the breathing amplitude is lower than the preset apnea amplitude threshold for a certain period of time before it is considered that an alarm is really needed. Specifically, when the activity state of the processor 10 is a large amount of activity, the statistical breathing amplitude is lower than the preset apnea amplitude threshold, And the activity state is the duration of substantial activity. After the duration exceeds the preset time threshold, the second priority suffocation alarm information is output (the alarm information is output after reducing the alarm priority), and/or, the delay is preset The suffocation alarm information will be output after the time; if the duration does not exceed the preset time threshold, the suffocation alarm information will not be output. Measurements of large-scale activities are usually less accurate, so adjusting the original alarm scheme can provide a more accurate alarm. Of course, in some embodiments, when the activity state is substantial activity, the processor 10 counts the duration during which the breathing amplitude is lower than the preset apnea amplitude threshold and the activity state is substantial activity, and when the duration exceeds the preset time After the threshold, the first priority suffocation alarm information is output; if the duration does not exceed the preset time threshold, the second priority suffocation alarm information is output, which is equivalent to reducing the first priority of the suffocation alarm information to the third. Output after the second priority.

Similarly, when the target object's activity status is slightly active or inactive, the target object's information does not meet the preset conditions. If the breathing frequency is lower than the preset apnea frequency threshold at this time, the processor 10 executes a pre-associated alarm. The solution is to output the first priority suffocation alarm information. When the activity state of the target object is a large amount of activity and the information of the target object meets the preset conditions, if the breathing frequency is lower than the preset apnea frequency threshold at this time, the processor 10 adjusts the pre-associated alarm scheme to obtain an alarm strategy, that is, no Output the suffocation alarm information, and/or, output the second priority suffocation alarm information, and/or, output the suffocation alarm information after delaying the preset time. The same as the breathing amplitude, the time factor can be considered on this basis. Specifically, when the activity state is a large amount of activity, the processor 10 counts the duration of the breathing frequency lower than the preset apnea frequency threshold and the activity state is a large amount of activity. , after the duration exceeds the preset time threshold, output the second priority suffocation alarm information, and/or, output the alarm information after delaying the preset time; if the duration does not exceed the preset time threshold, then no Output suffocation alarm information. Of course, in some embodiments, when the activity state is substantial activity, the processor 10 counts the duration during which the breathing frequency is lower than the preset apnea frequency threshold and the activity state is substantial activity, and when the duration exceeds the preset time After the threshold, the first priority suffocation alarm information is output; if the duration does not exceed the preset time threshold, the second priority suffocation alarm information is output.

Although the fluctuation data can remove interference through the above-mentioned filtering, anti-motion and other methods, some interferences are not easy to remove, such as large movements of newborns, medical intervention on newborns, and fluctuation data cannot remove interference (the frequency of the interference signal cannot be extracted, If an alarm message with measurement interference is generated, etc.), it means that the alarm message generated based on the fluctuation data may be inaccurate. This application adjusts the associated alarm scheme accordingly, thereby improving the accuracy of the alarm. It can be seen that the present invention can detect interference through the camera, and then take anti-interference measures. When the interference cannot be eliminated, it can also dynamically adjust the alarm scheme, effectively improving the accuracy of newborn breathing and heart rate measurement and the accuracy of the alarm, that is, It improves the accuracy of breathing and heart rate monitoring and improves the work efficiency of medical staff.

The preset condition of the target object's information may also include normal skin color. Although the alarm priority for abnormal skin color is not high, skin color can be mutually verified with alarm information related to respiratory amplitude and/or respiratory frequency. If the processor 10 determines that the skin color is abnormal, it means that the target object's information does not meet the preset conditions. At this time, if the breathing amplitude is lower than the suffocation amplitude threshold, the processor 10 executes the associated alarm plan, that is, directly outputs the first priority Apnea alarm information; similarly, if the breathing frequency is lower than the apnea frequency threshold at this time, the processor 10 directly outputs the first priority apnea alarm information.

If the processor 10 determines that the skin color is normal, the target object's information meets the preset conditions. At this time, if the breathing amplitude is lower than the apnea amplitude threshold, and/or the breathing frequency is lower than the apnea frequency threshold, the associated alarm scheme needs to be adjusted. Obtain a new alarm strategy, that is, the processor adopts at least one of the following three methods: does not output the suffocation alarm information corresponding to the alarm plan; outputs the second priority suffocation alarm information; delays the preset time and outputs the suffocation alarm information corresponding to the alarm plan. Suffocation alarm message. Time factors can also be considered on this basis, such as the processor counting the duration when the breathing amplitude is lower than the apnea amplitude threshold and/or the breathing frequency is lower than the apnea frequency threshold and the skin color is normal. If the duration exceeds the preset After the time threshold (that is, the target object's information meets the preset conditions), the second priority suffocation alarm information is output, and/or, the suffocation alarm information corresponding to the alarm scheme is output after a delay of the preset time; if the duration does not exceed the preset time, If the time threshold is set, the suffocation alarm information will not be output. Of course, in some embodiments, if the duration exceeds the preset time threshold, the first-priority suffocation alarm information may also be output; if the duration does not exceed the preset time threshold, the second-priority alarm information may be output. Suffocation alarm message.

The first method is to associate an alarm plan. If the target object's information and/or environmental information does not meet the preset conditions, the associated alarm plan will be directly executed. If the preset conditions are met, the alarm plan will be adjusted before execution.

In the second method, the breathing amplitude and/or breathing frequency are pre-associated with the first alarm plan and the second alarm plan. The processor 10 meets the initial alarm condition when the breathing amplitude and/or breathing frequency (ie, the breathing amplitude is lower than the apnea amplitude threshold, and/or, the breathing frequency is lower than the apnea frequency threshold), but the target object's information and/or environmental information do not satisfy When conditions are preset, the first alarm scheme pre-associated with respiratory amplitude and/or respiratory frequency is used as an alarm strategy and executed. The first alarm plan is the alarm plan associated with the first method, that is, the first priority alarm information is generated and output directly. The specific process is the same as the first method and will not be described again here.

When the breathing amplitude and/or breathing frequency meet the initial alarm conditions, and the target object's information and/or environmental information meet the preset conditions, the processor 10 uses the second alarm scheme pre-associated with the breathing amplitude and/or the breathing frequency as an alarm strategy. and execute. The second alarm solution includes: not outputting alarm information, outputting second priority suffocation alarm information, and outputting at least one of the alarm information after a preset time delay. It can be seen that the first alarm plan and the second alarm plan are obviously different.

Among them, in the second method, the judgment criteria of whether the target object's information and/or the environment information meet the preset conditions are the same as those in the first method, and will not be described again here.

The second alarm plan includes the adjusted alarm plan in the first method. The conditions for determining the second alarm plan as the alarm strategy are the same as the conditions for determining the adjusted alarm plan as the alarm strategy in the first method, and are not the same here. To elaborate. However, it should be noted that in the second method, the second alarm plan may also include other alarm plans besides the adjusted alarm plan in the first method.

In the second method, a mapping relationship is directly used to obtain the final alarm strategy, which simplifies the execution process of the processor and reduces the processing burden of the processor.

The processor 10 also determines whether the blood oxygen saturation is lower than a preset saturation threshold. In one embodiment, the processor 10 may output low blood oxygen saturation alarm information when the blood oxygen saturation is lower than a preset saturation threshold. In another embodiment, the processor 10 may also output low blood oxygen saturation alarm information only when the blood oxygen saturation is lower than a preset saturation threshold and the activity status is slight activity or inactivity. In other embodiments, the processor 10 can count the duration during which the blood oxygen saturation is lower than a preset saturation threshold and the activity state is substantial activity, and only output blood after the duration exceeds the preset time threshold. Alarm message for low oxygen saturation.

Skin color, blood oxygen saturation and respiratory status can corroborate each other to comprehensively judge the degree of development of respiratory diseases in children. Therefore, the processor 10 is also used to determine an alarm strategy corresponding to the respiratory amplitude and/or respiratory frequency based on the respiratory amplitude and/or respiratory frequency, skin color, and blood oxygen saturation. Specifically, the alarm information corresponding to the alarm policy includes first alarm information and second alarm information. The processor 10 determines whether the skin color belongs to a preset abnormal color; when the breathing amplitude and/or the breathing frequency are lower than the preset apnea threshold (that is, the breathing amplitude is lower than the apnea amplitude threshold, and/or the breathing frequency is lower than the apnea frequency threshold) ), when the blood oxygen saturation is lower than the preset saturation threshold and the skin color does not belong to the preset abnormal color, the first alarm information is output. The first alarm information may be, for example, alarm information used to prompt to keep the airway open and maintain normal breathing. That is, the newborn is asphyxiated and the blood oxygen saturation is low, but the skin color is normal, indicating that there is no obvious impact. At this time, the medical staff is reminded to keep the newborn's airway open and maintain normal breathing.

If suffocation occurs, blood oxygen saturation is low, and skin color is abnormal, such as cyanosis, special treatment is required, and ventilation treatment is required if necessary. Therefore, the processor 10 can operate when the breathing amplitude and/or breathing frequency is lower than the preset apnea threshold (that is, an apnea alarm message is generated), the blood oxygen saturation is lower than the preset saturation threshold, and the skin color is a preset abnormality. When the color is changed, the second alarm information is output. The second alarm information may be, for example, prompt information used to prompt the need for ventilation treatment. The priority of the second alarm information is higher than the priority of the first alarm information. For example, the second alarm information is a first-priority alarm, and the first alarm information is a second-priority alarm. It can be seen that the present invention can not only output suffocation alarm information to warn, but also provide alarm prompts of corresponding disposal methods, so as to facilitate better monitoring of children.

The processor 10 can transmit the images captured by the camera, respiratory data, heartbeat data, blood oxygen saturation data, and the above various alarm information to monitoring equipment (such as monitors, central stations, etc.) through the wireless communication module 80 . These images, data and information are displayed by the monitoring equipment.

In the above embodiment, the processor 10 is installed on the neonatal incubator. In some embodiments, the processor 10 of the monitoring system may be the processor of the monitoring device 90. As shown in Figure 6, the monitoring system includes the monitoring device 90, the above-mentioned camera 20, the above-mentioned radar sensor 30, and the above-mentioned adjustment device 40. , the above-mentioned blood oxygen module 50 and the above-mentioned wireless communication module 80. The monitoring device 90 is communicatively connected with the camera 20, the radar sensor 30, the adjustment device 40 and the blood oxygen module 50 through the wireless communication module 80. The above-mentioned processor 10 is the processor of the monitoring device 90, that is, the camera 20, the radar sensor 30 and the blood oxygen module. The data collected by the oxygen module 50 is transmitted to the monitoring device 90 by the wireless communication module 80, and the monitoring device 90 performs the functions of the above-mentioned processor 10 (see the above embodiments for details, which will not be described in detail). The monitoring device 90 also transmits the images captured by the camera, Respiration data, heartbeat data, blood oxygen saturation data, various alarm information and prompt information mentioned above are displayed through its display.

In summary, the monitoring system provided by the present invention can not only select an appropriate suffocation threshold according to the target object's posture state, but also perform filtering or anti-motion algorithms on the ups and downs data, and can also dynamically adjust the alarm strategy, comprehensive breathing status ( For example, the breathing amplitude is lower than the apnea amplitude threshold), blood oxygen saturation and skin color are cross-compared and verified, which greatly improves the accuracy of parameter measurement and the accuracy and reliability of alarms.

The invention also provides a non-contact physiological sign monitoring method for newborns, and a non-contact physiological sign monitoring method. The method is as shown in Steps 1 to 5 above, as well as the specific execution methods involved in the monitoring system, which will not be described again here.

Those skilled in the art can understand that all or part of the functions of various methods in the above embodiments can be implemented by hardware or by computer programs. When all or part of the functions in the above embodiments are implemented by a computer program, the program can be stored in a computer-readable storage medium. The storage medium can include: read-only memory, random access memory, magnetic disk, optical disk, hard disk, etc., through The computer executes this program to achieve the above functions. For example, the program is stored in the memory of the device, and when the program in the memory is executed by the processor, all or part of the above functions can be realized. In addition, when all or part of the functions in the above embodiments are implemented by a computer program, the program can also be stored in a storage medium such as a server, another computer, a magnetic disk, an optical disk, a flash disk or a mobile hard disk, and can be downloaded or copied to save it. into the memory of the local device, or performs a version update on the system of the local device. When the program in the memory is executed by the processor, all or part of the functions in the above embodiments can be realized.

This document is described with reference to various exemplary embodiments. However, those skilled in the art will recognize that changes and modifications can be made to the exemplary embodiments without departing from the scope herein. For example, the various operational steps, as well as the components used to perform the operational steps, may be implemented in different ways (e.g., one or more steps may be eliminated, modified or incorporated into other steps).

Additionally, as will be understood by those skilled in the art, the principles herein may be reflected in a computer program product on a computer-readable storage medium preloaded with computer-readable program code. Any tangible, non-transitory computer-readable storage medium may be used, including magnetic storage devices (hard disks, floppy disks, etc.), optical storage devices (CD-ROM, DVD, Blu Ray disks, etc.), flash memory and/or the like . These computer program instructions may be loaded onto a general-purpose computer, special-purpose computer, or other programmable data processing apparatus to form a machine, such that the instructions executed on the computer or other programmable data processing apparatus may generate a device that implements the specified functions. These computer program instructions may also be stored in a computer-readable memory, which may instruct a computer or other programmable data processing device to operate in a specific manner, such that the instructions stored in the computer-readable memory may form a Manufactured articles include devices that perform specified functions. Computer program instructions may also be loaded onto a computer or other programmable data processing device to perform a series of operating steps on the computer or other programmable device to produce a computer-implemented process such that the execution on the computer or other programmable device Instructions can provide steps for implementing a specified function.

Although the principles herein have been illustrated in various embodiments, many modifications of structure, arrangement, proportion, elements, materials and parts as are particularly suited to particular circumstances and operating requirements may be made without departing from the principles and scope of the disclosure. use. The above modifications and other changes or revisions are intended to be included within the scope of this document.

The foregoing detailed description has been described with reference to various embodiments. However, those skilled in the art will recognize that various modifications and changes can be made without departing from the scope of the disclosure. Accordingly, this disclosure is to be considered in an illustrative and not a restrictive sense, and all such modifications are to be included within its scope. Likewise, advantages, other advantages, and solutions to problems with respect to various embodiments have been described above. However, benefits, advantages, solutions to problems, and any elements that produce these, or make the solution more explicit, are not to be construed as critical, required, or necessary. As used herein, the term "comprises" and any other variations thereof are intended to be non-exclusively inclusive such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also those not expressly listed or otherwise not part of the process , methods, systems, articles or other elements of equipment. Furthermore, the term "coupled" and any other variations thereof as used herein refers to physical connection, electrical connection, magnetic connection, optical connection, communication connection, functional connection and/or any other connection.

Those skilled in the art will recognize that many changes may be made in the details of the embodiments described above without departing from the basic principles of the invention. Accordingly, the scope of the invention should be determined from the following claims.

Claims (36)

1. A neonatal physiological signs monitoring system, which is characterized by including:

   Box, used to accommodate newborns;

   A radar sensor is arranged on the box and is used to collect fluctuation data of a target part of a target object, and the target object is the newborn;

   An adjustment device, connected to the radar sensor, used to adjust the measurement position and/or viewing angle of the radar sensor;

   A processor, configured to obtain respiratory data and/or heartbeat data of the target object based on the fluctuation data.
2. The newborn physiological signs monitoring system according to claim 1, further comprising:

   A camera is installed on the box and is used to capture one or more frames of images of the target object.
3. The newborn physiological signs monitoring system according to claim 2, wherein the processor is also used to:

   According to the posture state of the target object, the measurement position and/or viewing angle of the radar sensor is adjusted by the adjustment device, so that the radar sensor is aligned with the target part of the target object.
4. The newborn physiological signs monitoring system according to claim 3, wherein the processor is also used to:

   The processor is further configured to obtain the posture state of the target object based on one or more frames of images of the target object captured by the camera.
5. The newborn physiological signs monitoring system according to claim 1, further comprising a blood oxygen module arranged in the box, and the blood oxygen module is used to measure the blood oxygen saturation of the target object.
6. The newborn physiological signs monitoring system according to claim 2, wherein the radar sensor is arranged on the top, bottom or side of the box, and the camera is arranged on the top or side of the box.
7. The newborn physiological signs monitoring system according to claim 1, further comprising:

   Processor for:

   Obtain the posture state of the target object; the posture state is divided into three types: supine state, prone state and side lying state. Each posture state is pre-associated with different suffocation thresholds. The suffocation thresholds include the suffocation amplitude threshold and /or suffocation frequency threshold;

   According to the fluctuation data, the breathing amplitude of the target part caused by breathing is obtained; it is judged whether the breathing amplitude is lower than the apnea amplitude threshold corresponding to the posture state, and if so, the apnea alarm information is output; or it is judged whether the breathing amplitude is lower than the apnea amplitude threshold value. The apnea amplitude threshold corresponding to the posture state, and the duration for which the breathing amplitude is lower than the apnea amplitude threshold corresponding to the posture state is greater than the preset time threshold, if so, the apnea alarm information is output; and/or,

   Obtain the respiratory frequency according to the fluctuation data; determine whether the respiratory frequency is lower than the apnea frequency threshold corresponding to the posture state, and if so, output apnea alarm information; or determine whether the respiratory frequency is lower than the apnea frequency threshold corresponding to the posture state. frequency threshold, and the duration for which the respiratory frequency is lower than the apnea frequency threshold corresponding to the posture state is greater than the preset time threshold, if so, apnea alarm information is output.
8. The neonatal physiological signs monitoring system according to claim 7, wherein the asphyxia threshold corresponding to the supine state is greater than the asphyxia threshold corresponding to the prone state, and the asphyxia threshold corresponding to the prone state is greater than the asphyxia threshold corresponding to the side lying state.
9. The newborn physiological signs monitoring system according to claim 2, wherein the processor obtains the respiratory data and/or heartbeat data of the target object based on the fluctuation data, including:

   The activity status of the target object is obtained according to the multi-frame images; the activity status is divided into three types: high-frequency activity, low-frequency activity and inactivity;

   If the activity state is high-frequency activity, high-frequency filtering is performed on the fluctuation data to filter out interference caused by the high-frequency activity, and the breathing data and/or of the target object are obtained based on the high-frequency filtered fluctuation data. heartbeat data;

   If the activity state is low-frequency activity, perform low-frequency filtering on the fluctuation data to filter out interference caused by the low-frequency activity, and obtain the respiratory data and/or heartbeat data of the target object based on the low-frequency filtered fluctuation data.
10. The newborn physiological signs monitoring system according to claim 2, wherein the respiratory data includes respiratory frequency and/or respiratory amplitude; the processor obtains the respiratory data of the target object according to the fluctuation data, including :

    The activity state of the target object is obtained according to the multi-frame images, and the activity state is divided into active and inactive;

    When the activity state is active, a preset breathing frequency is used as the breathing frequency of the target object during the activity period and/or a preset breathing amplitude is used as the breathing amplitude of the target object during the activity period; or;

    When the activity state is active, the breathing frequency before the activity is interpolated, and the interpolated breathing frequency is used as the breathing frequency of the target object during the activity period, and/or the breathing amplitude before the activity is interpolated, and the interpolated breathing frequency is used as the breathing frequency during the activity period of the target object. The subsequent breathing amplitude is used as the breathing amplitude of the target object during the activity period; or,

    When the activity state is active, smoothing the fluctuation data during the activity period, and obtaining the respiratory frequency and/or respiratory amplitude of the target object based on the smoothed fluctuation data; or,

    When the activity state is active, the breathing frequency before the activity is used as the breathing frequency of the target object during the activity period, and/or the breathing amplitude before the activity is used as the breathing amplitude of the target object during the activity period.
11. The newborn physiological signs monitoring system according to claim 2, wherein the processor is also used to:

    Obtain the respiratory amplitude and/or respiratory frequency of the fluctuation of the target part caused by breathing according to the fluctuation data;

    Obtain target object information and/or environmental information through the one or more frames of images,

    And determine an alarm strategy corresponding to the breathing amplitude and/or breathing frequency based on the target object's information and/or environmental information.
12. The newborn physiological signs monitoring system according to claim 11, wherein the environmental information includes information about whether adult hands are present in the environment, and the processor determines the breathing amplitude and/or breathing according to the environmental information. Alarm strategies corresponding to frequency include:

    When the breathing amplitude is lower than the preset suffocation amplitude threshold and there are no adult hands, the first priority suffocation alarm information is output; when the breathing amplitude is lower than the preset suffocation amplitude threshold and there are adult hands, no Output the suffocation alarm information, and/or, output the second priority suffocation alarm information, and/or, output the suffocation alarm information after delaying the preset time;

    and / or,

    When the respiratory frequency is lower than the preset suffocation frequency threshold and there are no adult hands, the first priority suffocation alarm information is output; when the respiratory frequency is lower than the preset suffocation frequency threshold and there are adult hands. , do not output the suffocation alarm information, and/or output the second priority suffocation alarm information, and/or output the suffocation alarm information after delaying the preset time;

    Among them, the first priority is higher than the second priority.
13. The newborn physiological signs monitoring system according to claim 11, wherein the information about the target object includes the activity status of the target object, and the processor is further configured to:

    The activity state of the target object is obtained according to the multi-frame images or the activity state of the target object is obtained according to the fluctuation data; the activity state is divided into three types: large activity, small activity and no activity;

    When the breathing amplitude is lower than the preset apnea amplitude threshold and the activity state is small activity or inactivity, the first priority apnea alarm information is output; when the breathing amplitude is lower than the preset apnea amplitude threshold , and when the activity state is a large amount of activity, output the second priority suffocation alarm information, and/or, output the suffocation alarm information after delaying the preset time;

    and / or,

    When the respiratory frequency is lower than the preset apnea frequency threshold and the activity state is small activity or inactivity, the first priority apnea alarm information is output; when the respiratory frequency is lower than the preset apnea frequency threshold , and when the activity state is a large amount of activity, output the second priority suffocation alarm information, and/or, output the suffocation alarm information after delaying the preset time;

    Among them, the first priority is higher than the second priority.
14. The newborn physiological signs monitoring system according to claim 11, further comprising a blood oxygen module arranged in the box, the blood oxygen module being used to measure the blood oxygen saturation of a target object; the target The object's information includes the skin color of the target object, and the processor is also used to:

    Obtain the skin color of the target object according to the one or more frames of images; and

    An alarm strategy corresponding to the respiratory amplitude and/or respiratory frequency is determined based on the respiratory amplitude and/or respiratory frequency, the skin color, and the blood oxygen saturation.
15. The newborn physiological signs monitoring system according to claim 14, wherein the alarm information corresponding to the alarm strategy includes first alarm information and second alarm information; The frequency, the skin color and the blood oxygen saturation determine the respiratory amplitude and/or the alarm strategy corresponding to the respiratory frequency includes:

    Determine whether the skin color belongs to a preset abnormal color;

    When the breathing amplitude and/or breathing frequency is lower than the preset suffocation threshold, the blood oxygen saturation is lower than the preset saturation threshold, and the skin color does not belong to the preset abnormal color, the first Alarm information; and/or,

    When the breathing amplitude and/or breathing frequency is lower than the preset suffocation threshold, the blood oxygen saturation is lower than the preset saturation threshold, and the skin color is a preset abnormal color, a second alarm is output information; wherein the priority of the second alarm information is higher than the priority of the first alarm information.
16. A non-contact physiological sign monitoring system, which is characterized by including:

    radar sensor;

    Camera, used to capture one or more frames of images of the target object;

    Processor for:

    The radar sensor collects the fluctuation data of the target part of the target object, and obtains the respiratory amplitude and/or respiratory frequency of the target part caused by breathing according to the fluctuation data; and obtains the target through the one or more frames of images. information about the object and/or environmental information,

    And determine an alarm strategy corresponding to the breathing amplitude and/or breathing frequency based on the target object's information and/or environmental information.
17. The system of claim 16, wherein:

    The respiratory amplitude and/or respiratory frequency are pre-associated with an alarm scheme corresponding to the initial alarm condition;

    Determining the alarm strategy corresponding to the respiratory amplitude and/or respiratory frequency based on the target object's information and/or environmental information includes:

    When the breathing amplitude and/or breathing frequency satisfy the initial alarm condition, and the information of the target object and/or the environmental information satisfy the preset conditions, the alarm scheme is adjusted to obtain the alarm strategy.
18. The system of claim 16, wherein determining the alarm strategy corresponding to the respiratory amplitude and/or respiratory frequency based on the target object's information and/or environmental information includes:

    When the breathing amplitude and/or breathing frequency meet the initial alarm condition, but the information of the target object and/or the environmental information do not meet the preset conditions, a first alarm is pre-associated with the breathing amplitude and/or the breathing frequency. The plan serves as the alarm strategy;

    When the breathing amplitude and/or the breathing frequency meet the initial alarm condition, and the information of the target object and/or the environmental information meet the preset conditions, a second alarm scheme that pre-associates the breathing amplitude and/or the breathing frequency. As the alarm strategy; the first alarm scheme and the second alarm scheme are different.
19. The system of claim 17, wherein:

    Adjusting the alarm scheme includes at least one of the following: not outputting alarm information, adjusting the priority of the alarm information corresponding to the alarm scheme before outputting it, and delaying output of the alarm information corresponding to the alarm scheme.
20. The system of claim 16, wherein the information about the target object includes at least one of the following: the activity state of the target object, the skin color of the target object; and the environment information includes whether there is Adult hand with information.
21. The system of claim 16, wherein the environmental information includes information on whether there are adult hands in the environment; the processor determines an alarm strategy corresponding to the breathing amplitude and/or breathing frequency based on the environmental information. ,include:

    When the breathing amplitude is lower than the preset suffocation amplitude threshold and there are no adult hands, the first priority suffocation alarm information is output; when the breathing amplitude is lower than the preset suffocation amplitude threshold and there are adult hands , do not output the suffocation alarm information, and/or, output the second priority suffocation alarm information, and/or, output the suffocation alarm information after delaying the preset time;

    and / or,

    When the respiratory frequency is lower than the preset suffocation frequency threshold and there are no adult hands, the first priority suffocation alarm information is output; when the respiratory frequency is lower than the preset suffocation frequency threshold and there are adult hands. , do not output the suffocation alarm information, and/or output the second priority suffocation alarm information, and/or output the suffocation alarm information after delaying the preset time;

    Among them, the first priority is higher than the second priority.
22. The system of claim 20, wherein the information about the target object includes the activity status of the target object, and the processor is further configured to:

    The activity state of the target object is obtained according to the fluctuation data and/or the one or more frames of images.
23. The system according to claim 20 or 22, characterized in that the activity status is divided into three types: large activity, small activity and no activity; the processor determines the breathing amplitude and/or according to the information of the target object. Or alarm strategies corresponding to respiratory frequency, including:

    When the breathing amplitude is lower than the preset apnea amplitude threshold and the activity state is small activity or inactivity, the first priority apnea alarm information is output; when the breathing amplitude is lower than the preset apnea amplitude threshold , and when the activity state is a large amount of activity, output the second priority suffocation alarm information, and/or, output the suffocation alarm information after delaying the preset time;

    and / or,

    When the respiratory frequency is lower than the preset apnea frequency threshold and the activity state is small activity or inactivity, the first priority apnea alarm information is output; when the respiratory frequency is lower than the preset apnea frequency threshold , and when the activity state is a large amount of activity, output the second priority suffocation alarm information, and/or, output the suffocation alarm information after delaying the preset time;

    Among them, the first priority is higher than the second priority.
24. The system of claim 23, wherein:

    When the breathing amplitude is lower than the preset apnea amplitude threshold and the activity state is a large amount of activity, the processor outputs the apnea alarm information of the second priority, and/or outputs the apnea alarm after delaying a preset time. information, including:

    When the breathing amplitude is lower than the preset apnea amplitude threshold and the activity state is substantial activity, count the duration during which the breathing amplitude is lower than the preset apnea amplitude threshold and the activity state is substantial activity, After the duration exceeds the preset time threshold, the second priority suffocation alarm information is output, and/or the suffocation alarm information is output after a delay of the preset time; if the duration does not exceed the preset time threshold, Then the suffocation alarm information will not be output;

    When the respiratory frequency is lower than the preset apnea frequency threshold and the activity state is a large amount of activity, the processor outputs the apnea alarm information of the second priority, and/or outputs the apnea alarm after delaying a preset time. information, including:

    When the respiratory frequency is lower than the preset apnea frequency threshold and the activity state is substantial activity, count the duration during which the respiratory frequency is lower than the preset apnea frequency threshold and the activity state is substantial activity, After the duration exceeds the preset time threshold, the second priority suffocation alarm information is output, and/or the suffocation alarm information is output after a delay of the preset time; if the duration does not exceed the preset time threshold, Then the suffocation alarm information will not be output.
25. The system of claim 20 or 21, wherein the processor is further configured to:

    Obtain respiratory frequency according to the fluctuation data;

    Determine whether the respiratory rate exceeds a preset shortness threshold;

    When the respiratory rate exceeds the preset shortness of breath threshold and there are no adult hands, output a shortness of breath alarm message;

    When the breathing frequency exceeds the preset shortness of breath threshold and an adult hand is present, the shortness of breath alarm message is not output.
26. The system according to claim 16, further comprising a blood oxygen module, the blood oxygen module being used to measure the blood oxygen saturation of the target object; the processor is also used to:

    Determine whether the blood oxygen saturation is lower than a preset saturation threshold;

    When the blood oxygen saturation is lower than the preset saturation threshold, an alarm message of low blood oxygen saturation is output; or,

    When the blood oxygen saturation is lower than the preset saturation threshold and the activity state of the target object obtained according to the one or more frames of images is small activity or inactivity, output a low blood oxygen saturation value. Alarm information; or,

    Count the duration during which the blood oxygen saturation is lower than the preset saturation threshold and the activity state of the target object obtained according to the one or more frames of images is substantial activity. When the duration exceeds the preset After the time threshold, an alarm message of low blood oxygen saturation is output.
27. The system according to claim 16, further comprising a blood oxygen module, the blood oxygen module is used to measure the blood oxygen saturation of the target object; the information of the target object includes the skin color of the target object, and the The processor is also used for:

    An alarm strategy corresponding to the respiratory amplitude and/or respiratory frequency is determined based on the respiratory amplitude and/or respiratory frequency, the skin color, and the blood oxygen saturation.
28. The system of claim 27, wherein the alarm information corresponding to the alarm strategy includes first alarm information and second alarm information; the processor determines the breathing amplitude and/or breathing frequency, the skin The alarm strategy corresponding to the color and the blood oxygen saturation determines the respiratory amplitude and/or respiratory frequency, including:

    Determine whether the skin color belongs to a preset abnormal color;

    When the breathing amplitude and/or breathing frequency is lower than the preset suffocation threshold, the blood oxygen saturation is lower than the preset saturation threshold, and the skin color does not belong to the preset abnormal color, the first Alarm information; and/or,

    When the breathing amplitude and/or breathing frequency is lower than the preset suffocation threshold, the blood oxygen saturation is lower than the preset saturation threshold, and the skin color is a preset abnormal color, a second alarm is output information; wherein the priority of the second alarm information is higher than the priority of the first alarm information.
29. A non-contact physiological sign monitoring method for newborns, which is characterized by including:

    The body position state of the target object is obtained by taking one or more frames of images of the target object located in the box through the camera. The body position state is divided into three types: supine state, prone state and side lying state. Each body position state is pre-set Different suffocation thresholds are associated, and the target object is a newborn; the suffocation threshold includes a suffocation amplitude threshold and/or a suffocation frequency threshold;

    Collect the fluctuation data of the target part of the target object through a radar sensor, and obtain the respiratory amplitude and/or respiratory frequency of the fluctuation of the target part caused by breathing based on the fluctuation data;

    Determine whether the respiratory amplitude and/or respiratory frequency are lower than the apnea threshold corresponding to the posture state, and if so, output an apnea alarm message, or determine whether the respiratory amplitude and/or respiratory frequency are lower than the apnea threshold corresponding to the posture state, and the The duration of the breathing amplitude and/or breathing frequency lower than the apnea threshold corresponding to the posture state is greater than the preset time threshold, and if so, an apnea alarm message is output.
30. A non-contact physiological sign monitoring method for newborns, which is characterized by including:

    Use a camera to capture one or more frames of images of a target object located in the box, and the target object is a newborn;

    Collect the fluctuation data of the target part of the target object through a radar sensor arranged on the box, and obtain the respiratory amplitude and/or respiratory frequency of the fluctuation of the target part caused by breathing based on the fluctuation data;

    The information of the target object and/or the environment information is obtained through the one or more frames of images, and the alarm strategy corresponding to the breathing amplitude and/or the breathing frequency is determined based on the information of the target object and/or the environment information.
31. The method of claim 30, further comprising:

    The respiratory amplitude and/or respiratory frequency are pre-associated with an alarm scheme corresponding to the initial alarm condition;

    Determining the alarm strategy corresponding to the respiratory amplitude and/or respiratory frequency based on the target object's information and/or environmental information includes:

    When the breathing amplitude and/or breathing frequency satisfy the initial alarm condition, and the information of the target object and/or the environmental information satisfy the preset conditions, the alarm scheme is adjusted to obtain the alarm strategy.
32. The method of claim 30, further comprising:

    Measuring the blood oxygen saturation of the target object and obtaining the skin color of the target object through the one or more frames of images;

    When the breathing amplitude and/or the breathing frequency meet the initial alarm condition, and the skin color and the blood oxygen saturation meet the preset conditions, the alarm strategy corresponding to the breathing amplitude and/or the breathing frequency is adjusted.
33. A non-contact physiological sign monitoring method, characterized by including:

    Take one or more frames of images of the target object through the camera;

    Collect the fluctuation data of the target part of the target object through a radar sensor, and obtain the respiratory amplitude and/or respiratory frequency of the fluctuation of the target part caused by breathing based on the fluctuation data;

    The information of the target object and/or the environment information is obtained through the one or more frames of images, and the alarm strategy corresponding to the breathing amplitude and/or the breathing frequency is determined based on the information of the target object and/or the environment information.
34. The method according to claim 33, wherein the breathing amplitude and/or breathing frequency are pre-associated with an alarm scheme corresponding to the initial alarm condition; and the method determined according to the information of the target object and/or the environmental information. Alarm strategies corresponding to the above respiratory amplitude and/or respiratory frequency include:

    When the breathing amplitude and/or breathing frequency satisfy the initial alarm condition, and the information of the target object and/or the environmental information satisfy the preset conditions, the alarm scheme is adjusted to obtain the alarm strategy.
35. The method of claim 33, wherein determining the alarm strategy corresponding to the respiratory amplitude and/or respiratory frequency based on the target object's information and/or environmental information includes:

    When the breathing amplitude and/or the breathing frequency meet the initial alarm condition, and the information of the target object and/or the environmental information do not meet the preset conditions, a first alarm is pre-associated with the breathing amplitude and/or the breathing frequency. The plan serves as the alarm strategy;

    When the breathing amplitude and/or the breathing frequency meet the initial alarm condition, and the information of the target object and/or the environmental information meet the preset conditions, a second alarm scheme that pre-associates the breathing amplitude and/or the breathing frequency. As the alarm strategy; the first alarm scheme and the second alarm scheme are different.
36. A computer-readable storage medium, characterized in that a program is stored on the medium, and the program can be executed by a processor to implement the method according to any one of claims 29-35.