System for analysing the ambient air of a room

Abstract

The invention relates to a system for analysing the ambient air of a room, comprising: • - an aeraulic module (1) configured to make some of the ambient air of the room flow through an analysis duct (10); • - a set of sensors (60-69) configured to measure physical and chemical parameters of the ambient air of the room, and / or of the air flowing through the duct (10); • - a module (2) for storing dynamic data output by the sensors (60-69) and static data; • - an analysis module (3) configured to detect the presence of one or more pathogens based on the data output by the sensors (60-69) and static data stored in the storage module (2); • - a communication module (4) configured to render the results of the analysis visually and / or audibly, and / or to transmit the data to an entity external to the analysis system.

Description

[0001] DESCRIPTION

[0002] TITLE: ROOM AMBIENT AIR ANALYSIS SYSTEM

[0003] technical field

[0004] The invention relates to a system for analyzing the ambient air in a room. The system can be implemented for the detection of airborne pathogens.

[0005] Previous technique

[0006] Pathogens can be transmitted via several routes, including contact or airborne transmission in the form of microdroplets. Due to their small size, infectious microdroplets can remain suspended in the air long enough to be inhaled by an individual or deposited on a wound, causing infection.

[0007] The detection of pathogens is based primarily on the identification of the pathogen through testing of patients who are already infected and symptomatic, thus excluding asymptomatic cases, followed by the study of its spread through the study of individual behavior and data collected by health networks.

[0008] However, airborne pathogens can spread very rapidly, especially in shared spaces, and this rapid spread makes them particularly well-suited to creating epidemics and pandemics.

[0009] Early detection of these airborne pathogens is therefore essential to rapidly implement effective measures to limit their spread.

[0010] Numerous solutions have been proposed for detecting the presence of pathogens in the air. These solutions implement methods such as measuring the electrical conductivity of aerosol deposits, measuring the weight of dehydrated aerosols, detecting the size of aerosols in the air after drying, or using antigen detection. However, these solutions do not allow for the real-time, continuous, and uninterrupted detection and identification of a wide range of pathogens, with good discrimination against pollutants. Indeed, methods using aerosol weight or conductivity cannot differentiate between pathogens and pollutants such as combustion dust. Antigen-based methods are more selective and do not allow for the study of a broad spectrum of pathogens; the use of reagents requires regular refilling and significant maintenance.Furthermore, these solutions require a significant amount of pathogens to initiate the reactions enabling their detection, which makes them, in practice, unusable given the low air charge (or requires extremely large quantities of air to be passed through, with significant consequences on noise, size, drying of reagents...).

[0011] There is therefore a need for a new alternative solution to detect and identify pathogens (such as bacteria, yeasts, viruses, spores, etc.) in ambient air.

[0012] Presentation of the invention

[0013] In this context, the present invention therefore aims to provide a system for analyzing the ambient air of a room, with a view to detecting pathogens.

[0014] The invention aims in particular at a suitable solution for the automatic detection of a wide range of pathogens in real time, continuously, without human intervention, with a good ability to discriminate against pollutants.

[0015] The invention relates to a system for analyzing the ambient air of a room, comprising:

[0016] - an aerodynamic module configured to circulate a portion of the ambient air from the room into an analysis duct;

[0017] - a set of sensors configured to measure physical and chemical parameters of the ambient air in the room, and / or the air circulating in the duct;

[0018] - a module for storing dynamic data from sensors and static data;

[0019] - an analysis module configured to identify and / or detect the presence of one or more pathogens from data from sensors and static data recorded in the storage module;

[0020] - a communication module configured to visually and / or audibly display the results of the analysis, and / or to transmit data and / or instructions to one or more entities external to the analysis system.

[0021] The system of the invention, through the use of various sensors, makes it possible to discriminate a target component, such as a pathogen, from other components present in the ambient air. In particular, the system of the invention makes it possible to identify and / or detect one or more target pathogens present in the ambient air of a room to be monitored and to limit detection errors induced, for example, by the presence of pollutants.

[0022] The sensor suite includes at least one or a combination of the following devices:

[0023] - room temperature sensor;

[0024] - relative humidity sensor for the room;

[0025] - module for measuring the level of ultraviolet rays present in the room;

[0026] - module for measuring the room's lighting level;

[0027] - module for measuring the CO2 level in the room air;

[0028] - module for measuring the level of nitrogen oxide NOx in the air of the room;

[0029] - volatile chemical compound measurement module;

[0030] - module for detecting and measuring sulfur and phosphorus in the air;

[0031] - module for detecting the chemical composition of aerosols or their fluorescence;

[0032] - module for determining the number and size distribution of droplets and dust particles suspended in the room;

[0033] - a spectrometry device configured to perform the decomposition of air constituents into excited simple elements and the analysis of the emission spectrum related to the de-excitation of the elements.

[0034] Static data can include one or a combination of the following data:

[0035] - the specific characteristics of the room, such as the geometry, the volume of the room, the presence of openings, ventilation, aeration, the reference conditions (temperature, humidity, etc.) of the room, the location of the room, the purpose of the room;

[0036] - Invariable data specific to each individual present in the room, such as age, medical history or particularities, ...; - Invariable data specific to each photosynthetic organism present in the room, such as plant type, growth type, CO2 absorption rate, chlorinated carbon (Ch) release rate,

[0037] - data relating to one or more known pathogens, such as their mode of transmission, their mode of propagation, their rate of multiplication, their average threshold of contagiousness...;

[0038] - data relating to one or more pollutants, dust, gas, organic compounds;

[0039] - data relating to one or more microorganisms considered to be non-pathogenic;

[0040] - etc.

[0041] Dynamic data can include data provided by sensors.

[0042] In practice, at least the aerodynamic module, the analysis duct, and the set of sensors are arranged in the same housing.

[0043] In practice, the analysis module includes a dynamic model enabling the detection of different types of pathogens from stored static and dynamic data.

[0044] For example, the dynamic model can be built by the analysis module, its construction including in particular an evolutionary learning phase during a predefined period and during which the analysis system is placed in real conditions and in the presence of a known target pathogen, the measured data being recorded at a regular frequency to constitute a dataset with which the model is built from one or more learning algorithms.

[0045] For example, the dynamic model may include a reference spectral signature of a target pathogen under real-world conditions.

[0046] Brief description of the figures

[0047] Other features and advantages of the invention will become clear from the description given below, by way of example and not limitation, with reference to the attached drawings, in which: [Fig 1] is a schematic representation of an analysis system according to one embodiment;

[0048] [Fig 2] is a schematic representation of an example of a spark-excited plasma spectroscopic sensor according to one embodiment.

[0049] It should be noted that in these figures, the same reference numerals designate identical or analogous elements, and the different structures are not drawn to scale. Furthermore, for the sake of clarity, only the elements essential to understanding the invention are shown in these figures.

[0050] Detailed description of the invention

[0051] An example of an ambient air analysis system for a room to be monitored is shown in Figure 1.

[0052] The system includes, in particular:

[0053] - an aerodynamic module (1) configured to circulate part of the ambient air of the room into an analysis duct (10);

[0054] - a set of sensors configured to measure in real time and continuously physical and chemical parameters of the ambient air in the room, and / or of the air circulating in the duct (10);

[0055] - a storage module (2) for dynamic data from sensors and static data;

[0056] - an analysis module (3) configured to detect in real time and continuously the presence of one or more pathogens from data from sensors and static data recorded in the storage module;

[0057] - a communication module (4) configured to visually and / or audibly display the results of the analysis, and / or to transmit the data to an entity external to the analysis system.

[0058] The sensors may include, for example: A temperature sensor (60). Indeed, temperature is a parameter used, in particular, to calibrate certain other sensors or to derive information from other sensors. Studying temperature variations makes it possible to detect events such as opening windows, turning on air conditioning, or other parameters involving sudden or gradual changes (day-night cycle, changes in the number of people in the room, etc.).

[0059] A humidity sensor (61). This data makes it possible to determine the survival capacity of pathogens in the room, but also to obtain "background" information, such as the presence of people (increase in ambient humidity), certain parameters such as the variation in size of particles emitted by an individual or fluorescence parameters that can be impacted by the humidity of the air.

[0060] An ambient light sensor (62), to differentiate for example between day and night, as well as to detect periods of presence and activity in the area, alone or in conjunction with other sensors.

[0061] An ultraviolet (UV) sensor (63) is used to determine the substantial external UV-A input. UV-A radiation is known to inactivate viruses and bacteria at sufficient intensity. Conversely, low UV-A intensity has a photoreparative effect on bacteria.

[0062] A CO2 sensor (64) to measure the absolute concentration of CO2 in the room, and to assess the physical effort developed in the area, as well as its variation.

[0063] A Volatile Organic Compound (VOC) sensor (65). Indeed, many diseases and infections alter the characteristics of exhaled air, enriching it in Volatile Organic Compounds (VOCs) resulting from the attack on human tissues by viruses and bacteria and the body's response. A Nitrogen Dioxide (NOx) sensor (66). Since the body emits nitrogen oxides as a result of physiological processes, studying variations in the amount of NOx in the air of the area can provide information on the health of an individual present in the room.

[0064] A particle sensor (67). Ambient air is laden with particles, including mineral particles (sand dust, sea spray, for example), organic particles (pollen, carbonaceous dust from fuel combustion, etc.), and viruses, bacteria, and spores. Studying the size and quantity of aerosols present in the air makes it possible to differentiate the droplets carrying these particles.

[0065] A UV-excited aerosol fluorescence spectroscopic sensor (68). An airflow from the room is directed into the duct (10) where this air is illuminated by one or more essentially monochromatic UV sources (680). The aerosols absorb some of the light, which they re-emit at longer wavelengths. This sensor can use several non-coherent light sources for excitation and generates the emission spectra(s) of the aerosols. In practice, these analyzed spectra can range from 200 nm to 2000 nm, preferably from 260 nm to 1000 nm. This sensor allows for a much finer analysis of fluorescence, and therefore a more precise determination of the chemical species present in the aerosols.

[0066] A near-infrared excitation aerosol fluorescence spectroscopic sensor (68). An airflow from the room is directed into the duct (10) where this air is illuminated by one or more essentially monochromatic near-infrared sources (680). The aerosols absorb some of the light, which they re-emit at longer wavelengths. This sensor can use several non-coherent light sources for excitation and generates the emission spectra of the aerosols. In practice, these analyzed spectra can range from 700 nm to 2000 nm, preferably from 700 nm to 1100 nm. This sensor allows for the analysis of infrared fluorescence, and therefore a more precise determination of the chemical species present in the aerosols.

[0067] A plasma spectroscopic sensor (69) is generated by a spark and / or laser and / or via a high-power LED. In practice, the airflow is concentrated towards the inlet of a spectrometry device, in a direction parallel to the direction of the inlet slit. The airflow is concentrated around a plasma-generating device located in front of the spectrometer slit. This plasma can be generated in several ways:

[0068] - By spark (690): a train of pulses is sent by a spark-generating device, onto electrodes, creating a plasma at the entrance of the spectrometer.

[0069] - By Laser: a laser beam is created by a nanosecond or femtosecond laser of power, and this laser is focused on the area in front of the spectrometer slit.

[0070] - By LED: this solution is identical to the laser, but the excitation system consists of an essentially monochromatic focused LED, driven by a pulsed current generator delivering a high intensity to the LED for a time on the order of ten nanoseconds.

[0071] This plasma emits radiation, including lines corresponding to the de-excitation of chemical elements composing air and aerosols. If the aerosols contain biological elements, these include sulfur-containing amino acids and phosphorus-containing organic material. The plasma breaks down these amino acids and / or organic material and / or nucleic acids into simpler chemical elements, including sulfur, calcium, oxygen, carbon, hydrogen, and phosphorus. The excitation of these elements produces specific lines, notably the line of doubly ionized sulfur (SU) at 545.38 nm and the line of ionized phosphorus (PI) at 253.56 nm. The detection of these lines indicates the presence of these elements in the aerosols. For example, since sulfur and phosphorus are naturally rare in air, the presence of these components is a significant indicator of the presence of viral, bacterial, pollen, and / or spore aerosols.To intensify the spectral lines, an electrostatic collection device can be implemented. A positively charged grid is placed upstream of the airflow, and a negative electrode is positioned downstream of the spectrometer slit. Particles are polarized by the current from the grid and attracted to the negative electrode, which then concentrates the charged species in the air. The negative electrode is thus used as a target for the spark or as the focal point of the laser / LED, and the collected particles are transformed into plasma during measurement. This collection mechanism greatly improves sensitivity by collecting and immobilizing charged particles containing pathogens, thereby increasing the system's accuracy and lowering the detection limit. An example of a spark-excited plasma spectroscopic sensor (7) is shown in Figure 2. This sensor comprises:

[0072] - an insulating tube (70) having a window (71) transparent to optical radiation to which a spectrometer (72) is sensitive;

[0073] - an air inlet (73) and an air outlet (74) through which a controlled flow of air flows;

[0074] - a negative electrode (75), which can be grounded, and a positive electrode (76) electrically coupled to an ionization grid (77), all three connected to a high voltage generator (78) capable of generating a spark (79) between the two electrodes, which are positioned so that the spark occurs in front of the window of the tube and the collection slit (720) of the spectroscope (72).

[0075] The analysis module (3) is configured to analyze in real time and continuously the ambient air of the room from the data from the sensors and static data recorded in the storage module and to interpret this data to determine if the air of the area contains aerosols carrying viruses or bacteria or other pathogens.

[0076] In practice, the analysis module may include a dynamic model, for example in the form of a digital twin, using the various stored data to predict the evolution of the physical and chemical parameters of the ambient air in the room and to alert in case of detection of viruses and / or bacteria potentially problematic for a human being.

[0077] As a non-limiting example, the construction of the dynamic model of the analysis module adapted for monitoring in a complex environment such as a hospital room may include an evolutionary learning phase and a real-time and continuous detection phase.

[0078] During the learning phase:

[0079] - The data from the various sensors are, after calibration and correction according to temperature and humidity, formatted, time-stamped and pre-processed to provide a uniform list of parameters. - An operator is asked to indicate the appearance of alterations in the patient's clinical health status (temperature, presence of fever, respiratory problems, etc.) as well as the tests and results of determining the presence of a pathogen (PCR, antigen test, etc.).

[0080] - These data are recorded at a regular frequency, and constitute a raw data catalog (or "dataset" in English) on which the model algorithm will be trained.

[0081] - A predetermined percentage (for example 10%) of the data is excluded from the dataset at this stage and will be used to validate the algorithm.

[0082] At the end of the data collection period, which can extend over several days, weeks, or months, the data is retrieved from the various systems used for training. The collection period will depend on the amount of data retrieved. It is important that the dataset be large and significant, and include data relating to diverse situations (presence / absence of patients, pathogens, infections, visits, etc.). Several learning algorithms can be successively applied to the dataset (neural networks, random forest, decision trees, etc.). The best algorithm is selected. The selection criteria are sensitivity, specificity, the rate of predicted positive values, the rate of predicted negative values, and the accuracy of the determination.

[0083] - The learning process can be regularly updated, particularly if the device allows healthcare teams to enter a patient's infectious status. Raw data can be retrieved via the communication module (4), and the refined model uploaded to the system by this same communication module (4). Furthermore, the dataset will be analyzed and processed by artificial intelligence (AI) to determine correlations and / or statistical patterns reflecting the probability and nature of the presence or absence of pathogens.

[0084] For example, one of the dataset's data points could be linked to the spectral signature of a target pathogen. This spectral signature can be stored as static data. During the training phase, the analysis system is placed in a real-world training environment, such as a hospital room with a patient exhibiting the target pathology. From all the static and dynamic data, including environmental data measured by sensors over a predefined period, a reference spectral signature of the target pathogen is reconstructed, for example, by extrapolation. This reference spectral signature corresponds to the spectral signature of the target pathogen under real-world conditions.

[0085] During the detection phase:

[0086] - The data from the sensors, previously calibrated, conditioned and formatted, are provided to the model obtained during the learning phase.

[0087] - The use of the model allows us to deduce various pieces of information such as the occupancy status of the room, the presence of pathogens and their type, and the risk of contamination if applicable.

[0088] - Users can be informed either by a visual and / or audible indicator or by a remotely transmitted alarm.

[0089] A second learning phase can be carried out simultaneously with the detection phase, in order to adjust the model according to the new dynamic and / or static data stored.

[0090] In another embodiment, the analysis module can be coupled to, or integrated with, an audio or sound analysis system incorporating at least one sound sensor, such as a microphone, configured to capture sound or audio signals present in the room. This sound system can be configured to detect the presence of individuals in the room by recognizing characteristic sounds, including speech or coughing noises. These sound signals can also be used by the analysis module to determine whether certain sound signals are associated with pathophysiological manifestations, such as coughing fits, and to correlate these events with physicochemical measurements recorded by the sensors, in order to refine the environmental diagnosis.The audio system can also be configured to recognize and interpret certain instructions or voice observations made by a healthcare professional present in the room, in order to automatically complete a patient record or dynamically adapt the analysis of environmental data based on the detected clinical context. The system thus enables the detection of microorganisms and the identification, in a single step, of one or more target pathogens present in the ambient air of a room under monitoring, even in the presence of asymptomatic individuals.

Claims

DEMANDS 1. Room ambient air analysis system comprising: - an aerodynamic module (1) configured to circulate part of the ambient air of the room into an analysis duct (10); - a set of sensors (60-69) configured to measure physical and chemical parameters of the ambient air in the room, and / or of the air circulating in the duct (10); - a storage module (2) for dynamic data from sensors (60-69) and static data; - an analysis module (3) configured to detect the presence of one or more pathogens from data from sensors (60-69) and static data recorded in the storage module (2); - a communication module (4) configured to display the results of the analysis visually and / or audibly, and / or to transmit data and / or instructions to one or more entities external to the analysis system; characterized in that: - the analysis module includes a dynamic model for detecting pathogens of different types from static and dynamic data; - the dynamic model is built by the analysis module, its construction including an evolutionary learning phase over a predefined period and during which the analysis system is placed in real conditions and in the presence of a target pathogen, the measured data being recorded at a regular frequency to constitute a dataset with which the model is built using one or more learning algorithms; the dynamic model includes a reference spectral signature of a target pathogen in real conditions.

2. Analysis system according to claim 1 in which the sensor assembly comprises at least one or a combination of the following devices: - room temperature sensor; - relative humidity sensor for the room; - module for measuring the level of ultraviolet rays present in the room; - module for measuring the room's lighting level; - module for measuring the CO2 level in the room air; - module for measuring the level of nitrogen oxide NOx in the air of the room; - volatile chemical compound measurement module; - module for detecting and measuring sulfur and phosphorus in the air; - module for detecting the chemical composition of aerosols or their fluorescence; - module for determining the number and size distribution of droplets and dust particles suspended in the room; - a spectrometry device configured to perform the decomposition of air constituents into excited simple elements and the analysis of the emission spectrum related to the de-excitation of the elements.

3. Analysis system according to claim 1 or 2, wherein the static data comprises one or a combination of the following data: - the specific characteristics of the room, such as the geometry, the volume of the room, the presence of openings, ventilation, aeration, the reference conditions of the room, the location of the room, the purpose of the room; - invariable data specific to each individual present in the room, such as age, history or medical condition; - data relating to one or more known pathogens, such as their mode of transmission, their mode of propagation, their rate of multiplication; - data relating to one or more pollutants, dust, gas, organic compounds.

4. Analysis system according to any one of claims 1 to 3, wherein at least the aerodynamic module (1), the analysis duct (10), and the sensor assembly (60-69) are arranged in the same housing.

Images