Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
AN AUTONOMOUS INTELLIGENT MATTRESS FOR AN INFANT
Document Type and Number:
WIPO Patent Application WO/2019/207570
Kind Code:
A1
Abstract:
An intelligent mattress for an infant, comprises a) a plurality of sensors adapted to provide data relative to physical parameters and vital signs of the infant; and b) a computing device configured to collect and process said provided data in order to obtain an actual behavior of the infant, to compare said processed data with predefined behavioral patterns, to respond according to response scripts associated with a detected predefined behavioral pattern, and to update the database of predefined behavioral patterns, while considering past events and parameters affected by the infant's development.

Inventors:
DAN ON JOSHUA (IL)
Application Number:
IL2019/050431
Publication Date:
October 31, 2019
Filing Date:
April 16, 2019
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
COMFORT SYSTEMS 2007 LTD (IL)
International Classes:
A61B5/00; A47C27/08; G08B21/04
Foreign References:
US20180102039A12018-04-12
US20120053424A12012-03-01
US20170100076A12017-04-13
US20150351693A12015-12-10
Attorney, Agent or Firm:
LUZZATTO, Kfir et al. (IL)
Download PDF:
Claims:
CLAIMS

1. An intelligent mattress for an infant, comprising

a) a plurality of sensors adapted to provide data relative to physical parameters and vital signs of the infant; and

b) a computing device configured to collect and process said provided data in order to obtain an actual behavior of the infant, to compare said processed data with predefined behavioral patterns, to respond according to response scripts associated with a detected predefined behavioral pattern, and to update the database of predefined behavioral patterns, while considering past events and parameters affected by the infant's development.

2. The mattress of claim 1, wherein the sensors are selected from the group consisting of heartbeat sensors, respiration sensors, temperature sensors, humidity sensors, blood pressure sensors, oxygen level sensors, pressure sensitive sensors and microphones.

3. The mattress of claim 1, including one or more sensors that are integral to the mattress, and one or more sensors that are external to it.

4. The mattress of claim 1, wherein the computing device comprises a storage device, a communication unit, and a processor unit coupled to storage device and communication unit.

5. The mattress of claim 1, wherein the storage device is configured to store predefined behavioral patterns of the infant, such as restlessness.

6. The mattress of claim 5, wherein the predefined behavioral pattern includes a set of correlated physical parameters and vital signs of an infant, such as body movements, crying sounds, heartbeats per minute and temperature.

7. The mattress of claim 1, in which the computing device is capable of activating one or more mechanical devices.

8. The mattress of claim 7, wherein the mechanical device is selected from a lifting or vibrating mechanism, embedded in said mattress or external thereto.

9. The mattress of claim 1, in which the computing device is capable of activating at least one audible device such as a music playing device used for playing relaxing tunes, or parent's voice.

10. The mattress of claim 1, in which the computing device is capable of generating a communication for wirelessly activating at least one remote mechanical or audible device, such as music player, alerting device and external vibrating and tilting devices.

11. The mattress of claim 1, in which the computing device is capable of generating a communication for messaging mobile devices such as tablets and smartphones.

12. The mattress of claim 1, further provided with a computing software for monitoring and controlling the mattress, by a remote device such as a smartphone, tablet and personal computer.

13. The mattress of claim 1, further provided with a sliding prevention device comprising a harness adapted to receive the infant when said mattress is tilted.

14. The mattress of claim 13, in which the harness is fastened to the mattress by straps that are attached to the mattress by detachable means such as by dedicated buttons or by a zipper.

15. The mattress of claim 13, in which the harness comprises at least one section adapted to receive the legs of the baby.

16. The mattress of claim 13, in which the harness comprises at least one sensor.

17. The mattress of claim 16, in which the straps comprises control and power supply wiring for the sensor.

18. The mattress of claim 13, in which the harness comprises a wireless communication device for communicating measured data.

19. A method of monitoring, responding to and learning an infant's behavior, comprising: a. detecting one or more physical events performed by a specific infant;

b. comparing the one or more detected events with previous stored data relating to similar events; c. responding to detected predefined behavioral patterns per predefined response scripts by mechanical, audible and communicative means for assisting and alerting in infant's distress situations; and

d. automatically updating a database of predetermined behavioral patterns with evolving behavioral patterns resulting from rapid infant's development and stored past events.

20. The method according to claim 19, in which the updating of the database is initiated or confirmed by the user.

Description:
AN AUTONOMOUS INTELLIGENT MATTRESS FOR AN INFANT

Field of the Invention

The present invention relates to infant sleeping systems. More particularly, the invention relates to an autonomous mattress for an infant that monitors and responds to the behavior of a specific infant. Furthermore, the invention relates to an intelligent mattress, that is capable of autonomously learning the evolving behavioral patterns of the rapid growing infant, and adapt itself accordingly.

Background of the invention

The life quality of parents to infants is affected by the presence of their child. Two parameters that significantly affect said life quality are the interrupted parents' night sleep (i.e., by a crying infant) and the common fear of failing to respond to infant's distress situation, in particular at night.

One common example of infants' night sleep interrupter is incidental awakening (i.e., of young infants that have not yet learnt to independently relax and return to sleep) followed by restlessness. A restless infant behavioral pattern may involve arms/legs movements combined with crying sounds that are produced by the infant. Accordingly, often infants calm down when they are picked up, gently rocked, and carried around the room. In many cases, infants experience an automatic calming reaction when they are being carried. For example, researchers from the RIKEN Brain Science Institute show that infants automatically and deeply relax when they are carried. Their study, published in the journal Current Biology, demonstrates that the infant calming response to maternal carrying is a coordinated set of nervous, motor and cardiac regulations.

Another significant disturbance to infants' night sleep is awakening or restlessness due to physical conditions such as reflux. Reflux is a well-known problem in babies, and the number of infants observed with reflux is on the rise. Holding the baby in an upright position or at list in a tilted position and not in a horizontal position for a while is very helpful, and assists to avoid or tame reflux. In addition, when babies suffer from a cold, or exhibit a respiratory distress situation, a tilted position is helpful to make them breathe better.

Flaving ways to autonomously identify behavioral patterns of sleeping infants and respond accordingly, would therefore significantly contribute to the life quality of parents of infants. The human newborn infant is equipped with a large set of interrelated abilities and competences, not only physiologically but also behaviorally. The sleep patterns of infants are markedly different from those of adults. Infants' behavioral states are not limited simply to wakefulness versus sleep: within sleep epochs, distinct states can also be observed, which have characteristic patterns of respiration, heart rate, electroencephalographic (EEG) activity, electromyography (EMG) of the chin muscle, eye movements, muscle activity, etc. However, although most infants may follow a few common general behavioral states, in particular during sleeping cycle, still each specific infant has his own unique behavioral patterns. Moreover, the unique behavioral patterns of each infant frequently change due to the relatively rapid mental and physical growth (e.g., from birth to about age 1 to 2 years in terms of increase in size of most organs, length/height and weight) and the different rate of the development states of each specific infant. For example, in general, length in normal-term infants increases about 30% by 5 months and bigger than 50% by 12 months; infants may grow about 25 cm during the 1st year. Weight may follow a similar pattern. Normal-term neonates generally lose 5 to 8% of birth weight in the days after delivery but regain their birth weight within 2 weeks. They then gain 14 to 28 g/day until 3 months, then 4000 g between 3 and 12 months, doubling their birth weight by 5 months, tripling it by 12 months, and almost quadrupling it by 2 yr.

It would be highly desirable, in order to address this problem, to detect the behavioral states of each specific infant while considering their relatively rapid grow.

It is an object of the present invention to provide a system (hereinafter referred to as "autonomous mattress" or "intelligent mattress") that is capable of responding to the behavior of an infant, during sleeping cycles as well as during waking times.

It is another object of the present invention to provide an intelligent mattress that is capable of learning the behavior of an infant and of adapting itself to evolving behavioral patterns corresponding to the relatively rapid growth of the infant. The terms "learning" and "adopting", as used herein, refer to the use of a variety of inputs, as will be further described below. It is yet another object of the present invention to provide an intelligent mattress, which is capable of reacting by changing the infant's state and/or of generating alerts, upon detection of abnormal behavior with respect to a specific infant.

Other objects and advantages of the invention will become apparent as the description proceeds.

SUMMARY OF THE INVENTION

An intelligent mattress for an infant, comprises

a) a plurality of sensors adapted to provide data relative to physical parameters and vital signs of the infant; and

b) a computing device configured to collect and process said provided data in order to obtain an actual behavior of the infant, to compare said processed data with predefined behavioral patterns, to respond according to response scripts associated with a detected predefined behavioral pattern, and to update the database of predefined behavioral patterns, while considering past events and affected parameters affected by the infant's development.

The sensors can be of any type suitable to collect data relative to the infant's status, such as heartbeat sensors, respiration sensors, temperature sensors, humidity sensors, blood pressure sensors, oxygen level sensors, pressure sensitive sensors and microphones. In one embodiment the mattress includes one or more sensors that are integral to the mattress, and one or more sensors that are external to it.

The computing device in one illustrative embodiment comprises a storage device, a communications unit, and a processor unit coupled to storage device and communications unit, the storage device being configured to store predefined behavioral patterns of the infant, such as restlessness. The predefined behavioral pattern typically includes a set of correlated physical parameters and vital signs of an infant, such as body movements, crying sounds, heartbeats per minute and temperature.

In one embodiment, the computing device is capable of activating at least one or more mechanical devices such as a lifting or vibrating mechanism, embedded in said mattress or external thereto. In a further embodiment the computing device is capable of activating at least one audible device such as a music playing device used for playing relaxing tunes, or parent's voice. According to another embodiment the computing device is capable of generating a communication for wirelessly activating at least one remote mechanical or audible device, such as music player, alerting device and external vibrating and tilting devices, or for messaging mobile devices such as tablets and smartphones. In one embodiment the mattress is provided with a computing software for monitoring and controlling the mattress by a remote device such as a smartphone, tablet and personal computer.

In some embodiments the mattress is further provided with a sliding prevention device comprising a harness adapted to receive the infant when said mattress is tilted, which harness is adapted to be fastened to said mattress, such that whenever said mattress is un-tilted said harness does not constrict the infant's movements. According to one embodiment the harness is fastened to the mattress by straps that are attached to the mattress by detachable means such as by dedicated buttons or by a zipper. In another embodiment the harness comprises at least one section adapted to receive the legs of the baby.

In some embodiments, the harness comprises at least one sensor and the straps comprise control and power supply wiring for the sensor. In other embodiments the harness comprises a wireless communication device for communicating measured data.

The invention also encompasses a method of monitoring, responding to and learning an infant's behavior, comprising:

a. detecting one or more physical events performed by a specific infant;

b. comparing the one or more detected events with previous stored data relating to similar events ;

c. responding to detected predefined behavioral patterns per predefined response scripts by mechanical, audible and communicative means for assisting and alerting in infant's distress situations; and

d. automatically updating a database of predetermined behavioral patterns with evolving behavioral patterns resulting from rapid infant's development and stored past events.

According to one embodiment, the updating of the database is initiated or confirmed by the user. Brief Description of the Drawings

In the drawings:

Fig. 1 schematically illustrates an intelligent mattress, according to an embodiment of the invention;

Fig. 2 (A - D) schematically illustrates a communication scenario between the intelligent mattress of Fig. 1 and a mobile device, according to an embodiment of the invention;

Fig. 3 is a block diagram generally illustrating a computing device associated with the intelligent mattress of Fig. 1, according to an embodiment of the invention; and Fig. 4 is a flow chart generally illustrating a method of autonomously monitoring, responding and learning the behavior of a specific infant, according to an embodiment of the invention.

Detailed Description of the Invention

Various terms are used throughout the description and the claims which have conventional meanings to those with a pertinent understanding of the technical field. Additionally, various descriptive terms are used in describing the exemplary embodiments in order to facilitate the reader's understanding. Flowever, while the description to follow may entail terminology which is perhaps tailored to certain artificial intelligence models that involve machine learning techniques that give computer systems the ability to "learn" with data without being explicitly programmed, such as progressively improve performance of learning the behavior of a specific infant, it will be appreciated by a person skilled in the art that such terminology is employed in a descriptive sense and not in a limiting sense. Where a confined meaning of a term is intended, it will be explicitly set forth or otherwise apparent from the disclosure.

The present invention provides an autonomous, intelligent mattress. According to an embodiment of the present invention, the mattress routinely monitors the behavior of a specific infant, compares the monitored behavior to predefined behavioral patterns, and responds accordingly. The response may include mechanical reactions (e.g., inclining at least part of the mattress, rocking the mattress, raising the mattress, and the like), audible/visual reactions (e.g., initiating an audible/video playing devices), communication reactions (e.g., notification signals transmitted to paired wireless devices), a combination thereof or no reaction (e.g., when the monitored behavior is of a relaxed or of a sleeping infant). The automatic response as will be illustrated in the following reference is intended to urge an automatic calming reaction of the infant, or to provide an early alert and preliminary support for infant distress situations such as reflux or respiratory distress. According to some embodiments of the present invention, the autonomous intelligent mattress learns the evolving behavior of a specific infant and correspondingly adapts its predefined behavioral patterns and corresponding response.

Reference will now be made to several embodiments of the present invention, examples of which are illustrated in the accompanying figures for the purpose of illustration only. One skilled in the art will readily recognize from the following description that alternative embodiments of the structures and methods illustrated herein may be employed without departing from the principles of the claimed invention. Moreover, the following discussion is intended to provide a brief, general description of a suitable computing environment adapted to be implemented in an infant's mattress. While part of the invention will be described in the general context of program modules or codes that execute in conjunction with an application program that runs on a computer system, those skilled in the art will recognize that the invention may also be implemented in combination with other program modules. The functions described herein may be performed by executable code and instructions stored in computer readable medium and running on one or more processor-based systems. Embodiments of the invention may be implemented as a computer process, e.g., a computer system that encodes a computer program of instructions for executing the computer process.

Fig. 1 schematically illustrates the elements of an intelligent mattress 1 for autonomously learning and monitoring the behavior of a specific infant, according to an embodiment of the invention. Mattress 1 comprises a plurality of sensors (e.g., thermometer 2a, pressure sensor 2b and microphone 2c) and a computing device 3, which may be integral with the mattress or located outside it and in wired or wireless communication therewith. Sensors 2a-2c shown in the figure may be replaced but other types of sensors, or used in together with different types of sensors, each of which is adapted to provide data relative to different physical expressions and vital signs of an infant, such as the infant's audible voice samples (e.g., crying, muttering, breathing sound), infant's movements (i.e., rate and amplitude), heartbeat readings (i.e., pulse), respiration, temperature, humidity, blood pressure, oxygen level, etc. Any number of sensors can be used and only three of them are shown in the figure for the sake of simplicity. Additionally, if a sensor is needed for any reason, which must be applied directly to the infant's body (e.g. a blood oxygen saturation sensor), its reading can be fed to computing device 3 via wired or wireless connection and it can then be used together with inputs from other, integral sensors.

The following descriptions refer to the term "behavioral pattern", which in the context of this description defines a combination of physical expressions and vital signs in specific amplitudes (e.g., pulse of 120-130 beats per minutes, or pressure amplitudes and frequency ranges for pressure fluctuations measured by specific sensors and induced by the moving limbs of an infant), where a "predefined behavioral pattern" refers to a behavioral pattern which is "known to mattress 1", i.e., stored in computing device 3, either from previous readings or as a preset.

Computing device 3 is provided with a storage device for handling a database of predefined behavioral patterns and corresponding prescribed response scripts. An initial database is provided with mattress 1 on its initial installation, containing average values related to infant age. Computing device 3 collects and processes the data in order to obtain the actual behavioral pattern of an infant and reacts with a corresponding response script, while in the background it compares the processed data by utilizing an artificial intelligence event correlation model that is based on a plurality of past events, to determine whether the actual behavioral pattern differs from predefined behavioral patterns. It is important to mention that due to the rapid growth of the infant, while monitoring the actual behavior of an infant processing the data, computing device 3 considers rapidly evolving behavioral patterns as the infant grows. For example, when monitoring the movements of a newborn infant, the rapid weight gaining of newborn infants has to be considered (i.e., heavier infant's limbs of a relaxed infant would produce higher amplitudes at the pressure sensors).

Computing device 3 routinely updates its database of predefined behavioral patterns either automatically or based on an indication from a user (e.g., input manually, by the infant's parents, via a mobile device associated with intelligent mattress 1). Additionally, computing device 3 generates a notification signal in the following cases:

- when identifying a predefined behavioral pattern, which requires response;

- when the actual behavioral pattern differs from any predefined behavioral pattern; and

- when there is a change in one or more vital signs of the infant (i.e., that were predefined as critical). According to an embodiment of the present invention, the notification signal may result in the generation of different reactions, such as wirelessly broadcasting an alert message to a remote device via Bluetooth, Wi-Fi, cellular network, and the like (e.g., sending a message to a mobile device 11 as shown in Fig. 2A, while an infant 10 is placed on intelligent mattress 1), activating mechanical arrangements associated with, or embedded within the body of intelligent mattress 1 (e.g., activating a vibration or a tilting mechanisms adapted to gently rock, or tilt mattress 1 as indicated by the wave lines in Fig. 2A and as shown in Fig. 2B), or activating a device that is adapted to play audible and/or visual content (e.g., a remotely controlled crib mobile), etc. Of course, combined reactions may also be initiated, for example, an identified reflux situation may be reacted to by a combined tilt of mattress 1 and alert message broadcasted to predefined mobile devices such as device 11 of Fig. 2A.

The desired reactions can be obtained by using common adjustable bed mechanism and/or vibrating arrangement, e.g., that may include a motor, a vibrating elements, an adjustable frame, an elevated arrangement, or any other mechanisms that can be controlled by computing device 3 and which are suitable to cause mattress 1 to rock, vibrate, tilt or change position/orientation. Computing device 3 may communicate with such mechanism by wired or wireless means, such Bluetooth, Wi-Fi, etc.

According to an embodiment of the invention, such mechanism can be embedded within the body of mattress 1. In other embodiments, such mechanism (or at least part of it) can be provided as an add-on device to mattress 1, or as a standalone arrangement (e.g., in form of a crib) that is adapted to receive mattress 1.

According to an embodiment of the invention, intelligent mattress 1 further comprises foam, innerspring, organic materials, etc. in order to provide suitable support for the infant as any other regular mattress. According to some embodiments of the invention, intelligent mattress 1 is provided in the form of an add-on layer that is adapted to be coupled to an existing mattress (e.g., as a top layer of another mattress).

Figs. 2C and 2D illustrate a sliding prevention device which is provided in a particular embodiment of the present invention, in which device 20 is comprised of harness 21, which is adapted to receive the infant upon tilting of mattress 1, and strap 23 which secures device 20 to mattress 1. Strap 23 may be a single strap with tightening means, or two coupled sections with buckles, while being designed to avoid a mechanical interference or limitations to the mechanical operation of mattress 1.

Harness 21 is designed to allow a free motion of the infant whenever mattress 1 is un-tilted, and to secure the infant whenever mattress 1 is tilted. Device 20 may be employed either as a safety means for extreme use in case an infant requires significant tilting (i.e., intended to prevent significant sliding), or as routine securement means in which the infant is loosely or firmly secured (e.g., an infant who is prone to reflux for some period). Harness 21 may further comprise cutout sections 22 (i.e., to receive the legs of the baby) which are substantially required for the use of device 20 as a securement device.

According to some embodiments of the present invention, harness 21 comprises at least one pressure sensor which is employed for assessing the infant's weight while mattress 1 is tilted at a known angle with respect to a leveled horizontal surface. The measured pressure is communicated through a hidden wire embedded in strap 23 (i.e., where device 20 is integrally attached with mattress 1), or wirelessly, to computer device 3. The adjacency of device 20 and computer device 3 enable the utilization of low energy communication device.

A person skilled in the art will appreciate the plurality of specific designs, materials, securement means and integration levels of device 20 with the autonomous intelligent operation of mattress 1, which can be used for producing numerous embodiments of the present invention. For instance, harness 21 may be made integral with a mattress cover such that straps 23 are not required, or different means to secure it to the mattress can be devised.

Fig. 3 schematically illustrates an exemplary configuration of computing device 3 of Fig. 1, according to an embodiment of the invention. Computing device 3 comprises a storage device 31, a communication unit 32, and a processor unit 33, coupled to storage device 31 and communication unit 32. Storage device 31 is configured to store predefined behavioral patterns of the infant, which can be updated routinely (i.e., considering the infant's rapid growth). The predefined behavioral pattern includes a correlation between pluralities of physical events performed by the infant. Processor unit 33 is configured to receive a plurality of events representing at least one physical event associated with the infant, wherein the physical event is obtained by processing the data received from the readings of one or more sensors 2a-2c, e.g., data relative to the voice of the infant, data relative to the movements of the infants and other physical responses, or combination thereof. Restlessness behavior, for example, includes a correlation between pluralities of physical events performed by the infant including body movements and sounds that are produced by the infant (e.g., crying patterns, data relative to body movements that may reflect pressure applied by the infant on mattress 1, etc.).

According to an embodiment of the invention, the detection of a restless behavior by processor unit 33 may involve the following steps:

obtaining the actual behavior of the infant as represented by sensors' inputs;

comparing the processed data with predefined behavioral patterns (e.g., by utilizing an artificial intelligence event correlation model that is based on a plurality of past events) to determine whether the actual behavior reflects a restless behavior while considering the infant's growth. For example, the predefined behavioral pattern may include threshold parameters, which are affected by the weight or height of the infant. The rapidly changing weight and/or height can be obtained automatically (i.e., by processing measured pressure at pressure sensors provided at different locations while mattress 1 is inclined at a known angle measured by an embedded leveling means), or adjusted manually (e.g., by the infant's parents, via an input device associated with mattress 1, e.g., a smart mobile device); and

updating the predefined behavioral pattern for restlessness. This step may require an external confirmation, e.g., modifying a behavioral pattern that relates to a distress situation may require parents' confirmation.

According to an embodiment of the invention, in case that a restless behavior is detected, computing device 3 automatically activates a mechanism that causes mattress 1 to vibrate and/or tilt (e.g., a mechanism that creates gentle, soothing vibrations). The vibrating intensity of mattress 1 may change according to the detected response of the infant. For example, if sensor 2c of Fig. 1 (i.e., the microphone) detects that the infant is continuously crying, then computing device 3 may instruct the relevant mechanism to increase the vibrating intensity and/or to change the position of mattress 1 (e.g., may incline mattress 1 by several degrees with respect to the floor level). In such a way, mattress 1 may automatically react (dynamically change position and/or vibration intensity) until the infant calms down.

The activation of the mechanism may result in the generation of one or more reactions that may adaptively change until the infant calms down, such as activating a mechanical arrangement associated or embedded within the body of mattress 1 (e.g., activating a vibration mechanism that is adapted to gently rock mattress 1 or a crib in which mattress 1 is located), activating a device that is adapted to play audible and/or visual content (e.g., a remotely controlled crib mobile), etc.

According to an embodiment of the invention, one or more computer readable media having computer-executable components are provided. The components may include an event processor component 34, a voice analysis component 34a, a complex event processing component 35, a machine learning component 36, and a decision support component 37. When executed by processor unit 33, the event processor component 34 and voice analysis component 34a causes the processor unit 33 to receive a plurality of inputs representative of events including one or more physical (i.e., both mechanical and vocal) events associated with the infant. When executed by processor unit 33, the complex event processing component 35 causes processor unit 33 to compare an actual behavior that is based on the physical events to one or more predefined behavioral patterns, to determine whether the actual behavior is correlated to a predefined behavioral pattern, associated with a corresponding response script. When executed by processor unit 33, the machine learning component 36 causes processor unit 33 to utilize an artificial intelligence event correlation model 38 that is based on a plurality of past events, for comparing the actual behavior to predefined behavioral patterns, considering the infant growth and the past learning, to determine whether a predefined pattern update is required, or a new pattern should be added to the database. When executed by processor unit 33, decision support component 37 causes processor unit 33 to generate a notification signal or a combination of notification signals, according to a response script, resulting with a mechanical/audible/communication reaction or a combination thereof. For example, when a new behavioral pattern is detected, such as an infant who turns on his back for the first time, processor unit 33 prepares a database update file (e.g., a file comprising the relevant set of physical and vital signs parameters), and in parallel generates a notification signal to communication unit 32 for messaging to mobile device 11 (of Fig. 2A) or to any other receiving device, and may request parents' confirmation. Alternatively, if the new pattern is suspected to represent a distress situation, processor unit 33 generates a different notification signal for communication unit 32 to further send an activation signal to a remote alerting device.

While certain references are made to certain example system components as indicated by numerals 34-38 in Fig. 3, other components can be used as well and/or the example components can be combined into fewer components and/or divided into further components.

The characterization of behavioral states of infants that are part of pre-defined data provided to intelligent mattress 1 can be based on information found in many brain researches, such as an article by Prechtl HF, "The behavioral states of the newborn infant (a review) Brain Research" 1974;76:185-212. For example, according to such brain researches, at state 1, quiet sleep is characterized by regular respiration, slow and regular heart rate, and the absence of eye movements and gross muscle movements, with the exception of transient movements and heart rate increases if the infant is startled. Neonates in quiet sleep show an EEG pattern known as trace alternant, defined by bursts of high amplitude slow wave activity interspersed with low voltage activity. In some brain researches it is found that while older infants also exhibit a quiet sleep state, the trace alternant disappears by 46-48 weeks postconceptional age. At state 2, active sleep is characterized by variable respiration and heart rate and the presence of both slow and rapid eye movements, along with continuous EEG activity of mixed amplitude. When sleeping, infants do not show the sustained, specific pattern of physiological and behavioral activity characteristic of either quiet or active sleep, they are said to be in indeterminate sleep. For example, the first year of the infant's life is characterized by a number of developmental changes in sleep patterns, including an increase in quiet sleep, increase in the number of sleep cycles and total sleep cycle length, decrease in active sleep, decrease in indeterminate sleep, decrease in total sleep time, development of a clear circadian rhythm, etc. In the first few months after birth, as the forebrain matures and exerts greater control over brainstem and cortical regions to organize sleep-wake rhythms, sleep episodes may become longer and more continuous.

Fig. 4 is a flowchart showing a method of autonomously learning the behavior of a specific infant, according to an embodiment of the invention. According to this particular embodiment the method involves the following steps:

receiving, by computing device 3, a plurality of detected events associated with the infant (i.e., according to readings received from one or more sensors) who lies on mattress 1 (Figs. 1, 2A-2D) (bloc 41). The events represent at least one physical event or behavior of the infant, wherein the events are obtained from the readings of one or more of sensors 2a-2c (Fig. 1);

determining an actual behavior of the infant (bloc 42). An actual behavior of the infant is determined by computing device 3 based on the readings of sensor 2a-2c (e.g., the events can be crying sounds and/or pressure generated by the infant on sensor 2a- 2c);

comparing, by computing device 3, the actual behavior to predetermined behavioral patterns, to determine whether the actual behavior is similar to, or differs from a predefined behavioral pattern (bloc 43). The predefined behavioral patterns include a correlation between a plurality of events associated with the infant;

if the actual behavior is similar to a predefined behavioral pattern, respond according to an associated response script (bloc 45a), with a mechanical, audible, or communication means, e.g., to support and alert of infant's distress situations, or to encourage the infant's relaxation and return to sleep and ;

if the actual behavior differs from predefined behavioral patterns (bloc 44), updating the database of predefined behavioral patterns (bloc 45b). An indication of whether the actual behavior is abnormal is received either from a user (e.g., a parent of the infant) or automatically from data relative to characterization of behavioral patterns.

According to some embodiments, while the infant is placed on the intelligent mattress 1, the present invention can be used to alert and/or prevent non-legitimate or abnormal behavior of the infant either during sleep (e.g., apnea, sleep disorders, etc.) or while the infant is awake (e.g., suffocation, overheating, etc.). As will be appreciated by the skilled person, the arrangement described in the figures results in an intelligent mattress which is capable of monitoring an infant behavior (i.e., actual behavior), and autonomously responding to predefined restless/distress situations (i.e., behavioral patterns), of learning the behavior of a specific infant while considering its rapid growth (i.e., the evolving physical and behavioral expressions of the infant), and automatically( and/or user controlled) updating the characteristics of predefined situations.

For example, it is well known that during apnea, there is no movement of the muscles of inhalation, and the volume of the lungs initially remains unchanged. In such scenario, intelligent mattress 1 processes data received from one or more of sensors 2a-2c (e.g., sensors adapted to provide heartbeat readings and respiration rate), and accordingly determines the actual behavior of the infant. Next, intelligent mattress 1 compares the actual behavior to predefined behavioral patterns, to determine whether the actual behavior differs from the expected behavior considering the infant's age, associated physical data, and behavioral state. Upon detection of abnormal behavior, e.g., that may result from breathing disorders, intelligent mattress 1 generates a notification signal that results in an alert message being sent to one or more predefined mobile devices (e.g., to the infant's parents, or other alerting devices). Alternatively, or in addition, intelligent mattress 1 may react by activating a mechanical arrangement (either embedded within intelligent mattress 1 or located externally to it) that is adapted to physically change the orientation/position of the intelligent mattress 1 and accordingly of the infant located on intelligent mattress 1), in order to help the infant to return to a normal breathing sequence. In case of an early abnormal breathing detection, changing the orientation/position of the intelligent mattress 1 may assist the infant to breath properly or even may prevent a case of sudden infant death syndrome (SIDS).

All the above description and examples have been given for the purpose of illustration and are not intended to limit the invention in any way. Many different mechanisms, autonomous learning methods, electronic and logical elements can be employed, all without exceeding the scope of the invention.