DEVICE FOR DETERMINING THE PHYSIOLOGICAL CONDITION OF BABIES AND TODDLERS

Description

Area

[0001] The present invention relates to a device for determining the physiological state of a baby or toddler, which is worn on the child's body and which contains several sensors. State of the art

[0002] Various devices and sensor systems for monitoring the vital parameters of newborns and infants are known in the state of the art. In a dedicated medical context, e.g., in a neonatal unit or intensive care unit of a hospital, these devices are often characterized by high precision in detecting particularly life-threatening physiological conditions. However, the sensor systems used for this purpose are often very expensive, and it takes time, experience, and often a multitude of complex and large pieces of equipment to correctly attach the sensors to the child's body. These devices, designed for a neonatal unit, are therefore not suitable for parents monitoring the physiological conditions of children in a home environment.

[0003] In the home environment, there are now several offers for parents to be able to monitor at least some of the physiological conditions of their children using relatively simple means.

[0004] Patent application US2016324466A1 describes a method, device, and system for local and environmental monitoring of risk factors for sudden infant death syndrome (SIDS). The device is used to monitor the sleeping environment of newborns and infants at home by a parent or other caregiver. The device is placed near the infant's face and monitors, for example, the CO2 content of exhaled air and the infant's sleeping position. In particular, sleeping position and the possible covering of the head by bedding are considered risk factors, as these can block the airways and impair breathing. Blood parameter values ​​are not monitored.

[0005] Patent application US2018000405A1 discloses a system and methods for health monitoring. The system records various vital parameters of the mother during the postpartum period as well as various parameters of the newborn, such as fetal heart rate and oxygenation. However, it does not describe the use of machine learning methods to predict physiological parameters, in particular, not to detect an increased risk of sudden infant death syndrome or to detect feelings of hunger.

[0006] Patent application US 2020 / 0060590 A1 describes a baby monitor consisting of a sensor unit and a receiving unit. The sensor unit contains various sensors, a processing unit, and a transmitting unit. The processing unit processes the raw data measured by the sensors, in particular, it formats it. The transmitting unit sends the formatted data to the receiving unit. The sensor unit is attached to the baby's foot and contains sensors for measuring heart rate, blood oxygen levels, and movement. Heart rate and oxygen levels are measured using pulse oximetry. The receiving unit (but not the sensor unit) analyzes the received data and triggers an alarm if necessary.

[0007] Many current devices for monitoring vital signs in the home environment have various problems. They often contain only a few sensors, as a large number of sensors is often difficult to integrate into clothing or accessories worn by babies and young children due to limited space or area for attaching the sensors. The small number of sensors often also results in a limited data base and poor quality predictions based on it. Incorporating additional sensors would also often significantly increase the cost of the device.

[0008] Another problem with some state-of-the-art devices is that the measurement data alone is often of only limited use to users. A change in breathing rate or a reduced oxygen concentration can have various causes, so these values ​​alone don't allow parents to determine whether a problem exists.

[0009] US patent application US 2006 / 0258921 A1 describes a physiological measurement system that can acquire a pulse oximetry signal, such as a photoplethysmogram, from a patient and then analyze the signal to measure physiological parameters, including respiration, heart rate, oxygen saturation, and movement. The system can be used as a general monitoring system or specifically for apnea in infants or adults and for protection against sudden infant death syndrome. The system comprises a pulse oximeter comprising a light-emitting device and a photodetector that can be applied to a person to receive a pulse oximetry signal; and an analog-to-digital converter device arranged to convert the pulse oximetry signal into a digital pulse oximetry signal.

[0010] Patent application DE 10 2011 077 515 A1 describes a device for electronic body monitoring, particularly for infants. The device comprises a sensor part and a display part. The sensor part has a sensor device for monitoring parameters of an infant, a fixing device for fixing the sensor part to the infant, and a first radio interface. The device comprises an evaluation device for evaluating the sensor signals and triggering an alarm. DE 19708297 A discloses a SIDS (Sudden Infant Death Syndrome) monitor for infants. WO 02 / 05702 A2 discloses a non-invasive sensor for measuring blood glucose and outputting a hunger index determined based on the blood glucose levels. Summary

[0011] The invention is based on the object of providing an improved device for detecting physiological conditions of a baby or toddler, which does not have the above-mentioned problems or does so to a lesser extent.

[0012] The objects underlying the invention are achieved with the features of the independent patent claims. Embodiments of the invention are specified in the dependent claims.

[0013] The invention comprises a device according to claim 1.

[0014] This can be advantageous because the aforementioned parameters have proven particularly predictive of a variety of relevant physiological conditions, including, in particular, physiological conditions associated with an increased risk of sudden infant death syndrome. The evaluation of said parameters is advantageous because it can be performed using sensors that can be attached directly to the body, so that the obtained measured values ​​are less susceptible to the infant's relative movements to external sensors and are less susceptible to various external influencing factors. The applicant has observed that the three aforementioned parameters are highly predictive of an increased risk of sudden infant death syndrome.Although accuracy can be increased further by taking additional parameters into account, the three parameters mentioned above already provide a sufficiently accurate prediction quality to, on the one hand, reliably warn of risk situations relating to SIDS and, on the other hand, to avoid triggering so many false alarms that parents would feel compelled to deactivate the function. External sensors such as external cameras or microphones for monitoring the child's position or breathing have the disadvantage that the child can move out of the sensor range, meaning critical situations may no longer be recorded. Another disadvantage is that setting up the sensor environment is so complex that in many situations, e.g. when on vacation or when the child is lying on the living room sofa and not in the cot, the sensor environment is not available at all. This creates gaps in protection.Because the device is designed as a wearable with the appropriate sensors and evaluation software, the effort involved in setting up the sensor environment is eliminated and it is also impossible for the child to move away from the area monitored by the external sensors.

[0015] The three minimum vital parameters recorded are also comparatively less susceptible to interference: for example, an elevated CO2 concentration in the outside air does not necessarily indicate a child's breathing problems. It is possible that the room air is generally stale. The evaluation of the acoustic signal from external microphones regarding breathing sounds can also be disrupted by background noise, such as renovation work, or by a blanket sliding in front of the microphone. These problems do not exist with the three vital parameters mentioned above.

[0016] In a further advantageous aspect, it is possible to record all three parameters using the same sensor or to derive them from the raw data of a single sensor, e.g. when a photoplethysmographic sensor, referred to here as PPG sensor, is used.

[0017] Embodiments of the invention may make it possible to predict the occurrence of problematic physiological conditions before they actually occur, so that parents or caregivers can take countermeasures in good time.

[0018] According to the invention, the device comprises at least the sensor(s) required for recording or deriving the three vital parameters mentioned. Optionally, the portable device can contain a series of additional sensors for other vital parameters and / or one or more environmental parameters. This means that no cabling is required for the child. Simply putting on or "putting on" the device is sufficient to bring the multitude of sensors into contact with the child's body. This way, the child's natural movement is not hindered by cables, and it is ensured that no gaps in protection arise due to the child being temporarily removed from a "monitored" environment or during a trip.

[0019] Unlike systems that measure the child's vital signs using external sensors, there's no risk of the measurements being distorted by the child's relative movements to the external measuring unit. Because the device is attached to the child's body, it also follows the child's movements.

[0020] In another advantageous aspect, the child's predicted physiological state is output as the result of the prediction or transmitted to the telecommunications device. The telecommunications device can be, for example, a parent's or caregiver's smartphone. Thus, the user does not have to interpret individual physiological parameters, but is directly informed about the child's likely physiological state.

[0021] Additionally or alternatively, some of the data collected or derived by the wearable device, the prediction results or intermediate prediction results may also be transmitted to a server computer system via a network. For example, the server computer system may further process the data received from the wearable device. The further processing may, for example, consist of performing more complex, computationally intensive analyses on the data and / or storing the raw data in a database. The further processing may involve combining the data from the wearable device with data from other external sensors to obtain a final prediction result regarding the at least one physiological state and storing this final prediction result and / or sending it to the parent's telecommunications device via the network.

[0022] In a further advantageous aspect, data processing takes place directly on the device, at least with regard to those physiological conditions that require immediate intervention by the caregivers.

[0023] According to embodiments, the prediction result, optionally supplemented by some of the parameter values ​​(raw data) detected by the sensors, is sent to the telecommunication device only if a current, critical physiological condition has been calculated or an acutely critical vital parameter value or environmental parameter value has been detected or if the caregiver has explicitly requested the data transmission (via pull function) via the telecommunication device.

[0024] This reduces data traffic over the network and can also extend battery life, as preparing the data for transmission and the transmission itself require computing power and therefore energy. Operating the radio module, especially in "normal radiation" mode, also requires energy. Predicting an increased risk of sudden infant death syndrome

[0025] According to the invention, the at least one physiological condition is a condition of increased risk of sudden infant death syndrome. The evaluation software is configured to use at least the heart rate, oxygen saturation, and respiratory rate as input to predict the acute or future presence of an increased risk of sudden infant death syndrome.

[0026] For example, the evaluation software may include a predictive model for predicting sudden infant death syndrome, referred to here as a "SIDS model." A SIDS model is defined here as a predictive model for predicting the increased risk of sudden infant death syndrome. The SIDS model is configured to use at least heart rate, oxygen saturation, and respiratory rate, as well as optionally some other vital and environmental parameters, as input to predict the presence of an increased risk of sudden infant death syndrome.

[0027] The SIDS model is preferably a machine learning-based model, in particular a neural network. However, alternative embodiments are also possible, such as a rule-based system.

[0028] The use of these parameters has the advantage of allowing detection of an increased risk of sudden infant death syndrome (SIDS) with greater sensitivity and specificity than was previously possible in the home device segment. High sensitivity is particularly important here, as sudden infant death syndrome is one of the most common causes of death in babies and young children. High specificity is also very important, as every false alarm is extremely stressful for parents, and an excessively high false alarm rate also carries the risk that an alarm will be ignored in an emergency.

[0029] The increased quality of the prediction is particularly due to the combined evaluation of the aforementioned parameters heart rate, oxygen saturation, and respiratory rate.

[0030] By combining heart rate, oxygen saturation, and respiratory rate, an increased risk of sudden infant death syndrome can be detected earlier, before respiratory arrest or unusual breathing patterns occur. Parents can be alerted sooner, gaining valuable time to prevent sudden infant death syndrome.

[0031] Abnormal breathing (apneas (pauses in breathing), irregularities in breathing rate) indicate an increased risk of SIDS even before hypoxemia develops. As the pathophysiology progresses, bradycardia (low heart rate) may develop. Finally, there is a sharp drop in the oxygen concentration in the blood (hypoxemia), and the child gasps for air. Normally, the autonomic nervous system would detect the oxygen deficiency and counteract it. In sudden infant death syndrome, however, this counter-reaction may not occur for as yet unknown reasons. A lack of maturity of the autonomic nervous system is suspected as a possible cause. This then leads to a further drop in oxygen concentration and SIDS.

[0032] According to embodiments of the invention, the evaluation software is designed to calculate an increased risk for the current or future occurrence of SIDS and to generate different alarm messages (e.g. coded in different colors according to urgency) graded according to urgency and to output them directly or indirectly (via the server computer system) to the telecommunications device of the caregiver: In case of irregular breathing patterns (apnea <ZEITDAUER) und ansonsten normaler Herzfrequenz und Sauerstoffkonzentration: Warnung: Kinderarzt aufsuchen! Bei unregelmäßigen Atemmustern (Apnoe > DURATION) and otherwise normal heart rate and oxygen concentration: Issue an acute alarm, level 1, orange; If irregular breathing patterns (apnea > DURATION) + hypoxia + normal heart rate: Issue an acute alarm, level 1, orange; If irregular breathing patterns (apnea > DURATION) + hypoxia + bradycardia: Issue an acute alarm, level 2, red.

[0033] The DURATION parameter is preferably a value in the range of 12 to 19 seconds, in particular a value of 14 to 17 seconds, e.g. 15 seconds, 16 seconds or 16.5 seconds.

[0034] The combined evaluation of the parameters enables earlier and more reliable warnings. Children at increased risk for SIDS can be identified early, allowing caregivers to recommend appropriate medical examination at an early stage. Since the recorded vital parameters (respiratory rate, heart rate, oxygen saturation) are preferably stored in the wearable device, the caregivers' telecommunications device, or the server computer system, parents can provide the doctor with meaningful long-term data on the child's vital parameters to facilitate diagnosis. In addition, the wearable device can detect exogenous stressors such as thermal stress (from prone positioning or excessively warm ambient temperature), obstruction due to prone positioning, covering the face due to prone positioning, blankets / pillows), so that caregivers can intervene immediately.

[0035] According to another embodiment, the wearable device includes a temperature sensor for measuring the child's skin temperature. Preferably, the wearable device also includes a temperature sensor for measuring the ambient temperature.

[0036] Body temperature in combination with ambient temperature can be used by the evaluation software to increase the accuracy of the prediction of the increased risk of SIDS. The combination of body temperature and ambient temperature allows for at least an approximate derivation of core body temperature. A significantly elevated core body temperature, for example, can indicate heat buildup, which can increase the risk of SIDS. By evaluating skin temperature in combination with ambient temperature, the quality of the prediction regarding the presence of an increased risk of SIDS can be further improved.

[0037] According to some embodiments, the device also contains a humidity sensor, whose measurements are also taken into account in the aforementioned forecast. In high humidity, the child is even less able to compensate for heat buildup through increased perspiration. By taking these risk factors (elevated body temperature, possibly in combination with the ambient temperature and optionally also the humidity level of the air surrounding the device) into account, the quality of the forecast is increased.

[0038] According to some embodiments, the evaluation software is configured to not only generate a prediction result indicating whether there is an increased risk of SIDS, but also to output the relevant risk parameters themselves (e.g., reduced oxygen concentration in the blood, altered heart or respiratory rate, excessively high body or ambient temperature, etc.). This gives parents the opportunity to specifically address the relevant risk factors. For example, the child's lying position can be changed, a blanket removed, or the room temperature lowered by opening windows.

[0039] Preferably, the thermometer for measuring the child's local skin temperature is located at the point where the device is worn, with the thermometer preferably having direct skin contact.

[0040] According to embodiments, the device has a thermometer for measuring a local skin temperature of the child and a thermometer for measuring the ambient temperature.

[0041] For example, the skin temperature sensor can be attached to the inside of a device configured as a band, which is in direct contact with the child's skin. The ambient temperature sensor can be attached to the outside of the band. However, according to some embodiments, the outside temperature sensor can also be configured as an external sensor that transmits the ambient temperature data to the wearable device and / or the server computer system via a base station.

[0042] This can be advantageous, as it further increases the prediction quality. For example, an elevated body temperature measured at the skin is less problematic when the ambient temperature is high, as the latter also directly influences the skin temperature. However, an elevated skin temperature at low outside temperatures is a clear sign of physiological overheating, e.g., due to too many blankets on the child.

[0043] According to embodiments, the evaluation software derives the core body temperature or changes in core body temperature from the measured skin temperature of the child. The body temperature thus derived, along with the ambient temperature, is then passed on as input to the evaluation software for predicting an increased risk of SIDS.

[0044] Methods for deriving core body temperature from skin temperature are known and are described, for example, for adult men in the following publication: Eggenberger P, et al.: "Prediction of Core Body Temperature Based on Skin Temperature", Heat Flux, and Heart Rate Under Different Exercise and Clothing Conditions in the Heat in Young Adult Males. Front Physiol. 2018;9:1780. Published 2018 Dec 10. doi:10.3389 / fphys.2018.01780. A corresponding data set can also be generated for children in which skin temperature, ambient temperature, and core body temperature measured simultaneously under different conditions are linked. By performing a regression analysis on this data, for example, a function specified as a formula or equation or a predictive model based on machine learning can be generated that is capable of deriving core body temperature from skin temperature.The applicant has observed that the use of derived body temperature instead of directly measured skin temperature may further increase the quality of prediction of an increased risk of sudden infant death syndrome, since core body temperature is less affected by environmental disturbance parameters and correlates more strongly with SIDS risks than skin temperature.

[0045] For example, by using the derived core body temperature instead of the skin temperature, the effect of blood centralization during fever as well as a temporal pattern of change in core body temperature can be detected.

[0046] According to embodiments, the evaluation software is also configured to detect the presence of and / or predict the future presence of fever. The evaluation software can use a profile of the change in the derived core body temperature and optionally one or more additional parameters as input to predict the current or future presence of fever. If fever is predicted, a corresponding message (fever alarm) is output directly by the device and / or transmitted to the telecommunications device via the interface.

[0047] Thus, by analyzing the entire data (e.g. heart rate, respiratory rate, blood oxygen concentration, derived core body temperature and optionally also ambient temperature), a better prediction quality with regard to sudden infant death can be achieved.

[0048] According to embodiments, the sensors comprise a photoplethysmographic sensor, referred to here as a PPG sensor. The evaluation software is configured to derive the child's heart rate, oxygen saturation, and respiratory rate from the signals detected by the PPG sensor and to provide them as input to the evaluation software for predicting a current or future increased risk of sudden infant death syndrome.

[0049] The use of a PPG sensor to derive the aforementioned vital signs from the PPG signals can be advantageous for several reasons: Firstly, it saves space, allowing additional sensors to be easily accommodated in the device. Furthermore, the device can be produced more cheaply, is lighter, and is less prone to failure, as it has fewer sensors than would be necessary if a separate sensor had to be installed for each of the parameters. The applicant has observed that the data generated by current PPG sensors contains sufficient information to derive the aforementioned parameters.

[0050] For example, the PPG sensor can be a photoplethysmographic probe with a light-emitting element and a light-detecting element. The light-emitting element can consist of a laser or a combination of several lasers. The spectrum and intensity of the reflected light from each laser provide information about the amount of blood being pumped through the vascular system near the PPG sensor at a specific time, and thus also allows the heart rate to be derived from the raw data. Since inspiration and expiration influence arterial blood flow, the respiratory rate can be derived from the PPG signal. Along with heart rate and oxygen saturation, abnormal respiratory rate is an important prognostic factor for the risk of SIDS.

[0051] The light signals captured by PPG allow fluctuations in the volume of blood transported per unit of time to be detected. Since these fluctuations are influenced by factors such as heart rate and respiration, the analysis software can also detect the heart rate and respiration rate from the PPG sensor data.

[0052] For example, heart rate can be measured or calculated using a PPG sensor as follows: The PPG sensor contains one or more light sources, e.g., LEDs of specific wavelengths, which emit light that passes through the skin and hits (among other things) blood vessels. The light is absorbed, scattered, and reflected by the tissue and the vessels within it. A photodetector measures the intensity of the transmitted or reflected light.

[0053] Because the absorption properties of blood and other tissue components differ, changes in blood vessel volume can be analyzed in the plethysmogram. The wave-like plethysmogram consists of the direct current (DC) and alternating current (AC) components. The DC component depends primarily on the tissue structure and the mean blood volume of the arterial and venous blood. Changes in venous capacitance can be detected as changes in the DC component. The AC component reflects the volume change during systole and diastole of the heart. Based on this pulsatility, the heart rate can be determined.

[0054] Preferably, the PPG sensor is used to record not only the heart rate but also the respiratory rate or to derive it from the raw data, since breathing and the cardiovascular system influence each other.

[0055] Inspiration and expiration lead to fluctuations in arterial and venous blood volume due to changes in intrathoracic pressure. Negative intrathoracic pressure during inspiration decreases venous pressure and increases venous flow to the heart. Systolic blood pressure, in particular, drops and heart rate increases. The opposite effect occurs during expiration.

[0056] These respiration-dependent fluctuations in blood pressure and heart rate lead to fluctuations in blood volume and thus to fluctuations in the measured intensity at the photodetector.

[0057] The PPG sensor can therefore also be used to determine the child’s respiratory rate.

[0058] The derivation of respiratory rate from a PPG signal can be performed, for example, as described in Nilsson LM. Respiration signals from photoplethysmography. Anesth Analg. 2013 Oct;117(4):859-65. doi: 10.1213 / ANE.0b013e31828098b2. Epub 2013 Feb 28. PMID: 23449854. Heart rate can also be derived from PPG data in a similar manner.

[0059] The values ​​can be influenced by other movements of the child, so movement represents a potential source of error. However, by using filters, fluctuations in blood flow caused by a child's movement (other than breathing movement!) can be detected and filtered out. Furthermore, according to embodiments of the invention, the PPG signal is used to determine the child's blood parameters, in particular oxygen saturation and preferably other blood parameters, which can improve the quality / accuracy of predicting an increased risk of sudden infant death syndrome and / or which can be used to predict or detect other physiological conditions.

[0060] The additional blood parameters that can be used to improve the quality / accuracy of the prediction of an increased risk of sudden infant death (thus serving as "control blood parameters") are, in particular, blood parameters that do not correlate with blood oxygen concentration, correlate negatively, or correlate positively with blood oxygen in a known and non-linear manner.

[0061] A blood parameter is a measured value that results from a specific property of the blood, e.g. the concentration of a specific molecule in the blood.

[0062] A blood parameter that is negatively correlated with the blood oxygen concentration is, for example, a blood parameter that decreases in strength with increasing blood oxygen concentration and increases in strength with decreasing blood oxygen concentration, e.g. the CO 2 concentration in the blood.

[0063] A blood parameter that is not correlated with blood oxygen concentration, for example, a blood parameter whose strength is at least approximately independent of the level of blood oxygen concentration. For example, the concentration of carboxyhemoglobin depends essentially on the carbon monoxide concentration in the air, not on the oxygen concentration, since carbon monoxide displaces oxygen from heme.

[0064] In general, however, other blood parameters can also be used as control blood parameters. These are derived from a blood component and have a known non-linear positive correlation with the oxygen concentration, e.g., according to an exponential or polynomial relationship. If, for example, a 30% drop in the oxygen concentration in the blood is detected, and a specific blood component that is known to rise or fall three times as much as the oxygen concentration depending on the oxygen concentration, can also be used as a control parameter. If the measured or derived concentration of this blood component also decreases by 30%, just like the oxygen concentration, then a measurement error can be assumed, e.g., because the PPG sensor, from whose raw data both the oxygen concentration and the control parameter are derived, has slipped.If the said blood component decreases by 90% when the oxygen concentration decreases by 30%, it can be assumed that there is actually a decrease in the oxygen concentration in the blood, since an error in the sensor technology, e.g. due to a lack of contact, is in most cases linearly and equally reflected in the measurements of all measured values ​​of these sensors.

[0065] According to some embodiments, the one or more sensors comprise a sensor for detecting at least one blood parameter of the child, wherein the blood parameter is, for example, a CO 2 concentration in the blood, a methemoglobin concentration and / or a carboxyhemoglobin concentration in the blood of the child.

[0066] For example, the sensor for measuring the blood parameter can be the PPG sensor, which is already used to measure blood oxygen levels, respiration rate, and heart rate. This is advantageous because no additional sensor is required and the same sensor that already measures the oxygen concentration in the blood or derives it from the raw data can be used.

[0067] The evaluation software is designed to use the at least one blood parameter (CO2 concentration, methemoglobin concentration and / or a carboxyhemoglobin concentration in the blood) as an additional input parameter in order to reduce the false positive rate of the prediction of the increased risk of sudden infant death.

[0068] In addition to or alternatively to the three minimum vital parameters mentioned above, namely heart rate, respiratory rate and blood oxygen concentration, one or more of the control blood parameters CO2 concentration, methemoglobin and / or carboxyhemoglobin concentration in the blood can also be used as control parameters in predicting an increased risk of SIDS.

[0069] Carboxyhemoglobin (HbCO) is formed by the reversible binding of carbon monoxide (CO) to the iron ion of the heme group. Carbon monoxide binds to hemoglobin at the same sites as oxygen, but about 200 times more strongly. As a result, HbCO can bind almost no oxygen. Since carbon monoxide does not occur in normal room air, or only in appreciable quantities, the CO concentration can be assumed to remain constant under normal conditions. If, in addition to an oxygen concentration that is too low, a carboxyhemoglobin concentration that is too low is measured, it can be assumed that there is a measurement error. If, on the other hand, the carboxyhemoglobin value remains constant, the evaluation software can assume that the oxygen concentration is indeed low.

[0070] Methemoglobin is a form of hemoglobin that also can no longer transport oxygen. It is formed by the oxidation of divalent iron (Fe2+) in the heme group to trivalent iron (Fe3+). The physiological concentration of methemoglobin in the blood is low at less than 1%, but the concentration can increase due to certain chemical compounds.

[0071] Methemoglobin, like carboxyhemoglobin, is a blood parameter that generally exists in a constant concentration in the blood and can therefore be used as a control parameter. According to embodiments, the methemoglobin level is passed to the evaluation software as an additional input parameter ("control parameter"), so that the software can determine, by comparing the entered blood oxygen concentration with this or other control values, whether a measurement error has occurred or whether the oxygen concentration is actually too low.

[0072] Measuring the concentration of the substances in question in the blood or deriving this concentration from the PPG sensor signals can be advantageous because the said blood parameters can each be used as control parameters to avoid false positive predictions and false alarms. If, for example, an oxygen concentration that is too low is measured, the evaluation software can use one or more of the said control blood parameters to determine whether there is really an increased risk of sudden infant death syndrome or whether a measurement error is the cause of the too low oxygen concentration. Such measurement errors can occur due to movement of the device when the child is moving. If the oxygen concentration in the blood is significantly low, but at the same time the CO2 concentration in the blood is within the normal range or even elevated, it is likely that the oxygen concentration in the blood is actually too low.If the concentration of CO2 (or another control substance such as carboxyhemoglobin or methemoglobin) in the blood is also low, it is likely that a false measurement is the cause. CO2, as a control substance, should rise in the event of a real decrease in oxygen saturation; in the event of a false alarm, CO2 would also decrease, just like O2.

[0073] Because the evaluation software uses and considers additional control parameters as input parameters, false alarms can be avoided, which is particularly important in the context of detecting a life-threatening physiological condition.

[0074] According to embodiments, the device comprises at least one sensor for determining at least one further vital parameter and / or environmental parameter.

[0075] In addition to or alternatively to the sensor for the environmental parameter, the device can also include an interface for receiving the additional vital parameter(s) and / or environmental parameters from one or more external sensors. For example, data can also be determined from another component installed in the room in which the child is located and transmitted to the device on the child's body and / or to the server computer system. The at least one additional environmental parameter can, in particular, be the CO2 concentration of the ambient air or the air humidity. The at least one additional vital parameter can include video data or movement data that characterize the child's movement activity.Acoustic data from a microphone built into the wearable device or configured as an external sensor may also be transmitted to the wearable device and / or the server computer system and used as additional input data in predicting an increased risk of SIDS.

[0076] The evaluation software is designed to use the at least one further vital parameter and / or environmental parameter as an additional input parameter in order to predict the presence of an increased risk of sudden infant death.

[0077] For example, one or more sensors can be attached to the mattress of a child's bed or as a sticker on the duvet cover, pajamas, or sleeping bag. The external sensors can be motion sensors or pressure sensors, for example, which measure the movement of the chest during sleep. Additionally or alternatively, a motion sensor, e.g. a gyroscope, can also be built into the wearable device. According to embodiments, this motion data is also used as a further input parameter by the evaluation software to reduce the false positive rate of the predictions and improve the quality of the prediction: in a child whose chest is moving, this can be an indication that breathing is functioning normally and a reduced oxygen concentration in the blood is probably due to a measurement error.Thus, in particular, the microphone and / or the video camera and their measurement data can be used to reduce the false positive rate of SIDS prediction.

[0078] According to another embodiment, an external or device-internal sensor measures the CO2 concentration of the ambient air. This parameter can be used as an additional input parameter by the evaluation software to increase prediction accuracy. A high CO2 concentration in the ambient air indicates that the ambient air is depleted. If the CO2 level is too high, this indicates unfavorable environmental conditions that can increase the risk of sudden infant death syndrome.

[0079] According to another embodiment, an external or device-internal acoustic sensor (a microphone) detects sounds from the child and the environment (since a microphone detects both the environment and the child's sounds, it is both an environmental sensor and a vital sign sensor). The detected acoustic signal can be used as an additional input parameter to increase prediction accuracy. If the child is crying, there may be other problems, but not oxygen deprivation or an increased risk of sudden infant death syndrome.

[0080] Additionally or alternatively, movement data from the wearable device's accelerometer and / or video data from an external camera aimed at the child can be used to capture the child's movement or movement patterns and provide this movement data as input to the analysis software. A child who is highly active can be assumed to be at no increased risk of sudden infant death syndrome.

[0081] According to one embodiment, the video camera is an infrared camera. This is particularly advantageous because images from an IR camera allow image analysis software to detect whether a child's face, which is usually clearly visible in an IR camera against the background of heat-insulating clothing or blankets, is facing up or down, thus indicating whether the child is in the prone or supine position. A prone position increases the risk of sudden infant death syndrome because the child breathes into the mattress and / or because temperature exchange may be restricted.

[0082] According to one embodiment, the prediction results of the portable device regarding the presence of an increased risk of sudden infant death syndrome are initially transmitted as an intermediate result to the server computer system. The server computer system is operatively linked to the IR camera via a network, e.g., directly or indirectly via a base station. The server computer system receives the IR images of the child from the external camera and evaluates them using image analysis software. The image analysis is comparatively computationally complex, so this analysis is preferably performed on the server rather than on the portable device, which has only limited computing power. The result of the image analysis is whether the child is in the prone or supine position.The server computer system is designed to calculate a final result regarding the presence of an increased risk of sudden infant death from the intermediate result of the portable device and the result of the image analysis and to send it to the caregiver's telecommunication device. Predicting hunger

[0083] According to the invention, at least one of the sensors for the vital parameters is designed to determine the child's blood sugar concentration in a non-invasive manner. The evaluation software is designed to detect a current or future feeling of hunger in the child. The evaluation software is configured to use at least the measured blood sugar concentration as input to predict the current or future presence of a feeling of hunger and / or the time of the onset of the feeling of hunger. The physiological state to be predicted is therefore a state in which the child is hungry. According to the invention, a feeling of hunger is predicted if the current or future blood sugar level is below a predefined limit.

[0084] For example, the evaluation software can be configured to calculate the child's current and / or future blood sugar level as a function of the flow properties of the blood and to predict a current or future feeling of hunger based on the calculated blood sugar level.

[0085] The flow properties of blood depend, among other things, on the blood glucose level. The blood glucose level is approximately proportional to the blood viscosity and inversely proportional to the flow rate. The analysis software can, for example, contain a convolutional neural network that can derive the blood glucose level from the PPG signal.

[0086] For example, the inference can be performed using said networks as described in S. Hossain, B. Debnath, S. Biswas, MJ Al-Hossain, A. Anika and SK Zaman Navid, "Estimation of Blood Glucose from PPG Signal Using Convolutional Neural Network," 2019 IEEE International Conference on Biomedical Engineering, Computer and Information Technology for Health (BECITHCON), 2019, pp. 53-58, doi: 10.1109 / BECITHCON48839.2019.9063187.

[0087] Alternatively, the blood glucose level can also be derived from the PPG sensor signal using a method described in Delbeck S et al.: "Non-invasive monitoring of blood glucose using optical methods for skin spectroscopy—opportunities and recent advances," Anal Bioanal Chem. 2019 Jan;411(1):63-77. doi: 10.1007 / s00216-018-1395-x. Epub 2018 Oct 3. PMID: 30283998. The PPG sensor performs pulsed measurements in the shortwave near-infrared spectral range using LEDs at 935, 950, and 1070 nm. The glucose concentration was predicted using an artificial neural network (ANN) after preprocessing the time-dependent signals with a neural network-based adaptive noise reduction filter (Adaline). After training the neural network, the network is used to predict blood glucose levels based on the spectral data collected by the PPG sensor.

[0088] According to embodiments, the evaluation software may include or be operatively coupled to another neural network or another prediction algorithm, wherein the other neural network or the other prediction algorithm is configured to predict the current or future feeling of hunger as a function of the calculated glucose concentration in the child's blood. For example, the other neural network may also be a convolutional neural network.

[0089] Correctly detecting or predicting a child's hunger early on can be beneficial and important for parents for many reasons: Children are not yet able to express themselves verbally, so it's often impossible for parents to determine whether a child's crying is caused by hunger, injury, illness, or some other reason. Using a wearable device that detects or predicts a child's hunger based on a measured blood glucose level allows parents to better understand their child's needs.

[0090] Another advantage can arise from detecting hunger early, i.e., at a time when the feeling hasn't yet become so strong that the child begins to cry. This can allow parents to prepare food early or, if the parents are traveling with the child, to find a place where the child can be fed early.

[0091] According to embodiments, the sensors comprise a photoplethysmographic sensor, referred to here as a PPG sensor. The evaluation software is configured to derive the child's measured blood glucose concentration from the signals recorded by the PPG sensor, in addition to the child's heart rate, oxygen saturation, and respiratory rate, and to provide at least the blood glucose concentration as input to the evaluation software.

[0092] This has the advantage that the blood sugar concentration can be measured non-invasively and very frequently, e.g. regularly, so that when the blood sugar level drops, the point at which a feeling of hunger occurs or becomes so strong that the child indicates this by crying can be predicted.

[0093] According to some embodiments, the blood parameters used as correction parameters in predicting the increased risk of SIDS are also used to detect erroneous blood glucose measurements.

[0094] According to the invention, the evaluation software is communicatively coupled to an electronic device for food preparation, either directly or via a software application of the telecommunications device or via the server computer system. The evaluation software and optionally also the software application of the telecommunications device are configured to activate the electronic device in response to the prediction that the feeling of hunger will occur now or in the future, thereby causing the electronic device to prepare food for the child.

[0095] For example, the electronic device may be a milk bottle warmer, a kettle, a microwave, or something similar. Further embodiments

[0096] According to embodiments of the invention, the evaluation software can be used to detect and / or predict a variety of physiological conditions. In addition to predicting an increased risk of sudden infant death syndrome and predicting hunger or the time of onset of hunger, the device can be used, for example, to detect fever and various acute or chronic diseases.

[0097] Thus, embodiments of the invention may allow the detection of abnormalities which may, for example, be an indication of congenital diseases, acute or chronic diseases.

[0098] Thanks to the multitude of parameters that can be recorded by the wearable device, a very broad database is created that allows for a high-quality prediction of the child's current and future physiological conditions.

[0099] According to some embodiments, the evaluation software is configured to selectively detect the current or future presence of a physiologically problematic condition in the child that requires immediate intervention. Such a physiological condition may, for example, be an increased risk of sudden infant death syndrome. The evaluation software forwards at least some of the vital signs measured or derived by the portable device or intermediate prediction results to the server computer system via a network without locally calculating a final prediction result, in order to enable the server computer system to predict physiological conditions that do not require immediate intervention.

[0100] This can be advantageous because the computing capabilities of the portable device are limited due to its small size.

[0101] By predicting critical physiological conditions by the device itself and sending non-critical conditions and / or the recorded raw data regularly or in bulk ("bulk upload"), e.g., during a charging process, to the server for server-side analysis, it can be ensured that an alarm signal can always be sent from the wearable device itself to the telecommunications device, regardless of whether a network connection to the server computer system exists. This also ensures that the child is monitored in every situation—whether sleeping on the living room sofa or traveling—with regard to truly critical physiological conditions and parameters (SIDS risk, blood oxygen concentration, etc.), since the core functionality of the wearable wristband is always available, independent of the server computer system and the availability of external sensors.Connecting to the server computer system and / or incorporating additional parameters provided by external sensors can further refine the prediction result. The prediction quality regarding critical system states will therefore be somewhat more accurate in the context of, for example, the child's usual sleeping place, where the camera, microphone, and base station may also be located. However, the child can be spontaneously moved to another location at any time where neither a network connection nor camera surveillance is possible; basic protection remains in place as long as the child attaches the wearable device to their body and the caregivers have instantiated the associated software on the telecommunications device.

[0102] The device's battery life is conserved, allowing it to operate for longer periods without changing or recharging the batteries. Processing the prediction results for transmission requires computing power, for example, for converting the data into the correct format for transmission or for establishing a communication channel. By not sending the measurement results or prediction results for every measurement and subsequent prediction, but rather only sending the prediction results when a critical or problematic physiological condition or parameter value is detected or predicted, computing power is saved.

[0103] Furthermore, according to some embodiments, the data traffic is reduced because the processing of the raw data relevant for the critical conditions already takes place on the portable device, so that instead of the raw data, only the prediction results and optionally a few more raw data relevant for the prediction result need to be transmitted, whereas the entirety of the sensor data or at least the sensor data that serve to predict another physiological condition are preferably transmitted collectively at a later time.

[0104] For example, data transmission to the telecommunications device can occur via a radio signal, in particular via a radio signal according to the Bluetooth protocol. However, it is also possible for the data transmission to occur via WLAN. According to some embodiments, at least some data is first transmitted from the portable device to the server computer system, e.g., directly via a WLAN connection to the Internet, or indirectly via radio or WLAN to a base station, and from there to the server computer system.

[0105] According to embodiments of the invention, the evaluation software is designed to detect a current or future presence of a physiologically problematic condition of the child, when a value of at least one vital parameter or environmental parameter is outside a pre-defined normal range; and when a pattern of values ​​of several vital parameters is detected which indicates a current or future problematic physiological condition of the child, whereby the pattern can also be detected if all vital parameters and / or environmental parameters are individually within their normal range.

[0106] The evaluation software is designed to send a message regarding the predicted problematic physiological condition to the mobile telecommunications device in response to the detection of a current or future physiologically problematic condition.

[0107] Thus, parents are not only warned if, for example, an increased risk of SIDS is detected, but also if, for example, the blood oxygen level has fallen below a minimum value or if the ambient temperature exceeds a predefined maximum value.

[0108] According to embodiments of the invention, the wearable device is a wristband or a band on the child's ankle or leg. This can be advantageous because the child's freedom of movement is not restricted, the device can be securely attached to these limbs, and, above all, because it is possible to adjust the contact pressure, e.g., by using an elastic material in the bands or by adjusting a fastener, so that the sensors rest against the child's body with a certain minimum pressure, which increases the quality of the measurements.

[0109] For example, the band can be designed to be worn on an arm, ankle, or leg with a circumference of approximately 7-15 cm (corresponding to the circumference of the corresponding limbs in babies and toddlers). For example, the band can be 7.5 to 20 cm long, including the closure mechanism.

[0110] According to embodiments, the sensors comprise one or more pressure sensors designed to detect the contact pressure of the device on the child's body. The evaluation software is designed to detect, based on the measured contact pressure, whether the contact pressure is within a predefined permissible contact pressure range, within which the one or more sensors for recording vital signs can function correctly and within which the band will not cause pressure pain in the child. The evaluation software is designed to output a warning to the user via a signal element of the device and / or to the telecommunications device via the interface if the measured contact pressure is outside the permissible contact pressure range.Additionally or alternatively, the evaluation software is designed to prevent the measurement of vital parameters by the one or more sensors until the contact pressure is again within the permissible contact pressure range.

[0111] This can be advantageous because it ensures that the sensors are always in sufficient contact with the body to take meaningful measurements. This reduces the number of false predictions and prevents the recording of measurement data that is meaningless due to the lack of contact with the body.

[0112] The portable device can, for example, send a warning regarding the lack of contact. For example, the warning can be issued directly via a speaker integrated into the portable device or via a light source, such as an LED lamp that illuminates or flashes. Additionally or alternatively, the portable device can also send the warning to software on the portable telecommunications device or to a base station, so that the warning regarding the lack of contact is issued by the telecommunications device and / or the base station.

[0113] By issuing a warning to the user, the wearable device can be repositioned at a suitable location on the child's body.

[0114] According to embodiments of the invention, the portable device is configured to automatically and regularly record the measured vital parameters and optionally also the at least one environmental parameter and evaluate them using the evaluation software. The data can then be collected and uploaded to the server computer, e.g., via push or pull functionality, and / or transmitted to the telecommunications device. For example, the transmission can occur when the portable device is being charged and / or when the caregivers send a request regarding the current data to the portable device via the telecommunications device.

[0115] According to embodiments, the interface for transmitting data to the telecommunications device is an interface for data transmission via a near-field signal, in particular via a radio signal, in particular a Bluetooth interface.

[0116] The portable device is configured to operate in a low-radiation and a normal-radiation operating mode.

[0117] The portable device is configured to operate in low-radiation mode in normal operating mode when no physiological condition is predicted and no vital or environmental parameter is measured that requires immediate intervention. If the evaluation software detects the current or future presence of a physiologically problematic condition, in particular an increased risk of sudden infant death syndrome and / or hunger, or the presence of a vital or environmental parameter in a health-critical value range, the portable device automatically switches to normal radiation mode. Preferably, the device automatically returns to low-radiation mode after transmitting the relevant data regarding the critical physiological condition and / or parameter.

[0118] This not only saves the portable device's battery, but also minimizes radio radiation, which some parents consider problematic.

[0119] Depending on the technology used, the low-radiation operating mode may be implemented slightly differently.

[0120] When using Bluetooth as a near-field communication technology, switching the portable device to low-radiation operating mode can be implemented, for example, as follows: Option 1: Change the "advertising rate" within the advertising operating mode

[0121] In this variant, the wearable device operates in both normal operating mode and low-radiation operating mode in the so-called "advertising" mode. In this operating mode of Bluetooth devices, the device in question is not permanently connected (paired) to other devices. The wearable device and the telecommunications device are therefore not paired in this operating mode, and no transmission of the data collected by the sensors takes place from the wearable device to the telecommunications device. In the "advertising" mode, the wearable device transmits a so-called "advertising" data packet via radio at regular intervals, e.g., every 10 seconds or 1 minute. This packet informs the wearable device that it exists but does not wish to establish a connection with the telecommunications device. The rate at which these "advertising" data packets are sent is referred to as the "advertising rate."A "low-radiation operating mode" is an "advertising" state of a Bluetooth-enabled device in which the advertising rate is below a predefined maximum value, e.g., a maximum of one advertising data packet per minute or a maximum of one advertising data packet per 10 seconds. A "normal radiation operating mode" is an "advertising" state of a Bluetooth-enabled device in which the advertising rate is above the predefined maximum value, e.g., more than one advertising data packet per minute or more than one advertising data packet per 10 seconds.

[0122] Within this "advertising" operating mode, the wearable device is normally in low-radiation mode unless a critical physiological condition is predicted. In this mode, vital signs are stored and analyzed locally in the wearable device, and an advertising data packet is sent at low frequency. This packet essentially only states that the wearable device exists but does not wish to be connected to other devices.

[0123] As soon as the evaluation software of the wearable device predicts a critical physiological condition requiring immediate intervention, the evaluation software increases the frequency of sending the advertising data packets so that the data packet can be transmitted as quickly as possible. Preferably, after transmitting one or more data packets containing an alarm and / or measured values ​​related to the predicted critical physiological condition and / or critical vital or environmental parameters, the wearable device switches from "low-radiation" to "normal radiation" operating mode. After sending the alarm data packet(s), the wearable device and its radio module return to low-radiation operating mode.

[0124] All devices within range of this advertising data packet that have previously connected to the portable device (that were paired with it), in particular all telecommunications devices for which this applies, can receive and further process this data packet and, if necessary, display it to the user on the telecommunications device's display. To provide greater certainty that the alarm message reaches the telecommunications device, the advertising request can also contain the information that the portable device now wishes to connect (pair) with the telecommunications device. Once this connection is established, the portable device can also recognize that the data packet containing the alarm has reached the recipient. Option 2: Changing the feedback rate within a coupled operating mode

[0125] According to this implementation variant, the wearable device is paired to a caregiver's telecommunications device in normal operating mode, thus maintaining an active connection between the wearable device and the telecommunications device. Bluetooth devices in the paired operating state typically send very frequent requests (e.g., approximately 100 times per second) to check whether the connected device is still present and expect confirmation from the connected device that this is the case.

[0126] In low-radiation mode, i.e., when the wearable device does not predict a physiological condition requiring intervention or measures environmental or vital parameters requiring intervention, the wearable device operates in a low-radiation mode in which the wearable device is coupled to the telecommunications device but reports back to the device that it will not respond for a predefined number of further feedback signals (e.g., the next 100 feedback signals). This means that fewer feedback data packets are sent from the wearable device to the telecommunications device. However, if a critical physiological condition, vital parameter, or environmental parameter is predicted or measured, the wearable device immediately sends a feedback signal containing information regarding the intervention-requiring condition to the coupled telecommunications device without waiting for the "cancelled" feedback cycles to expire.The portable communications device initially switches to normal radiation operating mode, as feedback signals are sent at the frequency typical for the paired Bluetooth state. However, as soon as the warning with information regarding the condition requiring intervention has been transmitted to the paired telecommunications device, the portable device automatically returns to low-radiation operating mode by sending a response that it will not respond for a predefined number of further feedback signals. The predefined number of further feedback signals is preferably more than 50, and furthermore preferably more than 100. Option 3: Reducing the transmission power when the connection is good

[0127] According to a third implementation variant, the portable device is paired with the telecommunications device in normal operating mode and continuously determines the quality of the connection. For example, it determines how often an expected feedback message is missed or the signal strength of the telecommunications device's Bluetooth signal.

[0128] If no critical system conditions are currently predicted and no critical environmental or vital parameters have been recorded or calculated, and if the connection quality to the telecommunications device is above a predefined minimum quality level, the wearable device reduces the transmission power of the Bluetooth radio module and thus switches to low-radiation operating mode. If the connection quality is poor, i.e., below the predefined minimum quality level, or if a physiological condition is predicted or an environmental or vital parameter has been recorded that requires immediate intervention, the wearable device's radio module maintains or increases the transmission power of the radio module.

[0129] This can save energy and increase battery life.

[0130] In addition to Bluetooth, other standards and / or protocols can be used for near-field data exchange, e.g. ZigBee.

[0131] When using Bluetooth or ZigBee, for example, the portable device has a radio module ("transmission module"). If there are no abnormalities that require the caregiver's immediate attention, the data is stored locally and the transmission module operates in a low-radiation mode. In this state, the portable device and the receiver device (i.e., the portable telecommunications device and, optionally, the base station) are not continuously synchronized, according to embodiments. If the evaluation software in the portable device determines that an abnormality and / or critical physiological condition exists, the radio module is switched to the radiation-normal operating mode, and data (measured values ​​and / or predicted results) are sent to the receiver device.

[0132] Since the range of a Bluetooth or ZigBee signal is often shorter in many households than, for example, a WLAN signal, WLAN or another suitable internet-based data communication between the evaluation software and the telecommunications device can be implemented as an alternative to Bluetooth or ZigBee.

[0133] According to embodiments, the portable device comprises one or more environmental parameter sensors selected from a group including: a thermometer for measuring the ambient temperature; a measuring device for measuring the ambient humidity; gases, in particular CO2; a microphone for detecting ambient noise and / or noise from the child; a UV sensor for detecting a cumulative UV radiation dose, in particular a daily cumulative UV radiation dose.

[0134] According to embodiments, the sensors for detecting vital parameters comprise further sensors selected from a group comprising: Acceleration sensor, e.g., for detecting the child's position (especially prone or supine position); thermometer for measuring the child's skin temperature; video camera (especially thermal imaging camera for detecting prone or supine position);

[0135] In a further aspect, the invention relates to a system comprising the device and one or more of the following additional components: the portable telecommunications device, wherein user software is instantiated on the portable telecommunications device, wherein the user software is interoperable with the evaluation software and is designed to display the prediction results received from the portable device via the interface to the user and / or to enable the user to configure the evaluation software; and / or the server computer system; and / or a base station to which one or more external sensors for measuring vital parameters of the child or environmental parameters of the child's environment are coupled; the base station is designed to forward the parameter values ​​measured by the external sensors to the server system in original or processed form.For example, the external sensors communicatively coupled or coupleable to the base station can comprise a video camera, in particular a thermal imaging camera, a microphone, an ambient temperature sensor, a sensor for air humidity and / or a sensor for detecting the CO2 concentration of the ambient air. According to some embodiments, the base station contains a module for charging the portable device, e.g. via an induction field. According to some embodiments, the base station contains a network interface, in particular a WIFI interface, for communicatively coupling the base station and / or the external sensors communicatively coupled thereto to the server computer system. The base station also contains software that is interoperable with the evaluation software on the portable device as well as with the server application; and / or one or more of the external sensors, in particular a video camera, in particular a thermal imaging video camera..

[0136] According to embodiments of the invention, the prediction software includes at least one predictive model for predicting the at least one physiological state. The at least one predictive model is a model generated by a machine learning method based on a training data set. In particular, the predictive model can be a neural network. Neural networks have proven particularly suitable for detecting and predicting the relationships between various vital parameters and / or environmental parameters, as well as various physiological states.

[0137] Predictive models based on machine learning allow complex dependencies between parameters, as well as with the physiological state to be predicted, to be identified and taken into account in the prediction. Especially in the field of physiology, vital parameters and environmental parameters often interact in a complex and nonlinear manner, reinforcing or weakening each other. Machine learning methods, such as neural networks, are capable of capturing these complex parameter dependencies and thus using them in predictions, making it possible to predict not only current physiological states, but also likely future states (and possibly even the time of their occurrence).

[0138] In a further aspect, the invention relates to a method for providing a portable device for monitoring the physiological state of a child according to claim 16. The method comprises providing a training data set. The training data set contains multiple data sets. Each data set specifies at least one physiological state of the child, which is stored in a linked manner with vital parameters of the child (in particular heart rate, oxygen saturation, and respiratory rate, possibly also skin temperature or derived core body temperature, movement patterns, video or audio data, etc.). Optionally, the data set can also contain one or more environmental parameters, e.g., ambient temperature, humidity, CO2 concentration of the ambient air, etc.Preferably, the dataset contains a plurality of data values ​​for each of the vital parameters and / or environmental parameters, each of which is stored linked to a timestamp, with the physiological state also being stored linked to a timestamp. This allows for the detection of not only correlations between multiple parameters and physiological states, but also their temporal dependencies.

[0139] The method further comprises performing a machine learning process on the training data to generate the at least one predictive model. The at least one predictive model is configured to predict the current or future physiological state of the child based on at least the heart rate, oxygen saturation, and respiratory rate, as well as optionally other vital parameters and / or environmental parameters. According to some embodiments, the predictive model also learns temporal dependencies, so that it is also capable of predicting the time of onset of the physiological state for a given set of parameter values.

[0140] The method further comprises installing evaluation software containing the at least one predictive model on the wearable device. The device is designed to be worn on a child's body, with the device being sized and shaped to be worn by a baby or toddler.

[0141] The device comprises one or more sensors for recording several of the child's vital parameters. The vital parameters include at least heart rate, oxygen saturation, and respiratory rate. The evaluation software is configured to use the at least one predictive model to predict the physiological state based on the measured values ​​recorded by the sensors.

[0142] The portable device further comprises an interface for transmitting a prediction result regarding the physiological state to a mobile telecommunication device of a user (a caregiver, e.g. a parent) and / or to a server computer system.

[0143] The interface for communication with the telecommunications device is preferably an interface for near-field communication, e.g. via radio or WLAN, but according to some embodiments the interface can also be a mobile radio connection.

[0144] The interface for communication with the server computer system can be, for example, a WLAN connection or a mobile phone connection.

[0145] According to embodiments of the invention, the at least one model comprises a SIDS model for predicting a current or future increased risk of dying from sudden infant death syndrome. Optionally, the at least one model can contain one or more additional predictive models, e.g., a hunger model for predicting whether and / or when the child will experience hunger.

[0146] During the training phase, the SIDS model is trained on training data that includes at least oxygen concentration, heart rate, and respiratory rate. In addition to oxygen concentration, the training data preferably also includes one or more other blood parameters that serve as control parameters, such as CO2 concentration, methemoglobin, and / or carboxyhemoglobin. The blood parameters in the training data are preferably recorded under realistic conditions, meaning that the training data also includes blood oxygen concentrations that are too low due to measurement errors and that are annotated as incorrect in the training data.

[0147] The training data set for training the hunger model preferably contains a large number of data sets, each of which includes, in addition to the time of onset of a feeling of hunger, some other time-stamped parameters, e.g., the blood sugar level.

[0148] Under a " portable device "Wearables" refers to an electronic device that is worn on the user's body during use. They are also referred to as "wearables." For example, the device can be attached to the body using certain fastening means (e.g., strap, especially Velcro, buckle, magnetic closure, etc.) or integrated into clothing. The device preferably comprises one or more sensors and a data processing unit.

[0149] Under a " Telecommunications device " is understood here to mean any portable data processing device capable of transmitting data via a network, in particular a mobile phone device, a smartphone, a smartwatch or a tablet computer or notebook.

[0150] Under a " child " is understood here as a toddler or baby. A " toddler" is understood here as a child in the second, third or fourth year of life, and a "baby" is understood as a child in the first year of life.

[0151] Under a " battery " is understood here as a non-rechargeable primary battery or a rechargeable secondary battery (commonly called accumulator).

[0152] Under a " predictive model" is understood here to mean an executable file, a set of parameters and / or a data structure which enables a software program or is itself designed to detect the current presence of a certain physical state of an entity and / or predict the future presence of this state. Typically, a predictive model uses historical data relating to the state to be detected or predicted for the calculation. For example, the historical data can be used as training data to extract the knowledge contained in this data during a machine learning process and to store it in the predictive model. The knowledge can in particular include knowledge of correlations between parameters.

[0153] Under " Machine learning" is understood here as a process by which knowledge about the relationships between several parameters, which is contained in training data, is transferred into a so-called "model", which can be used to automatically calculate predictions regarding the properties of entities and processes. This means that the examples are not simply learned by heart, but patterns and regularities in the learning data are recognized. In this way, the system can also evaluate unknown data (learning transfer). For example, the generated model can be a predictive model in the form of a trained artificial neural network, or other data structures such as support vector machines.

[0154] Under a "Vital parameters"Here, a "vital parameter" is understood to mean a data value, particularly a numerical value, that reflects a condition and / or a currently existing property of a person's body. A vital parameter can be a data value or raw data value obtained directly from a measurement, or a value mathematically derived from measured raw data.

[0155] Under a "Environmental parameters"Here, a data value, in particular a numerical value, is understood to be a data value, in particular a numerical value, which depends entirely or at least largely on entities outside the human body. For example, the intensity of the sun's UV radiation is an environmental parameter, but so is room temperature, since a certain amount of warming of a room by a person's body heat is possible, but the effect is generally negligible. An environmental parameter value can be a data value or raw data value, which is obtained as a measured value directly from a measurement, or a value mathematically derived from measured raw data.

[0156] Under a "physiological state"is used here to describe a biophysical state of certain vital processes in an organism, e.g., a child. This state can be, for example, a healthy state, a pathological state, or a risky state. For example, a state in which all vital parameters are within the normal range is generally considered healthy, while a state in which one or more important biophysical parameters deviate from the normal range and cause current symptoms is referred to as a pathological state. A "risky" state is one in which the affected person is not currently experiencing any noticeable health impairment, but in which the risk of developing a pathological state is significantly increased. Short description of the drawing

[0157] Embodiments of the invention are described below with reference to the drawing. In the drawing, Fig. 1 shows a method for providing a wearable device for predicting a child's physiological state; Fig. 2 shows an illustration of a variant of the device in the form of a wristband; Fig. 3 shows an illustration of the transmission of an alarm via a near-field signal from the wristband to a smartphone; Fig. 4 shows a block diagram of a section of a wearable device with multiple sensors; Fig. 5 shows a diagram relating to the derivation and use of various parameters for predicting an increased risk of SIDS; Fig. 6 shows a diagram relating to the derivation and use of a parameter for predicting feelings of hunger; Fig. 7 shows a system for monitoring the health of a child with multiple components.

[0158] Figure 1 shows methods for providing a wearable device for predicting a physiological state of a child.

[0159] In a first step 102, a training data set is provided. For example, the training data set can be provided on a storage medium or downloaded over a network.

[0160] The training dataset can, for example, be generated by attaching a wearable device with multiple sensors to several toddlers and babies in order to record and store multiple vital signs and / or environmental parameters over an extended period of time. Furthermore, the data obtained in this way is annotated with verified physiological states. If the number of children is sufficiently large and the observation period sufficiently long, various, sometimes critical situations and corresponding physiological states will arise. For example, colds and the associated fever may occur. Short-term feelings of hunger may arise if there are delays during a journey and the child's feeding times cannot be adhered to. Abnormal breathing patterns (e.g., apneas), hypoxemia, and bradycardia may also occur.These and other abnormalities can occur, particularly in premature babies, and can be collected and stored as a training dataset. The creation of the training dataset can also involve the use of external sensors in addition to the wearable device to record additional vital signs and / or environmental parameters in order to expand the training dataset.

[0161] In the next step 104, a predictive model is trained using a machine learning method. Various methods can be used, such as neural networks, support vector machines, and similar methods. However, neural networks have proven particularly advantageous in this context. The totality of the parameters and their respective timestamps represent the input parameters of the model to be trained. The annotated physiological states of the child represent the output data. During training, various parameters of the model, for example, weights of nodes of neural networks, are adjusted such that the output (physiological state) predicted by the model based on a set of input parameters is as identical or similar as possible to the physiological states that were actually observed and annotated in the training data set.This process may involve minimizing a so-called "loss function."

[0162] In a further step 106, the trained predictive model can be integrated into evaluation software, which can be installed on one or more portable devices and / or the server computer system. A software application interoperable with the evaluation software can be made available as an "app" via the app store of the respective operating system provider of the telecommunications device for download and installation on the telecommunications device.

[0163] Figure 2shows an illustration of a variant of the wearable device 200 designed as a wristband. In the variant shown here, all or the majority of the sensors are located within a central sensor block 202, to which two arms 204, 206 are attached. The size, shape, and material of the device are designed so that it can be worn on the wrists or ankles, or arms or legs, of a baby or toddler. For example, the arms can be made of a flexible material such as silicone or fabric. They have a locking mechanism that allows the device 200 to be securely attached to the child's body. Preferably, the material of the arms 204, 206 is elastic to ensure sufficient contact pressure.

[0164] In other embodiments, however, the sensors may also be distributed over one of the two arms or over both arms.

[0165] Figure 3shows an illustration of the transmission of an alarm via a near-field signal from the wristband to a smartphone. For example, the wristband 200 can have a module for near-field communication, such as a Bluetooth module. The radio signal of the Bluetooth standard is generally sufficient to penetrate one or two walls, so that a portable telecommunications device 302, such as a parent's smartphone, can receive warning signals from the portable device 200 even if the parents are, for example, briefly in another room but still in close proximity to the child. The telecommunications device 302 can contain software that generates an output based on the data received from the wristband 200 (in particular, predictive results regarding physiological states, but optionally also raw data or vital parameters derived from the raw data).For example, the output may include a GUI displayed on the smartphone screen, containing, for example, a warning regarding the predicted physiological state and / or a recommended course of action. The recommendation may, for example, be to turn the child over, feed the child, lower or raise the temperature in the room, or the like. In addition to or as an alternative to the visual output on the screen, the caregiver may also be warned acoustically, for example, through an alarm tone or by activating the smartphone's vibration function.

[0166] Figure 4 shows a block diagram of a portion of a wearable device with multiple sensors. The wearable device 200 includes a battery 406 and one or more processors 402, which may be implemented, for example, as microprocessors.

[0167] The device contains an interface 404 for exchanging data with the portable telecommunications device 302, for example, a radio interface. Preferably, it also contains an interface 403 for exchanging data with a server computer system. The interface 403 can, for example, be a cellular connection or a WLAN connection, enabling data to be exchanged with the server computer system over the Internet.

[0168] Evaluation software 408 is installed on the portable device. The software may contain one or more predictive models 410, each of which has been trained, for example, to predict a specific physiological condition (e.g., increased risk of sudden infant death syndrome, onset of hunger, onset of fever, etc.). However, rule-based algorithms may also be used instead of the models.

[0169] The sensor module 202 includes one or more sensors 418 for detecting vital parameters. In particular, the module 202 includes a PPG sensor 412, from whose raw data a variety of relevant vital parameters can be derived, including, for example, heart rate, respiratory rate, blood oxygen concentration, blood glucose concentration, and some other vital parameters or blood components used as a control in SIDS prediction. In some embodiments, the device 200 includes additional sensors for vital parameters, such as a skin temperature sensor 414, a gyroscope 416 for detecting the child's movements, and / or a microphone 418.

[0170] In addition, the sensor module 202 can include additional sensors 422 for detecting environmental parameters, for example, a humidity sensor 424, an ambient temperature sensor 426, and / or a daily or hourly UV radiation dose sensor 428. For example, the sensor 428 can be used to detect the UV light dose to which the child was exposed over the course of a day. If the recommended maximum dose is reached or exceeded, the evaluation software can send a warning to the smartphone app that the child needs to be protected from further sun exposure. However, recording the daily UV light dose over time can also help detect a sunlight deficiency.

[0171] Depending on the design, various sensors from different manufacturers can be used, some of which differ in the way they process the acquired measurement data. For example, temperature sensors typically indicate the temperature in degrees Celsius or degrees Fahrenheit. A PPG sensor signal 112, on the other hand, provides one or more light spectra, with one or more vital parameters, such as blood oxygen concentration or glucose concentration, being obtained only through subsequent processing of the spectra.

[0172] Figure 5shows a diagram regarding the derivation and use of various parameters for predicting an increased risk of SIDS. For example, the evaluation software on the portable device 200 can include a SIDS model 520, which requires at least the heart rate 504, the respiratory rate 506, and the oxygen concentration in the child's blood 508 as input parameters. These vital parameters can be obtained, for example, by a signal analysis 502 based on the raw data or spectral data of a PPG sensor 412. The three vital parameters are always available as long as the child is wearing the portable device. The SIDS model 520 is designed to evaluate further input parameters in order to make the prediction of an increased risk of sudden infant death even more accurate. These include, for example, some control parameters in the form of blood parameter values, which, for example,can also be derived from the raw data of the PPG sensor (not shown here). In addition, this can include other vital parameters such as the skin temperature 512 detected by a temperature sensor 414 of the device, respiratory sounds 510 recorded by a microphone 418, the carbon dioxide concentration in the ambient air detected by a carbon dioxide sensor 513, the air humidity 424 and / or the ambient temperature 426. An analysis of the respiratory sounds can, for example, be used to determine whether breathing is affected by obstruction, which would imply an increased risk of SIDS. Movement data and / or video data (not shown here) can also be included in the prediction, as they can detect, for example, whether the child is active or whether the child is lying on its stomach or back.

[0173] If the prediction indicates that the child currently has or will soon have an increased risk of sudden infant death syndrome, the device 200 sends an alarm message 522 either directly to the caregiver's smartphone or indirectly first to the server computer, where the prediction result can be further refined, if necessary, based on data provided by external sensors via the base station. The refined prediction result is then forwarded by the server computer system via the network to the caregiver's smartphone, where it is displayed, provided the refined prediction result also indicates an increased risk of sudden infant death syndrome.

[0174] Figure 6shows a diagram regarding the derivation and use of a parameter for predicting feelings of hunger. Here, too, the PPG sensor 412 can initially be used to capture one or more light spectra reflected by the child's skin and the vessels in the skin. From the raw data, the evaluation software determines at least one derived vital parameter, namely the blood sugar level 604, via signal analysis 602. At least the blood sugar level and optionally further parameters are used as input parameters in the prediction of a current or future feeling of hunger by a trained predictive "hunger" model 620. The result 622 of the prediction as to whether and, if so, when a feeling of hunger is present orwill be present, is sent either directly to the smartphone via a near-field signal or indirectly via the server computer system, whereby the server computer system serves in particular to refine the prediction result by incorporating further data from other sensors and / or by additional, possibly computationally intensive analyses.

[0175] Figure 7 shows a system for monitoring a child's health with multiple components. The system includes at least the wearable device 200, which here is designed, for example, as a bracelet for attachment to the wrist of a child 300.

[0176] The system may also include one or more portable telecommunications devices 302, typically caregivers' smartphones, on which software is installed that is interoperable with the evaluation software of the device 200 in order to exchange data with it. For example, the owners of the telecommunications device 302 can be informed about critical physiological conditions of the child via push notification from the evaluation software and / or can actively request status data or historical data regarding the physiological conditions of the child 300 from the portable device 200 via pull functionality.

[0177] The system may also include a server computer system 706, which is connected to the portable device 200 and the evaluation software 408 via a network 704, for example the Internet. For example, the data acquired and possibly derived by the device 200, as well as prediction results, can be transmitted to the server computer system via the network, depending on the urgency and configuration, immediately or, for example, during the battery charging process. The server computer system serves, in particular, to store the data received from one or more devices 200 or external sensors 712 in a database 708. In addition, prediction results received via the network 704 from the portable device 200 and its sensors can be refined and made more precise on the server computer system. This can be done, in particular, by using additional data acquired by external sensors 712 and transmitted directly via the network (e.g.Internet) or indirectly via a base station 710 to the server computer system, are additionally taken into account and / or by the server computer system performing complex, computationally intensive analyses. For example, a microphone 716 and / or a camera 712 (in particular a thermal imaging camera) or other additional sensors 714 can be installed as external sensors in or on the bed in which the child normally sleeps. These external sensors are communicatively coupled to the server computer system 706 either directly via the network or indirectly via a base station 710 and can send data to it. For example, a server application on the server computer system can perform an image analysis of the video data from camera 712, for example to determine whether the child is lying on its back or on its stomach, which is an important prognostic factor for the risk of sudden infant death syndrome.

[0178] According to one embodiment, the external sensor is a video camera, in particular a thermal imaging video camera, which is communicatively connected to the portable telecommunications device via a near-field communication interface (e.g. radio, in particular Bluetooth, or WLAN) in order to enable the caregivers to monitor the baby via a video signal. Preferably, the video camera is portable and can be set up anywhere and can be communicatively coupled to the server computer, e.g. via WLAN over the Internet, preferably even without a base station. This can have the advantage that parents can also install the camera in their child's environment without major installation effort, e.g. when they are traveling, thus increasing the parents' mobility.

[0179] According to the invention, the evaluation software of device 200 and optionally also the application on smartphone 302, which is interoperable with this evaluation software, are operatively coupled via network 704 to one or more devices 702, 703 that serve to prepare or cook food for the child. The devices 702, 703 can be, for example, a microwave, a kettle, a device for heating milk or baby food, etc. If the portable device 200 detects or predicts, using the evaluation software, that the child is currently or will be feeling hungry in the near future, the evaluation software can automatically send a control command to one or more of the devices 702, 703 to prompt them to begin preparing the food. Preferably, however, the control command is not sent directly to said devices 702, 703, but first to the software on smartphone 302.In response to receiving the control command, the smartphone software requests the user to authorize sending the control command to the relevant device. The smartphone then sends the control command to the relevant device 702, 703 after receiving the user's authorization. This ensures that the evaluation software does not automatically activate a device remotely without the knowledge of the caregivers, as this could pose a security risk.

[0180] The server computer system 706 may be a conventional, monolithic server computer. However, it may also be a distributed server architecture, in particular a cloud computer system.

Claims

1. A wearable device (200), - wherein the wearable device is configured to be worn on the body of a child (300), wherein the child is a baby or toddler; - wherein the device comprises: • one or more sensors (412, 414, 416) for sensing a plurality of vital parameters of the child, wherein the vital parameters comprise at least heart rate, oxygen saturation, and respiratory rat, wherein at least one of the sensors for the vital parameters is configured to measure the blood sugar concentration of the child non-invasively; and • an interface (403, 404) for transmitting a prediction result relating to at least a current or future physiological state to a mobile telecommunication device (302) of a user and / or to a server-computer system; • evaluation software (408) configured to predict at least a current or future physiological state of the child as a function of the heart rate, oxygen saturation, and respiratory rate measured by the sensors, wherein the at least one physiological state is a state of an increased risk of sudden infant death syndrome, wherein the evaluation software is configured to use at least the heart rate, oxygen saturation and respiratory rate as input to predict the presence of an increased risk of sudden infant death syndrome; wherein the evaluation software is configured to predict a further physiological state in the form of a current or future feeling of hunger in the child as a function of at least the measured blood sugar concentration, and / or to predict a future time of occurrence of the feeling of hunger, wherein the feeling of hunger is predicted when the current or future blood sugar level is below a predefined limit value, wherein the evaluation software is communicatively coupled to an electronic food preparation appliance (702) directly or by means of a software application of the telecommunication device, wherein the evaluation software and optionally also the software application of the telecommunication device is configured to cause the electronic appliance to prepare food for the child in response to the prediction that the child is feeling hungry now or will do in the future.

2. The wearable device according to claim 1, wherein the evaluation software is configured to predict the presence of an increased risk of sudden infant death syndrome as a function of one or more further parameters, wherein the one or more further parameters comprise: - the child's skin temperature; - the ambient temperature; and / or - the ambient air moisture.

3. The device according to any one of the preceding claims, - wherein the sensors comprise a photoplethysmographic sensor, referred to here as a PPG sensor, - wherein the evaluation software is configured to derive the heart rate, oxygen saturation and respiratory rate of the child from the signals detected by the PPG sensor and to make them available as input to the evaluation software.

4. The device according to any one of the preceding claims, - wherein the one or more sensors comprises a sensor for detecting at least one blood parameter of the child, wherein the at least one blood parameter is in particular a methaemoglobin concentration and / or a carboxyhaemoglobin concentration and / or a CO2 concentration in the blood of the child, wherein the sensor for detecting the blood parameter may be configured in particular as the PPG sensor according to claim 3; - wherein the evaluation software is configured to use the at least one blood parameter as an additional input parameter in order to reduce the false positive rate of the prediction of the increased risk of sudden infant death syndrome by the evaluation software.

5. The device according to any one of the preceding claims, - wherein the device comprises at least one sensor for determining at least one further vital parameter and / or environmental parameter, or an interface for receiving the further vital parameter and / or environmental parameter from an external sensor, wherein the at least one further environmental parameter is selected from a group comprising: • the CO2 concentration of the ambient air, • video data of the child, in particular video data of an infrared camera; • acoustic data detected by a microphone; and • movement data characterising the child's movement activity; - wherein the evaluation software is configured to use the at least one further vital parameter and / or environmental parameter as an additional input parameter in order to predict the presence of an increased risk of sudden infant death syndrome.

6. The device according to any one of the preceding claims, - wherein the sensors comprise a photoplethysmographic sensor, referred to here as a PPG sensor, - wherein the evaluation software is configured to derive the child's blood sugar concentration from the signals detected by the PPG sensor in addition to the heart rate, the oxygen saturation and the respiratory rate of the child and to provide at least the blood sugar concentration as input at least for predicting the feeling of hunger.

7. The device according to any one of the preceding claims, - wherein the evaluation software is configured to recognise the current or future presence of a physiologically problematic state of the child, • when a value of at least one vital parameter is outside a predetermined normal range; and / or • when a pattern of values of multiple vital parameters is detected that indicates a current or future problematic physiological state of the child, wherein the pattern may be recognised even if all vital parameters are individually within their normal range; and - wherein the evaluation software is configured to send a message regarding the predicted problematic physiological state to the mobile telecommunication device (302) and / or the server-computer system in response to the recognition of the current or future physiologically problematic state.

8. The device according to any one of the preceding claims, - wherein the evaluation software is configured to selectively recognise the current or future presence of a physiologically problematic state of the child requiring immediate intervention, wherein this physiologically problematic state comprises in particular the increased risk of sudden infant death syndrome; and - forwarding at least some of the vital parameters or intermediate prediction results measured or derived by the wearable device to the server-computer system via a network to enable the server-computer system to predict physiological states that do not require immediate intervention and / or calculate a refined final prediction.

9. The device according to any one of the preceding claims, wherein the device (200) is a bracelet or ankle strap or leg strap.

10. The device according to claim 9, - wherein the sensors comprise one or more pressure sensors which are configured to detect the contact pressure of the device on the child's body, - wherein the evaluation software is configured to recognise, on the basis of the measured contact pressure, whether the contact pressure is within a predefined permissible contact pressure range within which the one or more sensors for detecting the vital values are able to operate correctly, - wherein the evaluation software is configured to issue a warning via a signalling element of the device (200) to the user and / or via the interface to the telecommunication device if the measured contact pressure is outside the permissible contact pressure range; and / or - wherein the evaluation software is configured to prevent the measurement of vital parameters by the one or more sensors until the contact pressure is again within the permissible contact pressure range.

11. The device according to any one of the preceding claims, wherein the device is configured - to only send a message to the user device if the evaluation software has detected the current or future presence of a physiologically problematic state, in particular an increased risk of sudden infant death syndrome and / or a feeling of hunger, or the presence of a vital or environmental parameter in a health-critical value range; and - otherwise to automatically save the detected vital parameters and optionally detected environmental parameters without sending a message.

12. The device according to any one of the preceding claims, - wherein the interface (404) for transmitting data to the telecommunication device is an interface for data transmission via a near-field signal, in particular via a radio signal, in particular a Bluetooth interface or ZigBee interface, - wherein the wearable device is operable in a low-radiation operating mode and in a normal-radiation operating mode; - wherein the wearable device, in the normal operating mode, is configured to operate in the low-radiation operating state when no physiological state is predicted and no vital or environmental parameter requiring immediate intervention is measured; and wherein the wearable device is configured to automatically switch to the radiation-normal operating mode if the evaluation software has detected the current or future presence of a physiologically problematic state, in particular an increased risk of sudden infant death syndrome and / or a feeling of hunger, or the presence of a vital or environmental parameter in a health-critical value range.

13. The wearable device according to any one of the preceding claims, - wherein the wearable device comprises one or more environmental parameter sensors selected from a group including: • a thermometer that measures the ambient temperature; • a measuring device for measuring the ambient air moisture; • gases, especially CO2; • UV sensor for detecting a cumulative UV radiation dose, in particular a daily cumulative UV radiation dose; - and / or wherein the sensors of the wearable device for detecting the vital parameters comprise further sensors selected from a group comprising: • acceleration sensor for detecting the position of the child (supine position, prone position); • temperature sensor for detecting the skin temperature; • a microphone for detecting ambient noises and / or noises made by the child; • video camera, in particular a thermal imaging camera.

14. The wearable device according to any one of the preceding claims, wherein the prediction software comprises at least one predictive model for predicting the at least one physiological state, wherein the at least one predictive model is a model generated by a machine learning method on the basis of a training dataset.

15. A system comprising the wearable device according to any one of the preceding claims and one or more of the following further components: - the portable telecommunication device, wherein a user software is set up on the portable telecommunication device, wherein the user software is interoperable with the evaluation software and is configured to display the prediction results received from the wearable device via the interface to the user and / or to enable the user to configure the evaluation software; and / or - the server-computer system; and / or - a base station to which one or more external sensors are coupled for measuring vital parameters of the child or environmental parameters of the child's surroundings; and / or - one or more of the external sensors, in particular a video camera, in particular a thermal imaging video camera.

16. A method for providing a wearable device (200) for monitoring the physiological state of a child (300), comprising: - providing (102) a training dataset comprising a plurality of datasets, wherein in each dataset at least one physiological state of the child is stored linked to vital parameters of the child, wherein the vital parameters comprise at least the heart rate, oxygen saturation, and respiratory rate; - performing (104) a machine learning process on the training data to generate at least one predictive model (520, 620), wherein the model is configured to predict the physiological state of the child on the basis of at least the heart rate, oxygen saturation, and respiratory rate; - installing (106) evaluation software (408) including the at least one predictive model (410, 520, 620) on the wearable device, wherein the device is configured to be worn on the body of a child (300), wherein the child is a baby or toddler, wherein the device comprises: • one or more sensors (412, 414, 416) for detecting a plurality of vital parameters of the child, wherein the vital parameters comprise at least the heart rate, oxygen saturation, and respiratory rate, wherein the evaluation software is configured to use the at least one predictive model to predict the physiological state on the basis of the heart rate, oxygen saturation, and respiratory rate detected by the sensors, wherein at least one of the sensors for the vital parameters is configured to measure the child's blood sugar concentration non-invasively; and • an interface (403, 404) for transmitting a prediction result relating to the at least one physiological state to a mobile telecommunication device (302) of a user and / or to a server-computer system, wherein the at least one physiological state is a state of an increased risk of sudden infant death syndrome, wherein the evaluation software is configured to use at least the heart rate, oxygen saturation and respiratory rate as input to predict the presence of an increased risk of sudden infant death syndrome, wherein the evaluation software is configured to predict a further physiological state in the form of a current or future feeling of hunger in the child as a function of at least the measured blood sugar concentration, and / or to predict a future time of occurrence of the feeling of hunger, wherein the feeling of hunger is predicted when the current or future blood sugar level is below a predefined limit value, • wherein the evaluation software is communicatively coupled to an electronic food preparation appliance (702) directly or by means of a software application of the telecommunication device, wherein the evaluation software and optionally also the software application of the telecommunication device is configured to cause the electronic appliance to prepare food for the child in response to the prediction that the child is feeling hungry now or will do in the future.