WO2009064868A1 - Bioaerosol measuremeny system - Google Patents

Bioaerosol measuremeny system Download PDF

Info

Publication number
WO2009064868A1
WO2009064868A1 PCT/US2008/083383 US2008083383W WO2009064868A1 WO 2009064868 A1 WO2009064868 A1 WO 2009064868A1 US 2008083383 W US2008083383 W US 2008083383W WO 2009064868 A1 WO2009064868 A1 WO 2009064868A1
Authority
WO
WIPO (PCT)
Prior art keywords
particles
bioaerosol
particle size
size distribution
aerosol
Prior art date
Application number
PCT/US2008/083383
Other languages
French (fr)
Inventor
Darrel Baumgardner
Original Assignee
Droplet Measurement Technololies
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Droplet Measurement Technololies filed Critical Droplet Measurement Technololies
Publication of WO2009064868A1 publication Critical patent/WO2009064868A1/en

Links

Classifications

    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/02—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving viable microorganisms
    • C12Q1/04—Determining presence or kind of microorganism; Use of selective media for testing antibiotics or bacteriocides; Compositions containing a chemical indicator therefor
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
    • G01N15/02—Investigating particle size or size distribution
    • G01N15/0205—Investigating particle size or size distribution by optical means
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
    • G01N15/06—Investigating concentration of particle suspensions
    • G01N15/065—Investigating concentration of particle suspensions using condensation nuclei counters
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
    • G01N21/47—Scattering, i.e. diffuse reflection
    • G01N21/49—Scattering, i.e. diffuse reflection within a body or fluid
    • G01N21/53—Scattering, i.e. diffuse reflection within a body or fluid within a flowing fluid, e.g. smoke
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/64—Fluorescence; Phosphorescence
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N1/00—Sampling; Preparing specimens for investigation
    • G01N1/28—Preparing specimens for investigation including physical details of (bio-)chemical methods covered elsewhere, e.g. G01N33/50, C12Q
    • G01N1/44—Sample treatment involving radiation, e.g. heat
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
    • G01N15/01—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials specially adapted for biological cells, e.g. blood cells
    • G01N2015/019—Biological contaminants; Fouling
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
    • G01N21/47—Scattering, i.e. diffuse reflection
    • G01N2021/4792—Polarisation of scatter light
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/64—Fluorescence; Phosphorescence
    • G01N2021/6417—Spectrofluorimetric devices

Definitions

  • the present document relates to the field of measurement of aerosol particles containing biological materials.
  • Bioaerosols are defined as airborne particles that are living, contain living organisms, or which were released from living organisms.
  • bioaerosols may contain bacteria, bacterial spores, viruses, fungi, and fungal spores, some of which may be infectious.
  • Bioaerosols may be produced in a large number of ways, ranging from a common sneeze and contaminated swamp cooler through aerosol sprays, agricultural crop-dusting, and detonation of weapons.
  • bioaerosols may contain infectious material, hospitals, medical researchers, and food processors may desire to measure and monitor them. Similarly, since bioaerosols containing infectious material are known to have been developed as biological weapons and are known to be of interest to terrorists, measurement and monitoring of bioaerosols can be of intense interest to military and security authorities.
  • Bioaerosol particles can have nonspherical shapes. While some bacteria, including streptococcus and staphylococcus, have generally spherical or coccoidal forms, these are often found in clumps or chains. It is known that other bacteria, including Escherichia coli, have generally rodlike forms. Other shapes are known, including spirochetes with spiral forms. Bacteria of a single species may have vegetative forms of one shape, and environmentally tolerant spore forms of another shape. Mold spores are also often nonspherical in shape.
  • a bioaerosol measurement system receives air having entrained particles of an aerosol. It analyzes these particles by drying these particles and illuminates them as they flow through an analysis chamber. Fluorescence spectra of the particles are determined. The particles are also illuminated with polarized light, intensity and polarization of backscattered light is determined. A particle size distribution of the particles is also determined. Measured fluorescence spectra, backscatter data, and particle sizes are used to classify the aerosol. In embodiments, hygroscopicity of the particles is also determined and used to help classify the aerosol. In embodiments, the classifying is done using a neural network previously trained with data representative of a variety of aerosols.
  • FIG. l is a diagram showing one embodiment of an exemplary system for classifying aerosol particles by biogenic fluorescent activity, light scattering properties and hygroscopicity.
  • FIG. 2 is a diagram showing an exemplary embodiment of the biogenic analysis chamber shown in FIG. 1.
  • FIG. 3 is a diagram showing an exemplary embodiment of the condensation nuclei mixing chamber and optical particle measurement device of FIG. 1.
  • FIG. 4 is a flowchart of the measurement and classification process.
  • FIG. 1 An exemplary system 100 for classifying bioaerosol particles by biogenic fluorescent activity and light scattering properties is illustrated in FIG. 1.
  • system 100 hereinafter, hereafter the Aerosol Quality Classification System (AQCS)
  • AQCS Aerosol Quality Classification System
  • system 100 includes air inlets 101, a drying chamber 102, including a heater/dryer 103, a biogenic analysis chamber 105 for detecting light scattering and particle fluorescence, a condensation nuclei mixing chamber 107, a device 114 for optically measuring particle size, and an ambient air inlet valve 109 coupled between mixing chamber 107 and optical particle measurement device or particle sizer 114.
  • AQCS Aerosol Quality Classification System
  • a process controller and signal conditioning unit 115 connected to a digital processor 120, is electronically coupled to each of the chambers/devices 102, 105, 107, 109, 114, and 120.
  • Processor 120 is connected to a user input/display device 119, and to an artificial neural network 130.
  • an aerosol is sampled 402 by air inlet 101.
  • Some of the aerosol particles in air flowing into the system from air inlet 101 are dried 404 by drying chamber 102 to provide a stream 104 of dried particles in air.
  • Particles of this stream 104 are analyzed by biogenic analysis chamber 105 and passed as a stream 106 of measured, dried, particles to condensation nuclei mixing chamber 107 to provide a stream 108 to valve 109.
  • Valve 109 selects unhumidified, humidified, or raw untreated particles to the optical particle sizer 114.
  • Biogenic analysis chamber 105 measures 406 optical scattering properties of the dried particles using polarized light as hereinafter described, and also measures 408 optical fluorescence properties in ultraviolet (UV) light of individual particles.
  • Particle sizes of the dried particles are measured 410 by particle sizer 114, and particles are humidified 412, 414 by exposure to a first and a second supersaturation in condensation nuclei mixing chamber 107, the sizes of the dried and re-humidified particles are measured 416, 418 by the optical particle sizer 114.
  • the optical particle sizer 114 can also measure 420 raw particles taken directly from the incoming air inlet 101.
  • processor 120 measurements of scattering of the polarized light, fluorescence spectra in ultraviolet light, dried particle size, raw particle size, and humidified particle sizes are input 422 into a trained neural network 130. Particle classifications are then read 424 from the neural network and output from the system; these classifications may also be used to trigger alarms if required.
  • dried, airborne, particles in airstream 104 entering biogenic analysis chamber 105 from drying chamber 102 are illuminated with polarized, monochromatic light 203 of wavelength between 600 and 800, and in an embodiment of 680 nanometers wavelength, from laser 202 and polarizer 202a and also with pulsed UV light 207.
  • the polarized monochromatic light 203 is used to induce scattered light 208 from the particle and to detect presence of a particle in the analysis zone of the analysis chamber 105.
  • the pulsed UV light 207 is used to induce fluorescence emitted light 209 from biogenic particles.
  • the pulsed UV light 207 is preferably of wavelength equal to or shorter than 350 nanometers to provide proper stimulus for biogenic fluorescence, and is emitted by a UV source 206 that may be a metal cathode lamp, a laser, or an UV photodiode.
  • a beam-splitter 205 is used in the polarized monochromatic light 203 illumination path to avoid obstructing, and thereby allow measurement of, backscattered light 208.
  • Light is scattered from the particles in airstream 104 and some fraction of the forward and backward components of scattered light 208 are collected by optical assemblies 230 and 210.
  • the forward scattered optical assembly 230 focuses the light onto a photodetector 232 where the photon intensity is converted to an electrical pulse.
  • the backward scattered optical assembly 210 focuses the light onto an optical configuration 221 that divides the backward scattered light through polarized filters with separate angles of polarization and then onto P-polarization detector 223 and S-polarization detector 224.
  • P-polarization detector 223 and S- polarization detector 224 convert these backward scattered and filtered photon intensities to electrical signals 223 and 224.
  • the electrical signals from detectors 232, 223 and 224 are digitized and sent via process controller 115 to processor 120 for analysis.
  • Fluorescence spectrometer 216 has a dispersive device such as a prism or a diffraction grating and a linear array of photosensors (not shown) for rapid determination of spectra of emissions from the particles. Information from spectrometer 216 is also digitized and sent via processor controller 115 to processor 120 for signal analysis. This fluorescence emitted spectra measurement is performed on individual particles.
  • a second optical system 210 also collects the backscattered light 208 where S and P polarized components of the backscattered light 208 are separated by polarizer 221 and focused on respective photodetectors 224 and 223, the digitized signals from which are sent via controller 115 to processor 120 for signal analysis.
  • FIG. 3 illustrates an exemplary embodiment of the condensation nuclei mixing chamber 107 and optical particle measurement device 114 of the system of FIG. 1.
  • the air approaching the mixing chamber 305 is divided into two conduits, one that goes directly to the mixing chamber 305, the other that goes through a filter 307.
  • Filter 307 removes aerosol particles from the air.
  • Filtered air passes to a temperature-controlled humidifier 309 that saturates the air with water vapor at a constant temperature. Saturated air leaving the humidifier 309 also enters the mixing chamber 305 where rapid cooling as the humid air mixes with cooler air 106 bearing bioaerosol particles gives air that is supersaturated.
  • the level of supersaturation is controlled by the difference in temperature between the aerosol laden air and the air leaving humidifier 309.
  • small supersaturations i.e., when the water vapor pressure exceeds the saturation vapor pressure by only a small amount
  • only hygroscopic particles will grow by condensation prior to size measurement in the optical particle counter.
  • a much higher supersaturation is imposed, then almost all of the particles will grow in size by condensation; in typical operation of the system 100 two supersaturations are used.
  • the particles In the second stage of the particle hygroscopicity determining process, the particles, mixed with water-saturated air, leave mixing chamber 107 through an expansion chamber 305 where they and grow as water vapor condenses on their surface.
  • the particles then pass through valve 109 enter an aerodynamic nozzle 315 of optical particle sizer 114.
  • Sizer 114 determines the optical diameter of individual aerosol particles by measuring light that is scattered as the particle passes through a focused laser beam 321. In the detector 320, this scattered light 322 is collected by the optical components and focused on a photodetector 323. Photodetector 323 's output is digitized and sent, via process controller 115, to processor 120 where the output information is converted to particle diameter data.
  • Information sent to processor 120 from particle size measuring device 114 is integrated with the signals from the biogenic analysis chamber and processed to produce a detailed description of the size distribution, biogenic activity, hygroscopicity, light scattering coefficient and fluorescence of the aerosol population. All of this information is entered into the neural network 130. The patterns generated by the combination of all the sensors are evaluated by neural network 130 and alerts issued as necessary based on the nature of the evaluated patterns.
  • Neural networks are named after the cells in the human brain that perform intelligent operations.
  • the brain is made up of billions of neuron cells. Each of these cells is like a tiny computer with extremely limited capabilities; however, connected together, these cells form the most intelligent system known.
  • Neural networks are formed from simulated neurons connected together in much the same way as the brain's neurons.
  • neural network In a neural network model, simple nodes (called variously “neurons”, “neurodes”, 'processing elements' or 'units') are connected together to form a network of nodes, hence the term "neural network”. While a neural network does not have to be adaptive per se, its practical use is improved by the use of algorithms designed to alter the strength (weights) of individual connections within the network to produce a desired signal flow during a training phase.
  • neural networks In modern software implementations of artificial neural networks the approach inspired by biology has more or less been abandoned for a more practical approach based on statistics and signal processing. In some of these systems neural networks, or parts of neural networks (such as artificial neurons) are used as components in larger systems that combine both adaptive and non-adaptive elements.
  • a neural network 130 may be trained to recognize, classify and characterize aerosols of unknown origin with a reasonable degree of accuracy for applications directed to rapid real-time processing of aerosol properties.
  • a neural network may be applied to the recognition and classification of environmental, bacterial, fungal and industrial aerosols on the basis of their optical, hygroscopic, morphological and fluorescent properties.
  • input data to neural network 130 is constructed from the particle size distributions (PSD) obtained from different aerosol populations, as measured with the two single particle optical scattering detectors described above, the forward and backward scattering detectors in the Biogenic Analysis Chamber 105, and the optical scattering detector 114 at the exit of the condensation nuclei mixing chamber 107.
  • PSD particle size distributions
  • the input data to neural network 130 includes optical scattering detector 114 measurements of dry particles as well as particles that have been exposed to saturated air in condensation nuclei mixing chamber 107.
  • input data to neural network 130 is also constructed from particle fluorescence spectra as obtained from the fluorescence spectrometer 216.
  • the data acquisition system processes the raw signals by filtering noise, applying calibration coefficients and deriving various properties of the aerosol using a neural network that may be trained theoretically and by exposure of a prototype system to reference particles with known properties.
  • the network consists of at least three layers of neurons, input, hidden and output. Depending on the complexity of the training set and the number of input and output neurons, there can be more than a single hidden layer.
  • the output of each input neuron goes to every hidden neuron of at least one layer and the output of each hidden layer neuron of at least one layer goes to every output neuron.
  • the output of each hidden layer neuron depends on the sum of all the inputs, where each input has a weight that either reinforces or suppresses the response of the hidden neuron.
  • weights are determined during training using back propagation and a large data set.
  • Back propagation is a supervised learning method in which inputs are repeatedly provided to the network, an output is determined, and a difference of the output and a desired output is determined. An error signal is then fed through the network, altering the weights to minimize the error.
  • the network is trained by presenting with input and output pairs.
  • the inputs are from the outputs of the different sensors, the temperature at the inlet and the state of the condensation nuclei counter.
  • production aerosol surveillance and detection systems may receive weights from a trained neural network 130 but have back propagation disabled to prevent unexpected alteration of these weights.
  • the training is done by simulating the AQCS system with a computer model that generates size distributions of particles with varying composition, refractive indices and shapes and calculates what the AQCS would measure, based upon the known physical processes.
  • the network may be trained using data measured from a prototype system. The power of a neural network is that it does not depend upon exact, analytical relationships and in fact, the network will not be trained over all possible combinations but enough so that it can make correct inferences.
  • one training pattern would be a mixture of ammonium sulfate and some ideal PAH whose refractive index and vapor pressure is known so that we can calculate at what temperature only the ammonium sulfate fraction would remain.
  • Additional training of the neural network may incorporate comparison of the AQCS with other instruments that measure the specific chemical components of aerosol particles.
  • the AQCS will take measurements in ambient air that is also sampled with a commercial aerosol mass spectrometer (AMS).
  • AMS aerosol mass spectrometer
  • the measurements from the AMS will provide expected output data for the neural network of the AQCS, and back propagation is used to tune the neural network response by adjusting weights appropriately.
  • the neural network is trained to distinguish between an assortment of aerosols representing common organic and inorganic dusts and pollens, as well as at least one aerosol containing bacilli such that the system provides an alarm when exposed to anthrax bacilli and/or Legionella bacteria.

Landscapes

  • Chemical & Material Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Immunology (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Pathology (AREA)
  • Organic Chemistry (AREA)
  • Zoology (AREA)
  • Dispersion Chemistry (AREA)
  • Wood Science & Technology (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Engineering & Computer Science (AREA)
  • Toxicology (AREA)
  • Microbiology (AREA)
  • Molecular Biology (AREA)
  • Biotechnology (AREA)
  • Biophysics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • General Engineering & Computer Science (AREA)
  • Genetics & Genomics (AREA)
  • Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)

Abstract

A bioaerosol measurement system receives air having entrained particles of an aerosol. It analyzes these particles and drying these particles and illuminates them. Fluorescence spectra of the particles are determined. The particles are also illuminated with polarized light, amount and polarization of backscattered light is determined. A particle size distribution of the particles is also determined. Measured fluorescence spectra, backscatter data, and particle sizes are used to classify the aerosol. In embodiments, hygroscopicity of the particles is also determined and used to help classify the aerosol. In embodiments, the classifying is done by using a neural network previously trained with data representative of a variety of aerosols.

Description

BIOAEROSOL MEASUREMENT SYSTEM
RELATED APPLICATIONS
[0001] The present document claims benefit of priority to commonly- owned and copending U.S. Provisional Patent Application No. 60/988,013 filed 14 November 2007, which is incorporated herein by reference.
BACKGROUND Field
[0002] The present document relates to the field of measurement of aerosol particles containing biological materials.
[0003] Bioaerosols are defined as airborne particles that are living, contain living organisms, or which were released from living organisms. In particular, bioaerosols may contain bacteria, bacterial spores, viruses, fungi, and fungal spores, some of which may be infectious. Bioaerosols may be produced in a large number of ways, ranging from a common sneeze and contaminated swamp cooler through aerosol sprays, agricultural crop-dusting, and detonation of weapons.
[0004] Since bioaerosols may contain infectious material, hospitals, medical researchers, and food processors may desire to measure and monitor them. Similarly, since bioaerosols containing infectious material are known to have been developed as biological weapons and are known to be of interest to terrorists, measurement and monitoring of bioaerosols can be of intense interest to military and security authorities.
[0005] When measuring and monitoring bioaerosols, particle size, count, shape, solubility (hygroscopicity) and fluorescent properties can all be of interest. U.S. Patent No. 6,532,067 to Chang, et al., describes a monitoring system capable of determining a fluorescent emission spectra of individual particles of a bioaerosol when subjected to 266 nanometer radiation while measuring particle size; however this system can be confounded by particles containing substantial non-biogenic, volatile organic carbon materials, does not measure hygroscopicity, and is not able to distinguish particle shape. Jian Wang, et al., 'Fast Mixing Condensation Nucleus Counter: Application to Rapid Scanning Differential Mobility Analyzer Measurements', Aerosol Science and Technology, 36:6, 678 - 689 describes a Mixing Condensation Nucleus Counter (MCNC) for counting small particles after growth in supersaturated air formed by rapidly mixing heated, humidified air with lower temperature air containing an aerosol, but does not describe measurement of hygroscopicity and determining hygroscopic and nonhygroscopic particle subsets of the total aerosol population.
[0006] Bioaerosol particles can have nonspherical shapes. While some bacteria, including streptococcus and staphylococcus, have generally spherical or coccoidal forms, these are often found in clumps or chains. It is known that other bacteria, including Escherichia coli, have generally rodlike forms. Other shapes are known, including spirochetes with spiral forms. Bacteria of a single species may have vegetative forms of one shape, and environmentally tolerant spore forms of another shape. Mold spores are also often nonspherical in shape.
Summary
[0007] A bioaerosol measurement system receives air having entrained particles of an aerosol. It analyzes these particles by drying these particles and illuminates them as they flow through an analysis chamber. Fluorescence spectra of the particles are determined. The particles are also illuminated with polarized light, intensity and polarization of backscattered light is determined. A particle size distribution of the particles is also determined. Measured fluorescence spectra, backscatter data, and particle sizes are used to classify the aerosol. In embodiments, hygroscopicity of the particles is also determined and used to help classify the aerosol. In embodiments, the classifying is done using a neural network previously trained with data representative of a variety of aerosols.
BRIEF DESCRIPTION OF THE DRAWINGS
[0008] FIG. l is a diagram showing one embodiment of an exemplary system for classifying aerosol particles by biogenic fluorescent activity, light scattering properties and hygroscopicity.
[0009] FIG. 2 is a diagram showing an exemplary embodiment of the biogenic analysis chamber shown in FIG. 1. [0010] FIG. 3 is a diagram showing an exemplary embodiment of the condensation nuclei mixing chamber and optical particle measurement device of FIG. 1.
[0011] FIG. 4 is a flowchart of the measurement and classification process.
DETAILED DESCRIPTION OF THE EMBODIMENTS
[0012] An exemplary system 100 for classifying bioaerosol particles by biogenic fluorescent activity and light scattering properties is illustrated in FIG. 1. As shown in FIG. 1 , system 100, hereinafter, hereafter the Aerosol Quality Classification System (AQCS), includes air inlets 101, a drying chamber 102, including a heater/dryer 103, a biogenic analysis chamber 105 for detecting light scattering and particle fluorescence, a condensation nuclei mixing chamber 107, a device 114 for optically measuring particle size, and an ambient air inlet valve 109 coupled between mixing chamber 107 and optical particle measurement device or particle sizer 114. A process controller and signal conditioning unit 115, connected to a digital processor 120, is electronically coupled to each of the chambers/devices 102, 105, 107, 109, 114, and 120. Processor 120 is connected to a user input/display device 119, and to an artificial neural network 130.
[0013] With reference to FIG. 4 as well as FIG. 1 ; in operation an aerosol is sampled 402 by air inlet 101. Some of the aerosol particles in air flowing into the system from air inlet 101 are dried 404 by drying chamber 102 to provide a stream 104 of dried particles in air. Particles of this stream 104 are analyzed by biogenic analysis chamber 105 and passed as a stream 106 of measured, dried, particles to condensation nuclei mixing chamber 107 to provide a stream 108 to valve 109. Valve 109 selects unhumidified, humidified, or raw untreated particles to the optical particle sizer 114.
[0014] Biogenic analysis chamber 105 measures 406 optical scattering properties of the dried particles using polarized light as hereinafter described, and also measures 408 optical fluorescence properties in ultraviolet (UV) light of individual particles. Particle sizes of the dried particles are measured 410 by particle sizer 114, and particles are humidified 412, 414 by exposure to a first and a second supersaturation in condensation nuclei mixing chamber 107, the sizes of the dried and re-humidified particles are measured 416, 418 by the optical particle sizer 114. The optical particle sizer 114 can also measure 420 raw particles taken directly from the incoming air inlet 101.
[0015] Within processor 120, measurements of scattering of the polarized light, fluorescence spectra in ultraviolet light, dried particle size, raw particle size, and humidified particle sizes are input 422 into a trained neural network 130. Particle classifications are then read 424 from the neural network and output from the system; these classifications may also be used to trigger alarms if required.
[0016] As shown in FIG. 2, dried, airborne, particles in airstream 104 entering biogenic analysis chamber 105 from drying chamber 102, are illuminated with polarized, monochromatic light 203 of wavelength between 600 and 800, and in an embodiment of 680 nanometers wavelength, from laser 202 and polarizer 202a and also with pulsed UV light 207. The polarized monochromatic light 203 is used to induce scattered light 208 from the particle and to detect presence of a particle in the analysis zone of the analysis chamber 105. The pulsed UV light 207 is used to induce fluorescence emitted light 209 from biogenic particles. The pulsed UV light 207 is preferably of wavelength equal to or shorter than 350 nanometers to provide proper stimulus for biogenic fluorescence, and is emitted by a UV source 206 that may be a metal cathode lamp, a laser, or an UV photodiode. A beam-splitter 205 is used in the polarized monochromatic light 203 illumination path to avoid obstructing, and thereby allow measurement of, backscattered light 208.
[0017] Light is scattered from the particles in airstream 104 and some fraction of the forward and backward components of scattered light 208 are collected by optical assemblies 230 and 210. The forward scattered optical assembly 230 focuses the light onto a photodetector 232 where the photon intensity is converted to an electrical pulse. The backward scattered optical assembly 210 focuses the light onto an optical configuration 221 that divides the backward scattered light through polarized filters with separate angles of polarization and then onto P-polarization detector 223 and S-polarization detector 224. P-polarization detector 223 and S- polarization detector 224 convert these backward scattered and filtered photon intensities to electrical signals 223 and 224. The electrical signals from detectors 232, 223 and 224 are digitized and sent via process controller 115 to processor 120 for analysis.
[0018] The light 209 emitted by fluorescence, as stimulated by UV light 207, is collected with an optical collection/detection system 212 and transmitted 215 to fluorescence spectrometer 216. Fluorescence spectrometer 216 has a dispersive device such as a prism or a diffraction grating and a linear array of photosensors (not shown) for rapid determination of spectra of emissions from the particles. Information from spectrometer 216 is also digitized and sent via processor controller 115 to processor 120 for signal analysis. This fluorescence emitted spectra measurement is performed on individual particles.
[0019] A second optical system 210 also collects the backscattered light 208 where S and P polarized components of the backscattered light 208 are separated by polarizer 221 and focused on respective photodetectors 224 and 223, the digitized signals from which are sent via controller 115 to processor 120 for signal analysis.
[0020] The particles leave the biogenic analysis chamber 105 as a stream of measured, dried, particles 106 and enter condensation nuclei mixing chamber 107 where the hygroscopicity of the particles is determined. FIG. 3 illustrates an exemplary embodiment of the condensation nuclei mixing chamber 107 and optical particle measurement device 114 of the system of FIG. 1. As shown in FIG. 3, the air approaching the mixing chamber 305 is divided into two conduits, one that goes directly to the mixing chamber 305, the other that goes through a filter 307. Filter 307 removes aerosol particles from the air. Filtered air passes to a temperature-controlled humidifier 309 that saturates the air with water vapor at a constant temperature. Saturated air leaving the humidifier 309 also enters the mixing chamber 305 where rapid cooling as the humid air mixes with cooler air 106 bearing bioaerosol particles gives air that is supersaturated.
[0021] The level of supersaturation is controlled by the difference in temperature between the aerosol laden air and the air leaving humidifier 309. When small supersaturations occur (i.e., when the water vapor pressure exceeds the saturation vapor pressure by only a small amount), only hygroscopic particles will grow by condensation prior to size measurement in the optical particle counter. When a much higher supersaturation is imposed, then almost all of the particles will grow in size by condensation; in typical operation of the system 100 two supersaturations are used.
[0022] In the second stage of the particle hygroscopicity determining process, the particles, mixed with water-saturated air, leave mixing chamber 107 through an expansion chamber 305 where they and grow as water vapor condenses on their surface. The particles then pass through valve 109 enter an aerodynamic nozzle 315 of optical particle sizer 114. Sizer 114 determines the optical diameter of individual aerosol particles by measuring light that is scattered as the particle passes through a focused laser beam 321. In the detector 320, this scattered light 322 is collected by the optical components and focused on a photodetector 323. Photodetector 323 's output is digitized and sent, via process controller 115, to processor 120 where the output information is converted to particle diameter data. [0023] Information sent to processor 120 from particle size measuring device 114 is integrated with the signals from the biogenic analysis chamber and processed to produce a detailed description of the size distribution, biogenic activity, hygroscopicity, light scattering coefficient and fluorescence of the aerosol population. All of this information is entered into the neural network 130. The patterns generated by the combination of all the sensors are evaluated by neural network 130 and alerts issued as necessary based on the nature of the evaluated patterns.
Using a Neural Network to Measure Air Quality
[0024] Neural networks are named after the cells in the human brain that perform intelligent operations. The brain is made up of billions of neuron cells. Each of these cells is like a tiny computer with extremely limited capabilities; however, connected together, these cells form the most intelligent system known. Neural networks are formed from simulated neurons connected together in much the same way as the brain's neurons.
[0025] In a neural network model, simple nodes (called variously "neurons", "neurodes", 'processing elements' or 'units') are connected together to form a network of nodes, hence the term "neural network". While a neural network does not have to be adaptive per se, its practical use is improved by the use of algorithms designed to alter the strength (weights) of individual connections within the network to produce a desired signal flow during a training phase.
[0026] In modern software implementations of artificial neural networks the approach inspired by biology has more or less been abandoned for a more practical approach based on statistics and signal processing. In some of these systems neural networks, or parts of neural networks (such as artificial neurons) are used as components in larger systems that combine both adaptive and non-adaptive elements.
[0027] A neural network 130 may be trained to recognize, classify and characterize aerosols of unknown origin with a reasonable degree of accuracy for applications directed to rapid real-time processing of aerosol properties. A neural network may be applied to the recognition and classification of environmental, bacterial, fungal and industrial aerosols on the basis of their optical, hygroscopic, morphological and fluorescent properties.
[0028] In the present system, input data to neural network 130 is constructed from the particle size distributions (PSD) obtained from different aerosol populations, as measured with the two single particle optical scattering detectors described above, the forward and backward scattering detectors in the Biogenic Analysis Chamber 105, and the optical scattering detector 114 at the exit of the condensation nuclei mixing chamber 107. The input data to neural network 130 includes optical scattering detector 114 measurements of dry particles as well as particles that have been exposed to saturated air in condensation nuclei mixing chamber 107.
[0029] In the present system, input data to neural network 130 is also constructed from particle fluorescence spectra as obtained from the fluorescence spectrometer 216.
[0030] The data acquisition system processes the raw signals by filtering noise, applying calibration coefficients and deriving various properties of the aerosol using a neural network that may be trained theoretically and by exposure of a prototype system to reference particles with known properties. The network consists of at least three layers of neurons, input, hidden and output. Depending on the complexity of the training set and the number of input and output neurons, there can be more than a single hidden layer. The output of each input neuron goes to every hidden neuron of at least one layer and the output of each hidden layer neuron of at least one layer goes to every output neuron. The output of each hidden layer neuron depends on the sum of all the inputs, where each input has a weight that either reinforces or suppresses the response of the hidden neuron.
[0031] These weights are determined during training using back propagation and a large data set. Back propagation is a supervised learning method in which inputs are repeatedly provided to the network, an output is determined, and a difference of the output and a desired output is determined. An error signal is then fed through the network, altering the weights to minimize the error. The network is trained by presenting with input and output pairs. In the AQCS the inputs are from the outputs of the different sensors, the temperature at the inlet and the state of the condensation nuclei counter. In addition, because we are looking at how the properties change as the temperature and CN supers aturation changes, we also have input neurons that receive the sensor outputs at a previous time when the inlet temperature or supersaturation was different.
[0032] Once training is complete, production aerosol surveillance and detection systems may receive weights from a trained neural network 130 but have back propagation disabled to prevent unexpected alteration of these weights.
[0033] The training is done by simulating the AQCS system with a computer model that generates size distributions of particles with varying composition, refractive indices and shapes and calculates what the AQCS would measure, based upon the known physical processes. In addition, the network may be trained using data measured from a prototype system. The power of a neural network is that it does not depend upon exact, analytical relationships and in fact, the network will not be trained over all possible combinations but enough so that it can make correct inferences. For example, one training pattern would be a mixture of ammonium sulfate and some ideal PAH whose refractive index and vapor pressure is known so that we can calculate at what temperature only the ammonium sulfate fraction would remain. Likewise, we can have ideal mixture of ammonium sulfate and a hydrophobic organic material such that only the ammonium sulfate will be found hygroscopic. There are many publications on the volatility and hygroscopicity of ambient aerosols, accompanied by composition information. The results from these observations will be used in the simulations to train the network.
[0034] Additional training of the neural network may incorporate comparison of the AQCS with other instruments that measure the specific chemical components of aerosol particles. For example, the AQCS will take measurements in ambient air that is also sampled with a commercial aerosol mass spectrometer (AMS). The measurements from the AMS will provide expected output data for the neural network of the AQCS, and back propagation is used to tune the neural network response by adjusting weights appropriately.
[0035] In a particular embodiment, the neural network is trained to distinguish between an assortment of aerosols representing common organic and inorganic dusts and pollens, as well as at least one aerosol containing bacilli such that the system provides an alarm when exposed to anthrax bacilli and/or Legionella bacteria. [0036] While the forgoing has been particularly shown and described with reference to particular embodiments thereof, it will be understood by those skilled in the art that various other changes in the form and details may be made without departing from the spirit and hereof. It is to be understood that various changes may be made in adapting the description to different embodiments without departing from the broader concepts disclosed herein and comprehended by the claims that follow.

Claims

CLAIMSWhat is claimed is:
1. A bioaerosol measurement system, comprising means for drying particles contained in an aerosol; means for passing particles of the aerosol through an analysis chamber; means, located in the analysis chamber, for illuminating particles in the aerosol to cause fluorescence of particles of biogenic nature among the entrained particles, and for detecting the fluorescence to produce fluorescence information representative of the particles; means for determining a signal representative of a particle size distribution in the particles; means, operating in combination upon the fluorescence information and the signal representative of the particle size distribution, for determining one or more specific values for parameters selected from the group consisting of particle size distribution, biogenic activity, hygroscopicity, light scattering coefficient and chemical composition of the particles; and means for determining hygroscopicity of the particles.
2. The aerosol measurement system of claim 1, wherein the means for drying the entrained particles includes a heater.
3. The bioaerosol measurement system of claim 1 , wherein the means located on the optical pathway for illuminating the entrained particles includes a source of light of less than 350 nanometers wavelength, and a source of polarized light; and further comprising means for detecting and determining polarization of backscattered light.
4. The bioaerosol measurement system of claim 1, wherein the means located on the optical pathway for illuminating the particles includes a pulsed, polarized laser.
5. The bioaerosol measurement system of claim 1 , wherein the means located in the analysis chamber for determining particle size distribution includes means for collecting backscattered light from illuminating the particles with the pulsed, polarized laser, and means for analysis of respective S and P polarized components of the backscattered light for delivery to corresponding photodetectors to produce signals representative of the respective S and P components.
6. The bioaerosol measurement system of claim 1, wherein the means for determining particle size distribution includes: means located on the optical pathway for contacting the particles with water- saturated air and an expansion chamber that is configured for rapid cooling and growth of the particles as water vapor condenses on the particle surfaces; and an aerodynamic nozzle configured to direct the particles through a focused laser beam to produce scattered light components; and means for detecting the scattered light components to produce the signal representative of the particle size distribution of the particles.
7. The bioaerosol measurement system of claim 1, wherein the means operating in combination upon the fluorescence information and the signal representative of the particle size distribution includes a processor configured by program instructions to perform a neural network analysis indicating at least one of the particle size distribution, biogenic activity, hygroscopicity, light scattering coefficient and chemical composition of the particles
8. A bioaerosol measurement system, comprising apparatus for receiving, and drying particles contained in, an aerosol; an analysis chamber for receiving dried particles, the analysis chamber having apparatus for illuminating particles in the aerosol to cause fluorescence of particles of biogenic nature among the dried particles, apparatus for detecting and determining spectra of the fluorescence to produce fluorescence spectra information representative of the particles, apparatus for illuminating particles in the aerosol with polarized light, and apparatus for measuring polarization of backscattered light to give backscatter polarization information; apparatus for determining a signal representative of a particle size distribution in the particles; and apparatus for processing the fluorescence spectra information, backscatter polarization information, and particle size distribution to provide particle classification information.
9. The bioaerosol measurement system of claim 8 wherein the apparatus for illuminating particles in the aerosol comprises a source of light having wavelength less than 350 nanometers and the apparatus for illuminating particles in the aerosol with polarized light has wavelength between 600 and 800 nanometers.
10. The bioaerosol measurement system of claim 8 further comprising apparatus for exposing dried particles to humidified air to produce humidified particles, and apparatus for determining a signal representative of a particle size distribution in the humidified particles.
11. The bioaerosol measurement system of claim 8 wherein the apparatus for processing the fluorescence spectra information, backscatter polarization information, and particle size distribution to provide particle classification information comprises a neural network.
12. The bioaerosol measurement system of claim 11 wherein the neural network is trained to provide identification of specific aerosol properties and to provide an alarm on exposure of the system to particles comprising at least one bacillus.
13. A method of monitoring a stream of air for a bioaerosol comprising: passing a portion of the air through a dryer to provide dry air and to dry any bioaerosol particles entrained in the dry air; passing the dry air through an analysis chamber wherein any entrained bioaerosol particles in the dry air are illuminated with light to cause fluorescence; measuring the fluorescence; determining a signal representative of a particle size distribution in the entrained particles in the dry air; exposing at least a portion of the entrained particles in the dry air to humidified air to produce re-humidified particles; determining a particle size distribution of the re-humidified particles; and analyzing the fluorescence, the particle size distribution in the entrained particles in the dry air, and the particle size distribution of the re-humidified particles to determine presence and properties of the bioaerosol.
14. The method of claim 13 further comprising: illuminating the entrained particles in the analysis chamber with polarized light; and measuring intensity and polarization of backscattered light from the entrained particle illuminated with polarized light; and wherein the step of analyzing includes analyzing the intensity and polarization of backscattered light in determining the presence and properties of the bioaerosol.
15. The method of claim 14 wherein the light particles are illuminated with to cause fluorescence has wavelength less than 350 nanometers.
16. The method of claim 15 wherein the polarized light has wavelength between 600 and 800 nanometers.
17. The method of claim 15 wherein the step of analyzing is performed by apparatus comprising a neural network.
PCT/US2008/083383 2007-11-14 2008-11-13 Bioaerosol measuremeny system WO2009064868A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US98801307P 2007-11-14 2007-11-14
US60/988,013 2007-11-14

Publications (1)

Publication Number Publication Date
WO2009064868A1 true WO2009064868A1 (en) 2009-05-22

Family

ID=40639119

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2008/083383 WO2009064868A1 (en) 2007-11-14 2008-11-13 Bioaerosol measuremeny system

Country Status (1)

Country Link
WO (1) WO2009064868A1 (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102297824A (en) * 2011-05-18 2011-12-28 北京汇丰隆经济技术开发有限公司 Single light source bio-aerosol particle detecting apparatus
WO2012076474A1 (en) * 2010-12-05 2012-06-14 Markus Klotz Method and device for determining the water content of non-water-based liquids
WO2015088174A1 (en) * 2013-12-13 2015-06-18 코웨이 주식회사 Bio sensor and air purifier having same
EP2758767A4 (en) * 2011-09-20 2015-07-15 Tsi Inc Apparatus and system for characterizing an aerosol particle flow
US9803877B2 (en) 2013-12-13 2017-10-31 Coway Co., Ltd. Bio sensor and air cleaner having same
CN108287129A (en) * 2018-03-22 2018-07-17 中国计量大学 The detection device of multichannel fluorescence Spectra bioaerosol particle
EP3605059A1 (en) * 2018-07-31 2020-02-05 Koninklijke Philips N.V. Particle sensor and sensing method
CN111337392A (en) * 2020-02-28 2020-06-26 清华大学深圳国际研究生院 Suspended particle polarized fluorescence synchronous measurement device
WO2022114733A1 (en) * 2020-11-27 2022-06-02 삼성전자 주식회사 Method of determining particle size distribution and electronic device

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
PETER P. HAIRSTON ET AL.: "Design of an instrument for real-time detection of bioaerosols using simultaneous measurement of particle aerodynamic size and intrinsic fluorescence", J. AEROSOL SCI., vol. 28, no. 3, 1997, pages 471 - 482 *
STEVEN C. HILL ET AL.: "Real-Time Measurement of Fluorescence Spectra from Single Airborne Biological Particles", FIELD ANALYTICAL CHEMISTRY AND TECHNOLOGY, vol. 3, 1999, pages 221 - 239 *
T. A. REPONEN ET AL.: "Characteristics of Airborne Actinomycete Spores", APPLIED AND ENVIRONMENTAL MICROBIOLOGY, vol. 64, October 1998 (1998-10-01), pages 3807 - 3812 *
TUNA REPONEN ET AL.: "Effect of relative humidity on the aerodynamic diameter and respiratory deposition of fungal spores", ATMOSPHERIC ENVIRONMENT, vol. 30, no. 23, 1996, pages 3967 - 3974 *

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012076474A1 (en) * 2010-12-05 2012-06-14 Markus Klotz Method and device for determining the water content of non-water-based liquids
CN102297824A (en) * 2011-05-18 2011-12-28 北京汇丰隆经济技术开发有限公司 Single light source bio-aerosol particle detecting apparatus
EP2758767A4 (en) * 2011-09-20 2015-07-15 Tsi Inc Apparatus and system for characterizing an aerosol particle flow
US9880097B2 (en) 2011-09-20 2018-01-30 Tsi Incorporated Apparatus and system for simultaneously measuring particle concentration and biocontaminants in an aerosol particle flow
WO2015088174A1 (en) * 2013-12-13 2015-06-18 코웨이 주식회사 Bio sensor and air purifier having same
US9803877B2 (en) 2013-12-13 2017-10-31 Coway Co., Ltd. Bio sensor and air cleaner having same
CN108287129A (en) * 2018-03-22 2018-07-17 中国计量大学 The detection device of multichannel fluorescence Spectra bioaerosol particle
EP3605059A1 (en) * 2018-07-31 2020-02-05 Koninklijke Philips N.V. Particle sensor and sensing method
CN111337392A (en) * 2020-02-28 2020-06-26 清华大学深圳国际研究生院 Suspended particle polarized fluorescence synchronous measurement device
WO2022114733A1 (en) * 2020-11-27 2022-06-02 삼성전자 주식회사 Method of determining particle size distribution and electronic device

Similar Documents

Publication Publication Date Title
WO2009064868A1 (en) Bioaerosol measuremeny system
US9851291B2 (en) Realtime optical method and system for detecting and classifying biological and non-biological particles
Wu et al. Label-free bioaerosol sensing using mobile microscopy and deep learning
EP2758767B1 (en) Systems for characterizing an aerosol particle flow
KR101418295B1 (en) Pathogen detection by simultaneous size/fluorescence measurement
Hairston et al. Design of an instrument for real-time detection of bioaerosols using simultaneous measurement of particle aerodynamic size and intrinsic fluorescence
Cochrane Using vegetation reflectance variability for species level classification of hyperspectral data
Kawashima et al. Automated pollen monitoring system using laser optics for observing seasonal changes in the concentration of total airborne pollen
JP4598766B2 (en) Multispectral optical method and system for detecting and classifying biological and non-biological microparticles
CN101479592B (en) Pathogen detection by simultaneous size/fluorescence measurement
US20090027674A1 (en) Aerosol trigger device and methods of detecting particulates of interest using an aerosol trigger device
KR20110005677A (en) Pathogen detection by simultaneous size/fluorescence measurement
JP2004037370A (en) Method and apparatus for quickly detecting and identifying microscopic fine particle for quantitative and qualitative measurement
Njalsson et al. Design and optimization of a compact low-cost optical particle sizer
Dick et al. Optical shape fraction measurements of submicrometre laboratory and atmospheric aerosols
WO2014201264A1 (en) Methods for detecting parasites, viruses, bacteria and drugs in human and animal blood and cerebral spinal fluid, using laser-induced breakdown spectroscopy
Matsuda et al. Relationship between laser light scattering and physical properties of airborne pollen
CN112903543B (en) Light scattering-based aerosol particle ellipticity measurement method and system
CN110987736B (en) Aerosol particle spectrum and concentration measuring device and method
Lu et al. Accurate measurement of airborne biological particle concentration based on laser-induced fluorescence technique
WO2018115052A1 (en) Particle sensor and sensing method
Lattanzio Particulate Matter Sensing for Air Quality Measurements
Arnold et al. Measurements of elastic light-scattering patterns and images of single, oriented, optically trapped particles
Zhu et al. Development of a bioaerosol detection system based on 280-nm LED for classifying aerosol types
Sivaprakasam et al. Wavelength resolved polarized elastic scatter measurements from micron-sized single particles

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 08849423

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 08849423

Country of ref document: EP

Kind code of ref document: A1