FR3076201A1 - SYSTEM FOR MONITORING THE HEALTH STATUS OF A PERSON, AND METHOD OF MONITORING THE SAME - Google Patents

Abstract

The present invention relates to a system for monitoring the state of health of a person, comprising at least two housings (1, 3, 4) intended to be placed in the home (9) of the person, and a remote server ( 5). According to the invention, each of the housings (1, 3, 4) comprises at least two environmental data sensors, measuring environment data in the room in which the housing (1, 3, 4) is placed, and sending them to the remote server (5). The remote server (5) analyzes the environment data, associating a combination of environment data from different sensors with an activity of that person. The invention also relates to a monitoring method implementing such a system.

Description

SYSTEM FOR MONITORING THE HEALTH OF A PERSON, AND METHOD FOR

CORRESPONDING MONITORING

Field of the Invention The present invention relates to a system for monitoring the state of health of a person, and more particularly of an elderly person or a person with fragile health.

In particular, the invention relates to a monitoring system capable of verifying that an elderly or vulnerable person living in autonomy retains a state of health compatible with this autonomy.

In particular, the invention relates to a monitoring system able to identify data representative of a person's activities, and to process this data to deduce therefrom information on the person's state of health.

The invention also relates to a monitoring method implementing such a monitoring system.

PRIOR ART [005] The aging of the population, associated with the breakdown of traditional family structures, means that more and more elderly people remain alone in their homes. Depending on their level of dependence, these elderly people can benefit from varying degrees of assistance. They can thus receive regular visits from health personnel, nurses or nursing assistants, home helpers for cleaning or preparing meals, or their relatives. These various parties can regularly observe changes in the state of health of the person being supported, if the person being accompanied does not conceal the symptoms of this development from them. However, they do not systematically share information concerning these developments. In addition, since the workers are in regular contact with the person being supported, they are not necessarily aware of changes in their lifestyle which may be signs of a change in their state of health.

Numerous monitoring systems have been devised to help monitor the state of health of the people supported when they are alone at home. Most of these monitoring systems are designed to trigger an alert, alerting caregivers that the caregiver has experienced a sudden incident, such as falling or feeling unwell. This alert can be triggered, for example, by an alert button that the supported person wears. It can also be triggered by a detection system capable of detecting such an incident. Such detection systems most often use remote movement detection devices, remote presence detection devices, and cameras which are installed in the home of the supported person. Such systems can be effective in triggering rapid response in the event of an incident or one-time accident. However, they are not generally intended to measure changes in the state of health of the person supported over a long period.

[007] Furthermore, systems comprising remote motion or presence detectors and cameras are difficult to accept by accompanied persons, who consider them to be a hardly acceptable intrusion into the privacy of their home. For this reason, such surveillance systems are generally reserved only for very dependent people, whose general state of health gives reason to fear the imminence of a fall or discomfort which they could not alert themselves. companions.

Such systems also require a significant installation, in the home of the accompanied person, consisting notably in fixing and adjusting the cameras and motion or presence detectors remotely and in connecting the corresponding cabling. In addition to its significant cost, such an installation is itself experienced as an additional intrusion into the home of the accompanied person which can be difficult to accept.

To make the observation system less intrusive for the accompanied person, it has been imagined to place, in the home of the accompanied person, a plurality of sensors making it possible to collect data representative of the activities taking place there. These sensors can thus include sensors for opening or closing doors and windows, motion or presence detectors remotely, as well as sensors measuring the consumption made in the housing. All these data can be further processed to deduct the activities practiced in the accommodation.

If such systems, which do not require the installation of cameras, may appear to be less intrusive for accompanied persons, their implementation however requires a significant installation, during which speakers will install sensors in multiple places of residence. In addition to its significant cost, this intervention can be traumatic for the supported person who will have the impression that his entire home is riddled with sensors leaving him no privacy.

[011] There is therefore a need to ensure observation of the development over time of the state of health of a person accompanied in as discreet a manner as possible, avoiding the costly and traumatic installation of a multitude of sensors in the home of the supported person.

Objectives of the invention [012] The present invention aims to overcome these drawbacks of the prior art.

In particular, the present invention aims to allow observation, over time, of the evolution of the state of health of an accompanied person.

[014] In particular, the present invention aims to allow regular observation of the activities practiced daily in the home of the accompanied person, without this observation being perceived as intrusive by the accompanied person.

A particular objective of the invention is to allow such observation without the implementation of the observation system requiring long, expensive and intrusive installation in the home of the accompanied person.

DESCRIPTION OF THE INVENTION These objectives, as well as others which will appear more clearly below, are achieved using a system for monitoring the state of a person's health, comprising at least two boxes intended to be placed in the home of said person, and a remote server. This system is characterized, according to the invention, in that each of said boxes includes at least two environmental data sensors, measuring environmental data in the room in which said box is placed, and sending them to said remote server. , and in that said remote server analyzes said environment data, by associating a combination of environment data coming from different sensors with an activity of said person.

This system therefore makes it possible, by measuring environmental data in rooms in the home of the supported person, to deduce the information useful for monitoring the supported person. By "environmental data" is meant data which can be felt at the level of the housing itself, without the need for remote measurement. Thus, among the environmental data, the data relating to the air surrounding the housing (temperature, composition, etc.) are understood, the data relating to the light illuminating the housing, the data relating to the audible sounds at the housing level, and the data concerning the movements of the case itself. On the other hand, the environmental data does not include capturing images remotely from the housing, nor detecting a movement or a remote presence of the housing, which are data measured remotely.

Unlike surveillance systems based on image capture or remote detection of presence or movement, the simple capture of environmental data does not require that the boxes be positioned at a position or in a precise orientation . The installation of the system is therefore particularly simple, and is not intrusive. In addition, the surveillance system prevents any intrusion into the person's privacy by capturing images in their home.

[019] Preferably, at least one of said boxes comprises at least one microphone.

[020] Indeed, among the environmental data, the sounds emitted by daily activities are particularly valuable indicators helping to identify the activities of the person being supported.

[021] According to a preferred embodiment, said boxes include at least two environmental data sensors from:

a microphone, a light sensor, a thermometer, a hygrometer, a sensor representing information representative of the movements of said housing.

Combinations of such sensors, which can be implemented easily, make it possible to identify numerous activities of the person being supported. It is moreover particularly advantageous that at least one of said boxes includes at least one microphone and at least three environmental data sensors from a light sensor, a thermometer, a hygrometer and an information sensor representative of the movements of said housing. Thus, it is possible that a small number of boxes installed in a home, typically between two and four, is sufficient to collect enough environmental data to ensure effective monitoring of a person's state of health.

[023] Advantageously, at least one of said boxes comprises at least one environment data sensor from among:

a CO2 sensor, a VOC sensor.

[024] Such sensors can, combined with other environmental data sensors, help to identify the activities of the supported person.

Advantageously, at least one of said boxes comprises both a thermometer and an information sensor representative of the movements of said box.

[026] Such a housing can advantageously be used as a specific refrigerator housing. Installed in the refrigerator door, it allows both to identify the opening and closing movements of the refrigerator, and to verify that no break in the cold chain occurs in the refrigerator.

[027] According to a preferred embodiment, said boxes do not include any image capture device in the home of said person.

Likewise, preferably, said boxes do not include any device for the remote detection of movement or of presence in the home of said person.

[029] The boxes, which therefore preferably only include environmental data sensors, can be installed very easily, for example by being placed on a piece of furniture, and can be moved without this preventing their operation.

Of course, it is possible to add to the monitoring system according to the invention other types of sensors than environmental sensors, which will produce data which can improve the identification of activities. However, the system according to the invention is capable of identifying the activities of the accompanied person based solely on the sensors of the environmental data, which avoids having to collect other data.

[031] Preferably, one of said boxes collects the environmental data measured by the others of said boxes, and sends to said remote server all of the environmental data measured in said housing.

In this case, advantageously, said boxes include a radio communication module allowing them to communicate with each other.

These radio communication modules can communicate via any protocol known to those skilled in the art, such as for example the protocols known to those skilled in the art under the names Bluetooth, Wifi, LoRa, LAN (Local Area Network), ZigBee or Z-Wave.

[034] According to a preferred embodiment, said remote server comprises first calculation means analyzing said environment data by comparing them with data from an activity database, in which combinations of environment data coming from different sensors are associated, with a given level of probability, with activities constituting the timetable for said person.

This first means of calculation, or first calculator, therefore makes it possible to recognize the activities of the accompanied person.

[036] Preferably, said first calculation means modify the data of said activity database, as a function of the environmental data coming from different sensors associated with information on the timetable of said person.

[037] The activity database can therefore evolve, by enriching or modifying its data, by continuous learning often known by the English expression "machine learning". This learning can use information on the timetable of the accompanied person, which can be information declared by the accompanied person or by an accompanying person, information on his / her past schedule, representative of the habits of the accompanied person, who can help interpret environmental data, or information about the environment of the cared-for person, such as weather information about the location of that person.

[038] Preferably, said remote server comprises second calculation means analyzing the timetable determined by said first calculation means, by comparing them with data from a database of changes in timetable, in which changes in a person's schedule are associated, with a given level of probability, with life situations.

[039] Such life situations of the accompanied person are changes, rapid or slow, in the state of health or the environment of the accompanied person. They can for example correspond to a loss of autonomy, a change in the living environment, or an early sign of an illness or loss of autonomy. The identification of these life situations advantageously makes it possible to carry out curative or preventive actions to correct or mitigate the corresponding drawbacks.

[040] According to a particularly advantageous embodiment, said remote server comprises third calculation means analyzing the data from the databases of the schedule modifications of a plurality of people, associated with information on life situations and / or the evolution of the state of health of these people, in order to modify or complete the associations between changes in a person's schedule and a life situation, to modify or supplement the data in said databases schedule change data.

[041] The databases of timetable modifications can therefore evolve, by enriching or modifying their data, by continuous learning ("machine learning"), in order to allow ever more precise identification of the life situations of people supported. Thus, life situations corresponding to the beginnings of a disease may for example, in certain cases, be identified even before the appearance of its visible symptoms, which will allow the initiation of early preventive or curative actions.

[042] The present invention also relates to a method for monitoring a person, implementing a monitoring system as described above, and comprising:

a step of measuring environmental data in at least two rooms of said person's home, a step of sending said environmental data to said remote server, a step of analyzing said environmental data by said remote server, associating a combination of environmental data from different sensors with an activity of said person.

[043] Preferably, said analysis step comprises a first sub-step of analysis of said environment data during which they are compared with data from an activity database, in which combinations of data environment from different sensors are associated, with a given level of probability, with activities constituting the timetable of said person.

[044] Preferably, during said first analysis sub-step, the data in said activity database are modified as a function of the environmental data coming from different sensors associated with information on the timetable of said person.

Preferably, said analysis step comprises a second sub-step for analyzing the timetable determined during said first sub-step, by comparing it with data from a database of modifications d 'timetable, in which changes in a person's timetable are associated, with a given level of probability, with a life situation.

[046] According to a particularly advantageous embodiment, said analysis step comprises a third sub-step of analyzing data from the databases of the changes in the timetable of a plurality of people, associated with information on the life situations or the evolution of the state of health of these people, in order to modify or supplement the associations between changes in a person's timetable and a life situation, to modify or supplement the data of said databases of timetable modifications.

List of Figures [047] The invention will be better understood on reading the following description of preferred embodiments, given by way of simple figurative and nonlimiting example, and accompanied by the figures among which:

Figure 1 is a schematic representation of a system for monitoring the good health of a person, according to one embodiment of the invention;

Figure 2 is an exploded view of the main housing of the monitoring system of Figure 1;

Figure 3 is a schematic representation of the acquisition card implemented in the main housing of Figure 2;

Figure 4 is an exploded representation of a secondary housing implemented in the monitoring system of Figure 1;

Figure 5 is an exploded representation of a specific secondary refrigerator housing, implemented in the monitoring system of Figure 1;

FIG. 6 is a diagram representing the processing of the data in the surveillance system of FIG. 1.

Detailed description of embodiments of the invention

Monitoring system [048] FIG. 1 schematically represents a system for monitoring the state of health of a person according to an embodiment of the invention. The expression “monitoring of the state of health”, in the present patent application, signifies the monitoring of a person's physical and psychological capacities, enabling him to maintain autonomy in his daily life, in particular the capacity to get up , to wash, to dress, to move, to eat, etc., and the monitoring of its maintenance in an environment allowing it this autonomy. This monitoring of the state of health therefore does not include any medical examination of the person being supported, and does not alone enable a medical diagnosis to be made.

[049] As shown in this figure, the monitoring system comprises, on the one hand, elements located in the home 9 of the accompanied person, and on the other hand elements located outside this home.

[050] In the home 9 of the accompanied person, the monitoring system comprises at least one main unit, and may include one or more secondary units. In the embodiment shown, it comprises a main housing 1, a secondary housing 3 and a specific secondary housing for refrigerator 4. Advantageously, the main housing 1 is installed in the living room of the home 9, the secondary housing 3 is installed in the bathroom and / or kitchen, and the specific secondary box 4 is installed in the refrigerator door.

[051] Each of the secondary boxes 3 and 4 communicates with the main box 1, by means of short-range communications 91. The main box 1 communicates with a remote server 5 via a long-range communication 92. Finally, this remote server 5 communicates with monitoring terminals 6, via communication networks 93 such as the Internet.

Main unit [052] Figure 2 is an exploded representation of a main unit 1 implemented in the health monitoring system of a person represented by Figure 1. This main unit 1 comprises a box 11 closed by a cover 12 and in which are placed a control card 13, a long-range communication module 14 and an acquisition card 2.

Acquisition card [053] The acquisition card 2 is shown in detail in FIG. 3. This electronic card comprises a plurality of sensors intended to measure the environmental data, at the level of the box in which it is placed.

[054] Thus, this card includes a thermometer 21 capable of measuring the ambient temperature of the room in which the housing is placed. This thermometer is preferably placed in a position in which it does not undergo a temperature variation generated by the other components present in the housing, and in which it can effectively measure the temperature outside the housing 1.

[055] This acquisition card 2 also includes a hygrometer 22 for measuring the relative humidity of the air surrounding the housing. This hygrometer 22 is also placed in the housing in a position such that it is in contact with outside air which can circulate in the housing.

[056] The acquisition card 2 also comprises a microphone 23. In the embodiment shown, this microphone 23 is not intended to record the surrounding sounds but only to detect the parameters making it possible to identify the activities carried out in the vicinity of the housing. For this, the microphone is configured to detect the sound volume (in decibels) at the housing. It can also, in certain embodiments, identify the main sound frequency ranges. This discrimination in sound frequencies can, for example, allow the microphone to record a different signal for a conversation conducted by people near the housing and for noise emitted by a television set or a loudspeaker.

[057] Thus, this microphone 23 can collect data, the subsequent analysis of which will make it possible to identify, by the noise level and possibly by the frequency ranges emitted, the sounds generated during the use of the various pieces of equipment of a housing and constituting the sound signatures of this equipment: operation of a household appliance, opening of a tap, use of a shower, opening or closing of shutters, etc.

[058] Similarly, the subsequent analysis of the sounds picked up by the microphone 23 can, by analyzing the sound level, allow an evaluation of the number of people present in the accommodation. It can also make it possible to identify a cry from the person being accompanied, which will be characterized by a sudden sound, louder than normal conversations, and in a higher frequency range.

[059] The acquisition card 2 also includes a brightness sensor 24, or light sensor. This light sensor is advantageously placed, in the main housing 1, opposite an opening, micro-perforations or a translucent portion of the upper cover 12, in order to receive light from outside the housing. In the embodiment shown, this light sensor is a luxmeter for measuring the intensity of light. It may also, in other possible embodiments, be able to analyze the light spectrum of the light emitted. This measurement of the light intensity, and possibly of the light spectrum, makes it possible to measure the lighting in the room in which the main housing 1 is placed. The subsequent analysis of the measured data will make it possible to identify the nature of this lighting, in particular to determine whether it is natural or artificial.

[060] The acquisition card 2 also includes a CO2 sensor 25, making it possible to measure the level of CO2 in the air surrounding the housing, and a volatile organic compound sensor (usually designated by the expression "VOC sensor"). ) 26 for measuring the level in the air of the main volatile organic compounds that may be present.

[061] Finally, the acquisition card 2 comprises a component for detecting movements of the housing 27, presenting information sensors representative of the movements of the housing, in particular an accelerometer and a gyroscope, and making it possible to measure the movements carried out by the main unit 1.

[062] The various data collected by the acquisition card 2 are called “environment data”, because they are representative of the environment, atmospheric, sound and light, around the box comprising the card without being focused on a person. nor on an area of the housing other than that where this box is placed. Thus, the exact position of the box in a room is not decisive for the collection of this data, as long as it is placed in the same environment as the person being supported.

Control card [063] The box 1 also includes the control card 13, which is connected to the acquisition card 2 so as to be able to receive all the data measured by the sensors of the acquisition card 2.

[064] This control card 13 includes a short-range communication component. This component can for example be a radio transceiver using a known protocol such as Bluetooth, Wifi, LoRa, LAN (Local Area Network), ZigBee or Z-Wave. This component makes it possible to communicate data between the main box 1 and the secondary boxes.

[065] The control card 13 also includes a processing module which is capable of monitoring and controlling the system made up of the main unit and the secondary units connected to it. It can thus monitor the proper functioning of the various components and sensors, the state of charge of the batteries in these various boxes and generate appropriate messages or alerts.

[066] Furthermore, this processing module collects all the data received from the acquisition card 2 and the data received, via the short-range communication component, from the secondary boxes connected to the main box. Each data is time-stamped and carries the information of the box from which it comes.

[067] In the embodiment shown, the processing module can perform a first processing of this data, for example a qualification of the data, to determine if it is relevant, a selection of the relevant data and an aggregation of this data to form a message. In other possible embodiments, however, the processing module can only perform an aggregation of the data coming from the sensors, without prior processing, to form the message.

Long-range communication module [068] The main unit 1 also includes a long-range communication module 14. In the embodiment shown, this module is a radio transceiver which can communicate, for example, according to GSM standards, LTE, 4G, M2M (term for machine-to-machine communications), LoRa, Sigfox or any other protocol suitable for long-range communications. In other possible embodiments, this component may be a short-range radio communication component or a wired communication interface able to communicate with an external communication box, such as a modem present near the main box 1 .

[069] The long-range communication component 14 makes it possible to send the remote server 5 the messages which are prepared by the processing module.

Power supply [070] The main unit 1 also includes a power supply module (not shown in the figures), intended to power the various components electrically. In the embodiment shown, this power supply module is an assembly formed by a transformer and a battery, the battery being able to ensure continuity of the power supply in the event of short power outages. In this case, the power cut by the transformer can be detected by the processing module, which can generate, if necessary, an appropriate alert signal. In other possible embodiments, the power supply module can be constituted solely by a transformer which is connected to the mains, or only by a cell or a battery.

Secondary unit [071] FIG. 4 is an exploded representation of a secondary unit 3 implemented, with the main unit 1, in the system for monitoring the state of health of a person represented by FIG. 1.

[072] This secondary housing 3 comprises a box 31 closed by an upper cover 32 covering a plexiglass ring 33. The upper cover 32 includes a translucent window 321 which, when the upper cover 32 is closed, allows outside light to enter the housing 31.

[073] The housing 3 contains an acquisition card 2, of the same type as the acquisition card 2 of the main housing 1. The components of these acquisition cards are therefore identical, but they can be configured differently. It is, for example, possible that some of these components, which are unnecessary in the secondary housing or which risk consuming too much energy, are deactivated. In the embodiment shown, for example, the CO2 sensor and the VOC sensor of the card 2 are deactivated, in order to reduce the electrical consumption of the secondary unit 3.

This secondary unit 3 also includes a short-range communication component (not shown), using the same protocol as the short-range communication component of the control card 13 of the main unit 1. The secondary unit 3 can thus communicate with the main unit 1, in particular to communicate to the control card 13 the data collected by the secondary unit 3.

[075] Finally, this secondary box 3 also includes a power supply module which, in the embodiment shown, consists of a battery 34. Thus the secondary box 3 can, like the main box 1, collect data from environment representative of the activities that are carried out nearby. Furthermore, since this box does not have to be plugged in, its installation is particularly easy.

Specific secondary refrigerator housing [076] Figure 5 is a diagram of a specific secondary refrigerator housing 4. This housing is a secondary housing designed specifically to be placed in the door of a refrigerator. For this, it has an egg shape, conferred by the assembly of its lower cover 41 and its upper cover 42, allowing it to be easily accommodated in one of the locations intended for eggs which are generally provided in the doors of the refrigerators. This specific secondary housing 4 comprises an acquisition card 2 of the same type as the acquisition card 2 of the main housing 1, the components of which are configured in a different manner.

[077] The accelerometer and gyroscope in the specific refrigerator housing, detecting the opening and closing movements of the door of the refrigerator which receives the housing, can not only measure the use of the refrigerator but also measure representative data the quality of the opening and closing movement, and in particular its speed, whose subsequent analysis will make it possible to deduce useful information to monitor the state of health of the person being supported. The subsequent analysis of the temperature data collected by the thermometer will make it possible to verify that the refrigerator remains permanently at a temperature making it possible to avoid any break in the cold chain, and any deterioration of the food obtained.

Operation of the monitoring system [078] FIG. 6 schematically represents the method of processing data by the monitoring device represented by FIG. 1. In the home 9 of the accompanied person, the main unit 1 collects data from environment 61 in a room, the secondary boxes 3 collect environment data 62 from other rooms and the specific secondary box 4 of the refrigerator collects data 63 representative of the use of the refrigerator. These various data, which are all time-stamped, are gathered in a data message 64 which is sent to the server 5 via a long-range communication 92.

[079] It is possible, in other embodiments, to supplement the environmental data collected by the main unit 1 and the secondary units 3 and 4 with other data, coming from other sensors. It is thus possible to complete these data with consumption information collected in real time by meters, electricity, gas, water, etc. However, obtaining these data, which generally requires specific connections, is not necessary for the implementation of the invention.

[080] It is also possible, in a specific embodiment, that the control card 13 of the main unit 1 is capable of directly identifying manifestly problematic situations, such as an abnormal temperature, an excessively high level of CO2, etc. and to issue an alarm, for example audible or visual, to allow a correction of this situation. In the preferred embodiment, however, the main unit 1 does not process the collected data, which is sent for processing to the server 5.

First computer [081] In the server 5, this data is analyzed by a first computer 51, or first calculation means, intended to identify the activities which correspond to the environmental data 64 collected. For this, the data collected can be compared to a database of activities 510 linking, with a certain level of probability, combinations of environmental data with activities. The term "activities" in this application designates both the activities carried out by the accompanied person and other events that can be identified in their accommodation, such as an absence, the presence of a third person, living conditions, etc. .

[082] For example, an activity such as a shower can be identified by the combination of a particular sound signature, associated with an increase in temperature and humidity in the room. Taking a meal can be identified by an increase in temperature and humidity, an increase in the noise level including the sound signatures corresponding to the use of kitchen appliances, and openings and closings of the refrigerator detected by the specific secondary refrigerator housing.

[083] The rising of the accompanied person may be identified, by a box located outside the bedroom, by an increase in the sound level, an increase in the brightness and an increase in the level of CO2 due to the opening. from the bedroom door, the CO2 having accumulated in the bedroom overnight spreading into the other rooms. Television usage can be identified by a combination of brightness and sound data. Furthermore, the sensors of the specific refrigerator case can identify the movements of the case corresponding to a prolonged opening of the refrigerator and the increase in temperature which can generate a break in the cold chain in the refrigerator.

[084] The activities database 510 may contain, as soon as the monitoring system is installed, combinations of data corresponding to the main events and activities of the accompanied person's daily life.

However, the data corresponding to each event being very variable depending on the accommodation and the people, this activity 510 database will be updated constantly, or at regular intervals, to better recognize the activities in a given accommodation. This 510 activity database, containing the correspondences between the data collected and the activities or events, will thus be specific to each accommodation, or to each accompanied person, and will evolve by a machine learning method (usually known as name of "machine learning").

[085] To accelerate this learning, it is possible that the first computer 51 questions the accompanied person or his accompanying persons on certain aspects of his timetable (times for getting up, going to bed, meals, etc.). This interrogation can in particular be carried out when a set of data cannot be associated, by the first computer 51, with sufficient probability, with a particular event or activity. The answer to this query thus makes it possible to remove the ambiguity and to enrich the database of 510 activities.

[086] These interrogations on the activities of the accompanied person could advantageously be made via a monitoring terminal 6, constituted by a dedicated application which can be installed on a computer, a tablet or a telephone. This monitoring terminal 6 can be held by the person accompanied, by his companions or by his carers.

[087] Finally, the identification of the activities by the first computer 51 can also be made according to weather information near the home of the accompanied person, given by an external data source 52. The temperature data, outdoor humidity and sunshine can thus be compared with the temperature, humidity and brightness data in the accommodation 9 of the accompanied person, which allows for example to help identify the opening of a door, d 'a window, the operation of heaters, or the difference between natural light and artificial light.

[088] A displacement of one of the boxes of the monitoring system can be detected by the data from the information sensors representative of the movements of the box. Since such a displacement can lead to a modification of the captured environmental data, for example of the noise or light level, the first computer 51 can, after such a displacement, recalibrate the correspondences recorded in the activity database 510 .

[089] When the first computer has identified the activities of the accompanied person, it records these activities, which are time-stamped, in a timetable database 511.

Second computer [090] A second computer 52, or second means of calculation, analyzes the data contained in this timetable database 511, in order to identify anomalies or changes, abrupt or progressive, in the job of the person's time.

[091] These modifications of the timetable can be analyzed by comparing them to a database of modifications of the timetable 520 which establishes a link, with a certain level of probability, between modifications of the timetable and life situations that may be abnormal or pathologies.

[092] Thus, this second computer 52 can identify abnormal activities or absences of activity, which can justify the immediate visit of an accompanying person to the accompanied person. Such life situations can be, for example, an abnormal temperature in the housing, too high or too low, an absence of getting up, an absence of food, etc.

This second computer 52 can also identify a life situation resulting in a sudden change in habits, for example reflecting a drop in activities, which may be the sign of an illness, frailty or loss of autonomy of the supported person.

[094] Finally, this second computer 52 can also identify life situations resulting in signals of lesser importance, called “weak signals”, such as a progressive shift in the time of getting up, a decrease in the frequency of toilets. , a decrease in brightness in the accommodation or a less frequent opening of windows. Such weak signals, spread over time, can be the warning signs of a pathology, a depressive state, or a loss of autonomy.

[095] To correctly identify the abnormal life situations or the evolutions of the timetable, the database of the modifications of the timetable 520 is enriched, progressively, by the information concerning the habits of the accompanied person contained in the time use database 511.

[096] For each of the identified life situations which can justify an intervention, curative or preventive, a message can be sent to the accompanying persons of the accompanied person, via the monitoring terminal 6, in order to allow them to take the measures appropriate. This message can also ask them about the life situation corresponding to the changes in timetable identified (for example on a pathology or a change in habits identified by the carers), so that this information can enrich the database of changes in timetable 520.

This analysis of the weak signals can make it possible to detect early signs of pathologies or loss of autonomy even before the appearance of any sensitive symptom. This prediction made by the second computer 52 can thus lead to the sending of a message to the accompanying persons of the accompanied person, to draw their attention to the identified risk and to enable them to implement a preventive action or an early curative action.

Third calculator [098] Advantageously, a third calculator 53, or third calculating means, analyzes the data contained in the databases of timetable modifications 520 associated with a large number of different accompanied persons. This statistical analysis, which can take into account known information on the life situations of each accompanied person, allows the third calculator 53 to identify changes in timetable which correspond to a life situation such as a pathology or a loss of autonomy, or even the warning signs of a pathology or a loss of autonomy, in order to recognize these life situations in a person with similar changes in schedule. The analysis by the third calculator 53 of data of a large number of people thus makes it possible to enrich the databases of the modifications of timetable 520, by matching, with a determined level of probability, the evolutions of timetable with potential life situations of the supported person.

In the embodiment represented by FIG. 6, the analysis of the evolution of the timetable is done on the basis of the information of activities identified, and stored in the employment database of the time 511. In other possible embodiments, it is also possible that this analysis is carried out by the second computer directly on the basis of the environmental data coming from the various sensors.

[100] Furthermore, although FIG. 6 represents different computers, the different computer processing operations can obviously be implemented differently, without departing from the invention. The different computer processing operations can, for example, be implemented by different programs of the same computer.

[101] Advantageously, a main box and the secondary boxes which are intended to be used in the same housing are configured and paired before their delivery in this housing. Thus, their installation is particularly simple. It suffices to plug the main box 1 into a main living room (for example the kitchen), the secondary box or boxes in the other rooms, for example in the bathroom and in the living room, and the sensor secondary specific 4 in the refrigerator.

[102] Installation of the system is therefore particularly easy, and makes it possible, in a very non-intrusive manner, to observe both the events in the life of the accompanied person that require the intervention of the accompanying persons, and the weak signals may require preventive action or early curative action.

Claims (10)

1. A system for monitoring the health of a person, comprising at least two boxes (1, 3, 4) intended to be placed in the home (9) of said person, and a remote server (5), characterized in that each of said housings (1, 3, 4) comprises at least two environmental data sensors (21, 22, 23, 24, 25, 26, 27), measuring environmental data (64) in the room in which said box is placed, and sending them to said remote server (5), and in that said remote server (5) analyzes said environment data (64), by associating a combination of environment data coming from different sensors (21, 22, 23, 24, 25, 26, 27) to an activity of said person. 2. Monitoring system according to the preceding claim, characterized in that at least one of said boxes (1, 3, 4) comprises at least one microphone (23). 3. Monitoring system according to any one of the preceding claims, characterized in that said boxes (1, 3, 4) comprise at least two environmental data sensors from: a microphone (23), - a light sensor (24), - a thermometer (21), - a hygrometer (22), - An information sensor representative of the movements of said housing (27). 4. Monitoring system according to the preceding claim, characterized in that at least one of said boxes (4) comprises both a thermometer (21) and an information sensor representative of the movements of said box (27). 5. Monitoring system according to any one of the preceding claims, characterized in that one of said boxes (1) collects the environmental data measured by the others of said boxes (3, 4), and sends to said remote server ( 5) all of the environmental data (64) measured in said housing (9). 6. Monitoring system according to any one of the preceding claims, characterized in that said remote server (5) comprises first calculation means (51) analyzing said environment data (64) by comparing them with data from a activity database (510), in which combinations of environment data (64) from different sensors (21, 22, 23, 24, 25, 26, 27) are associated, with a given probability level , to activities constituting the timetable of the said person. 7. Monitoring system according to the preceding claim, characterized in that said first calculation means (51) modify the data of said activity database (510), as a function of the environmental data coming from different sensors (21 , 22, 23, 24, 25, 26, 27) associated with information on the timetable of said person. 8. Monitoring system according to any one of claims 6 and 7, characterized in that said remote server (5) comprises second calculation means (52) analyzing the schedule determined by said first calculation means (51 ), by comparing it with data from a schedule change database (520), in which changes in a person's schedule are associated, with a given level of probability, with a life situation. 9. Monitoring system according to the preceding claim, characterized in that said remote server comprises third calculation means (53) analyzing the data from the databases of schedule modifications (520) of a plurality of people, associated with information on the life situations of these people, in order to modify or supplement the associations between changes in a person's timetable and a life situation, to modify or supplement the data in said databases of schedule changes (520). 10. A method of monitoring the state of health of a person, characterized in that it implements a monitoring system according to any one of the preceding claims, and in that it comprises: a step of measuring environmental data in at least two 5 rooms in the home (9) of said person, a step of sending said environment data (64) to said remote server (5), a step of analyzing said environment data by said remote server (5), combining a combination of environment data 10 from different sensors to an activity of said person.