Infant feeding monitoring system

Abstract

A monitoring system for monitoring an infant during bottle feeding is provided. Based on bottle orientation information and motion information about a feeding bottle during feeding, infant orientation information about the infant is obtained.

Description

Technical Field

[0001] This invention relates to feeding bottles, and more particularly to a system for monitoring an infant's orientation when drinking from a feeding bottle. Background Technology

[0002] When bottle-feeding an infant, it is desirable to know how well the infant is drinking. Monitoring drinking performance and providing feedback to parents is known. One known example is a sleeve for infant feeding bottles that includes a pressure-sensing element to measure the weight of milk contained in the bottle before and after feeding, thereby calculating the amount of milk consumed by the infant. The sleeve also includes an accelerometer to provide feedback to parents about the correct bottle angle, and to monitor the child's drinking behavior by observing bottle movement (e.g., to identify drinking bursts and pauses). The system also allows data to be sent to an accompanying application for analysis and visualization.

[0003] The quality of bottle feeding is affected by the baby's position or orientation while drinking. For example, it is recommended to keep the baby relatively upright. This will prevent milk from flowing into the inner ear, where it can cause infection. The baby's position can also affect the chances of excessive air intake, the chances of reflux, and the risk of choking.

[0004] Therefore, automatically determining the baby's position during feeding is of interest. This information can be used to provide feedback during feeding on whether the position is appropriate, and if so, to suggest repositioning (e.g., if the baby is restless). Tracking the baby's orientation over time (optionally combined with other feeding cues) can help parents understand what is most comfortable for their child. Camera systems can be used for orientation detection, but this is not a convenient or desirable solution.

[0005] One parameter of interest is the infant's tilt angle about the vertical axis. This tilt can occur when the infant leans against the parent's chest while drinking. However, estimating the infant's tilt angle from an accelerometer coupled to the bottle presents several challenges, particularly due to the rotational degrees of freedom in the system.

[0006] Another parameter of interest is the angle between the baby and the horizontal, i.e., how upright the baby is sitting during feeding. The bottle angle is the same as the tilt angle if the bottle's longitudinal axis is perpendicular to the baby's longitudinal body axis (i.e., an approximate straight line representation of the spine). However, in reality, these axes are not perfectly perpendicular to each other.

[0007] We also hope to be able to monitor one or two of these body orientation parameters accurately in real time during feeding.

[0008] CN 110339067 discloses a smart bottle base that monitors the position and movement of the feeding bottle to provide feeding analysis and feeding measurement. Summary of the Invention

[0009] This invention is defined by the claims.

[0010] According to an example of one aspect of the present invention, a monitoring system for determining infant orientation information during bottle feeding is provided, comprising:

[0011] A sensor arrangement for acquiring bottle orientation and movement information about the feeding bottle during feeding; and

[0012] A processor adapted to identify infant orientation information about the infant from signals from the sensor arrangement; and

[0013] The output interface is used to provide output information based on the baby's orientation information.

[0014] The output information includes the tilt angle of the baby's body axis relative to the vertical and / or horizontal, and thus in real-world 3D coordinate space.

[0015] This invention is based on the understanding that, although an infant's orientation is not directly associated with (and therefore readily derived from) the orientation of the bottle from which they are fed, the infant's orientation can be derived from a combination of orientation and motion information associated with the bottle. Specifically, the process of drinking from the bottle causes the bottle to move in a direction that depends on the infant's orientation and the infant's orientation relative to the bottle. Therefore, orientation and motion information associated with the bottle can be used to obtain information about the infant.

[0016] Sensor arrangements may include, for example, three-axis motion sensors, such as a three-axis accelerometer and / or a three-axis gyroscope.

[0017] The output interface may include a wireless transmitter for sending output information to a remote device to be presented to the user. This information could be, for example, objective feedback about the baby's location and could be used to help parents understand what is most comfortable for their child.

[0018] In the first example, the processor can be adapted to recognize infant orientation information, which includes the tilt angle of the infant's body axis about a vertical axis. This is the infant's left-right tilt. The body axis is a general axial representation of the infant's body, such as being aligned end-to-end with the spine.

[0019] To address this, the processor can be adapted to transform the sensor placement signals to a reference coordinate system to compensate for rotation about the bottle's longitudinal axis during feeding. If biaxial acceleration is monitored in the plane at the bottom of the bottle, it's initially unknown how the bottle is being held. Specifically, the rotational position about the bottle's longitudinal axis is unknown. This ambiguity is resolved by transforming to a reference coordinate system.

[0020] The processor can then be adapted to identify motion components corresponding to longitudinal movements parallel to the infant's body axis caused by jaw movements, thereby determining the orientation of the infant's body axis (relative to a known reference coordinate system). These longitudinal movements can be detected and then interpreted using the reference coordinate system.

[0021] The processor is adapted, for example, to identify motion components by finding the minimum correlation between motion components in orthogonal orientations in a plane perpendicular to the longitudinal axis of the bottle. By finding the orientations with the lowest correlation, the orientations of the principal orientations of these motion components (within the reference coordinate system) are then determined. This subsequently identifies the baby's orientation in the reference coordinate system.

[0022] In another example, the processor is adapted to identify infant orientation information, which includes the tilt angle of the infant's body axis about a horizontal axis. This is the infant's forward or backward tilt. This is typically close to being orthogonal to the infant's body axis. However, for greater precision, the processor is preferably adapted to identify the offset between the bottle's longitudinal axis and an axis perpendicular to the infant's body axis.

[0023] The processor can be adapted to identify offsets by using a regression model that simulates how the bottle's motion changes during feeding based on the offset angle.

[0024] The monitoring system is, for example, arranged as a sleeve to be installed around the feeding bottle.

[0025] The present invention also provides a computer-implemented method for determining infant orientation information during bottle feeding, the method comprising:

[0026] Obtain information about the orientation and movement of the feeding bottle during feeding;

[0027] Identifying infant orientation information about the infant from bottle orientation and motion information; and

[0028] Provides output information that depends on the baby's orientation.

[0029] The output information includes the tilt angle of the baby's body axis relative to the vertical and / or horizontal, and thus in real-world 3D coordinate space.

[0030] The method may include:

[0031] Identify infant orientation information, which includes the tilt angle of the infant's body axis about the vertical axis; and / or

[0032] Identify infant orientation information, which includes the tilt angle of the infant's body axis about the horizontal axis.

[0033] The present invention also provides a computer program including computer program code components, wherein when the program is run on a computer, the computer program code components are adapted to implement the above-described method.

[0034] These and other aspects of the invention will become apparent from the embodiments described below. Attached Figure Description

[0035] To better understand the invention and to more clearly illustrate how to implement it, reference will now be made to the accompanying drawings by way of example only, wherein:

[0036] Figure 1 A feeding bottle is shown, which is mounted in a sleeve that serves as a monitoring system;

[0037] Figure 2 A set of possible signals from a combination of a three-axis accelerometer and a three-axis gyroscope is shown;

[0038] Figure 3 The images show front views of the baby in a vertical position (left image) and in an angled position (right image);

[0039] Figure 4 The bottom of the bottle is shown, and the x and y axes are indicated in a plane perpendicular to the longitudinal axis of the bottle.

[0040] Figure 5 Used to explain coordinate system translation;

[0041] Figure 6 An example of actual feeding of accelerometer signals is shown;

[0042] Figure 7 It shows how drinking causes repetitive bottle movements;

[0043] Figure 8 A graph showing the correlation versus matrix rotation from real feeding is shown;

[0044] Figure 9 A scatter plot showing the estimated tilt versus the actual applied tilt is presented;

[0045] Figure 10 This represents the tilt angle α and the corresponding bottle angle β;

[0046] Figure 11 This demonstrates how to use the gravity vector to determine the angle of the bottle;

[0047] Figure 12 An example is shown where the bottle is placed at a slightly more upright angle (by angle δ) relative to the baby.

[0048] Figure 13 The test results are shown to illustrate the procedure for determining the tilt angle;

[0049] Figure 14 The processed acceleration and gyroscope signals are shown, in which the gravitational effect has been filtered out;

[0050] Figure 15 The contributions of accelerometer and gyroscope signals are presented in statistical form; and

[0051] Figure 16 The results of determining the tilt angle for the test data are shown. Detailed Implementation

[0052] The invention will be described with reference to the accompanying drawings.

[0053] It should be understood that while the detailed description and specific examples indicate exemplary embodiments of the apparatus, system, and method, they are for illustrative purposes only and are not intended to limit the scope of the invention. These and other features, aspects, and advantages of the apparatus, system, and method of the present invention will become more readily apparent from the following description, the appended claims, and the accompanying drawings. It should be understood that the drawings are merely schematic and not drawn to scale. It should also be understood that the same reference numerals are used in all the drawings to denote the same or similar parts.

[0054] This invention provides a monitoring system for monitoring an infant during bottle feeding. Based on bottle orientation information and motion information about the bottle during feeding, orientation information about the infant (referred to herein as "infant orientation information") is obtained.

[0055] Figure 1 A feeding bottle 10 is shown mounted in a sleeve 12, which serves as a monitoring system. The sleeve 12 surrounds the feeding bottle 10.

[0056] In this example, the monitoring unit 16 is disposed in the base of the sleeve 12 and includes a sensing device 18 for sensing bottle orientation and movement (of the bottle coupled to the monitoring unit), and an output interface 20. The monitoring unit 16 can be incorporated anywhere within or on the sleeve.

[0057] The base of the sleeve may include, for example, a battery, and optionally, a device for providing visual feedback to the user via an LED. The output interface 20 may include such an LED device. However, a preferred implementation alternatively (or additionally) has an output interface that wirelessly transmits the results to the illustrated mobile phone 24 or tablet computer.

[0058] Processor 22 determines the infant orientation relevant to the infant and then provides information to the user. This information may be the infant orientation itself, or it may be suggestions or information derived from knowledge of the infant orientation.

[0059] In the example shown, processor 22 is the processor of mobile phone 24 that wirelessly communicates with monitoring unit 16. Therefore, the sleeve locally detects movement, and the remote processor analyzes the movement to derive the baby's orientation. Thus, parents feeding their baby can monitor feeding-related information, particularly the baby's orientation, on their mobile phones. This is, of course, just an example. Processor 22, which analyzes the movement data, could also be located on the sleeve and integrated with monitoring unit 16. In this case, only the baby's orientation information needs to be sent to the mobile phone. This eliminates the need to send raw motion data to the phone, saving time and battery life.

[0060] The sensing device 18 preferably includes a 3-axis accelerometer and / or a 3-axis gyroscope.

[0061] Figure 2 A set of possible signals from a combination of a 3-axis accelerometer and a 3-axis gyroscope is shown. An image of the feeding bottle 10 is shown to illustrate the 3-axis orientation.

[0062] This arrangement provides three linear acceleration signals Xacc, Yacc, and Zacc, and three angular velocity signals Xgyro, Ygyro, and Zgyro. The sensor arrangement is typically an inertial measurement unit and / or a force or acceleration measurement unit.

[0063] Different implementations of this system may use only an accelerometer, or only a gyroscope sensor, or both.

[0064] The processor 22 is typically programmed to determine infant orientation information about the infant during feeding.

[0065] The following explains two types of infant orientation information. The first type of infant orientation information is the tilt angle with respect to the vertical axis.

[0066] Figure 3 The image shows a front view of an infant in an upright position (left image) and an inclined position at an angle α relative to the vertical orientation (right image). The infant may be tilted while leaning against the parent's chest while drinking.

[0067] By determining this tilt angle, information and guidance can be provided to mothers regarding the baby's positioning, and it can also be used to enhance detailed analysis of the baby's drinking behavior over time.

[0068] Due to the several rotational degrees of freedom in the system, estimating the baby's tilt angle using an accelerometer presents several challenges. First, there is no unique mapping from the bottle's position to the baby's tilt angle. For example, when looking at... Figure 3 When the bottles are in the same position, the two babies have different tilt angles.

[0069] Second, the orientation of the accelerometer's axis is not uniquely defined. Sleeve 12 is a more or less rotationally symmetric circular object, thus it can rotate freely about the bottle (and therefore the x and y axes can rotate freely about the z-axis). Parents can attach the sleeve to the bottle differently each feeding. Furthermore, even if only one way the sleeve is attached to the bottle, the bottle-sleeve combination is more or less rotationally symmetric. Parents can hold the bottle differently during each feeding, and the bottle can rotate about its longitudinal axis (z-axis) during feeding. This was frequently observed in the data obtained by the applicant during the family study.

[0070] It should be noted that this applies not only to the sleeve, but to any installation where the sensor is arranged; the bottle can always be held in a different rotational position.

[0071] The present invention is based on the understanding that motion information can be used to determine the infant’s orientation, including the tilt angle.

[0072] Figure 4 The bottom of the bottle is shown, and the x and y axes are indicated, which lie in a plane perpendicular to the longitudinal axis of the bottle (e.g., the bottom of the bottle). Figure 4 As shown, the x and y axes can be freely rotated from x, y to x', y' around the longitudinal axis of the bottle. The orientation of the bottle can be changed between feedings or during feeding.

[0073] The first part of the process involves translation to a fixed orientation system, where the y-values ​​of the translation are vertically aligned and the x-values ​​are horizontally aligned. This transforms the sensor arrangement signals into a reference coordinate system to compensate for rotation about the bottle's longitudinal axis during feeding.

[0074] Figure 5 This is used to explain the translation of the coordinate system. In this example, the x-axis and y-axis are rotated counterclockwise by an angle β relative to the desired horizontal and vertical axes. Gravitational acceleration generates signals Sx and Sy in the x and y directions, respectively. It is necessary to determine the angle between x and the desired reference x-axis. Figure 5 The angle β in the equation can be determined using the gravitational acceleration g measured by an accelerometer. The gravity vector is offset from the Sx signal by an angle α. Based on trigonometric rules, the angle β can be determined as follows:

[0075]

[0076] The components Sx and Sy can be obtained through low-pass filtering (because the gravity vector is constant), thus excluding the baby's motion when determining the coordinate system translation.

[0077] Gravity will cause the accelerometer axis to deflect. Therefore, the goal is to extract these deflections (which have low frequencies). Faster motion caused by drinking is eliminated due to low-pass filtering.

[0078] Subsequently, the signals in the x and y directions can be rotated so that they reflect horizontal and vertical accelerations due to mapping to the reference coordinate system. The signals can be rotated using the following counter-clockwise rotation matrix:

[0079]

[0080] The rotation angle may need to be adjusted depending on the quadrant in which the x-axis and y-axis are located.

[0081] Figure 6 An example of actual feeding is shown. The top graph shows the three accelerometer signals. The top curve is the x-axis, the middle curve is the z-axis, and the bottom curve is the y-axis.

[0082] The figure below shows the angles of the x and y directions relative to the horizontal axis during feeding. The top curve represents the x-axis angle relative to the horizontal axis, and the bottom curve shows the y-axis angle relative to the horizontal axis.

[0083] As can be seen, the bottle orientation angle can change significantly throughout the feeding process. Therefore, angle determination and axis rotation can be performed for each sample to obtain a fixed reference coordinate system throughout the feeding process.

[0084] The infant's tilt angle can be derived using knowledge of the regular bottle movements induced during feeding. Drinking results in repetitive bottle movements. Sucking behavior involves forward and backward movements of the tongue, which result in movement along the bottle's longitudinal axis (z-axis). Palate movements also cause up and down movements, typically along the infant's longitudinal axis.

[0085] Figure 7 This illustrates how drinking causes these repetitive bottle movements, primarily visible along the bottle's longitudinal axis and the infant's body axis. The infant's longitudinal movement is particularly relevant in determining the aforementioned tilt angle. The infant's body axis corresponds to the vertical axis of rotation for the tilt angle to be determined.

[0086] The way the baby tilts can be understood by considering different scenarios. If the y-axis is aligned with the longitudinal axis of the baby, the movement of the bottle will be clearly visible in that direction, while only a limited amount of movement (mainly noise) will be visible in the vertical direction (x-axis). Therefore, there is no correlation between the x-signal and y-signal. If the y-axis is not aligned with the longitudinal axis of the baby, the movement of the bottle is expected to be visible in both the x- and y-directions, and they will have some correlation.

[0087] This principle can be used to find the tilt angle. Specifically, the components of the motion can be obtained by finding the minimum correlation between the motion components in these orthogonal x and y directions (perpendicular to the longitudinal axis of the bottle). By applying different rotation matrices to the acceleration signal, the rotation angle where the correlation coefficient becomes minimum (or zero) is obtained, and the baby's tilt angle (relative to a now-known reference coordinate system) can be determined from this.

[0088] Figure 8 A graph showing the correlation with the matrix rotation angle from the actual feeding is shown. The correlation between the time series signals of the x and y accelerometer signals is plotted for different rotations.

[0089] The amount of data required to establish a correlation is related to the sucking frequency, typically between 1 and 2 Hz. Sufficient variation in the data is needed to detect the correlation. After a few sucks from the infant, movement along the infant's longitudinal axis can be captured, and then the correlation can be detected. Therefore, a few seconds of data during the infant's drinking is sufficient. During long interruptions, it is, of course, impossible to update the tilt angle because there is no motion information.

[0090] Zero-correlation rotation provides an estimate of the baby's tilt angle. The figure shows the results of a real-world feeding at a tilt angle of 24 degrees.

[0091] Experiments were conducted in a laboratory setting to test the tilt estimation method described above. In these experiments, a bottle sleeve was attached to a retainer system representing an infant. A tilt angle was applied to the retainer system, which then caused the bottle to move to mimic sucking behavior. The algorithm was then applied to the measured data to estimate the applied tilt angle. The experiments were repeated for different tilt angles (and different bottle angles).

[0092] Figure 9 A scatter plot showing the estimated tilt versus the actual applied tilt is presented. Circles correspond to measurements where the infant's head and body are in the same plane, i.e., the head is not rotated laterally. These estimates are fairly close to the actual tilt angle. In cases where the head is rotated laterally relative to the body, the head is in a different tilt position, which introduces error. Crosses correspond to measurements with the head rotated 45 degrees, resulting in an underestimation of the tilt angle. However, a strong correlation still exists between the applied tilt angle and the estimated tilt angle.

[0093] Therefore, the estimated tilt angle usually matches the imposed tilt angle well. There are two conditions under which a less accurate estimate is obtained. First, if the bottle is almost vertical, an inaccurate estimate is obtained. This is because the x and y axes experience the same gravitational components. However, this would mean the baby is lying flat, which is an uncommon drinking position. As mentioned above, another source of noise is introduced when the baby's head turns laterally relative to the body, because the bottle then moves in a different plane than the body, leading to an underestimation or overestimation of the body tilt angle.

[0094] The example above estimates type 1 infant orientation information; the infant is tilted at a vertical angle. Type orientation information of interest is the infant's vertical position angle (tilt relative to the horizontal plane) during bottle feeding.

[0095] Figure 10 This represents the tilt angle α and the corresponding bottle angle β. Estimating the tilt angle and inclination angle, for example, provides a more complete picture of the baby's position / orientation during feeding.

[0096] Determining how upright an infant should be during feeding is also challenging. If the bottle's longitudinal axis is perpendicular to the infant's body axis, then the angle β of the bottle relative to the vertical orientation is equal to the tilt angle α. However, in reality, these axes are not perfectly perpendicular to each other. Variations in how the bottle is placed in the infant's mouth will introduce underestimation or overestimation of the tilt angle.

[0097] Figure 11 This illustrates how the bottle angle β can be determined using the gravity vector. Depending on the bottle angle, the acceleration caused by gravity will be distributed differently across the three components of the accelerometer. Based on the trigonometric rules, the bottle angle can be calculated as follows:

[0098]

[0099] As stated above, if the longitudinal axis of the bottle is perpendicular to the baby's body axis, then the bottle angle β is equal to the tilt angle α. However, it should be noted that this only holds true if the baby is not rotating around its body axis (i.e., not rotating laterally). Lateral rotation of the baby will cause a slight overestimation of the tilt angle (e.g., ≈ 1.5% for a 10-degree rotation).

[0100] In reality, there will be no perfect vertical alignment, which will introduce an underestimation or overestimation of the tilt angle.

[0101] Figure 12An example is shown where the bottle is placed at a slightly more upright angle (by an angle δ) relative to the baby. In this case, describing the tilt angle as a function of the bottle's angle would lead to an underestimation. Instead, the tilt angle α is given by the sum of β and δ. The challenge is determining δ, as it cannot be directly derived from the accelerometer.

[0102] To determine the tilt angle, the angle δ is estimated by analyzing the bottle motion during feeding caused by the infant's sucking behavior, and thus the total tilt angle is estimated. This method is based on the understanding that the relative magnitudes of linear acceleration and angular velocity change in three different directions when the bottle is positioned in the infant's mouth at different angles. Therefore, features can be derived from the motion data to estimate the angle δ. Regression-like modeling techniques can be used to define the relationship between the motion features and the angle δ.

[0103] To demonstrate the feasibility of the method, a test was conducted in which the bottle nipple was repeatedly moved to simulate an infant's sucking behavior. The test consisted of three phases, during which the bottle angle changed while the tilt angle remained constant.

[0104] Figure 13 The results are shown. Figure 13 The top plot shows the acceleration along all three axes during the three phases. The top curve is the z-axis, the middle curve is the y-axis, and the bottom curve is the x-axis. The bottom plot shows the calculated bottle angle.

[0105] Based on the acceleration data, the angle of the bottle was calculated, as shown in the figure below. The first stage ended at approximately t = 105s, and the second stage ended at approximately t = 150s.

[0106] Figures 14 to 16 It shows how to calculate the tilt angle. Figure 14 The diagram shows processed acceleration and gyroscope signals, with gravitational effects filtered out. Several motion components change during different phases. The top plot shows the three-axis acceleration signals; they often overlap. The bottom plot shows the three-axis gyroscope signals, which also often overlap.

[0107] Figure 15 The statistical form shows the Figure 14 Analysis of the data. For each of the three phases 1, 2, and 3, accelerometer (top three bars) and gyroscope (bottom three bars) signals are shown, where the mean and standard deviation (as a percentage) of the relative contribution of each motion component are displayed. Thus, the height of the bar represents the mean of that axis relative to the other axes, while the boundary bars represent the standard deviation.

[0108] For example, a gradient can be observed in Yacc, Zacc, and Xgyro. These features can be combined with precisely measured bottle angles to develop regression models that map these inputs to the correct baby tilt angle.

[0109] Figure 16 The results of the test data are shown. Line 160 shows the actual bottle angle, and the gray area represents the time periods for which the tilt angle was set for different stages. Curve 162 shows the predicted tilt angle of the baby obtained through linear regression. The predicted baby angle clearly adapts to the changes at each stage and therefore deviates from the bottle angle. If the tilt angle is determined solely based on the bottle angle, it leads to a significant underestimation.

[0110] The obtained tilt angle, tilt angle, or both information can be used to provide parents with feedback on the baby's position. Optionally, other feeding cues can also be collected, such as whether feeding is restless or steady, whether the baby is having cramps, etc., which will provide an opportunity to identify the optimal conditions for successful feeding.

[0111] The example above shows a sleeve, but any sensing device fixed in position relative to the bottle can be used. It can be integrated into the bottle or cap.

[0112] By studying the accompanying drawings, the disclosure, and the appended claims, those skilled in the art can understand and implement variations of the disclosed embodiments in practicing the claimed invention. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality.

[0113] A single processor or other unit can implement the functions described in the claims.

[0114] The fact that certain measures are described in mutually different dependent claims does not imply that combinations of these measures cannot be used advantageously.

[0115] Computer programs can be stored / distributed on suitable media, such as optical or solid-state media provided with or as part of other hardware, but can also be distributed in other forms, such as via the Internet or other wired or wireless telecommunications systems.

[0116] If the term “suitable” is used in the claims or specification, it should be noted that the term “suitable” is intended to be equivalent to the term “configured as”.

[0117] Any reference numerals in the claims should not be construed as limiting the scope.

Claims

1. A monitoring system for determining infant orientation information during bottle feeding, comprising: Sensors are arranged to obtain information about the orientation and movement of the feeding bottle during feeding; as well as The processor (22) is adapted to identify infant orientation information about the infant from the signals arranged by the sensors; as well as Output interface (20) for providing output information dependent on the infant orientation information, characterized in that the processor (22) is adapted to identify the infant orientation information, the infant orientation information including: (i) the angle of inclination of the infant's body axis about the vertical axis; or (ii) The angle of inclination of the infant's body axis about the horizontal axis; or (iii) The angle of inclination of the infant's body axis about the vertical axis and the angle of inclination of the infant's body axis about the horizontal axis.

2. The monitoring system according to claim 1, wherein the sensor arrangement (18) includes a triaxial motion sensor.

3. The monitoring system according to claim 2, wherein the sensor arrangement (18) includes a triaxial accelerometer and / or a triaxial gyroscope.

4. The monitoring system according to any one of claims 1 to 3, wherein the output interface (20) includes a wireless transmitter for transmitting the output information to a remote device (24) for presentation to a user.

5. The monitoring system according to any one of claims 1 to 3, wherein the processor (22) is adapted to convert the sensor arrangement signal to a reference coordinate system to compensate for rotation about the longitudinal axis of the bottle during feeding.

6. The monitoring system according to claim 5, wherein the processor (22) is adapted to identify motion components corresponding to longitudinal movements parallel to the infant's body axis caused by jaw movements, and thereby determine the orientation of the infant's body axis.

7. The monitoring system according to claim 6, wherein the processor (22) is adapted to identify the motion components by finding the minimum correlation between motion components in orthogonal directions in a plane perpendicular to the longitudinal axis of the bottle.

8. The monitoring system of claim 1, option (ii), wherein the processor (22) is adapted to identify the offset between the longitudinal axis of the bottle and the axis perpendicular to the body axis of the infant.

9. The monitoring system of claim 8, wherein the processor (22) is adapted to identify the offset by using a regression model, the regression model modeling the bottle motion changes during feeding based on the offset.

10. The monitoring system (12) according to any one of claims 1 to 3 and 6 to 9 is arranged for mounting a sleeve around the feeding bottle.

11. A computer-implemented method for determining infant orientation information during bottle feeding, comprising: Sensors are used to obtain information about the orientation and movement of the feeding bottle during feeding. as well as Infant orientation information about the infant is identified from the bottle orientation information and motion information; as well as The output interface provides output information dependent on the infant orientation information, characterized in that the infant orientation information includes: (i) the angle of inclination of the infant's body axis about the vertical axis; or (ii) The angle of inclination of the infant's body axis about the horizontal axis; or (iii) The angle of inclination of the infant's body axis about the vertical axis and the angle of inclination of the infant's body axis about the horizontal axis.

12. The method of claim 11, comprising: Identify infant orientation information, which includes the tilt angle of the infant's body axis about the vertical axis; and / or Identify infant orientation information, which includes the tilt angle of the infant's body axis about the horizontal axis.

13. A computer program product including computer program code components, wherein when the program is run on a computer, the computer program code components are adapted to control the system according to claim 1 to implement the method according to claim 11 or 12.

Images