Room air analysis system

Abstract

A room air analysis system comprising: - an air handling module (1) configured to circulate a portion of the room's ambient air through an analysis duct (10); - a set of sensors (60-69) configured to measure physical and chemical parameters of the room's ambient air and / or the air circulating in the duct (10); - a storage module (2) for dynamic data from 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 from the sensors (60-69) and the static data recorded in the storage module (2); - a communication module (4) configured to display the analysis results visually and / or audibly, and / or to transmit the data to an entity external to the analysis system. Figure for the abstract: Fig 1

Description

Title of the invention: System for analyzing the ambient air in a room. Technical field

[0001] The invention relates to a system for analyzing the ambient air of a room. The system can be implemented for the detection of pathogens in the air. Previous technique

[0002] Pathogen transmission can occur via several routes, including contact or airborne transmission in the form of microdroplets, for example. Due to their small size, infectious microdroplets are able to remain suspended in the air for a sufficient time to allow inhalation by an individual or deposition on a wound, causing infection.

[0003] The detection of pathogens is essentially based on the identification of the pathogen through testing of patients already infected and symptomatic, which excludes asymptomatic cases, followed by the study of its spread through the study of the behavior of individuals, and data collected by health networks.

[0004] However, airborne pathogens can spread very rapidly, especially in shared spaces, and this rapid spread makes them particularly suited to the creation of epidemics and pandemics.

[0005] Early detection of these airborne pathogens is therefore essential to rapidly implement effective measures to limit their spread.

[0006] Numerous solutions for detecting the presence of pathogens in the air have been proposed. These solutions implement the measurement of the electrical conductivity of the aerosol deposit, or the measurement of the weight of dehydrated aerosols, or the detection of the size of aerosols in the air after drying, or detection by reaction to antigens. However, these solutions do not allow for the detection and identification of a wide range of pathogens in real time, continuously, without human intervention, with good discrimination against pollutants. Indeed, methods using the weight or conductivity of aerosols 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 wide range of pathogens; the use of reagents implies 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 pressure (or which requires extremely large quantities of air to pass through, with significant consequences for noise, size, drying of reagents...)

[0007] There is therefore a need for a new alternative solution for detecting and identifying pathogens (such as bacteria, yeasts, viruses, spores, etc.) in ambient air. Presentation of the invention

[0008] 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.

[0009] 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.

[0010] The invention relates to a system for analyzing the ambient air of a room comprising:

[0011] - an aerodynamic module configured to circulate a portion of the ambient air from the part in an analysis conduit; - 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; - a module for storing dynamic data from sensors and static data; - 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; - 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.

[0012] 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.

[0013] 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.

[0014] Static data may include 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 (temperature, humidity, etc.) 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, ...; - the invariable data specific to each photosynthetic organism present in the room, such as plant type, growth type, CO2 absorbed rate, O2 released rate, ...; - 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...; - data relating to one or more pollutants, dust, gas, organic compounds;

[0015] - data relating to one or more microorganisms considered as not pathogens; - etc.

[0016] Dynamic data may include data provided by sensors.

[0017] In practice, at least the aerodynamic module, the analysis duct, and the set of sensors are arranged in the same housing.

[0018] In practice, the analysis module includes a dynamic model enabling the detection of pathogens of different types from the stored static and dynamic data.

[0019] For example, the dynamic model can be built by the analysis module, its construction including in particular an evolutionary learning phase over a predefined period and during which the analysis system is placed in a condition real and in the presence of a known target pathogen, the measured data are recorded at a regular frequency to constitute a dataset with which the model is built from one or more learning algorithms.

[0020] For example, the dynamic model may include a reference spectral signature of a target pathogen under real-world conditions. Brief description of the figures

[0021] Other features and advantages of the invention will become clear from the following description, which is by way of example and not limitation, with reference to the accompanying drawings, in which:

[0022] [Fig.1] is a schematic representation of an analysis system according to one embodiment;

[0023] [Fig.2] is a schematic representation of an example of a spark-excited plasma spectroscopic sensor according to one embodiment.

[0024] It should be noted that in these figures, the same reference numerals designate identical or analogous elements, and the different structures are not to scale. Furthermore, only the elements essential to understanding the invention are represented in these figures for reasons of clarity. Detailed description of the invention

[0025] An example of an ambient air analysis system for a room to be monitored is shown in [Fig.1].

[0026] The system includes, in particular: - an aerodynamic module (1) configured to circulate part of the ambient air of the room into an analysis duct (10); - 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); - a storage module (2) for dynamic data from sensors and static data; - 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; - 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.

[0027] The sensors may include, for example:

[0028] A temperature sensor (60). Indeed, temperature is a parameter that allows, in particular, the calibration of certain other sensors or the derivation of information from other sensors. The study of temperature variation makes it possible to detect events such as the opening of windows, the switching on of air conditioning or other parameters of sudden or slow variation (day-night rhythm, variation in the number of people in the room, etc.).

[0029] 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.

[0030] 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.

[0031] A UV ultraviolet ray sensor (63) to determine the substantial input of external UV-A. UV-A rays are known for their ability to inactivate viruses and bacteria if their intensity is sufficient. Conversely, if the UV-A intensity is low, there is a photoreparative effect on bacteria.

[0032] A CO2 sensor (64) to measure the absolute concentration of CO2 in the room, and to evaluate the physical effort developed in the area, as well as its variation.

[0033] A sensor for Volatile Organic Compounds VOCs (65). Indeed, many pathologies and infections modify the characteristics of exhaled air, enriching it in particular with Volatile Organic Compounds (or VOCs in English for "Volatile Organic Compounds") resulting from the attack of human tissues by viruses and bacteria and the response of the body.

[0034] A nitrogen dioxide sensor Nox (66). Since the body emits nitrogen oxides as a result of physiological processes, studying the variations in the amount of NOx in the air of the area can provide information on the health status of an individual present in the room.

[0035] A particle sensor (67). Ambient air is laden with particles, such as mineral particles (sand dust, sea spray, for example), organic particles (pollen, carbonaceous dust from fuel combustion, etc.), or it may be laden with viruses, bacteria, and spores. Studying the size and quantity of aerosols present in the air makes it possible, in particular, to differentiate the droplets carrying these particles.

[0036] A UV-excited aerosol fluorescence spectroscopic sensor (68). An airflow from the room is directed to the duct (10) in which 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 The analyzed spectra can range from 200nm to 2000nm, preferably between 260nm and 1000nm. This sensor allows for a much finer analysis of fluorescence, and therefore a more precise determination of the chemical species present in aerosols.

[0037] A near-infrared excitation spectroscopic aerosol fluorescence 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(s) 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.

[0038] A plasma spectroscopic sensor (69) generated by 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: - By spark (690): a train of pulses is sent by a spark generating device to electrodes, creating a plasma at the inlet of the spectrometer. - By laser: a laser beam is created by a high-power nanosecond or femtosecond laser, and this laser is focused on the area in front of the spectrometer slit. - 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.

[0039] This plasma emits radiation, and in particular lines corresponding to the de-excitation of the chemical elements composing air and aerosols. If the aerosols include biological elements, these include sulfur-containing amino acids and phosphorus-containing organic material. The plasma decomposes these amino acids and / or organic material and / or nucleic acids into simple chemical elements, including sulfur, calcium, oxygen, carbon, hydrogen, and phosphorus. The excitation of these elements produces specific lines, including the line of doubly ionized sulfur (SII) 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 The presence of these components is a significant indicator of the presence of viral, bacterial and / or pollen and / or spore aerosols.

[0040] 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. The 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 pathogen-laden particles, thereby increasing the system's accuracy and lowering the detection limit.

[0041] An example of a spark-excited plasma spectroscopic sensor (7) is illustrated in [Fig. 2]. This sensor comprises: - an insulating tube (70) having a window (71) transparent to optical radiation to which a spectrometer (72) is sensitive; - an air inlet (73) and an air outlet (74) through which a controlled flow of air flows; - 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).

[0042] 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 the 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.

[0043] In practice, the analysis module may include a dynamic model, for example in the form of a digital twin, using the different stored data to predict the evolution of the physical and chemical parameters of the ambient air of the room and to alert in the event of detection of viruses and / or bacteria potentially problematic for a human being.

[0044] By way of 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.

[0045] During the learning phase: - The data from the different sensors are, after calibration and correction according to temperature and humidity, formatted, time-stamped and pre-processed in order to provide a uniform list of parameters. - An operator is asked to indicate the appearance of alterations in the patient's clinical health (temperature, presence of fever, respiratory problems...) as well as the tests and results of determining the presence of a pathogen (PCR, antigen test...). - These data are recorded at a regular frequency, and constitute a raw data catalogue (or “dataset” in English) on which the model algorithm will be trained. - 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. 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. - 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.

[0046] By way of example, one of the dataset's data points can be linked to the spectral signature of a target pathogen. The spectral signature specific to the target pathogen can be stored as static data. During the learning phase, the analysis system is placed in a real-world learning environment, such as a hospital room with a patient exhibiting the target pathology. From all the static and dynamic data, and in particular the environmental data measured by the 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.

[0047] During the detection phase: - The data from the sensors, previously calibrated, conditioned and formatted, are provided to the model obtained during the learning phase. - 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. - Users can be informed either by a visual and / or audible indicator or by a remotely transmitted alarm.

[0048] A second learning phase can be carried out simultaneously with the detection phase, so as to adjust the model according to the new dynamic and / or static data stored.

[0049] The system thus makes it possible to detect microorganisms and to identify in a single step, one or more target pathogens present in the ambient air of a room to be monitored, even in the presence of asymptomatic individuals.

Claims

Demands

1. A room ambient air analysis system comprising: - an air handling module (1) configured to circulate a portion of the room ambient air through an analysis duct (10); - a set of sensors (60-69) configured to measure physical and chemical parameters of the room ambient air, and / or of the air circulating in the duct (10); - a storage module (2) for dynamic data from the sensors (60-69) and static data; - an analysis module (3) configured to detect the presence of one or more pathogens from the data from the sensors (60-69) and the static data recorded in the storage module (2); - a communication module (4) configured to display the analysis results visually and / or audibly, and / or to transmit the data and / or instructions to one or more entities external to the analysis system.

2. 2. An analysis system according to claim 1, wherein the sensor assembly comprises at least one or a combination of the following devices: - room temperature sensor; - room relative humidity sensor; - module for measuring the level of ultraviolet radiation present in the room; - module for measuring the level of illuminance in the room; - module for measuring the CO2 level in the room air; - module for measuring the level of nitrogen oxide (NOx) in the room air; - module for measuring volatile chemical compounds; - 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. 3. Analysis system according to claim 1 or 2, wherein the static data comprise one or a combination of the following data: - the specifics of the part, such as the geometry, the volume of the part, the presence of openings, ventilation, aeration, the reference conditions of the part, the location of the part, the purpose of the part; - the invariant data specific to each individual present in the part, such as age, history or medical particularity; - the data relating to one or more known pathogens, such as their mode of transmission, their mode of propagation, their rate of multiplication; - the data relating to one or more pollutants, dust, gas, organic compounds.

4. 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.

5. 5. Analysis system according to any one of claims 1 to 4, wherein the analysis module includes a dynamic model for detecting pathogens of different types from static and dynamic data.

6. Analysis system according to claim 5, wherein the dynamic model is built by the analysis module, its construction comprising 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.

7. Analysis system according to claim 6, wherein the dynamic model can include a reference spectral signature of a target pathogen under real-world conditions.

Images