The present invention relates to a wearable alarm device for alerting a third party to a wearer’s condition and to a method of use thereof.

Wearable devices to detect and report on a user’s condition are known. They are typically used as alarm devices to alert a third party when some form of accident or fall has befallen the wearer. Such“smart technology” exists in many forms.

A problem with known devices is the prevalence of“false positives” in the triggering of alarms. When an alarm is triggered and in fact no fall or accident has befallen the wearer, then it is frustrating for the third party that has been called to have to respond. The repeated triggering of false alarms can lead to a dangerous reaction from the third party. For example, the third party could become de-sensitized to the alarms, always believing that it is likely to be false. In some cases, users could even be led to switch off the alarm or certain aspects of it so as to minimise the chance of false alarms being generated. If this happens in the event of a true occurrence then the wearer can be left in serious trouble.

A number of examples of devices and technology exist to enable alarms to be raised in the event of some occurrence to a user. One example is US-A-9,572,503, which discloses a software app for a mobile device for alerting a custodian of a person to be protected of an emergency situation involving the person to be protected. The app includes software instructions for carrying out a method including establishing a range of normal heart rates for the person using a heart rate monitor; detecting a heart rate for the person that is outside of the established range; activating at least one of a camera, a microphone, an accelerometer and location indicator on the mobile device carried by the person and establishing a wireless data connection between the mobile device and a communication network. The device further provides for the transmission of data to a custodian from the activated camera, microphone, accelerometer, or location indicator via the communication network. Further examples are described in any or all of US-A-8,956,293,

US-A-9,492,092, US-A-8,956,294, US-A-9,055,928, US-A-9,566,077, US-A-8, 180,440, US-A-8,909,330, US-A-8, 594, 776. The documents contain descriptions and examples of various alarm systems that require the presence of at least two sensors positioned on a body. Typically, continuous monitoring of the sensors is utilised to identify, using various rules, when an alarm situation is encountered.

A further example is disclosed in US-A-8, 647, 268 which discloses an entire monitoring system for a person in a building that includes a plurality of wireless bases positioned in the building. A body mounted temperature sensor and a body mounted heart signal sensor together with a wearable device coupled to the sensors is provided. The wearable device is able to communicate with the wireless bases within the building and thereby monitor a user’s status. Typically, such a system might be utilised in a hospital or care home.

According to a first aspect of the present invention, there is provided a wearable alarm device for detecting an individual’s condition, the device comprising: a controller; one or more movement sensors coupled to the controller, and one or more sensors for detection of human vital signs also coupled to the controller, wherein in normal use the movement sensor is operative and the one or more sensors for detection of human vital signs are switched off, the device being arranged such that upon receipt by the controller of a trigger signal from the movement sensor, activating the one or more sensors for detection of human vital signs. The device can be used in embodiments for alerting a third party to a wearer’s condition.

A device is provided for detecting a wearer’s condition. The device is arranged such that in normal use many of its components are not activated or turned on such that they do not draw any power. In the event of detection of some a parameter associated with a user’s condition the one or more sensors for detection of human vital signs are activated, i.e. turned or switched on, such that they are able to detect human vital signs.

In one example, the movement sensor or sensors are 6 axis accelerometers. In one example, the movement sensor is 6-axis inertial sensor, including a digital, triaxial 12bit acceleration sensor and a digital, triaxial 16bit, ±2000 s gyroscope. In one example, the one or more sensors for detection of human vital signs include a detector for measuring heart rate.

In one example the one or more sensors for detection of human vital signs include a detector for measuring blood pressure.

In one example, the one or more sensors for detection of human vital signs include a detector for measuring SP02 levels.

In one example, one or more of the sensors for detection of human vital signs are optical sensors.

In one example, the wearable alarm device comprises a communication module to enable communication from the device to a third party via a communications network.

In one example, the communication module is a 3G/4G wireless communication module.

In one example, the controller is arranged to make an initial determination that an event has occurred and in response thereto to activate the one or more sensors for detection of human vital signs.

In one example, the wearable alarm device comprises a power management unit arranged to provide power to the one or more sensors for detection of human vital signs responsive to that an event has occurred, and to not provide power to the one or more sensors for detection of human vital signs unless such determination is made.

In one example, the power management unit is coupled to the controller and is arranged to provide power to the one or more sensors for detection of human vital signs through the controller.

In one example, the wearable alarm device comprises an audio processing unit to enable the device to receive and transmit audio signals. In one example, the wearable alarm device comprises a manual alarm activation button arranged upon activation thereof to cause the device to communicate with a 3 ^rd party to raise an alarm.

In one example, the wearable alarm device is configured as a watch for wearing on a user’s wrist.

According to a second aspect of the present invention, there is provided a method of detecting an individual’s condition, the individual wearing a wearable device comprising a controller, one or more movement sensors coupled to the controller, and one or more sensors for detection of human vital signs also coupled to the controller, the method comprising; wherein in normal use the movement sensor is operative and the one or more sensors for detection of human vital signs are switched off, the device being arranged such that upon receipt by the controller of a trigger signal from the movement sensor, activating the one or more sensors for detection of human vital signs. The method, in embodiments includes the step of alerting a third party to an individual’s condition.

In an example, the step of activating the one or more sensors for detection of human vital signs comprises activating one or more of a detector for measuring heart rate, a detector for measuring blood pressure, and a detector for measuring SP02 levels.

In an example, upon activation of the movement sensors, issuing an invitation to a wearer to confirm status of condition.

In an example, upon confirmation of status de-activating the one or more sensors for detection of human vital signs. This provides a significant advantage in terms of minimising power usage. Indeed, the step of turning off or deactivating the sensors for detection of human vital signs if a user indicates that no event has occurred ensures that power consumption is minimised since the sensors for detection of human vital signs will only have been turned on for a short period of time, i.e. the period of time before the potential event was detected and receipt of the user’s indication that is a false alarm. In one example, the method comprises learning the pattern of movement that has given rise to the false alarm and optionally storing this learned pattern for use in subsequent processing. This provides the advantage that detected patterns of movement which might appear to be the sign of a possible event, but are in fact simply normal (and safe) movement patterns of the wearer, can be recognised and in some cases when the same pattern of movement is subsequently detected the controller does not activate or turn on the sensors for detection of vital signs.

Embodiments of the present invention will now be described in detail with reference to the accompanying drawings, in which:

Figure 1 is a schematic block diagram showing the functional components of a device for alerting a third party to a wearer’s condition;

Figure 2A is a flow diagram showing an exemplary method of operation for a device such as that of Figure 1 ;

Figure 2B is a flow diagram showing another exemplary method of operation for a device such as that of Figure 1 ; and

Figure 3 is a schematic view of an assembled device according to figure 1 configured as a wrist wearable“watch device”.

A wearable arm device is provided for detecting an individual’s condition and, in embodiments, and if required, alerting a third party to a wearer’s condition. As will be described below, the device includes a controller and one or more movement sensors coupled to the controller. The device is arranged such that the movement sensors are operative to determine when a likely fall has occurred and in response to such detection, to cause the controller to switch on one or more other sensors arranged in the device for detecting human vital signs and optionally to switch on the communications module to enable the device to send data to a call centre. The one or more other detectors might include, for example, a light sensor (or plural light sensors) arranged to detect and report on heart rate, blood pressure or SP02 levels. A common single light sensor can be used to detect and report on all of heart rate, blood pressure or SP02 levels or alternatively separate dedicated sensors can be provided for detecting each of the parameters (or a selected pair).

In addition a temperature sensor is preferably provided to detect a user’s body temperature, which can be an indication of the need to raise an alarm. One example of temperature sensor includes a metallic disc (such as a brass disc) built into the housing of the wearable device and arranged in use in thermal contact with a wearer’s skin so as to be able to measure the wearer’s temperature. The thermal contact can be direct or indirect i.e. through another layer of material. The temperature sensor can be used to provide data to the microcontroller which in turn can store the data or via the

communication module send it on to some third party for processing. By selective timed measurements a continuous profile of body temperature of a wearer can be acquired which in turn can be used for example to determine sleep patterns of a wearer. Body temperature of a wearer can be used to model and determine sleep patterns of a user.

Thus, the power usage of the device is managed extremely efficiently since in normal use, the sensors for detecting the human vital signs and the communications module are not configured to draw any power from the device. They are only used, i.e. activated or turned on and then used to take measurements of various parameters, in the event that a fall has been detected. Indeed, at this point they may be turned on and function to provide readings to the controller which, in turn, as will be explained below, will make a decision as to whether or not an alarm is triggered.

Referring to Figure 1 , a schematic block diagram is shown illustrating the various components that are included in an exemplary wearable alarm device 2. A controller 4 is provided which might typically be a microcontroller or microprocessor as is well known.

A movement detector such as a 6-axis inertial sensor, consisting of: A digital, triaxial 12bit acceleration sensor and a digital, triaxial 16bit, ±2000 s gyroscope is provided coupled to the controller. A power management module 8 is provided which could typically comprise a rechargeable battery or a replaceable battery unit to provide power to the device. In a preferred example the power management module includes a battery which is rechargeable via connection to power lead such as a USB charging lead such that he device 2 can be charged in a manner similar to other USB rechargeable devices. One or more sensors 10 are provided which could be in the form of optical sensors for measuring human vital signs such as heart rate, blood pressure and SP02 levels. An audio processing unit 12 is provided which includes a microphone and speaker. The microphone and speaker can be an available integrated off the shelf component which enable a user to speak into the device for communication and also to hear third party communications received by the device 2.

Finally, a communications unit 14 is provided which enables the device 2 to effectively operate as a mobile telephone. The device 2 is thus able to communicate with external devices, such as via communication links including the internet, 3G/4G networks and any other appropriate communications network. The communications unit 14 is, in normal operation in a sleep mode so that it is not using power (similar to the sensors 10). It is only activated or turned on if the controller determines that it is needed to communicate with a 3 ^rd party such as a telecare centre or a user’s family member.

As will be explained below, the individual components might be assembled on a PCB or hardwired into a processor. In use, the device operates generally in that if there is an event that the controller of the movement sensor believes is a fall, then the other sensors 10 are activated and begin to accumulate data. The device controller 4 will assess the movement pattern just before the“fall”, as well as the subsequent movement. A typical fall, for example, could cause the sensors to report a sudden movement followed by an abrupt impact and very small subsequent movement. This could be generated, by the device initially falling to the ground on the arm of a user as they fall, and then the slow movements on the user’s wrist as they are in a state of shock and/or semi-consciousness.

Once light sensors 10 have been activated and are reporting on parameters such as heart rate, blood pressure and SP02 levels, and monitoring the trends, the outputs from the sensors 10 are provided to the controller 4. An algorithm operating in software on the controller, is arranged to make a decision as to whether or not a fall has taken place based on data from all of the sensors, i.e. both the movement sensor 6 and the optical sensors 10. In other words, this includes both the movement sensor that initially detected the likely fall and activated the other sensors 10 and the other sensors 10 themselves. The process typically takes such a few seconds. In the case of a suspected fall, the device is then arranged to speak to the wearer and ask them to confirm or cancel an alert. In the absence of any response from the wearer, the device is arranged to automatically call a response service which will involve a human being attempting to speak to the wearer through the device before calling first responders such as an ambulance.

The algorithm is one that receives as input variables, values from the various sensors and based on these make a determination of the likelihood of a fall or alarm worthy event having occurred. By utilising both the movement sensor and the sensors 10 for measuring human vital signs such as heart rate, blood pressure and SP02 levels the risk of a false positive is significantly reduced.

The data generated and received by the various sensors 6 and 10 is collected and stored in memory (not shown in Figure 1) associated with the controller 4. The data can be forwarded via the communications system 14 to a nominated first responder. Furthermore, the system is able to learn patterns of behaviour based on the collected data and use this in future determinations as to the likelihood of a fall having occurred.

Overall, then the device 2 is able to reduce the occurrence of false positives, i.e. an alarm being raised when in fact no event worthy of an alarm has occurred. The learning of behaviour pattern and correlation of user responses to enquiries as to whether or not a fall has happened means that the occurrence of false positives can be reduced. Furthermore, the selected activation of the optical sensors 10 means that power requirements for the device can be reduced and therefore the likelihood of the power source running flat when it is needed is correspondingly reduced.

Figure 2A is a flow diagram showing the typical steps in the operation of a method for detecting a user’s fall. Initially, at step 16 the movement sensor 6 of the device is operated and based on a determined movement or pattern of movement, if a fall is detected at step 18, sensors 10 such as optical sensors integrated into the device are activated at step 20. The optical sensors operate and collect data from a user and at step 22 provide these to the controller 4. At step 24, an algorithm is applied and a decision is made, based on the received data from the sensors, as to whether or not it is likely that fall has occurred and so whether or not an alarm should be triggered and the communications centre 14 switched on. If a determination is made that in all likelihood a fall has not occurred, the optical sensors can be turned off at step 26. As described above, before the sensors are turned off, an audible signal such as a recorded question can be played from the audio processing unit 12 to the wearer. The enquiry can be as to whether or not a fall has occurred and/or whether help is needed. Alternatively, if sufficient confidence exists, the sensors 10 can simply be turned off at step 26 without a question being asked. If a clear determination is made at step 24 that a fall has occurred, an alarm can be triggered.

The alarm could immediately communicate, e.g. using the communications system 14, with a first responder or could simply generate the question to the user and await a response therefrom.

Thus, it can be seen that critically, at step 20 the sensors 10 for measuring the vital signs are turned on and indeed before this stage in the method, the sensors 10 are turned off so that power usage of the device during normal operation, i.e. before a fall has actually been detected (or suspected), is minimal.

Figure 2B shows another example of a flow diagram showing a method of operation for a device such as that of Figure 1. Initially, the movement sensors in the device are in operation. At step 17 this is shown, leading on to step 19 at which a possible event is detected. Upon detection of the possible event, the sensor(s) for detection of human vital signs are activated or switched on, as is the communications module 14. Next, at step 23, the wearer is issued with an invitation to cancel the alarm if no event has actually occurred. In other words, if the movement sensors have detected some movement which has been identified as a possible event but in actual fact is a false alarm due to some other movement that the alarm device has undertaken, then to avoid raising any alarm and involving a third party, this can be cancelled at an early stage by the wearer.

Furthermore, the controller is arranged to learn the pattern of movement that has given rise to the false alarm and can use this in subsequent processing. For example, it could learn after a number of such false alarms that some detected pattern of movement which might appear to be the sign of a possible event, is in fact simply a normal (and safe) movement pattern of the wearer. The notification or invitation to the wearer to cancel if there is no event generated at step 23 is done locally by the controller within the device 2 and does not involve any third party.

If a user, at this stage, indicates that no event has occurred, then the method returns (via step 25 in which the sensors are turned off) to step 17. The step of turning off or deactivating the sensors for detection of human vital signs if a user indicates that no event has occurred ensures that power consumption is minimised since the sensors for detection of human vital signs will only have been turned on for a short period of time.

Next, if a user does not cancel the alarm at step 23, then the readings from the sensors turned on at step 21 are provided to the controller and an associated algorithm can be activated. At step 29, a telecare sensor is called by the device 2 and

simultaneously sends data relating to the vital signs of the wearer, i.e. as detected by the sensors turned on at step 21 , automatically to the telecare centre.

Next, an operator at the telecare centre at step 31 , is able to talk to the wearer of the device 2 through the device itself and is able, in real time, to obtain status information from the user and to view vital signs data which can, optionally, continue to be sent to the telecare operator during the conversation. If during conversation it becomes clear that the event can be cancelled and that the user is in fact OK, then the method returns to step 25 at which point the sensors are turned off and the method ultimately returns back to step 17. If, during communication with the wearer it becomes clear that a response is actually needed, then the telecare operator can do this manually and send assistance in the form of first aid or an ambulance, to assist the wearer.

Figure 3 is a schematic view of the actual wearable device that could be configured as a watch. The device 30 comprises a strap in the form of a conventional watch strap 32. The main body 34 includes the various components shown in and described with reference to Figure 1. An alarm button 36 is provided which, when pressed, can generate automatically a distress signal which will be sent via the communications system 14 to a telecare response service. The alarm button 36 is arranged on the surface of the watch in association with raised or textured features 38 which enable a user to navigate the watch surface by touch alone even if they cannot see the surface of the watch. This could be the case if they had suffered a fall and are lying in a prone position.

It will be appreciated that the watch 30 operates both as an alarm detection device and also as a mobile communication system such as a mobile telephone. With the incorporation of the communications centre 14, it is able to communicate with networks to enable a user to speak through the watch to a responder in a telecare centre. The number of the telecare centre could for example be hardwired into the device or the controller could be a programmable unit to enable the number to be changed. In addition, in one example further numbers are programmed into the controller such that as well as contacting the telecare centre, in the event of an event occurring, a family member or friend can also be immediately alerted.

In one example, as well as the general call being made the data stored by the device 2 is sent to the telecare centre. This means that in the event of a likely fall, the data from the sensors is provided, effectively in real time, to a human operator at a telecare centre which means that a human decision can be made to confirm (or overturn) the initial decision made by the algorithm operating in the device 2 itself.

After use for a period of time by a specific user, the controller and algorithm operate to learn behaviour and signal or data patterns that correspond to a fall. This means that with use, the risk of false positives is reduced further since the watch effectively knows for the user in question what patterns of movement or sets of sensor data are most likely to correspond to an actual fall and which data sets are likely not to.

The watch strap and body 32 and 34 are preferably formed of an anti-bacterial plastic or polymer material

Embodiments of the present invention have been described with particular reference to the examples illustrated. However, it will be appreciated that variations and modifications may be made to the examples described within the scope of the present invention.