GB2627246A - Activity monitor - Google Patents
Activity monitor Download PDFInfo
- Publication number
- GB2627246A GB2627246A GB2302231.2A GB202302231A GB2627246A GB 2627246 A GB2627246 A GB 2627246A GB 202302231 A GB202302231 A GB 202302231A GB 2627246 A GB2627246 A GB 2627246A
- Authority
- GB
- United Kingdom
- Prior art keywords
- electronic device
- data
- wearable electronic
- movement
- movement data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000000694 effects Effects 0.000 title abstract description 40
- 230000033001 locomotion Effects 0.000 claims abstract description 226
- 238000000034 method Methods 0.000 claims abstract description 87
- 238000012544 monitoring process Methods 0.000 claims abstract description 44
- 230000004044 response Effects 0.000 claims abstract description 32
- 230000003442 weekly effect Effects 0.000 claims abstract description 10
- 238000004891 communication Methods 0.000 claims description 124
- 230000008569 process Effects 0.000 claims description 29
- 230000005540 biological transmission Effects 0.000 claims description 10
- 238000010801 machine learning Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 4
- 230000000977 initiatory effect Effects 0.000 claims description 4
- 230000003612 virological effect Effects 0.000 claims description 4
- 238000013264 cohort analysis Methods 0.000 claims description 3
- 230000000630 rising effect Effects 0.000 claims description 3
- 238000005259 measurement Methods 0.000 description 15
- 230000007704 transition Effects 0.000 description 13
- 230000006870 function Effects 0.000 description 10
- 230000001413 cellular effect Effects 0.000 description 9
- 230000008859 change Effects 0.000 description 7
- 238000005265 energy consumption Methods 0.000 description 7
- 230000006399 behavior Effects 0.000 description 6
- 238000001514 detection method Methods 0.000 description 5
- 230000001965 increasing effect Effects 0.000 description 5
- 235000019580 granularity Nutrition 0.000 description 4
- 230000036541 health Effects 0.000 description 4
- 230000007774 longterm Effects 0.000 description 4
- 230000002618 waking effect Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 238000013480 data collection Methods 0.000 description 3
- 230000005802 health problem Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000008867 communication pathway Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 206010025482 malaise Diseases 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 206010000117 Abnormal behaviour Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000001154 acute effect Effects 0.000 description 1
- 230000002547 anomalous effect Effects 0.000 description 1
- 230000009118 appropriate response Effects 0.000 description 1
- 230000036760 body temperature Effects 0.000 description 1
- ZPUCINDJVBIVPJ-LJISPDSOSA-N cocaine Chemical compound O([C@H]1C[C@@H]2CC[C@@H](N2C)[C@H]1C(=O)OC)C(=O)C1=CC=CC=C1 ZPUCINDJVBIVPJ-LJISPDSOSA-N 0.000 description 1
- 230000001010 compromised effect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000003116 impacting effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 208000015181 infectious disease Diseases 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 230000036642 wellbeing Effects 0.000 description 1
- 210000000707 wrist Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0407—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis
- G08B21/0423—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis detecting deviation from an expected pattern of behaviour or schedule
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0438—Sensor means for detecting
- G08B21/0446—Sensor means for detecting worn on the body to detect changes of posture, e.g. a fall, inclination, acceleration, gait
Landscapes
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Gerontology & Geriatric Medicine (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
A method of monitoring user activity comprises: measuring a plurality of movement values of a user S206; storing data S208 based on the plurality of movement values to form a stored dataset S210 comprising daily or weekly historical data; subsequently monitoring movement data S212 of a wearable electronic device (110,fig.1) via at least one sensor (106,fig.1), such as an accelerometer; comparing the movement data with the stored dataset to determine a current metric S214; and transmitting a warning signal S218 via a transmitter (116,fig.1) of the wearable device to one or more electronic devices (140,150,fig.1) in response to determining that the current metric meets a predetermined condition S216. The method may further comprise determining that the wearable device is being worn by a user, for example via a capacitive skin sensor (108,fig.1), and monitoring movement data of the wearable device when it is determined that the user is wearing the wearable device. Movement data may be transmitted from the wearable device to a remote server (120,fig.1) whereby the current metric may be determined at the remote server.
Description
ACTIVITY MONITOR
FIELD OF THE INVENTION
The present invention relates to a method, device and system for monitoring user activity.
In particular, but not exclusively, the invention relates to a wearable electronic device for monitoring user activity and providing a warning signal to the wearer, or to third parties such as care providers.
BACKGROUND
Recent developments in personal warning systems have enabled vulnerable people, who may require an elevated level of care, to live autonomously in their own homes without constant in-person supervision. For example, improvements in long-life battery technologies and energy management have enabled the development of wearable electronic devices that do not require charging or can run for long periods without charging. Such devices provide a means of raising an alarm if a user of the wearable electronic device encounters a situation where assistance is required, at any time of the day.
Such devices are valuable in acute situations where a user is able to initiate an alarm and also in situations when an alarm is generated in response to a specific event, such as the detection of a fall by an accelerometer. However, not all immediate, emerging or pending health problems are known to a user, or signalled in such an overt manner. This means that known personal warning systems are not directed to, or sufficiently sensitive to, raise warnings, such as alarms or alerts, in such situations, for which earlier intervention could be beneficial. Additionally, increasing the functionality of known devices generally requires further complexity and energy consumption, which in turn drains the battery cells, making them unsuitable for situations where constant, long-term monitoring of vulnerable people is required.
SUMMARY OF INVENTION
In order to mitigate for at least some of the above-described problems, there is provided a method of providing a warning signal, the method comprising: measuring a plurality of movement values of a user; storing data based on the plurality of movement values to form a stored dataset comprising historical data for at least part of one or more daily 2 374 560v 1 and/or weekly cycles of the user; subsequently monitoring movement data provided by at least one sensor of a wearable electronic device comprising at least one sensor and a transmitter; comparing the movement data with the stored dataset to determine a current metric; and transmitting a warning signal to one or more electronic devices in response to determining that the current metric meets a predetermined warning metric trigger condition. Advantageously, continuous monitoring of a user's movement and comparison with their typical routines enables the identification of emerging issues and early intervention.
Preferably, the method comprises determining that the wearable electronic device is being worn by a user; and monitoring movement data provided by the sensor of the wearable electronic device when it is determined that the user is wearing the wearable electronic device. Advantageously, data is monitored when it is known that a user is wearing the electronic device, thereby providing improved identification of issues and aiding energy conservation.
Optionally, the movement data is compared with the stored dataset to determine a current metric at the wearable electronic device. Beneficially, local comparison of data reduces the complexity of the system.
Optionally, the method comprises transmitting the movement data from the wearable electronic device to a remote server; comparing the received movement data with the stored dataset at the remote server; and determining the current metric at the remote server. Advantageously, the centralised processing of data enables multiple devices and technologies to share resources and improves the adaptability of the devices to changes in conditions that may otherwise impact on reliability.
Preferably, the method comprises determining one or more data subsets indicative of an awake portion and/or a sleep portion within the stored dataset. Preferably, determining one or more data subsets indicative of an awake portion and/or a sleep portion comprises determining a rising edge and/or a falling edge indicative of the awake portion and/or the sleep portion within the stored data. Advantageously, the sleep and awake portions of daily cycles provide distinct physical states for the improved accuracy in identification of issues requiring intervention.
Preferably, a rate at which the movement data is monitored is determined based on correlating a current time with a corresponding time within a daily cycle based on the 2 374 560v 1 stored dataset. Advantageously, energy consumption is conserved based on reduced risk.
Preferably, the method comprises updating the stored dataset with the movement data from the wearable electronic device and/or with movement data from one or more other wearable electronic devices. Beneficially, the accuracy of the stored dataset is dynamically improved and supplemented with further data.
Preferably, the method comprises updating the stored dataset with movement data from a cohort of users, preferably wherein the cohort comprises one hundred or more users over a period of many months, for example at least one month, three months or six months.
Preferably, the cohort of users is categorised based on at least one of life stage, medical condition and geography. Advantageously, the use of cohort data improves the accuracy in identifying issues and providing warnings in response to issues arising within the cohort.
Preferably, comparing the movement data with the stored dataset to determine the current metric comprises filtering one or more data points in the movement data based on one or more conditions indicative of events for which a warning signal is not to be transmitted. Beneficially, explicable events that may occur from time to time are identified and false warning signals are not transmitted.
Preferably, transmitting the warning signal to one or more electronic devices in response to determining that the current metric meets a predetermined warning metric trigger condition, comprises: determining if the current metric meets a further predetermined warning metric trigger condition; and determining an electronic device of a plurality of electronic devices to which the warning message is transmitted based on whether the current metric meets a predetermined warning metric trigger condition and/or meets a further predetermined warning metric trigger condition. Advantageously, warning signals are directed for the appropriate response, such that the response is initiated in a most efficient manner.
Preferably, the movement data is based on a plurality of movement values, preferably wherein the movement data is an average of a plurality of movement values in a predetermined time window or an average of a predetermined number of movement values. Preferably, measuring a plurality of movement values of a user comprises 2 374 560v 1 recording the plurality of movement values based on a plurality of respective inputs at a sensor of a wearable electronic device, at regular time intervals. Beneficially, continuously monitored data is provided that enables accurate and reliable responses.
Preferably, measuring a plurality of movement values of a user comprises recording the plurality of movement values based on a plurality of respective inputs at a sensor of a wearable electronic device at regular time intervals continuously for at least seven days, preferably for three consecutive months, more preferably for at least six consecutive months and yet more preferably for at least twelve consecutive months. Advantageously, a stream of uninterrupted data provides improved detection of issues.
Preferably, the movement data is compared with the stored dataset to determining the current metric based on at least one of the day of the week and the month of the year. Advantageously, routine occurrences that could otherwise be interpreted differently are included in a comparison of currently monitored data with historic data.
Preferably, the method comprises determining whether the wearable electronic device is in communication range of a first intermediate device; transmitting monitored movement data to a first intermediate device when it is determined that the wearable electronic device is in communication range of the first intermediate device; determining whether the wearable electronic device is in communication range of a second intermediate device when it is determined that the wearable electronic device is not in communication range of the first intermediate device; and transmitting monitored movement data to a second intermediate device when it is determined that the wearable electronic device is not in communication range of the first intermediate device. Preferably, the method further comprises: temporarily storing monitored movement data at a memory of the wearable electronic device when it is determined that the wearable electronic device is not in communication range with the first intermediate device and the second intermediate device; and subsequently transmitting monitored movement data to the first intermediate device or the second intermediate device when at least one of the first intermediate device and the second intermediate device is determined to be in communication range. Preferably, the rate of monitoring movement data is based on whether the wearable electronic device is transmitting monitored movement data to the first intermediate device or to the second intermediate device. Advantageously, the system manages energy and reliability requirements to ensure accurate, reliable and efficient decision making.
2 374 560v 1 Preferably, the method comprises determining the predetermined warning metric trigger condition and/or further predetermined warning metric trigger condition using cohort analysis and/or machine learning. Advantageously, the accuracy of the comparison, and hence early warning signal, is improved.
Preferably, the method comprises providing an indication of the state of the battery power of the wearable electronic device, wherein the rate of monitoring movement data is based on the indication of the state of the battery power of the wearable electronic device and/or wherein the method further comprises initiating a process to ensure battery continuity based on the indication of the state of the battery power of the wearable electronic device, wherein the process comprises rapid battery replacement or rapid battery charging such that the battery is replaced or charged with in a time period less than an interval between transmissions of monitored data. Beneficially, continuity of measurements and/or monitoring is provided, thereby enhancing the ability to determine issues resulting in warning signals and early intervention.
Preferably, the method comprises measuring a plurality of temperature values of a user; storing data based on the plurality of temperature values to form the stored dataset comprising historical data for at least part of one or more daily cycles of the user; subsequently monitoring temperature data provided by the wearable electronic device; and comparing the movement data and temperature data with the stored dataset to determine the current metric. Beneficially, the combination of temperature and movement measurements provides improved identification of issues.
Preferably, determining that the current metric meets a predetermined warning metric trigger condition comprises comparing the monitored movement data of the user with the stored dataset and monitored movement data from a cohort of one or more users and/or one or more inputs based on an event, as a weather event, sporting event, television broadcast, religious event, bank holiday, viral outbreak, news event, national security event and /or political event. Beneficially, the use of corresponding monitored movement data from a cohort and/or input from events known to affect user behaviour improves triggering of warning signals by reducing the number of false positives.
There is also provided a wearable electronic device comprising: at least one sensor; a processor; a memory; and a wireless transmitter, wherein the wearable electronic device is configured to perform the method described herein.
2 374 560v 1 There is also provided a system comprising: a wearable electronic device; a server; and one or more electronic devices, wherein the system is configured to perform the method described herein.
There is also provide a computer readable medium comprising instructions which, when executed by a computer, cause the computer to carry out the method described herein.
Further aspects of the invention will be apparent from the description and the appended claims.
BRIEF DESCRIPTION OF THE FIGURES
The invention will now be further and more particularly described, by way of example only, and with reference to the accompanying drawings, in which: Figure 1 shows a system for monitoring user activity; Figure 2 shows a flowchart of a method for providing a warning signal; and Figures 3 to 6 show plots of sensed movement against time for a monitored data and a stored dataset, for different scenarios.
DETAILED DESCRIPTION OF FIGURES
A user's activity levels and physiological movement reflect various aspects of their wellbeing. Identifying changes in daily and weekly activity levels, and the correlation between waking and sleeping hours, helps to identify immediate, imminent or pending health problems. Early identification of such changes provides early identification of the problems, enabling intervention at an early stage.
For example, many elderly people have a routine that does not change from week to week. For such people, a change in their daily activity results from a particular problem. For example, an elderly person who is housebound may feel unwell and stay in bed longer than usual. They may have to get up several times in the night to use the toilet.
During the day they may feel unwell and sit and watch television, rather than going about their normal domestic tasks.
In such situations, the users may not realise that they are less active and/or may be reluctant to ask for help. This can often lead to complications which could be avoided if the problem is treated at an earlier stage. For example, an infection which may otherwise 2 374 560v 1 be minor if treated early could cause serious problems if left untreated. Accordingly, early detection of such problems has significant benefits.
System for monitoring user activity Figure 1 shows a system 100 for monitoring user activity. There is shown a wearable electronic device 110 having a first sensor 106, a second sensor 108, an additional sensor 105, a memory 112, a processor 114 and an interface 116. There is also shown a server 120 having a memory 122, a processor 124 and an interface 126. There is also shown a first electronic device 140 having a memory 142, a processor 144 and an interface 146, and a second electronic device 150 having a memory 152, a processor 154 and an interface 156. There is also shown a communication device 101 having an interface 103 and a further communication device 102 having an interface 104. In further examples, the devices shown in the system 100 have different and/or additional components enabling the functionality described herein.
The wearable electronic device 110 is a device that is used to measure and record a person's daily movement. In an example, the wearable electronic device 110 is a wrist worn device that is waterproof, shock proof, comfortable to wear, does not require overnight charging and can operate for many months or years. The wearable electronic device 110 is worn by a user at all times, providing an effective monitoring and warning system. In further examples, the wearable electronic device 110 is arranged to be attached to a user directly or indirectly in any suitable manner for providing movement measurements for an extended period of time (for example, over many daily cycles of 24 hours).
The wearable electronic device 110 is configured to operate continuously, twenty-four hours a day for at least three months without being removed, for example without being removed in order to charge a battery. Continuous monitoring involves collecting sensor data at regular intervals, for example at a predetermined rate of data collection. The predetermined rate of data collection is configurable. However, the data collection does not stop, for example to allow a battery to be recharged. In further examples, the wearable electronic device 110 is configured to operate continuously, twenty four hours a day for at least six months and, preferably, for at least twelve months. Beneficially, the use of a wearable electronic device that does not need to be removed, for example, in order to charge a battery, meaning that data can be gathered in an uninterrupted manner for an 2 374 560v 1 extended period of time, enabling long-term changes to be determined whilst ensuring continuous around-the-clock monitoring.
The wearable electronic device 110 provides an indication of the state of the battery that provides power to the wearable electronic device 110. The provision of such an indication means that the battery can be replaced before it expires. This is important, as the continuous monitoring of a user ensures that there is no device downtime, during which a warning could otherwise be signalled. The replacement of a battery or the electronic device, or the rapid recharging of the battery, is preferably completed within a time period of fifteen minutes. This means that the provision of continuous monitoring is not compromised and/or that the rate of continuous monitoring remains substantially constant.
The first sensor 106 of the wearable electronic device 110 is a 3-axis accelerometer that provides inputs to determine motion of the wearable electronic device 110 and hence motion of the user to whom the wearable electronic device is attached. In further examples, the first sensor 106 is a different type of sensor configured to provide motion information.
The second sensor 108 of the wearable electronic device 110 is a capacitive skin sensor configured to determine the presence of a user and hence to confirm that the wearable electronic device 110 is being worn by the user. In further examples, the second sensor 108 is a different type of proximity device configured to determine that the wearable electronic device 110 is being worn by the user, for example, based on the determination of at least one of movement, orientation, proximity to the body and temperature. In an example, the second sensor 108 is a mechanical switch used to determine if the wearable electronic devices is being worn by a user. In some examples, an additional sensor 105 is a temperature sensor configured to measure body temperature and/or to detect if the wearable electronic device 110 is being worn by a user. In further examples, the wearable electronic device 1 10 comprises one or more further sensors.
The first sensor 106 and the second sensor 108 of the wearable electronic device 110 provide inputs which are processed at the processor 114 of the wearable electronic device 110. Processed inputs are optionally stored in the memory 112 of the wearable electronic device 110 and/or transmitted to one or more other devices via the communication interface 116 of the wearable electronic device 110.
2 374 560v 1 Beneficially, the combination of a first sensor 106 for detecting motion a second sensor 108 for detecting that the wearable electronic device 110 is being worn by a user means that activity is only analysed for when the wearable electronic device 110 is being worn. This provides information that enables the determination that a device has been removed from a user. Further, there is a relative reduction in power requirements that arises from only analysing data arising from measurements when the wearable electronic device 110 is being worn. This contrasts with known sports/fitness trackers, which operate on the assumption that the user is alive and thus the absence of movement/heartbeat etc., means that the wearable device is not being worn.
Advantageously, wearable electronic devices with long battery life (e.g., 12 months or more) can be used for long-term, continuous monitoring of users. Such devices do not require frequent charging, as is the case for commercial wearable smart devices and thus make it possible to build a full picture of a person's movement patterns, both within their home and further afield. In contrast to devices that monitor for immediate changes in a user's situation (e.g., a fall, or significant change in heart rate), long-life wearable electronic devices enable a longer-term picture to be developed, thus enabling slow declines in health to be detected at the earliest opportunity.
Whilst the wearable electronic device 110 is described with reference to a first sensor 106, a second sensor 108 and an additional sensor 105, in further examples there are yet further additional or alternative sensors.
The wearable electronic device 110, communication device 101, server 120, further communication device 102, first electronic device 140 and second electronic device 150 form part of a network 130. In an example, the network 130 is the internet. In further examples, the network 130 is a local area network or a cellular network. In further examples, the network 130 is a network of two or more interconnected sub-networks, such as the internet and a cellular network.
The wearable electronic device 110 is in communication with the server 120 via a wireless communication path 132 between the interface 116 of the wearable electronic device 110 and the interface 103 of the communication device 101. As such, the communication device 101 is a first intermediate device between the wearable electronic device 110 and the server 120. The communication device 101 is shown to have a communication path 109 to the network 130 via the interface 103. The communication device 101 is a base station. In an example, the communication device 101 communicates with the network 2 374 560v 1 via a further device, such as an internet router. Alternatively, or additionally, the communication device 101 is in direct communication with the server 120 through a further communication device (not shown) between the interface 103 of the communication device 101 and the interface 126 of the server 120.
Alternatively, or additionally, the wearable electronic device 110 is in communication with the server 120 via a wireless communication path 132 between the interface 116 of the wearable electronic device 110 and the interface 104 of a further communication device 102. As such, the further communication device 102 is a second intermediate device between the wearable electronic device 110 and the server 120. The further communication device 102 is shown to have a communication path 135 to the network via the interface 104. The communication device 102 is a smartphone. In an example, the communication device 102 communicates with the network 130 via a further device, such as a cellular base station forming a cellular sub-network 130 of the network. In a further example, the communication device 102 communicates with the network 130 via a further device, such as an internet router, to which it may form a direct or indirect communication path.
Alternatively, or additionally, the communication device 102 is in direct communication with the server 120 through a further communication path (not shown) between the interface 104 of the communication device 102 and the interface 126 of the server 120.
The interface 116 of the wearable electronic device 110 provides communication between the wearable electronic device 110 and one or more further devices. The interface 116 comprises a transmitter. In some examples, the interface comprises a plurality of transmitters, enabling transmission at different frequencies and in accordance with different protocols. In some examples, the interface comprises one or more receivers, enabling two way communication at two or more different frequencies and in accordance with two or more different communication protocols.
Alternatively, in a further example, the wearable electronic device 110 communicates directly via communication path 133 or with the network 130, which in turn facilitates communication with the server 120 via the interface 126 of the server. In further examples, additional and/or alternative devices are used to enable communication between the wearable electronic device 110, the server 120 and/or the first electronic device 140 and the second electronic device 150. For example, additional devices such 2 374 560v 1 as access points, mesh points, repeaters and/or gateways are used to facilitate communication within the system 100.
The communication device 101 is a base station connected via an Ethernet connection to the Internet. In further examples, the communication device 101 is any type of device forming an intermediate step along a communication pathway. The further communication device 102 is a smartphone connected to the Internet by a Wi-Fi connection or via a cellular connection. The wearable electronic device 110 is connected to the smartphone 102 by a Bluetooth® communication link. In further examples, the further communication device 102 is any type of device forming an intermediate step along a communication pathway. In an example, the wearable electronic device 110 communicates with the communication device 101 using a different communication protocol compared with the communication protocol that the wearable electronic device 110 uses to communicate with the further communication device 102. Accordingly, the wearable electronic device 110 selectively communicates on different frequencies via a plurality of transmitters and/or receivers forming part of the interface 116 of the wearable electronic device 110. In an example, the energy consumption varies as a function of the communication protocol and the wearable electronic device 110 selects the communication protocol based on the lowest energy consumption protocol that is available. In further examples, the communication protocol is selected based on the reliability of the connection, and/or the range of connection to be used.
The first electronic device 140 and the second electronic device 150 are in communication with the server 120. The first electronic device 140 and the second electronic device 150 are smartphones. Alternatively, in further examples the first and/or second electronic devices 140, 150 are any suitable device, such as a mobile phone, landline, computer terminal, tablet, VoIP enabled device or other suitable device for providing warnings, such as alarms and alerts, to one or more responders.
The server 120 is a remote server 120 capable of analysing and storing information. The server 120 has a memory 122. In further examples, the server 120 is in communication with one or more further devices, such as memory databases.
The first electronic device 140 and the second electronic device 150 communicate directly with the server 120 via the network 130. In further examples, there are additional and/or alternative intermediate steps between the first electronic device 140 and the server 120, and between the second electronic device 150 and the server 120. For example, the first 2 374 560v 1 electronic device 140 and/or the second electronic device 150 are in communication with a cellular network forming a sub-network of the network 130.
The server 120 communicates with the network 130 via an interface 126 of the server 120, which provides a communication path 134 to the network 130 and the first electronic device 140 also communicates with the network 130 via an interface 146 providing a communication path 137. The second electronic device 150 also communicates with the network 130 via an interface 156 providing a communication path 138. Alternatively and/or additionally, the first and/or second electronic devices 140, 150 communicate directly with the server 120. For example, the second electronic device 150 is shown to communicate with the server 120 via a communication path 136 between the interface 156 of the second electronic device 150 and the interface 126 of the server 120.
The communication paths 107, 109, 131, 132, 133, 134, 135, 136, 137, 138 are wireless communication paths between wireless interfaces 116, 103, 104, 126, 146, 156. However, in further examples, at least some of the communication paths 107, 109, 131, 132, 133, 134, 35, 136, 137, 138 are wired communication paths between interfaces 116, 103, 104, 126, 146, 156. In further examples the system 100 is arranged to provide the functionality described herein through the direct and/or indirect communication between devices using wireless and/or wired communication paths.
The first electronic device 140 and the second electronic device 150 are devices associated with a first responder and a second responder, respectively. Responders are health professionals, or friends or family of the user, for example. The first electronic device 140 and the second electronic device 150 comprise outputs, such as visual and/or audio outputs, which can be used to provide warning signals, such as alarms, alerts and other decisive actions to the responders.
Examples of a warning signal include sounding an alarm, SMS text messages, e-mails and telephone calls. Further examples of warning signals are computer implemented instructions directing one or more physical actions in response to determining that a warning metric trigger condition has been met. For example, the warning signal is an instruction to an Internet of Things (loT) device at the location of the user in direct response to determining that a warning metric trigger condition has been met. For example, physical conditions within a home are altered in response to the warning signal.
2 374 560v 1 In further examples, the first electronic device 140 and/or second electronic device 150 are configured to control one or more further technical systems in response to receiving a warning signal. Whilst one wearable electronic device 110 is shown, in further examples there are additional wearable electronic devices forming part of the system 100. The additional wearable electronic devices operate in an analogous manner to the wearable electronic device 110 and provide data to the server 120, which in turn can determine warning signals based on a combination of data from different wearable electronic devices.
Process for monitoring user activity and initiating a warning signal Figure 2 shows a flowchart showing a process 200 for establishing a user's daily and weekly sleep-awake cycle, detecting problems and initiating a warning signal. The method is implemented in the system 100 described with respect to Figure 1. In further examples, the method is implemented in any appropriate computing system comprising a wearable electronic device and further electronic devices configured to receive warning signals. Beneficially, a wearable electronic device is used continuously to monitor a user, who may be vulnerable, over an extended period of time, in order to initiate a warning signal raising an alarm/alert at one or more remote locations, in the event that untoward behaviours are measured. The process 200 is at least partially implemented at a wearable electronic device, such as the wearable electronic device 110 described with reference to Figure 1, in order to transmit warning signals to one or more remote electronic devices, such as the first electronic device 140 and the second electronic device 150 described with reference to Figure 1. Whilst the wearable electronic device 110 is used to take measurements of a user's movement, the analysis of such measurements is performed at a remote server, such as the server 120 described in relation to Figure 1. Alternatively, the analysis is performed at the wearable electronic device 110 and/or any further devices implemented in the system 100.
The process 200 involves the generation of a stored dataset and the subsequent monitoring of data for comparison with the stored dataset. Advantageously, the same wearable electronic device 110 is used to form a stored dataset and to subsequently monitor a user's movements. Accordingly, the same data gathering process is used for providing a historical, stored dataset and for subsequently monitoring a user's movements. Alternatively, a stored dataset of a user's typical movements is generated independently using the same wearable electronic device or a different wearable 2 374 560v 1 electronic device. For example, data is gathered at a first rate in order generate a stored dataset and, subsequently, data is monitored at a second, different rate in order to compare a user's current daily cycle of movement with their typical daily cycle of movement.
Forming a stored dataset of historical data A wearable electronic device 110 is used to provide a stored dataset and the process 200 is initiated at step S202. The process 200 then moves to step 5204, where, optionally, it is determined that a device, such as the wearable electronic device 110 of the system 100 described with reference to Figure 1 is being worn by a user. Alternatively, the process moves to step S206, where movement of the wearable electronic device 110 is sensed.
If data is to be sensed in response to determining that the device is being worn at step S204, it is determined whether the wearable electronic device 110 is being worn by a user, for example in response to a measurement by the second sensor 108. When it is determined that the wearable electronic device 110 is being worn by a user, the process 200 moves to step S206, where movement of the wearable electronic device 110 is sensed.
The wearable electronic device 110 measures user movement using the first sensor 106. In an example, the first sensor 106 is a 3 axis accelerometer that samples movement at a rate of 12 Hertz. The movement of the 3 axes are resolved to a single vector. The magnitude of this vector is calculated at the processor 114 and compared with a previously recorded value stored in the memory 112 such that the difference is accumulated to provide a measure of movement. The processor 114 determines an average and the result is transmitted to the communication device 101, which in turn relays the information to the server 120 via the network 130, every 15 minutes.
Accordingly, a measure of user activity as a function of time at a predetermined granularity is acquired, such that a plot of user movement values at fifteen minute intervals for a twenty four hour cycle is provided. The measure of movement is configurable to provide different granularities of movement as a function of time. In further examples, the first sensor determines movement based on different sensed inputs. In further examples, the rate of sampling and transmission is different and/or the acquired measurement data is processed differently to provide a value indicative of activity associated with a time of day such that a daily cycle of user movement is recorded.
2 374 560v 1 Whilst the wearable electronic device 110 transmits the data at fifteen minute intervals to the server 120 via the base station communication device 101, in the event that it is determined that the base station communication device 101 is not within range (for example, if the user has left their home, where the base station communication device 101 is fixed), the wearable electronic device 110 instead transmits the data at fifteen minute intervals to the further communication device 102 via a Bluetoothe communication link. The further communication device 102 then relays the data to the server 120 via the network 130 (which may include sub-networks, such as a cellular network and the internet). Advantageously, the wearable electronic device 110 continues to provide user movement data through a sequential hierarchy of preferable communication routes and, if no communication is possible, stores the data and transmits the data at a later time. Beneficially communicating data via the base station communication device 101 may provide one or more advantages, such as being more reliable/robust, using less energy and or providing an improved data transfer rate compared with communicating data via the smartphone further communication device 102, for example.
Optionally, the rate at which data is communicated from the wearable electronic device 110 varies as a function of the manner in which it is communicated. For example, when a user is within range of a base station, the rate at which data is communicated may be different from when a user is not within range of the base station, but is within range of their smartphone. This is beneficial as less energy may be used for communication methods that have higher energy requirements, for example by reducing the rate of data communication (and the reduced rate of data communication may be less likely to be problematic when the user has left their home). The rate at which data is monitored may also be variable in response to the communication methods available. In an example, the rate at which data is monitored is maintained and data is stored locally in the memory 112 of the wearable electronic device 110 until such a time that a communication path is within range and/or available.
The process 200 then moves to step S208, where the movement sensed by the wearable electronic device 110 is recorded to form a dataset. For example, the information received at the server 120 from the wearable electronic device 110 is stored in the memory 122 of the server 120. The process 200 then moves to step S210 where a stored dataset is formed based on the information received at the server 120 from the wearable electronic device 110. Accordingly, the stored data is based on a plurality of movement values to form a stored dataset. The stored dataset comprises historical data for at least 2 374 560v 1 part of one or more daily and/or weekly cycles of the user and/or one or more other users. When forming a stored dataset, any interruptions in the continuous monitoring of a user are identified and excluded from the dataset.
Steps S204 to S208 can be repeated in order to provide further data in the dataset. The stored dataset comprises an average movement value for each given time of a daily and/or weekly cycle based on the number of repeated measurements. The dataset is enhanced by repeated daily and/or weekly cycles of the measurements such that the stored dataset reflects a user's daily and/or weekly routine. Monitoring a user's movement routine over a period of a few days enables an average of the user's movement routine to be calculated. Preferably, the stored dataset comprises historical data that is an average of data recorded over a period of one or more months. However, differences between a user's current daily movements and their historical daily movements may be determined based on historical data recorded over a shorter time period, such as at least one day. Whilst the stored dataset reflects a daily cycle, in further examples, the stored dataset comprises different daily cycles for each day of the week, such that the daily cycle in the stored dataset is subsequently compared against a user's current daily cycle as a function of the day of the week. In further example, additionally or alternatively, the stored dataset comprises different daily cycles for each month of the year, such that the comparison of the stored dataset with the current daily cycle is made as a function of the day and/or month of the year.
Whilst the stored dataset is formed based on measurements from the wearable electronic device 110, in further examples, the stored dataset is formed based on measurements from additional or alternative sources, such as different wearable electronic devices of the user. In some examples, the stored dataset is supplemented with data from different users. Advantageously, the use of data from a large cohort, for example six or more months of continuously monitored data, from 100 or more users provides data with further nuances associated with typical user movement. The use of information from other users can be used to determine the warning trigger metrics for anomalous situations. A cohort of users can be categorised based on different measures, such as life stage, medical condition and geography. Advantageously, supplementing the stored dataset based on a cohort of users within a same category as a user improves the determination of issues related to that same category.
2 374 560v 1 The metrics and algorithms used to trigger a warning can be continually improved, for example by using machine learning to train a system using historical movement data from a cohort group of users where these users have provided periods when warnings should have been detected. Figure 3 shows a plot 300 of sensed movement against time for a monitored data for a 24 hour period and a stored dataset, obtained in accordance with steps S206 and S208 of Figure 2. The plot is indicative of normal activity. Movements are shown on the y axis 304 and time is shown on the x axis 302. There is shown a dashed line trace of the averaged historical data 308 and a further line trace of the monitored data 306 for one daily cycle (24 hour period) starting at midnight on one day and ending at midnight of the next day. The line trace of the monitored data 306 for one daily cycle is incorporated into the historical stored dataset when the historical stored dataset is being generated.
The 24 hour period can be divided into different portions. There is a first portion corresponding to a sleep portion 314, where there is a low level of activity between 0 hours (midnight, for example) and approximately 7 hours, demarcated by a first line 310.
There is a second portion corresponding to an awake portion 318, demarcated by a second line 312 and a transition portion 316 between the sleep portion 314 and the awake portion 318. In further examples, the transition portion 316 is determined to form part of the sleep portion 314 and/or awake portion 318.
Advantageously, determining one or more data subsets indicative of an awake portion and/or a sleep portion within the stored dataset provides an effective, reliable and objective measure against which deviations can be determined.
Beneficially, the identification of one or more data subsets indicative of an awake portion and/or a sleep portion is further enhanced by determining a rising edge and/or a falling edge indicative of the awake portion and/or the sleep portion within the stored data. For example, determination of the gradient of the trace within the transition portion 316 shown at Figure 3 is a helpful indicator of the different data subsets within a daily cycle for both average historical data and recently accumulated, monitored, data.
Once a stored dataset of a user's typical daily movement has been provided, subsequently monitored data can be compared against the stored dataset in order to determine variances and hence to trigger warning signals when issues are identified. Steps 204 to S210 provide a stored dataset for use in providing a warning signal, such as an alarm or an alert signal. The process of forming a dataset involves sensing movement 2 374 560v 1 when the device is being worn. Whilst the stored dataset is provided by movements sensed and recorded by the wearable electronic device 110, alternatively a stored dataset for use in providing warning signal is provided by one or more additional devices. If the stored dataset has already been formed, the process for providing a warning signal can be initiated at step S212.
Monitoring data At step S212 of the process S200, the wearable electronic device 110 monitors data provided by a sensor, such as the first sensor 106 or the second sensor 108 of the wearable electronic device 110. The monitored movement data is sent to the remote server 120 for further processing. The process for monitoring movement data is the same as that used to form a stored dataset. Alternatively, the parameters for monitoring a user's movement are adapted to provide further advantages, such as reduced energy consumption. In further examples, alternatively, the monitored movement data is processed locally at the wearable electronic device 110.
The wearable electronic device 110 records a plurality of movement values based on a plurality of respective inputs at the sensor, at regular time intervals. Optionally and advantageously, the plurality of movement values are sensed and recorded in response to determining that the wearable electronic device is being worn by a user. The monitoring of data is performed in the same way in which data is gathered to form a historical stored dataset of daily cycles of movement. The first sensor 106 is a 3 axis accelerometer that samples movement at a rate of 12 Hertz. The movement of the 3 axes are resolved to a single vector. The magnitude of this vector is calculated at the processor 114 and compared with a previously recorded value stored in the memory 112 such that the difference is accumulated to provide a measure of movement. The processor 114 determines an average and the result is transmitted to the communication device 101, which in turn relays the information to the server 120 via the network 130 every 15 minutes. Accordingly, a measure of user activity as a function of time at a predetermined granularity is acquired, such that a plot of user movement values at fifteen minute intervals for a twenty four hour cycle is provided. The measure of movement is configurable to provide different granularities of movement as a function of time. In further examples, the first sensor 106 determines movement based on different sensed inputs. In further examples, the rate of sampling and transmission of is at a different rate and/or the acquired measurement data is processed differently to provide a value 2 374 560v 1 indicative of activity associated with a time of day such that a daily cycle of user movement is recorded. In further examples, movement data sensed at the first sensor 106 of the wearable electronic device 110 is processed locally by the processor 114 of the wearable electronic device 110. Processed data can be transmitted to the server 120 via one or more communication routes. Processed data can be stored locally in the memory 112 and accumulated processed data can be transmitted to the server 120 at a later time.
Alternatively, or additionally, the rate at which the measurement data is acquired by the sensor 106 is dependent on the variation in movements sensed. For example, if there low variation in movement sensed this causes the rate at which movements are sensed to reduce. For example, measurements are sensed at a rate of one per minute or higher.
Whilst the wearable electronic device 110 transmits the data at fifteen minute intervals to the server 120 via the base station communication device 101, in the event that it is determined that the base station communication device 101 is not within range (for example, if the user has left their home, where the base station communication device 101 is fixed), the wearable electronic device 110 instead transmits the data at fifteen minute intervals to the further communication device 102 via a Bluetooth® communication link. The further communication device 102 then relays the data to the server 120 via the network 130 (which may include sub-networks, such as a cellular network and the internet). In the event that there are no available communication paths, the wearable electronic device stores the data locally in the memory 112 and transmits it at a later time.
Advantageously, the wearable electronic device 110 continues to provide user movement data through a sequential hierarchy of preferable communication routes. Beneficially, the energy used when communicating data via the base station communication device 101 is less than when used communicating data via the smartphone further communication device 102.
Beneficially, the connection from the wearable electronic device 110, such as a wristband, to the remote server 120 is reliable, allowing reliable transfer of the daily movement data to the remote server 120. In an example, the connection enables communication at least once an hour. The wearable electronic device 110 can implement multiple wireless connections to achieve this reliability The first could be a proprietary implementation that provides sufficient range to link the wearable electronic device to a base station, such as the communication device 101, for coverage throughout the user's house and garden. A secondary connection could utilise a short range link such as Bluetooth to connect the 2 374 560v 1 wearable electronic device 110 to a mobile phone, such as the further communication device 102. An alternative solution is to use the cellular network forming part of the network 130.
Optionally, the rate at which data is communicated from the wearable electronic device 110 varies as a function of the manner in which it is communicated. For example, when a user is within range of a base station, the rate at which data is communicated may be different than when a user is not within range of the base station, but is within range of their smartphone. This may be beneficial as less energy is used for communication methods that have higher energy requirements (for example, by reducing the rate of data communication, where the reduced rate of data communication may be less likely to be problematic when the user has left their home). The rate at which data is monitored may also be variable in response to the communication methods available. In an example, the rate at which data is monitored is maintained and data is stored locally in the memory 112 of the wearable electronic device 110 until such a time that a communication path is within range and/or available.
The movement values provided by the sensor 106, 108 of the wearable electronic device 110 when it is determined that the user is wearing the wearable electronic device 110 are used to provide monitored data.
Movement data is based on a plurality of movement values, preferably wherein the movement data is an average of a plurality of movement values in a predetermined time window or an average of a predetermined number of movement values.
Advantageously, the rate at which the movement data is monitored is determined based on correlating a current time with a corresponding time within a daily cycle based on the stored dataset. For example, the rate at which movement data is monitored may reduce in response to determining that a current time falls within a sleep portion of a sleep-awake cycle and the rate at which movement data is monitored may increase in response to determining that a current time falls within an awake portion of a sleep-awake cycle. Alternatively, the rate at which movement data is monitored may increase in response to determining that a current time falls within a sleep portion of a sleep-awake cycle and/or the rate at which movement data is monitored may reduce in response to determining that a current time falls within the awake portion of a sleep-awake cycle. As the ability to determine anomalies and/or risks of particular problems are different within different 2 374 560v 1 portions of a daily cycle, beneficially, energy consumption is responsively reduced, thereby prolonging battery life with minimal impact on the effectiveness of the alarm.
Alternatively, the rate at which the movement data is monitored is determined based on a movement threshold. Beneficially, when a user's activity level is low with respect to a movement threshold, the rate at which the movement data is monitored by the wearable electronic device 110 is reduced. This means that energy is selectively conserved in a responsive, dynamic, way to changes in activity/movement that result in reduced risk and/or ability to detect anomalies. Conversely, alternatively, the rate at which the movement data is monitored by the wearable electronic device 110 is increased in response to determining that a user's activity is low with respect to a movement threshold.
Increased monitoring during such periods may be used to improve detection of otherwise difficult to detect problems.
In some examples, beneficially, the stored dataset is dynamically updated with the movement data from the wearable electronic device and/or with movement data from one or more other wearable electronic devices. Advantageously, the stored dataset is improved with data indicative of normal movement, preferably wherein the updated data falls within a predetermined deviation of data. Advantageously, data that might otherwise be indicative of a slow change in behaviour, or data that might result from common deviations in movement behaviour, do not contribute to the stored dataset, thereby improving the accuracy and the reliability of the warning.
The monitored data generated at step S212 is transmitted from the wearable electronic device 110 to a base station connected to the internet, such as the communication device 101 described with respect to Figure 1. The monitored data is transmitted at regular intervals to the base station. The base station is used to transmit the monitored data to a remote server, for example by the internet. In further examples, the monitored data generated at step S212 is transmitted from the wearable electronic device 110 to the remote server 120 via any suitable communication route. For example, the wearable electronic device 110 transmits the monitored data generated at step S212 via its interface 116 to the interface 104 of the further communication device 102, which may be a smartphone. The further communication device 102 then relays the data to the remote server 120 either directly, or indirectly. For example, the further communication device 102 transmits the monitored data to the remote server 120 via the network 130, which may comprise one or more sub-networks. Alternatively, the wearable electronic device 2 374 560v 1 transmits the monitored data directly to the remote server 120, or indirectly, for example via one or more devices forming part of the network 130.
The use of a base station with a communication range covering a predefined location, such as a user's home, provides improved reliability of connection. In some examples, the use of a base station enables energy consumption requirements to be reduced, thus extending the battery lifetime. In order to provide continuous monitoring of a user's activity and movements, the system dynamically responds when a user moves out of communication range of the base station. Accordingly, the wearable electronic device 110 is configured to select transmission of monitored data in response to determining that it is in communication range of the base station. If the wearable electronic device 110 is not in communication range of the base station, it is determined if the wearable electronic device 110 is in communication range with a secondary device, such as a smartphone with Bluetooth capability. Advantageously, the smartphone is used to transmit data to a remote server.
If it is determined by the wearable electronic device 110 that it is not in communication range of the base station or the smartphone, it temporarily stores data for transmission at a later time. Beneficially, transmission of data based on hierarchical energy requirements ensures that continuous monitoring of daily cycles is possible using the lowest amount of energy from a battery source, thus prolonging the life of the battery.
Determining a current metric The monitored data generated at step S212 is then used at step S214 to determine a current metric by comparing the movement data with the stored dataset to determine the current metric.
The comparison of monitored movement data with the stored dataset is used to determine differences based on the comparison of one or more data points at equivalent times in a daily cycle. For example, the comparison is based on the identification of one or more subsets of data relating to sleep and awake portions of a daily cycle, where the rate of change of movements as a function of time is used to determine the sleep and awake portions. In further examples, additional and/or alternative metrics are used when determining differences between the monitored movement data and the stored data set.
For example, the movement level at an equivalent time of the daily cycle is used to 2 374 560v 1 determine differences between monitored movement data and the stored data set in order to determine a current metric.
The current metric is reflects a difference between the currently monitored movement data and the stored dataset. The current metric may comprises multiple components reflecting different aspects of variances between the currently monitored movement data and the stored dataset. For example, the current metric may comprise a value indicative of a difference in a waking time and a value indicative of a difference in activity during a particular portion of the day, such as the sleep portion or the awake portion. Advantageously, the current metric is formed of different components that are used to identify issues where any one different component in isolation would not otherwise be indicative of the issue.
The monitored data is compared with the stored dataset at regular intervals in order objectively to determine variance within the data resulting from changes. For example, at some time during the day, for example mid-morning, the previous 24 hours of activity of monitored data is compared against the long term average of the stored dataset. A comparison is made by taking a measure of the difference in activity during the normal sleep and active periods, such as the sleep portion 314 and the awake portion 318 described with reference to Figure 3. The measure is a basic sum of differences. Alternatively, in further examples, the measure is a more complex analysis. For example, typical movement anomalies such as an overnight visit to the toilet are filtered out of the data. The sleep portion 314 of a daily cycle and the awake portion 318 of a daily cycle reflect distinct physical states in a user's routine. Accordingly, the movement patterns of a user during each of these portions are also distinct. Movement activity during a sleep portion 314 can be quantified to provide a metric for comparison with a typical routine.
Where the movement activity during the sleep portion 314 is unusually high, this is a flag of an issue. Similarly, movement activity during an awake portion 318 can be quantified to provide a metric for comparison with a typical routine, however, conversely with a sleep portion 314, where movement activity during an awake portion 318 is unusually low, this is a flag of an issue. Flagged issues can be used to instigate warning signals.
In further examples, additionally, or alternatively, in order to provide improved differentiation between explicable, low risk anomalies in movement data, which should not give rise to a warning signal, and those which should give rise to a warning signal, machine learning is used in the comparison of historical stored data and monitored data.
2 374 560v 1 Historical movement data is supplemented with inputs from a cohort of users for an extended period of time. For example, data from more than 100 users, collected for a period of six months or more is used, for which information regarding health issues occurring throughout the extended period of time is collected. Accordingly, the use of the collected data as an input in a machine learning leads to improved detection of anomalies in monitored movement data. Further, machine learning and cohort analysis are used to provide an improved system as more data are accumulated.
Beneficially, comparing the movement data with stored dataset to determine a current metric comprises filtering one or more data points in the movement data based on one or more conditions indicative of events for which a warning signal is not to be transmitted.
Common deviations from the stored dataset therefore are not misinterpreted and the overall provision of the warning signal is more accurate and reliable.
Comparison of current metric and warning metric trigger condition The process 200 enables warning signals to be sent when it is determined that current metrics associated with particular issues meet warning metric trigger conditions, and hence identify issues of concern. The warning metric trigger condition is a predetermined warning metric trigger condition and is determined based on the problem for which a warning alarm or alert is to be raised. For example, warning metric trigger conditions are based on the identification of patterns within the stored dataset and the current metrics generated by monitoring movement data are compared with the stored dataset in order to determine significant deviations. The identification of patterns may include a predetermined warning metric trigger condition comprising a degree of tolerance within which similarities are not statistically significant.
Once a current metric has been determined at step S214, the process moves on to step S216, where it is determined whether or not the current metric meets a predetermined warning metric trigger condition. For example, if a comparison of a current metric and a predetermined warning trigger condition shows that one or more data values of a current metric derived from the monitored movement data are outside a predetermined range of one or more respectively corresponding data values in the stored dataset, the current metric meets the conditions of a predetermined warning metric trigger, resulting in a warning signal being transmitted. The predetermined warning metric trigger condition is based on one or more patterns identified in the stored dataset. In further examples, the predetermined warning metric trigger condition is based on one or more thresholds such 2 374 560v 1 that a comparison of a current metric with the stored dataset results in the predetermined warning metric trigger condition when one or more data values of the current metric exceed the one or more thresholds.
In an example, where a warning is directed to determining abnormal sleep, the warning metric trigger condition can be directed to determining high activity during a sleep portion, or a delayed awake portion. The use of a single predetermined warning metric trigger value provides an efficient mechanism for signalling a warning when a single value current metric is used to identify an issue in the comparison of currently monitored movement data and a stored dataset. In a further example, low activity levels determined during the awake portion of a daily sleep-awake cycle is used to provide a warning metric trigger condition for determining a health problem.
In further examples, predetermined warning metric trigger conditions comprise multiple sub-triggers based on different current metrics. The conditions under which a warning signal is transmitted are then based on the combination of different current metrics being compared against different predetermined warning metric sub-trigger combinations.
For example, a current metric derived from the monitored movement data at a particular time within a sleep portion may not meet a predetermined warning metric trigger condition and a current metric derived from the rate of change of the monitored movement data at a particular time of day may not meet a predetermined warning metric trigger condition.
However, each of the current metrics may individually result in a sub-trigger condition being met, with the sub-trigger combination resulting in a predetermined warning metric trigger condition being met and the transmission of a subsequent warning signal. Therefore, combinations of anomalies detected in the data might be used to raise a warning signal, such as an alarm and/or alert.
The warning metric trigger condition is predetermined in order to enable a warning signal to be raised in the event that particular events are detected. Alternatively, in further examples, the warning metric trigger condition is dynamically updated in order to provide improved determination of problems. For example, monitored data is used dynamically to update the stored dataset in a continually improving manner, such that explicable deviations are not erroneously used in determining if a current metric meets a predetermined warning metric trigger condition. Additionally or alternatively, warning metric trigger conditions are dynamically enhanced using machine learning algorithms 2 374 560v 1 based on one or inputs of identified issues, monitored user movement data from one or more users and/or input based on events occurring for one or more durations.
Alternatively or additionally, determining that the current metric meets a predetermined warning metric trigger condition, comprises determining if the current metric meets a further predetermined warning metric trigger condition. Advantageously, having multiple warning metric triggers conditions for different levels of deviation from the stored dataset enables different levels of warning for alarms and/or alerts to be determined. For example, where a current metric comprises multiple components, the comparison of a current metric and a warning metric trigger involves the comparison of different facets of the current metric and different warning metric triggers. In further examples, the determination of the magnitude of the current metric by determining that the conditions for both a predetermined warning metric trigger condition and further predetermined warning metric trigger condition have been met provides information relating to the type and seriousness of the problem. For example, a one hour delay in the start of the wake portion of a monitored user's current daily cycle causes a low level warning signal in contrast to a four hour delay in the start of the wake portion of a monitored user's current daily cycle causes a high level warning signal to be transmitted. Accordingly, the electronic device (and hence responder) to which message warning signal is sent is dependent on the current metric.
Alternatively and/or additionally, the warning metric trigger condition is predetermined based on the stored dataset comprising user's personal historical movement and monitored data from one or more users in a cohort of users. Monitored movement data is received at the remote server 120 from one or more cohort users in the same manner as the movement data monitored by the user of the wearable electronic device 110. The current metric determined at step S214 is compared with the stored dataset and monitored movement data from one or more cohort users in order to determine if the current metric of the user meets a predetermined warning metric trigger condition. Advantageously, the use of current metrics of a user and the equivalent data from other users provides improved filtering of abnormal behaviours which are explicable. For example, where the same activity anomalies are shown by a significant proportion of the cohort, the warning trigger will be tempered. For instance, people have very different behaviour at different times of the year, such as Christmas and New Year, or for major sporting events such as a world cup final on late at night in a particular region, or for particular meteorological conditions, such as a blizzard. In those circumstances it may be 2 374 560v 1 preferable not to trigger a warning, whereas the personal data without additional context would lead to a warning. Further examples, of events can be evident from currently monitored data from a cohort of users includes weather events, sporting events, television broadcasts, religious events, bank holidays, viral outbreaks, news events, national security events and political events can impact the movement of users. Accordingly, the use of cohort data corresponding to a user's monitored data and current metric results in an improved trigger with fewer false positive results.
Additionally, or alternatively, warning metric trigger conditions are based on events known to impact large groups of people. For example, events such as weather events, sporting events, television broadcasts, religious events, bank holidays, viral outbreaks, news events, national security events and political events can impact the movement of users. Where these events are known, they can be used as inputs identifying specific time periods when the predetermined metric trigger conditions factor in the inputs and are adjusted accordingly. Therefore, warning triggers may be tempered during event windows such that low user activity does not trigger a warning as it might outside of the event window, for example. In further examples inputs based on events are used in a machine learning algorithm to enhance predetermined warning metric trigger conditions. For example, the use of inputs based on events impacting a large group of people can be used to understand the impact on real world user behaviour from one or more user in order to improve the predetermined warning metric trigger conditions such that future warning signals are sent in response more accurately to identifying issues when comparing a current metric of a user with a predetermined warning metric trigger condition. This results in improved and earlier intervention in response to identifying issues.
Warning signals If the current metric meets the necessary predetermined warning metric trigger condition, the process moves to step S218 and a warning signal is transmitted to one or more electronic devices, such as the first electronic device 140 and/or the second electronic device 150.
The first electronic device 140 and the second electronic device 150 are smartphones.
Alternatively, in further examples, the first and/or second electronic devices 140, 150 are mobile phones, landlines, VoIP enabled devices, computer terminals, tablets or any other suitable device for raising a warning alarm or alert. Transmitting a warning signal to one 2 374 560v 1 or more electronic devices means that responders can intervene and attend to a user, who may be vulnerable, at an early stage, or particular processes are instigated based on the warning signal, such as sounding an alarm or sending control instructions to devices forming part of an loT system, for example in a user's home. In further examples, a warning signal comprises transmitting encrypted information, such as passcode information enabling access to a user's home and/or to access to areas within a user's home, such as to safely stored medicines. Such warning signals may automatically enable access to locked systems, or provide responders with the necessary details to access locked systems. In further examples, the warning signals alert health services to ensure early stage intervention in response to detecting issues that may otherwise not be detected until too late. In some examples, the warning signals comprise secure information, such as patient identification thereby enabling one or more secure processes to begin with increased efficiency in response to determining an issue identified by comparing the current metric against a predetermined warning metric trigger condition.
The process then ends at step S220. If it is determined at step S216 that the current metric does not meet a predetermined warning metric trigger condition, the process moves back to step S212 and the wearable electronic device 110 continues to monitor data based on sensed movement values.
Examples of monitored movement data and historical stored datasets Figures 4 to 6 show plots of sensed movement against time for a monitored data and a stored dataset for different scenarios, illustrating examples where continuously monitored movement data is compared with stored datasets to determine when one or more current metrics meet one or more respective predetermined warning trigger conditions to trigger a warning signal.
Abnormal user activity during sleeping hours Figure 4 shows a plot 400 of sensed movement against time for a monitored data for a 24 hour period and a stored dataset. Movements are shown on the y axis 404 and time is shown on the x axis 402. There is shown a dashed line trace of the averaged historical data 408 and the trace line of the monitored data 406 for one daily cycle (24 hour period).
The 24 hour period can be divided into different portions. There is a first portion corresponding to a sleep portion 414, where there is a low level of activity between 0 hours (midnight, for example) and approximately 7 hours, demarcated by a first line 410 in 2 374 560v 1 the typical dashed line trace of the averaged historical data 408. There is a second portion corresponding to an awake portion 418, demarcated by a second line 412 and a transition portion 416 between the sleep portion 414 and the awake portion 418. In further examples, the transition portion 416 is determined to form part of the sleep portion 414 and/or awake portion 418. A comparison of the trace line of monitored movement data 408 with the stored dataset 406 shows unusually active movements during the sleep portion 414, including noticeable peaks 413, 415. In this case, the application of the method described with reference to Figures 1 and 2 is such that different metrics for comparison within the sleep portion 414 enable a comparison of a current metric and a warning metric trigger condition. For example, accumulated movement values and/or peak movement values within the sleep portion 414 are examples of metrics that are used for comparison and for determining that a warning signal is transmitted.
User delay in waking Figure 5 shows a plot 500 of sensed movement against time for a monitored data for a 24 hour period and a stored dataset. Movements are shown on the y axis 504 and time is shown on the x axis 502. There is shown a dashed line trace of the averaged historical data 508 and the trace line of the monitored data 506 for one daily cycle (24 hour period). The 24 hour period can be divided into different portions for each of the stored historical data 508 and the monitored data 506 for one daily cycle.
For the historical data 508 there is a first portion corresponding to a sleep portion 514, where there is a low level of activity between 0 hours (midnight, for example) and approximately 7 hours, demarcated by a first line 510. There is a second portion corresponding to an awake portion 518, demarcated by a second line 511 and a transition portion 516 between the sleep portion 514 and the awake portion 518. In further examples, the transition portion 516 is determined to form part of the sleep portion 514 and/or awake portion 518.
For the monitored data 506 there is a first portion corresponding to a sleep portion 514', where there is a low level of activity between 0 hours (midnight, for example) and approximately 9 hours, demarcated by a third line 512. There is a second portion corresponding to an awake portion 518', demarcated by a fourth line 513 and a transition portion 516' between the sleep portion 514' and the awake portion 518'. In further examples, the transition portion 516' is determined to form part of the sleep portion 514' and/or awake portion 518'.
2 374 560v 1 A comparison of the monitored movement data 508 with the stored dataset 506 shows an unusual delay in the awake portion 518'. Accordingly the determination of the start of the awake portion 518' is used as a metric for determining the current metric. Where this current metric meets a predetermined warning metric trigger condition based on this metric, a warning signal is initiated.
User inactivity during waking hours Figure 6 shows a plot 600 of sensed movement against time for a monitored data for a 24 hour period and a stored dataset. Movements are shown on the y axis 604 and time is shown on the x axis 602. There is shown a dashed line trace of the averaged historical data 608 and a line trace of the monitored data 606 for one daily cycle (24 hour period).
The 24 hour period can be divided into different portions for each of the stored historical data 608 and the monitored data 606 for one daily cycle.
For the historical data 608 there is a first portion corresponding to a sleep portion 614, where there is a low level of activity between 0 hours (midnight, for example) and approximately 7 hours, demarcated by a first line 610. There is a second portion corresponding to an awake portion 618, demarcated by a second line 611 and a transition portion 616 between the sleep portion 614 and the awake portion 618. In further examples, the transition portion 616 is determined to form part of the sleep portion 614 and/or awake portion 618.
For the monitored data 606 there is a first portion corresponding to a sleep portion 614', where there is a low level of activity between 0 hours (midnight, for example) and approximately 7 hours, demarcated by the first line 610. There is a second portion corresponding to an awake portion 618', demarcated by the second line 611 and a transition portion 616' between the sleep portion 614' and the awake portion 618'. In further examples, the transition portion 616' is determined to form part of the sleep portion 614' and/or awake portion 618'. Within the awake portion 618' of the monitored movement data 606 there is a trough between a third line 612 and a fourth line 613. The trough is indicative of a period of significantly reduced movement during the awake portion 618'. The sub portion of the awake period 618' between the third line 612 and the fourth line 613 is indicative of a significant change in daily routine.
A comparison of the monitored movement data 608 with the stored dataset 606 shows an unusual period of reduced movement in the awake portion 618', along with an unusual 2 374 560v 1 decrease in movement throughout the awake portion 618'. A comparison of the monitored daily movement data and the stored dataset in accordance with the method described with reference to Figures 1 and 2 results in a current metric meeting a predetermined warning metric trigger condition and a warning signal being transmitted as a consequence. Whilst one warning metric trigger condition based on average movement in the awake portion 618' may or may not be met, this is combined with a significant drop in movement for a significant period of time within the awake portion 618', which meets a further warning metric trigger condition. The combination of warning metric trigger conditions being met by current metrics arising from the comparison of monitored movement data and the stored dataset results in the warning signal being transmitted. In further examples, a warning signal is transmitted in response to only one predetermined warning metric trigger condition being met by a current metric based on one predetermined metric. In yet further examples, any number of appropriate warning metric trigger conditions and current metrics are compared to make appropriate decision based on patterns of differences identified in the data, with warning signals being transmitted accordingly.
Whilst the stored dataset is based on movement data, in further examples, the stored dataset comprises data sourced from different sensors such as the additional sensor 105 of the wearable electronic device 110, which may be a temperature sensor. The use of further metrics can further enhance the determination of different issues and the continuous monitoring of such metrics can enhance early warning signally when issues are detected in the daily cycle data. This provides an improved system where issues are not evident to users and are not reliably determined and sensed without continuous monitoring.
Temperature data In further examples, the use of a temperature sensor, such as additional sensor 105, provides information that is measured analogously to movement data and synergistically supplements the movement data. Temperature data is used to form the stored dataset. Subsequent comparisons of historical stored data are made with monitored data based on both temperature data and movement data. The combination of temperature data and movement data provides further current metrics that are used provide warning signals. For example, the combination of abnormally low temperature data combined with abnormally active movement data, or the combination of abnormally high temperature 2 374 560v 1 data combined with abnormally inactive movement data are situations where a current metric may meet the predetermined warning metric trigger condition. Temperature data may provide an independent metric that can result in a warning signal being transmitted, for example, where a user's temperature is identified as being abnormally high or low.
Indication of the state of the battery As described herein, the wearable electronic device 110 provides an indication of the state of the battery. Such an indication is indicative of the remaining power that the battery can provide. An indication is transmitted to a third party device by the interface 116 of the wearable electronic device 110. The indication may be transmitted to the server 120 via the network 130, or any intermediary device in order to initiate a process for battery replacement or rapid recharging. The indication may be transmitted directly to a device within the system 100 to initiate a process for battery replacement or rapid recharging. The indication initiates a process for battery replacement or rapid recharging in response to determining that the state of the battery is below a battery level threshold.
The rate at which an indication of the state of the battery is transmitted varies in response to the determined state of the battery. In an example, the rate at which an indication of the state of the battery is transmitted increases in response to determining that the state of the battery has reached a predetermined level. Accordingly, a third party operating a third party device is caused to ensure battery replacement or rapid recharging with a degree of flexibility and increased urgency as the state of the battery reduces. The use of an indication of the state of the battery ensures battery continuity by rapid battery replacement or recharging. Such rapid replacement or recharging is performed with minimal disruption to the collection of data, for example such that the battery is replaced or charged with in a time period less than an interval between transmissions of monitored data.
Whilst a particular method and system have been described herein, with reference to particular devices, and method steps in a particular order, the skilled person understands that in further examples the method is implemented in any order suitable to provide a warning signal based on monitoring a user, using appropriate devices and that the steps described herein can be combined.
Various further aspects and embodiments of the present invention will be apparent to those skilled in the art in view of the present disclosure.
2 374 560v 1 "and/or" where used herein is to be taken as specific disclosure of each of the two specified features or components with or without the other. For example "A and/or B" is to be taken as specific disclosure of each of (i) A, (H) B and (iii) A and B, just as if each is set out individually herein.
Unless context dictates otherwise, the descriptions and definitions of the features set out above are not limited to any particular aspect or embodiment of the invention and apply equally to all aspects and embodiments which are described.
It will further be appreciated by those skilled in the art that although the invention has been described by way of example with reference to several embodiments. It is not limited to the disclosed embodiments and that alternative embodiments could be constructed without departing from the scope of the invention as defined in the appended claims.
2 374 560v 1
Claims (25)
- CLAIMS1. A method of providing a warning signal, the method comprising: measuring a plurality of movement values of a user; storing data based on the plurality of movement values to form a stored dataset comprising historical data for at least part of one or more daily and/or weekly cycles of the user; subsequently monitoring movement data provided by at least one sensor of a wearable electronic device comprising at least one sensor and a transmitter; comparing the movement data with the stored dataset to determine a current metric; and transmitting, a warning signal to one or more electronic devices in response to determining that the current metric meets a predetermined warning metric trigger condition.
- 2. The method according to claim 1 comprising: determining that the wearable electronic device is being worn by a user; and monitoring movement data provided by the sensor of the wearable electronic device when it is determined that the user is wearing the wearable electronic device.
- 3. The method according to claim 1 or 2, wherein the movement data is compared with the stored dataset to determine a current metric at the wearable electronic device.
- 4. The method according to claim 1 or 2, comprising: transmitting the movement data from the wearable electronic device to a remote server; 2 374 560v 1 comparing the received movement data with the stored dataset at the remote server; and determining the current metric at the remote server.
- 5. The method according to any preceding claim, comprising: determining one or more data subsets indicative of an awake portion and/or a sleep portion within the stored dataset.
- 6. The method according claim 5, wherein determining one or more data subsets indicative of an awake portion and/or a sleep portion comprises determining a rising edge and/or a falling edge indicative of the awake portion and/or the sleep portion within the stored data.
- 7. The method according to any preceding claim, wherein a rate at which the movement data is monitored is determined based on correlating a current time with a corresponding time within a daily and/or weekly cycle based on the stored dataset.
- 8. The method according to any preceding claim, comprising: updating the stored dataset with the movement data from the wearable electronic device and/or with movement data from one or more other wearable electronic devices.
- 9. The method according to any preceding claim, comprising: updating the stored dataset with movement data from a cohort of users, preferably wherein the cohort comprises one hundred or more users over a period of six months or more, more preferably wherein the cohort is categorised based on at least one of life stage, medical condition and geography.
- 10. The method according to any preceding claim, wherein comparing the movement data with the stored dataset to determine the current metric comprises filtering one or more data points in the movement data based on one or more conditions indicative of events for which a warning signal is not to be transmitted.
- 2 374 560v 1 11. The method according to any preceding claim, wherein transmitting the warning signal to one or more electronic devices in response to determining that the current metric meets a predetermined warning metric trigger condition, comprises: determining if the current metric meets a further predetermined warning metric trigger condition; and determining an electronic device of a plurality of electronic devices to which the warning message is transmitted based on whether the current metric meets a predetermined warning metric trigger condition and/or a further predetermined warning metric trigger condition.
- 12. The method according to any preceding claim, wherein the movement data is based on a plurality of movement values, preferably wherein the movement data is an average of a plurality of movement values in a predetermined time window or an average of a predetermined number of movement values.
- 13. The method according to any preceding claim, wherein measuring a plurality of movement values of a user comprises recording the plurality of movement values based on a plurality of respective inputs at a sensor of a wearable electronic device, at regular time intervals.
- 14. The method according to claim 13, wherein measuring a plurality of movement values of a user comprises recording the plurality of movement values based on a plurality of respective inputs at a sensor of a wearable electronic device at regular time intervals continuously for at least three consecutive months, preferably for at least six consecutive months and more preferably for at least twelve consecutive months.
- 15. The method according to any preceding claim, wherein the movement data is compared with the stored dataset to determining the current metric based on at least one of the day of the week and the month of the year.
- 16. The method according to any preceding claim, comprising: 2 374 560v 1 determining whether the wearable electronic device is in communication range of an first intermediate device; transmitting monitored movement data to a first intermediate device when it is determined that the wearable electronic device is in communication range of the first intermediate device; determining whether the wearable electronic device is in communication range of a second intermediate device when it is determined that the wearable electronic device is not in communication range of the first intermediate device; and transmitting monitored movement data to a second intermediate device when it is determined that the wearable electronic device is not in communication range of the first intermediate device.
- 17. The method according to claim 15, further comprising: temporarily storing monitored movement data at a memory of the wearable electronic device when it is determined that the wearable electronic device is not in communication range with the first intermediate device and the second intermediate device; and subsequently transmitting monitored movement data to the first intermediate device or the second intermediate device when at least one of the first intermediate device and the second intermediate device is determined to be in communication range.
- 18. The method according to any of claim 16 and 17, wherein the rate of monitoring movement data is based on whether the wearable electronic device is transmitting monitored movement data to the first intermediate device or to the second intermediate device.
- 19. The method according to any preceding claim, comprising determining the predetermined warning metric trigger condition and/or further predetermined warning metric trigger condition using cohort analysis and/or machine learning.
- 20. The method according to any preceding claim, comprising providing an indication of the state of the battery power of the wearable electronic device, wherein the rate 2 374 560v 1 of monitoring movement data is based on the indication of the state of the battery power of the wearable electronic device and/or wherein the method further comprises initiating a process to ensure battery continuity based on the indication of the state of the battery power of the wearable electronic device, wherein the process comprises rapid battery replacement or rapid battery charging such that the battery is replaced or charged with in a time period less than an interval between transmissions of monitored data.
- 21. The method according to any preceding claim further comprising: measuring a plurality of temperature values of a user; storing data based on the plurality of temperature values to form the stored dataset comprising historical data for at least part of one or more daily cycles of the user; subsequently monitoring temperature data provided by the wearable electronic device; and comparing the movement data and temperature data with the stored dataset to determine the current metric.
- 22. The method according to any preceding claim, wherein determining that the current metric meets a predetermined warning metric trigger condition comprises comparing the monitored movement data of the user with the stored dataset and monitored movement data from a cohort of one or more users and/or one or more inputs based on an event, such as a weather event, sporting event, television broadcast, religious event, bank holiday, viral outbreak, news event, national security event and /or political event.
- 23. A wearable electronic device comprising: at least one sensor; a processor; a memory; and 2 374 560v 1 a wireless transmitter, wherein the wearable electronic device is configured to perform the method of any of claims 1 to 22.
- 24. A system comprising: a wearable electronic device; a server; and one or more electronic devices, wherein the system is configured to perform the method of any of claims 1 to 22.
- 25. A computer readable medium comprising instructions which, when executed by a computer, cause the computer to carry out the method of any of claims 1 to 22.2 374 560v 1
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
GB2302231.2A GB2627246A (en) | 2023-02-16 | 2023-02-16 | Activity monitor |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
GB2302231.2A GB2627246A (en) | 2023-02-16 | 2023-02-16 | Activity monitor |
Publications (2)
Publication Number | Publication Date |
---|---|
GB202302231D0 GB202302231D0 (en) | 2023-04-05 |
GB2627246A true GB2627246A (en) | 2024-08-21 |
Family
ID=85772545
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
GB2302231.2A Pending GB2627246A (en) | 2023-02-16 | 2023-02-16 | Activity monitor |
Country Status (1)
Country | Link |
---|---|
GB (1) | GB2627246A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050181771A1 (en) * | 2004-02-04 | 2005-08-18 | Cuddihy Paul E. | System and method for determining periods of interest in home of persons living independently |
US20080139899A1 (en) * | 2005-05-04 | 2008-06-12 | Menachem Student | Remote Monitoring System For Alzheimer Patients |
US20150164377A1 (en) * | 2013-03-13 | 2015-06-18 | Vaidhi Nathan | System and method of body motion analytics recognition and alerting |
US20190236923A1 (en) * | 2017-12-30 | 2019-08-01 | Philips North America Llc | Method for tracking and reacting to events in an assisted living facility |
US20200111345A1 (en) * | 2016-12-15 | 2020-04-09 | Goertek. Inc, | User behavior monitoring method and wearable device |
US20200375505A1 (en) * | 2017-02-22 | 2020-12-03 | Next Step Dynamics Ab | Method and apparatus for health prediction by analyzing body behaviour pattern |
-
2023
- 2023-02-16 GB GB2302231.2A patent/GB2627246A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050181771A1 (en) * | 2004-02-04 | 2005-08-18 | Cuddihy Paul E. | System and method for determining periods of interest in home of persons living independently |
US20080139899A1 (en) * | 2005-05-04 | 2008-06-12 | Menachem Student | Remote Monitoring System For Alzheimer Patients |
US20150164377A1 (en) * | 2013-03-13 | 2015-06-18 | Vaidhi Nathan | System and method of body motion analytics recognition and alerting |
US20200111345A1 (en) * | 2016-12-15 | 2020-04-09 | Goertek. Inc, | User behavior monitoring method and wearable device |
US20200375505A1 (en) * | 2017-02-22 | 2020-12-03 | Next Step Dynamics Ab | Method and apparatus for health prediction by analyzing body behaviour pattern |
US20190236923A1 (en) * | 2017-12-30 | 2019-08-01 | Philips North America Llc | Method for tracking and reacting to events in an assisted living facility |
Also Published As
Publication number | Publication date |
---|---|
GB202302231D0 (en) | 2023-04-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9277870B2 (en) | Infant monitoring system and methods | |
US9526421B2 (en) | Mobile wireless customizable health and condition monitor | |
US8618930B2 (en) | Mobile wireless customizable health and condition monitor | |
US20160246259A1 (en) | Method, appararus and wearable apparatus for automatically reminding user to sleep | |
CN107734487B (en) | Method for controlling wearable electronic equipment, central device and equipment | |
JP2004529704A (en) | Adaptive selection of warning limits in patient monitoring | |
US11176798B2 (en) | Anomaly notification system and anomaly notification method | |
CN108042140A (en) | A kind of Old Age Homes' monitor system based on Internet of Things and fall down detection method | |
WO2017118184A1 (en) | Position monitoring method, intelligent wear apparatus, mobile terminal apparatus and monitoring apparatus | |
CN203493628U (en) | Intelligent detection device for activity statuses and health conditions of old people based on wireless network | |
KR20180106583A (en) | Care device and care system for the old and the infrim | |
JP2017027574A (en) | Safety information management device using short-range communication device | |
CN105249945A (en) | Intelligent old-age care monitoring system | |
CN110197732A (en) | A kind of remote health monitoring system based on multisensor, method and apparatus | |
CN113077883A (en) | Health care system for old people | |
CA3188477A1 (en) | Devices, systems and methods for fall detection and preventing false alarms | |
CN206075496U (en) | A kind of intelligent home care system | |
CN205163050U (en) | Intelligence endowment monitor system | |
CN209091352U (en) | Care system | |
CN108903923B (en) | Health monitoring device, system and method | |
GB2627246A (en) | Activity monitor | |
CN107833431A (en) | A kind of intelligent SOS alarming methods and system based on dynamic adjustment mechanism | |
CN103654744A (en) | Sleep quality monitoring method and system | |
Salem et al. | Detection and isolation of faulty measurements in medical Wireless Sensor Networks | |
JP7088397B1 (en) | Data collection system, data collection device, data acquisition device and data collection method |