U.S. patent application number 15/515499 was filed with the patent office on 2017-08-03 for organism identification.
The applicant listed for this patent is Purdue Research Foundation. Invention is credited to Euiwon Bae, Arun K. Bhunia, Edwin Daniel Hirleman, Huisung Kim, Valery Patsekin, Bartlomiej P. Rajwa, Joseph Paul Robinson.
Application Number | 20170219485 15/515499 |
Document ID | / |
Family ID | 55631762 |
Filed Date | 2017-08-03 |
United States Patent
Application |
20170219485 |
Kind Code |
A1 |
Bae; Euiwon ; et
al. |
August 3, 2017 |
Organism Identification
Abstract
A system for the identification of micro-organisms includes an
irradiation unit adapted to sequentially provide coherent
electromagnetic radiation of one or more wavelengths along a common
optical path. A holder is adapted to retain a substrate having a
surface adapted for growth of a micro-organism colony. A
beamsplitter is adapted to direct the coherent electromagnetic
radiation from the common optical path towards the retained
substrate. An imager is arranged opposite the beamsplitter from the
retained substrate and is adapted to obtain images of
backward-scattered light patterns from the micro-organism colony
irradiated by the respective wavelengths of the directed coherent
electromagnetic radiation. Some examples provide radiation of
multiple wavelengths and include an imager arranged optically
downstream of the retained substrate to obtain images of
forward-scattered light patterns from the micro-organism colony
irradiated by the wavelengths of radiation. Organism identification
methods are also described.
Inventors: |
Bae; Euiwon; (West
Lafayette, IN) ; Bhunia; Arun K.; (West Lafayette,
IN) ; Hirleman; Edwin Daniel; (Merced, CA) ;
Kim; Huisung; (West Lafayette, IN) ; Rajwa;
Bartlomiej P.; (West Lafayette, IN) ; Robinson;
Joseph Paul; (West Lafayette, IN) ; Patsekin;
Valery; (West Lafayette, IN) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Purdue Research Foundation |
West Lafayette |
IN |
US |
|
|
Family ID: |
55631762 |
Appl. No.: |
15/515499 |
Filed: |
October 1, 2015 |
PCT Filed: |
October 1, 2015 |
PCT NO: |
PCT/US2015/053553 |
371 Date: |
March 29, 2017 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
62058478 |
Oct 1, 2014 |
|
|
|
62058734 |
Oct 2, 2014 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G01N 2021/4709 20130101;
G01N 21/255 20130101; G01N 2021/4707 20130101; G01N 21/51 20130101;
G01N 21/47 20130101 |
International
Class: |
G01N 21/47 20060101
G01N021/47 |
Goverment Interests
STATEMENT OF FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
[0002] This invention was made with Government support under
Contract No. 59-1935-2-279 awarded by the United States Department
of Agriculture--Agricultural Research Service. The government has
certain rights in the invention.
Claims
1-15. (canceled)
16. A system for the identification of micro-organisms, the system
comprising: an irradiation unit adapted to sequentially provide
coherent electromagnetic radiation of multiple wavelengths along a
common optical path; a holder adapted to retain a substrate having
a surface adapted for growth of a micro-organism colony; a
beamsplitter adapted to direct the coherent electromagnetic
radiation from the common optical path towards the retained
substrate; and an imager arranged opposite the beamsplitter from
the retained substrate and adapted to obtain images of
backward-scattered light patterns from the micro-organism colony
irradiated by the respective wavelengths of the directed coherent
electromagnetic radiation.
17. The system according to claim 16, further comprising a stage
adapted to translate the retained substrate or the beamsplitter
with respect to each other so that the directed coherent
electromagnetic radiation irradiates the micro-organism colony.
18. The system according to claim 17, further comprising a
controller configured to: operate the stage and the irradiation
unit to irradiate a first colony of a plurality of micro-organism
colonies on the retained substrate; operate the imager to obtain a
first image and a second image of backward-scattered light patterns
from the first colony, the first image corresponding to a first
wavelength and the second image corresponding to a second,
different wavelength; subsequently, operate the stage and the
irradiation unit to irradiate a second colony of the plurality of
micro-organism colonies on the retained substrate; and operate the
imager to obtain a third image and a fourth image of
backward-scattered light patterns from the second colony, the third
image corresponding to a third wavelength and the fourth image
corresponding to a fourth wavelength different from the third
wavelength.
19. The system according to claim 16, wherein the irradiation unit
comprises: multiple sources for the respective wavelengths of the
coherent electromagnetic radiation; and one or more source
beamsplitters configured to direct the coherent electromagnetic
radiation from the sources to the common optical path.
20. The system according to claim 19, wherein the source
beamsplitters comprise respective pellicle beamsplitters.
21. The system according to claim 16, wherein the irradiation unit
further comprises a sensor configured to detect a level value of
the coherent electromagnetic radiation.
22. The system according to claim 21, wherein the controller is
further configured to: determine respective level values of the
multiple wavelengths using the sensor; and adjust respective output
levels of the coherent electromagnetic radiation of the respective
ones of the wavelengths based at least in part on the respective
level values and a selected set point.
23. The system according to claim 16, further comprising: a second
imager arranged opposite the retained substrate from the
beamsplitter and adapted to obtain images of forward-scattered
light patterns from the micro-organism colony irradiated by the
respective wavelengths of the directed coherent electromagnetic
radiation.
24. The system according to claim 23, further comprising: a first
sensor arranged optically upstream of the retained substrate and
configured to detect a first level value of the coherent
electromagnetic radiation; and a second sensor arranged optically
downstream of the retained substrate and configured to detect a
second level value of the coherent electromagnetic radiation.
25. A system for the identification of micro-organisms, the system
comprising: an irradiation unit adapted to provide coherent
electromagnetic radiation of a selected wavelength along an optical
path; a holder adapted to retain a substrate having a surface
adapted for growth of a micro-organism colony; a beamsplitter
adapted to direct the coherent electromagnetic radiation from the
optical path towards the retained substrate; and an imager arranged
opposite the beamsplitter from the retained substrate and adapted
to obtain an image of a backward-scattered light pattern from the
micro-organism colony irradiated by the directed coherent
electromagnetic radiation.
26. The system according to claim 25, further comprising: a stage
adapted to translate the retained substrate or the beamsplitter
with respect to each other so that the directed coherent
electromagnetic radiation irradiates the micro-organism colony; and
a controller configured to: operate the stage and the irradiation
unit to successively irradiate ones of a plurality of
micro-organism colonies on the retained substrate; and operate the
imager to obtain a plurality of images of backward-scattered light
patterns from the successively-irradiated micro-organism colonies,
the plurality of images including at least first and second images
of a first colony at respective, different wavelengths, and third
and fourth images of a second, different colony at respective,
different wavelengths.
27. The system according to claim 25, wherein: the irradiation unit
further comprises a sensor configured to detect a level value of
the coherent electromagnetic radiation; and the system further
comprises a controller responsive to the level value and a selected
set point to adjust an output level of the coherent electromagnetic
radiation.
28. A system for the identification of micro-organisms, the system
comprising: an irradiation unit adapted to sequentially provide
coherent electromagnetic radiation of multiple wavelengths along a
common optical path; a holder adapted to retain a substrate having
a surface adapted for growth of a micro-organism colony in
operative arrangement to receive the coherent electromagnetic
radiation along the common optical path; and an imager arranged
optically downstream of the retained substrate and adapted to
obtain images of forward-scattered light patterns from the
micro-organism colony irradiated by the respective wavelengths of
the directed coherent electromagnetic radiation.
29. The system according to claim 28, further comprising: a stage
adapted to translate the retained substrate or irradiation unit
with respect to each other so that the directed coherent
electromagnetic radiation irradiates the micro-organism colony; and
a controller configured to: operate the stage and the irradiation
unit to irradiate a first colony of a plurality of micro-organism
colonies on the retained substrate; operate the imager to obtain a
first image and a second image of backward-scattered light patterns
from the first colony, the first image corresponding to a first
wavelength and the second image corresponding to a second,
different wavelength.
30. The system according to claim 29, wherein the controller is
further configured to: after operating the imager to obtain the
first image and the second image, operate the stage and the
irradiation unit to irradiate a second colony of the plurality of
micro-organism colonies on the retained substrate; and operate the
imager to obtain a third image and a fourth image of
backward-scattered light patterns from the second colony, the third
image corresponding to a third wavelength and the fourth image
corresponding to a fourth wavelength different from the third
wavelength.
31. The system according to claim 28, wherein the irradiation unit
comprises: multiple sources for the respective wavelengths of the
coherent electromagnetic radiation; and one or more source
beamsplitters configured to direct the coherent electromagnetic
radiation from the sources to the common optical path.
32. The system according to claim 31, wherein the sources comprise
respective lasers and each source beamsplitters comprises at least
one of a pellicle beamsplitter or a cage mount.
33. The system according to claim 28, wherein: the irradiation unit
further comprises a sensor configured to detect a level value of
the coherent electromagnetic radiation; the level value corresponds
to a selected one of the wavelengths; and the system further
comprises a controller responsive to the level value and a selected
set point to adjust an output level of the coherent electromagnetic
radiation of the selected one of the wavelengths.
34. The system according to claim 33, wherein: the sensor is
arranged substantially upstream of the retained substrate along the
common optical path; and the system further comprises a second
sensor arranged optically downstream of the retained substrate and
configured to detect a second level value of the coherent
electromagnetic radiation.
35. The system according to claim 34, further comprising a
computation unit configured to determine an optical density of the
micro-organism colony irradiated by the directed coherent
electromagnetic radiation based at least in part on the level value
and the second level value.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This international application claims priority to, and the
benefit of, U.S. Patent Application Ser. No. 62/058,478, filed Oct.
1, 2014, and entitled "Multispectral Forward Scatterometer for
Microbial Colony Interrogation," and U.S. Patent Application Ser.
No. 62/058,734, filed Oct. 2, 2014, and entitled "Scatterometer for
Microbial Colony Interrogation," the entirety of each of which is
incorporated herein by reference.
TECHNICAL FIELD
[0003] The present application relates to characterizing,
classifying, or identifying microscopic structures. Various aspects
relate to such structures including, e.g., colonies of
micro-organisms, clusters of cells, or organelles.
BACKGROUND
[0004] Rapid identification and classification of microbial
organism is a useful task in various areas, such as
biosurveillance, biosecurity, clinical studies, and food safety.
There is, for example, a need for methods for monitoring and
detecting pathogenic micro-organism such as Escherichia coli,
Listeria, Salmonella, and Staphylococcus.
BRIEF DESCRIPTION
[0005] A system for the identification of micro-organisms includes
an irradiation unit adapted to sequentially provide coherent
electromagnetic radiation of one or more wavelengths along a common
optical path. A holder is adapted to retain a substrate having a
surface adapted for growth of a colony of micro-organisms. A
beamsplitter is adapted to direct the coherent electromagnetic
radiation from the common optical path towards the retained
substrate. An imager is arranged opposite the beamsplitter from the
retained substrate and is adapted to obtain images of
backward-scattered light patterns from the micro-organism colony
irradiated by the respective wavelengths of the directed coherent
electromagnetic radiation. Some examples provide radiation of
multiple wavelengths and include an imager arranged optically
downstream of the retained substrate to obtain images of
forward-scattered light patterns from the micro-organism colony
irradiated by the wavelengths of radiation. Organism identification
methods are also described.
[0006] This brief description is intended only to provide a brief
overview of subject matter disclosed herein according to one or
more illustrative embodiments, and does not serve as a guide to
interpreting the claims or to define or limit scope, which is
defined only by the appended claims. This brief description is
provided to introduce an illustrative selection of concepts in a
simplified form that are further described below in the Detailed
Description. This brief description is not intended to identify key
features or essential features of the claimed subject matter, nor
is it intended to be used as an aid in determining the scope of the
claimed subject matter. The claimed subject matter is not limited
to implementations that solve any or all needs or disadvantages
noted in the Background.
BRIEF DESCRIPTION OF THE DRAWINGS
[0007] The above and other objects, features, and advantages will
become more apparent when taken in conjunction with the following
description and drawings wherein identical reference numerals have
been used, where possible, to designate identical features that are
common to the figures, and wherein:
[0008] FIG. 1A shows a schematic diagram of components of an
example system for the identification of micro-organisms;
[0009] FIG. 1B shows a schematic diagram of components of an
example system for the identification of micro-organisms;
[0010] FIG. 2 shows a mechanical drawing of components of an
example system for the identification of micro-organisms;
[0011] FIG. 3A is a graphical representation of a photograph of
components of a constructed scattering pattern measurement
instrument according to various examples;
[0012] FIG. 3B is a graphical representation of a photograph of
components of a constructed scattering pattern measurement
instrument according to various examples;
[0013] FIG. 3C is a graphical representation of a photograph of
components of a constructed scattering pattern measurement
instrument according to various examples;
[0014] FIG. 3D is a graphical representation of a photograph of an
example beamsplitter and related components useful with various
examples such as that shown in FIG. 3A;
[0015] FIG. 3E is a schematic showing optical components of an
example scattering pattern measurement instrument;
[0016] FIG. 4A shows an example of an optical path according to
various aspects;
[0017] FIG. 4B shows an example of an optical path according to
various aspects;
[0018] FIG. 5A is a graphical representation of a photograph of
components of a constructed scattering pattern measurement
instrument according to various examples;
[0019] FIG. 5B shows an example measured backward scattering
pattern of an example bacterial colony;
[0020] FIG. 5C shows an example measured forward scattering pattern
corresponding to the example bacterial colony of FIG. 5B;
[0021] FIG. 5D shows an example measured backward scattering
pattern of an example bacterial colony;
[0022] FIG. 5E shows an example measured backward scattering
pattern of an example bacterial colony;
[0023] FIG. 5F shows an example measured backward scattering
pattern of an example bacterial colony;
[0024] FIG. 5G shows an example measured forward scattering pattern
corresponding to the example bacterial colony of FIG. 5D;
[0025] FIG. 5H shows an example measured forward scattering pattern
corresponding to the example bacterial colony of FIG. 5E;
[0026] FIG. 5I shows an example measured forward scattering pattern
corresponding to the example bacterial colony of FIG. 5F;
[0027] FIGS. 6A-6I show graphical representations of photographs of
measured reverse scattering patterns for example bacterial
colonies;
[0028] FIG. 7 shows a coordinate system and schematic diagram of an
embodiment of forward optical scatter for a bacterial colony;
[0029] FIG. 8A is a graphical representation of a visible-light
photograph of S. aureus colonies on a BHI agar plate;
[0030] FIG. 8B is a graphical representation of a 3D morphology map
of an example S. aureus colony measured using the Integrated Colony
Morphology Analyzer (ICMA);
[0031] FIG. 8C is a graphical representation of an optical density
(OD) map of the example colony;
[0032] FIG. 8D is a graphical representation of a phase contrast
microscope (PCM) image of the example colony;
[0033] FIG. 8E is a graphical representation of a measured 405 nm
forward scatter pattern image of the example colony;
[0034] FIG. 8F is a graphical representation of a measured 635 nm
forward scatter pattern image of the example colony;
[0035] FIG. 8G is a graphical representation of a measured 904 nm
forward scatter pattern image of the example colony;
[0036] FIG. 8H is a graph showing cross sectional morphology and OD
profile of a center region of the example colony;
[0037] FIG. 9 is a graph showing a simulated intensity profile of a
scattering pattern;
[0038] FIG. 10A is a graph showing simulated numbers of rings and
half diffraction angles for various incident wavelengths;
[0039] FIG. 10B is a graph showing simulated locations of first to
third minima;
[0040] FIG. 10C is a graph showing simulated dimensions of the ring
width;
[0041] FIG. 10D is a graph showing simulated dimensions of the ring
gap;
[0042] FIG. 11A is a graph showing simulated intensity profiles of
forward scattering patterns at various wavelengths;
[0043] FIG. 11B is a graph showing measured intensity profiles of
forward scattering patterns of an example bacterial colony at
various wavelengths;
[0044] FIG. 11C is a graphical representation of a quarter view of
the simulated scatter patterns of FIG. 11A; and
[0045] FIG. 11D is a graphical representation of a quarter view of
the measured scatter patterns of FIG. 11B;
[0046] FIG. 12 is a graph showing a comparison of simulated and
measured spectral half diffraction angles;
[0047] FIG. 13 is a graph showing a comparison of intensity
profiles of simulated spectral diffraction patterns;
[0048] FIG. 14A shows an example of an optical path according to
various aspects;
[0049] FIG. 14B shows an example of an optical path according to
various aspects;
[0050] FIG. 15A is a graph showing optical properties of a measured
pellicle beamsplitter;
[0051] FIG. 15B is a graph showing relative intensities of various
wavelengths of a tested example irradiation system;
[0052] FIG. 16A is a graph showing optical density measurements
that were collected of example bacterial colonies;
[0053] FIG. 16B is a graph showing relative optical density of the
measurements of FIG. 16A;
[0054] FIG. 17A shows a tested experimental configuration;
[0055] FIG. 17B is a graph showing experimental data of optical
density of agar plates;
[0056] FIG. 17C is a graph showing experimental data of optical
density of bacterial colonies on agar plates;
[0057] FIG. 17D is a graph showing optical density of measured
bacterial colonies;
[0058] FIG. 18A is a graphical representation of a simulated
forward scattering pattern for an example bacterial colony
illuminated with 405 nm-wavelength light;
[0059] FIG. 18B is a graphical representation of a simulated
forward scattering pattern for an example bacterial colony
illuminated with 635 nm-wavelength light;
[0060] FIG. 18C is a graphical representation of a simulated
forward scattering pattern for an example bacterial colony
illuminated with 904 nm-wavelength light;
[0061] FIG. 18D is a graph showing intensity profiles of simulated
scatter patterns;
[0062] FIG. 19 shows graphical representations of photographs of
measured scattering patterns for example bacterial colonies at
various wavelengths;
[0063] FIG. 20A shows graphical representations of polar intensity
plots of measured scattering patterns for example bacterial
colonies at various wavelengths;
[0064] FIG. 20B shows average intensity as a function of radius for
scatter patterns of a measured bacterial colony at various
wavelengths;
[0065] FIG. 21 is a graph showing classification accuracy for
various numbers of features;
[0066] FIGS. 22A-22D are three-dimensional linear-discriminant
analysis plots showing feature-space distances between colonies of
different types of micro-organisms;
[0067] FIG. 23 is a flowchart showing example methods for
identifying micro-organisms; and
[0068] FIG. 24 is a high-level diagram showing components of a
data-processing system.
[0069] The attached drawings are for purposes of illustration and
are not necessarily to scale.
DETAILED DESCRIPTION
[0070] This application is related to U.S. Pat. No. 7,465,560,
issued Dec. 16, 2008, and U.S. Pat. No. 8,787,633, issued Jul. 22,
2014, the contents of which are incorporated herein by reference in
their entirety.
[0071] Various aspects herein advantageously relate to scalar
diffraction modeling of multispectral forward scatter patterns from
bacterial colonies. While conventional culture based methods are
still used, utilizing laser scattering phenomena from bacterial
colonies has provided a possible label-free discrimination
methodology, named Bacteria Rapid Detection using Optical
scattering Technology (BARDOT). Various aspects herein relate to a
multispectral domain which provides additional optical
characteristics such as spectral absorption and spectral forward
scattering patterns. Various aspects permit classifying bacterial
colonies as to the subspecies of bacteria in the colony, e.g., at
the serovar level.
[0072] Compared to conventional detection methods, label-free
optical diagnostics delivers fast and accurate results, and
provides cost-effective and non-destructive evaluation of the
samples, allowing for secondary confirmation with further
verification.
[0073] Owing to the wide range of the spectral region that is
available for optical diagnostics, different optical windows
ranging from UV to IR for the detection and classification of
micro-organisms have been used in the art. In the area of food
inspection, numerous uses of hyperspectral imaging to classify the
quality of harvested vegetables, fruits, meats, and poultry can be
used. Spectral imaging can also be used in biomedical applications
such as skin cancer detection, heart disease diagnostics, and
detection of retinal diseases.
[0074] Spectral techniques used in the art rely on standard
far-field imaging. However, cells and bacterial colonies are three
dimensional objects, and optically interrogating the whole volume
can provide better classification accuracy. A label free,
non-destructive, and automated detection technique, based on
elastic light scatter (ELS) patterns of bacteria colonies from a
single-wavelength laser, has been used in the art for rapid
detection and classification of microbial organisms. It is
applicable and effective for a limited number of genera or species
of various different organisms. While the interrogation photons
interact with the whole volumes of the colonies, thus imprinting
better phenotypic characteristics than simple reflective imaging,
classification performance suffers when large number of species and
strains are analyzed simultaneously.
[0075] Various aspects relate to scalar diffraction imaging of
multispectral forward scattering patterns for bacterial colonies
and multispectral bacterial phenotyping. According to an aspect of
the invention, there is provided a new design and validation of a
multispectral forward scatter phenotyping instrument called
MultiSpectral BActerial Rapid Detection using Optical scattering
Technology (MS-BARDOT) which combines multiple wavelength diode
laser sources. A variety of embodiments of the invention provide an
optical density (OD) measurement unit with the conventional BARDOT
system. Various embodiments advantageously provide the simultaneous
measurement of both multiple wavelengths of forward scattering
pattern and OD of a bacteria colony. Various embodiments
advantageously provide a series of coordinate matched and
correlated bio-optical characteristics of colonies, consequently
improving the classification accuracy of previously introduced
standard BARDOT system. Various experiments were performed in which
scattering patterns of four pathogenic bacteria were measured and
analyzed. Various embodiments of MS-BARDOT can advantageously
perform in-situ measurement of three different wavelength forward
scattering patterns of a bacterial colony within four seconds
without moving the specimen. Various embodiments of the invention
can include a reflection scatterometer providing reflection
patterns, e.g., of opaque samples such as bacterial colonies on
opaque agar.
[0076] Various embodiments advantageously can simultaneously detect
three-wavelength scatter patterns and the associated optical
density from individual bacterial colonies, overcoming a limitation
of prior instruments that used a single wavelength for signal
collection. Various examples can use absorption measurement of
liquid bacterial samples in addition to spectroscopic information
to distinguish samples. Various examples use optical components
such as pellicle beam splitter and optical cage system for robust
acquisition of multispectral images. Various embodiments
advantageously can perform scatter pattern classification by
combining the features collected at all three wavelengths and
selecting the best features via feature selection mechanisms,
thereby providing better classification rates than the same number
of features at a single wavelength.
[0077] Optical interrogation of biological samples is popular in
diverse fields from agricultural to biomedical applications. Due to
the inherent wide spectral window of the optical interrogation,
strategic selection of appropriate wavelengths is useful for
enhanced resolution and proper classification of the biological
sample. In biomedical applications, multispectral technique has
been widely used in skin diagnostics and microscopic dark-field
imaging. In the agricultural and food science fields, multispectral
spectral reflectance measurements can be used to automatically
detect and monitor the quality of the harvested fruits. In some
examples, acquiring spectral reflection images from bacterial
colonies on the surface of food can permit label-free
classification and identification of such colonies.
[0078] Recently, there has been developed a label-free colony based
bacterial classification system which utilizes the single 635 nm
wavelength for interrogation. Various examples of the system can be
used for classifying genus and species levels and some cases down
to serovar levels. Bacterial colonies can be modeled as a
biological spatial light modulator which changes the amplitude and
phase of the outgoing wave and the characteristics of the scatter
patterns to the morphological trait of the individual colonies were
closely investigated. Various colonies have profiles such as convex
shapes with different radii of curvature and a Gaussian profile.
For example, a profile of a Staphylococcus Aureus (S. aureus)
colony can closely match a Gaussian curve, which is similar to a
bell curve with a tailing edge with smaller aspect ratio (colony
height to diameter ratio). In a tested example, a measured colony
generated a concentric circular diffraction pattern. Various
aspects herein permit measuring the 3D morphology of each colony
and 3D Optical Density (OD) map simultaneously without moving
specimen. Staphylococcus is a common micro-organism and can reside
on the human skin and other organisms, and has a relatively simple
colony morphology and a substantially concentric circular
diffraction pattern.
[0079] Various aspects herein describe a multiple wavelength
interrogation instrument which permits determining scatter patterns
from different laser wavelengths. Various aspects of the
multispectral approach provide: 1) capability to provide ELS
patterns in multiple wavelengths, 2) acquisition of spectral
optical density, and 3) leverage of different spectral response via
wavelength-dependent refractive indices. Various aspects herein use
scalar diffraction theory to model the ELS patterns across visible
range of spectrum. Detailed simulation and prediction of the
multispectral ELS patterns can be performed. For experimental
verification, an example MS-BARDOT system was constructed. The
example system included stackable cage type pellicle beam splitter
units. Staphylococcus aureus was chosen as a model organism and the
spectral ELS patterns from three different interrogation
wavelengths were compared.
[0080] FIGS. 1A and 1B show schematic diagrams of measurement
device 100, e.g., an MS-BARDOT instrument, according to various
aspects. The illustrated example instrument includes a
multispectral forward scatterometer 102 and a sequence controller
104. Some embodiments can also include a two-dimensional stage, as
discussed below with reference to FIG. 2. The optics shown in FIG.
1A can include one, two, three, or more beamsplitters 106, e.g.,
cage type R45:T55 pellicle beam splitters (e.g., supplied by
Thorlabs Inc., NJ, USA), which can be selected to reduce ghost
effects which can arise with some glass type beam splitters. In
FIG. 1A, arrows show paths of light travel. In the illustrated
example, two pellicle beam splitters 106A and 106B for light
sources 108 are positioned above a target, e.g. a petri dish 110,
at a distance of, e.g., 67 mm from the top of the petri dish 110 to
the center of the bottom pellicle beam splitter 106B. In some
examples of light sources, two collimated 1 mW laser diode modules
with round beams, one emitting at 405 nm and the other at 635 nm
(Coherent Inc., CA, USA), and a 904 nm laser diode module
(Lasermate Group Inc., CA, USA), can be used as 108A, 108B, 108C,
respectively, and mounted to the ports of cage mounted pellicle
beam splitter units 106, e.g., as shown. In various embodiments,
the choice of individual wavelengths can be selected based on the
spectral absorption characteristics of the desired target bacterial
genera, and the availability of specific spectral lines from a
diode laser or other light source. Some examples using a stacked
pellicle beam splitter unit design permit measuring multispectral
ELS patterns from a bacterial colony in less than, e.g., 4-5
seconds without moving the specimen target. As used herein, the
light from sources 108A, 108B, 108C travels in a "downstream"
direction, in this example through the colony to the imager 112.
The opposite direction, towards the sources, is an "upstream"
direction.0
[0081] To capture a forward scattering pattern, in a tested
configuration, a monochromatic CMOS camera 112 (Pixelink, PL-B741,
ON, Canada) with 1280 (H).times.1024 (V) pixels and 6.7-.mu.m-unit
pixel size was located under the petri dish 110 at a distance of,
e.g., 9.7 mm or 39 mm from the bottom of the petri dish to the
surface of the image sensor. In addition to the pellicle beam
splitters, some embodiments include an additional port and a
spectral intensity monitor (see FIG. 2). A Si photodiode 114
(Thorlabs Inc., CA, USA) (PD) with an active wavelength range of
from 400 nm to 900 nm was mounted to bottom pellicle beam splitter
106B. CMOS camera 112 is one example of an imager; other imagers
can be used, e.g., CCD imagers, film capture devices, or
latent-image sensors such as those used in computed radiography. In
some examples, the sources 108A, 108B, 108C can be illuminated
sequentially. In other examples, at least two of the sources 108A,
108B, or 108C can be illuminated simultaneously.
[0082] FIG. 2 shows a schematic diagram of one embodiment of
optical components useful, e.g., with electric and mechanical
components illustrated in FIG. 1B. Using two caged R45:T55 pellicle
beam splitters 106, three different wavelength LDs (Laser Diodes)
208 as light sources, a PIN PD (Photo Diode) 114 as a LD intensity
monitor, and a CMOS sensor 112 as a forward scattering pattern
capture unit, induced forward scattering pattern of a bacteria
colony can be measured at multiple different wavelengths without
moving the specimen. In an experiment, measurements of a colony
were collected at three different wavelengths within 4 seconds.
[0083] FIG. 2 also shows a holder having two arms supporting the
petri dish 110. In some examples, the system can include this or
another configuration of holder adapted to retain, directly or via
one or more plates, petri dishes, or other supports, a substrate
such as an agar gel having a surface adapted for growth of one or
more colonies of micro-organisms. Example agars are described
herein. The holder can include one or more forks, arms, pins, or
other retention features or mechanisms. As used herein, references
to positioning or orientation of components with respect to a
retained substrate apply whether or not a substrate is retained at
any particular time. As used herein, references to positioning of
parts opposite one another or opposite other parts do not constrain
the tolerances or imply any requirement of coaxiality unless
otherwise specified.
[0084] In some embodiments, e.g. for optical density (OD)
measurement, an additional pellicle beam splitter 214 can be
positioned between the petri dish 110 and the CMOS camera 112, and
two Si photodiodes (PD) 114, 216, e.g., with an active wavelength
range of from 400 nm to 900 nm, can be operationally arranged with
respect to the middle and bottom pellicle beam splitters 106, 214.
The PD 114 attached to the middle beam splitter 106 can monitor the
intensity of incident light, while the PD 216 integrated to the
bottom beam splitter 214 can measure that of light transmitted
through a sample.
[0085] The sequence controller 104, FIG. 1B, can include a
microcontroller unit (MCU) (e.g., an Atmel AVR128) as a data
acquisition unit, and a personal computer 120 (PC) as a master
controller. Using, e.g., the MCU's internal 10 bit A/D conversion,
signals from variable non-inverting amplified and 2.sup.nd low pass
filtered photodiode (system 122) are captured, and transferred to
the PC 120, e.g., through a data connection such as a USB or
RS-232C interface. Three intensity-tunable diode laser drivers 124
are connected to the digital I/O of the MCU. The CMOS camera 112 is
connected to PC 120 with, e.g., IEEE 1394 or another high-speed bus
or interface, and is controlled using the software development kit
(SDK) from manufacturer (Pixelink, ON, Canada). The PC 120 can
control sequences and log synchronized information, tagged with the
incident wavelength. OD of the sample can be determined for each
wavelength, as will be described below.
[0086] In a variety of embodiments, a reflection type scatterometer
can be included as shown in FIGS. 3A-3E. These embodiments can be
useful for opaque samples, such as fungi, mold, or yeast. As shown
in FIG. 3A-3C, the reflection scatterometer utilizes a laser source
302 directed to a beam splitter 304 (R25:T75 BS in the diagram),
and a large area CMOS (or other) detector 306 (referred to in FIG.
3A as a backward scattering pattern grabber).
[0087] As discussed below, four genera of bacteria (Escherichia
coli O157:H7 EDL933, Listeria monocytogenes F4244, Salmonella
enteritidis PT21 and Staphylococcus aureus ATCC 25923) were
measured using instruments such as discussed above with reference
to FIGS. 1A, 1B, 2, and 3A-3D. The acquired information showed
differences in scatter characteristics between the tested
organisms. In addition, colony-based spectral optical density
information was also collected. Optical modeling performed using
diffraction theory correctly predicted wavelength-related
differences in scatter patterns which were matched by the
experimental results.
[0088] FIG. 4A shows an example of an optical path according to
various aspects. FIG. 4A illustrates geometry of example
reflective/transmissive imaging systems such as discussed below
with reference to FIG. 5A.
[0089] FIG. 4B shows an example of an optical path according to
various aspects. FIG. 4B illustrates geometry of example reflective
imaging systems such as discussed above with reference to FIG. 3A.
Table 1 shows example characteristics of imagers that can be used,
e.g., as imagers 112, 306, according to various examples.
TABLE-US-00001 TABLE 1 Unit (mm) Width Height Diagonal(calculated)
Active area PL-B741F 8.57 6.86 10.98 58.79 Nikon 1/CX 13.20 8.80
15.86 116.16 Four Third 17.30 13.00 21.64 224.90 Foveon(Sigma)
20.70 13.80 24.88 285.66 APS-C(Canon) 22.20 14.80 26.68 328.56
APS-C(Nikon) 23.60 15.70 28.35 370.52 APS-H(Canon) 28.70 19.00
34.42 545.30 Full Frame 36.00 24.00 43.27 864.00 Medium-Format
48.00 36.00 60.00 1728.00
[0090] FIG. 5 shows a comparison of sample reflective (backward)
scattering patterns to forward scattering patterns, e.g., captured
as described above with reference to FIGS. 3A-3E. The reflective
scattering patterns are clearer than the forward patterns for the
measured bacterial colony.
[0091] FIGS. 6A-6I show measured reverse scattering patterns
measured on TAS blood agar (sheep blood 5%). FIGS. 6A-6C show
patterns for E. coli K12. FIGS. 6D-6F show patterns for Listeria
PU12. FIGS. 6G-6I show patterns for Citrobacter PU89. As shown, the
reverse scattering patterns are qualitatively different between
organisms and similar within an organism.
[0092] FIG. 7 shows a coordinate system and schematic diagram of
forward scattering from a bacterial colony. The bacterial colony
and a semi-solid media are located at the aperture plane, and the
forward scattering pattern is captured at the image plane, which
are defined as (x.sub.a,y.sub.a) and (x.sub.i,y.sub.i),
respectively. Light propagation direction is defined as the z axis,
and the distance between the aperture plane and image plane is
defined as Z.sub.i. According to optical theory, the diffraction
pattern in the image plane is the Fourier transform of the field in
the aperture plane. Even though it is a combined result of macro
scale (the colony's morphological characteristics) and micro scale
(each individual bacteria cell), a macroscopic-only approach is
adapted to this modeling whereby the colony is considered as
amplitude and phase modulator. A bacteria colony is modeled as the
Gaussian-like profile (bell curve shape with tailing edge) as Eq.
1:
Colony ( x a , y a ) = H 0 exp [ - ( x a 2 + y a 2 ) r c 2 ] = H (
x a , y a ) ( Eq . 1 ) ##EQU00001##
where H.sub.0 and r.sub.c are defined as height of center and
radius of the colony, respectively. The ratio between H.sub.0 and
2.times.r.sub.c is defined as aspect ratio. In a simulated example,
1:7 was selected as being a representative aspect ratio for S.
aureus.
[0093] Based on the Fresnel approximation formula, a TEM00 mode of
an incident laser beam induces an electrical field Ea at the
aperture plane, as in Eq. 2:
E a ( x a , y a , z ) = E 0 exp [ - ( x a 2 + y a 2 ) .omega. 2 ( z
) ] exp ( ikz ) exp [ ik ( x a 2 + y a 2 ) 2 R ( z ) ] ( Eq . 2 )
##EQU00002##
where E.sub.0 is on-axis strength and the three exp( ) terms are
known as the amplitude of the field, the longitudinal phase, and
the radial phase respectively.
[0094] The quantities .omega.(z) and R(z) are beam waist and radius
of the wave front, respectively, and are defined as Eq. 3:
.omega. 2 ( z ) = .omega. 0 2 [ 1 + ( z z 0 ) 2 ] , R ( z ) = z [ 1
+ ( z 0 z ) 2 ] ( Eq . 3 ) ##EQU00003##
where z.sub.0 is defined as the Z location where the 1/e.sup.2
radius has expanded to {square root over (2)} times the beam waist
.omega..sub.0.
[0095] Using the Huygens-Fresnel principle in rectangular
coordinates, the Fresnel-Kirchhoff diffraction formula, and the
First Rayleigh-Sommerfeld solution, the electric field at the image
plane E.sub.i induced by the Ea is derived as Eq. 4:
E i ( x i , y i ) = 1 i .lamda. .intg. .intg. t ( x a , y a ) E a (
x a , y a ) exp [ ik .PHI. ( x a , y a ) ] exp [ ikr ai ] r ai cos
.theta. dx a dy a ( Eq . 4 ) ##EQU00004##
where t(x.sub.a,y.sub.a) is the 2D transmission coefficient,
.PHI.(x.sub.a,y.sub.a) is the 2D phase modulation factor, r.sub.ai
is the distance between the aperture plane and a point on the image
plane, and .theta. is angle between vectors {right arrow over (z)}
and {right arrow over (r.sub.ai)}, which is calculated as
r.sub.ai/Z.sub.2.
[0096] With the Fresnel approximation based on a binomial expansion
of the square root, r.sub.ai can be as in Eq. 5:
r ai = [ z i 2 + ( x a - x i ) 2 + ( y a - y i ) 2 ] 1 2 .cndot. z
i [ 1 + 1 2 ( x a - x i z i ) 2 + 1 2 ( y a - y i z i ) 2 ] ( Eq .
5 ) ##EQU00005##
[0097] Accordingly, in the illustrated example, the electric field
at the image plane is expressed as Eq. 6:
E.sub.i(x.sub.i,y.sub.i)=C.intg..intg.T(x.sub.a,y.sub.a)exp[i.PHI..sub.r-
]exp[i.PHI..sub.c]exp[i.PHI..sub.g]exp[-2.pi.i(f.sub.xx.sub.a+f.sub.yy.sub-
.a)]dx.sub.ady.sub.a (Eq. 6)
where T is the amplitude modulator; f.sub.x and f.sub.y are defined
as x.sub.i/(.lamda.Z.sub.2) and x.sub.i/(.lamda.Z.sub.2), known as
a spatial frequency; .PHI..sub.r, .PHI..sub.q, and .PHI..sub.g are
radial-, quadratic-, and Gaussian-phase components, respectively.
The latter are defined as in Eqs. 8-10, below. The summation of
these phase components is functions as phase modulator for the
propagating light.
[0098] The amplitude modulator T is as in Eq. 7:
T ( x a , y a ) = exp [ - ( x a 2 + y a 2 ) .omega. 2 ( z ) ] E out
E 0 = exp [ - ( x a 2 + y a 2 ) .omega. 2 ( z ) ] ( 1 - r air - bac
) ( 1 - r k ) 2 l ( 1 - r bac - agar ) ( Eq . 7 ) ##EQU00006##
[0099] The model includes the amplitude modulator
T(x.sub.i,y.sub.i,.lamda.) and a phase modulator .PHI..sub.overall,
the latter of which comprises .PHI..sub.c, .PHI..sub.q, and
.PHI..sub.r, which are defined as the colony-, quadratic-, and
Gaussian-phase components, respectively, in Eqs. 8-10:
.PHI. r ( x a , y a ) = k ( x a 2 + y a 2 ) 2 R ( Eq . 8 ) .PHI. q
( x a , y a ) = k ( x a 2 + y a 2 ) 2 z i ( Eq . 9 ) .PHI. g ( x a
, y a ) = k ( n 1 - 1 ) H 0 exp [ - ( x a 2 + y a 2 ) r c 2 ] ( Eq
. 10 ) ##EQU00007##
[0100] A cross section of bacteria colony accumulates with densely
packed multiple layers of bacteria cell, and is covered with
extracellular materials with an overall thickness of .DELTA.. In
reality, propagating light is attenuated by both reflections and
absorptions, however, only normal incident reflection is assumed to
be a major contributor of the intensity loss in this modeling.
[0101] The coefficient of Eq. 6, C, is derived as in Eq. 11:
C = E 0 exp ( ikn agar .DELTA. agar ) exp ( ikH 0 ) exp [ ik ( z +
z i ) ] exp [ ik ( x i 2 + y i 2 ) / 2 z i ] i.lamda.z i ( Eq . 11
) ##EQU00008##
where .DELTA..sub.agar and n.sub.agar are defined as the thickness
of agar and the refractive index of agar, respectively.
[0102] The attenuation of the k.sup.th layer of bacteria cells is
modeled as Eq. 12:
E.sub.k+1=E.sub.k(1-r.sub.k) (Eq. 12)
where r.sub.k is the reflection coefficient for the k.sup.th layer
and is assumed to be identical for all the cells.
[0103] The other reflective coefficients for the air-bacteria cell
interface, r.sub.air-bac and the bacteria cell-agar interface,
r.sub.bac-agar are defined as in Eq. 13:
r air - bac = n air - n bac n air + n bac , r bac - agar = n bac -
n agar n bac + n agar , r k = n bac - n ec n bac + n ec ( Eq . 13 )
##EQU00009##
where n.sub.air, n.sub.bac, n.sub.ec, n.sub.agar are the refractive
indices of air, the bacteria cell, the extracellular material, and
agar, respectively.
[0104] As the bacteria colony is modeled as a stacked layer
structure, l is defined as Eq. 14:
l = ColonyProfile ( x a , y a ) .DELTA. ( Eq . 14 )
##EQU00010##
[0105] The intensity of the electric field is calculated via Eq.
15,
I=1/2.epsilon.c|E.sub.i|.sup.2 (Eq. 15)
where c is speed of light in a vacuum, and .epsilon. is vacuum
permittivity. In this example, all components of the forward
scattering modeling, such as amplitude modulator, phase modulator,
and coefficient, are influenced by the incident wavelength, the
wavelength-induced forward scattering prediction for S. aureus
colony is complete.
[0106] The multispectral forward scattering pattern of a bacterial
colony is modeled as in Eq. 16:
E.sub.i(x.sub.i,y.sub.i,.lamda.)=C.intg..intg.T(x.sub.a,y.sub.a,.lamda.)-
exp[i.PHI..sub.overall(x.sub.a,y.sub.a,.lamda.)]exp[-2.pi.i(f.sub.x(.lamda-
.)x.sub.a+f.sub.y(.lamda.)y.sub.a)]dx.sub.ady.sub.a (Eq. 16)
[0107] FIGS. 8A-8H shows examples of different measurement
modalities for S. aureus. Staphylococcus, in the illustrated
example, forms a Gaussian profile shaped colony with smaller aspect
ratio (colony height to diameter ratio, 1:6.7), and can provide a
circularly-symmetric diffraction pattern, as discussed below. To
predict the multispectral forward scattering pattern of
Staphylococcus Aureus (S. aureus) colony, morphological and optical
bacterial colony characteristics were modeled, and multispectral
forward scattering pattern modeling for S. aureus colony using
Elastic Light Scattering (ESL) theory was introduced. The
scattering pattern modeling included two components, an amplitude
and a phase component. Since S. aureus colony has different optical
characteristics such as reflective index and Optical Density (OD)
depending on the incident wavelength, both amplitude and phase
components of the model included wavelength effects on forward
scattering patterns. These were predicted, in a simulated example,
as a combination form of these components. As described herein,
diffraction ring width, gap, maximum, minimum, and the first
deflection point, which were directly related with pattern size
were defined. In various tested and simulated examples, longer
incident wavelength induced wider ring width and gap, and smaller
pattern and smaller numbers of rings, while, shorter incident
wavelength formed smaller ring width and gaps with larger pattern
size, and larger numbers of rings.
[0108] FIG. 8A shows a camera image of a tested plate. FIG. 8B
shows a plot of 3D colony morphology. FIG. 8C shows colony optical
density. FIG. 8D shows a phase contrast microscope image (PCM) of
the colony. FIG. 8E to 8G show measured spectral forward scattering
patterns of an S. aureus colony for 405 nm, 635 nm, and 904 nm,
respectively. Diameter of the measured S. aureus colony was 792 um,
and verified by both PCM and ICMA. As FIGS. 8B and 8H show, height
of sample S. aureus colony was 120 .mu.m, and the aspect ratio of
the colony was computed as 1:6.7. These spectral scattering
patterns display periodic and circularly symmetric ring patterns,
while the size of the pattern, and the width and gaps of each ring
vary depending on the interrogating wavelengths.
[0109] FIG. 8H shows the cross sectional morphology and OD profile
at the center region of the sample S. aureus colony. The cross
sectional morphology of the colony was similar to a bell curve
shape, while OD showed a Gaussian shape with a raised center area,
a shoulder near the center area, and a sharply increasing slope
near edge area, which means optical properties of the tested colony
are not proportional to the morphology. The OD of BHI agar was
measured to be approximately 0.2 using ICMA light source (675 nm),
while the maximum OD of the colony was measured as 0.64.
[0110] FIG. 9 shows example simulation data labeled with
definitions of components for multispectral forward scattering
pattern analysis. Several parameters of scatter patterns can be
defined, such as the first deflection point, ring gap, and ring
width (FIG. 9). First deflection point 902 is defined as the
outermost point of the pattern and correlates with the diameter of
the scatter patterns. Ring gap 904 is defined as the distance
between adjacent maxima and minima, e.g., as shown. Due to the
asymptotically decreasing intensity at the edge, it is difficult to
choose a first deflection point in a consistent manner. Therefore,
the location of the first deflection point 902 can be computed by
adding the distance between the first maximum point 906 (determined
starting from the outside of the pattern and moving towards the
center) and the first minimum point 908 (likewise defined) to the
location of the first maximum point 906.
[0111] For the simulation, the diameter of the colony was assumed
to be 800 .mu.m, the height of the colony was assumed to be 120
.mu.m, the distance in between the aperture and image planes was
assumed to be 46 mm, and the diameter of the beam at incidence on
the colony was assumed to be 1 mm. Refractive index of the S.
aureus cell along the incident wavelength was assumed to be that of
a thin cellulose film, a polymeric organic compound with the basic
(monomeric) formula C.sub.6H.sub.10O.sub.5, for 400 nm to 900 nm
incident light, i.e. from 1.4850 to 1.4524, respectively. The
reference refractive index was curve fitted using a two term power
equation, e.g. Eq. 17:
f(x)=ax.sup.b+c (Eq. 17)
with coefficients of 2.486E-08, -1.789, and 1.455 for a, b, and c,
respectively, with the goodness of fit according to SSE, R-square,
and RMSE being 9.226E-07, 0.9995, and 0.0001386, respectively.
[0112] FIGS. 10A-D show example analyses of spectral forward
scattering pattern for a bacteria colony. In particular, FIG. 10A
shows the computed results for the number of rings (squares, Eqs.
19, 21) and half-diffraction angle (triangles, Eq. 18, 20) of the
predicted model plotted against wavelength. The number of the rings
decreases from 113 at 400 nm to 47 at 900 nm (a 57% decrease).
Similar to the location of the first minimum point (FIG. 10B), the
computed half-maximum diffraction angle also decreases from 0.048
to 0.0462 rad (a 3.7% decrease) for the incident wavelength shift
from 400 nm to 900 nm. Both the number of rings and half
diffraction angle are inversely proportional to the incident
wavelength.
.theta. / 2 max = 1 k ( d .DELTA. .PHI. dr ) max ( Eq . 18 ) N ring
= .DELTA..PHI. max 2 .pi. ( Eq . 19 ) .theta. ( .lamda. ) / 2 ma x
= 1 k ( d .DELTA. .PHI. overall ( .lamda. ) dr ) max ( Eq . 20 ) N
ring ( .lamda. ) = .DELTA. .PHI. overall ( .lamda. ) max 2 .pi. (
Eq . 21 ) ##EQU00011##
[0113] FIG. 10B shows the spectral dependence of the location of
the first to third minima, while FIG. 10C shows the widths of the
first through third rings, and FIG. 10D shows the ring gap. Both
the number of rings and half diffraction angle were inversely
proportional to the incident wavelength, while ring width and gap
were proportional to the incident wavelength. Each local maximum
and minimum point was automatically found from the predicted
patterns, and the ring width and the ring gap were computed based
on the location of these local maxima and minima. The locations of
the minimum points decrease with increasing wave length, while the
ring width and ring gap increase with longer wavelength. Since the
first minimum point is defined as the first deflection point, and
that determines the pattern size, the pattern size decreases by
4.4% from 400 nm to 900 nm, and is inversely proportional to the
incident wavelength. Ring width and ring gaps increase by 40.1%,
and 40.8%, respectively, with increasing wavelength from 400 nm to
900 nm of the incident light, and are proportional to the
wavelength. Second and third ring widths and ring gaps showed
similar trends to those of the first ring case.
[0114] In an experiment that was performed, Staphylococcus aureus
ATCC 25923 was inoculated and grown on a BHI agar for 14.5 hours at
37.degree.. While 10-30 colonies appeared on the surface of the
agar, 5-10 colonies were selected that had grown closer to 1 mm
diameter. Then multispectral BARDOT captured the forward scattering
patterns in all three wavelengths and spectral OD simultaneously,
e.g., as described above with reference to FIG. 2.
[0115] FIG. 11 shows an example comparison of spectral forward
scattering pattern prediction and experimental results of S. aureus
ATCC25923 as measured by one embodiment of a measurement instrument
as described herein. FIGS. 11A and 11B show 1D cross sections of
spectral diffraction patterns while FIGS. 11C and 11D show 2D
patterns. FIGS. 11A and 11C show simulation results; FIGS. 11B and
11D show experimental results. Considering the pixel width of the
CMOS camera, the x coordinates of FIG. 11B were converted from
pixel to mm scale, and that of y coordinate was converted to
normalized intensity by considering the quantum efficiency of CMOS
along the incident wavelength. The cross sections of predicted
simulation results show wider and sparser periods of patterns for
the longer incident wavelength, while the pattern size (the
location of the first deflection point) decreases from 4.46 mm to
4.25 mm. The prediction and experimental results showed good
agreement. The average of percent error between the spectral
prediction and experimental results were computed as 3.54% (pattern
size), 0.04% (location of the first to third maxima), and 7.66%
(ring width).
[0116] Theoretical calculation of the forward scatter patterns were
conducted using a diffraction model based on Rayleigh-Sommerfeld
and Fresnel diffraction theory. The bacteria colonies were modeled
as bell curves with tailing edge profiles, which is modeled as an
amplitude and phase modulator. Staphylococcus aureus ATCC 25923 is
known for its concentric ring patterns, and high aspect ratio of
1:5 for center height to colony diameter ratio. Here the spectral
effect was analyzed via utilizing the variation of the refractive
index of bacteria versus the interrogating wavelength. The results
are shown in FIG. 11A, which displays the comparison of the theory
(11A) and experiment (11B) from S. aureus. The general trend shows
that as the wavelength increases from 405 to 904 nm, the overall
pattern size decreases by 4.376% and 4.35% for model and
experiment, respectively. The ring gap and ring width increased
44.17% and 45.02% for the theoretical model, and 40.16% and 41.27%
in the experimental results. The number of peaks and their
locations showed excellent agreement between the theory and
experiment.
[0117] FIG. 12 shows the comparison of spectral half diffraction
angle which compares experimental results with two different
theoretical results. The first value utilizes the half diffraction
angle formula (Eqs. 18, 20, triangles), the second value is the
calculated pattern diameter from the proposed multispectral model
(square), and the third is experimental results from S. aureus
(hexagon). As FIG. 12 indicates, all three methods show that longer
wavelength induces a smaller diffraction angle and thus smaller
diameter than the short wavelength patterns.
[0118] FIG. 13 shows a comparison of cross sectional views of the
spectral diffraction pattern near the outermost boundary of the
predicted patterns. The X axis represents the lateral direction of
the pattern, the Y axis represents the wavelength, and the vertical
offset and shading represent the intensity of the pattern (lighter
shading=higher intensity). Longer wavelength incident light induces
a smaller pattern size, wider ring width and gaps, and fewer rings
in the diffraction pattern for S. aureus colony.
[0119] FIGS. 14A and 14B depict beam paths from light sources 106
to photodiode 114 and sensor 112 for forward scattering mode, and
from light sources 106 to photodiodes 114, 216 for OD measurement
mode, respectively (all FIG. 1). FIGS. 14A and 14B show example
beam paths, and show how the beam is reflected and transmitted from
a light source to each sensor (photodiode or imager) through the
pellicle beam splitter. Since pellicle beam splitters, photodiodes,
CCD sensors, and CMOS sensors used in cameras can have inherent
optical response characteristics such as spectral reflectance,
transmittance and quantum efficiency, the light intensities and
sensitivity of the each sensors can be calibrated and compensated
for using both optical and electric means to maintain similar input
intensities for each wavelength.
[0120] FIG. 15A shows experimental reflectance (R) and
transmittance (T) ratios of a nominally R45:T55 pellicle beam
splitter. The data are R56.8:T43.1, R44.3:T55.7, and R40.9:T59.0
for 405 nm, 635 nm, and 904 nm respectively.
[0121] Overall attenuation by the beam splitter units from 405 nm,
635 nm, and 904 nm laser sources to CMOS was experimentally
determined to be 0.0805, 0.2468, and 0.1428; to PD #1 was 0.2454,
0.5570, and 0.1679; to PD #2 was 0.1060, 0.1963, and 0.0991
respectively. These values were determined using a commercial laser
power meter (Coherent, Fieldmate, CA, USA).
[0122] FIG. 15B shows wavelength-resolved intensity compensation
factors for the sensors. These compensation factors can be used to
compensate for differences between the quantum efficiencies of
various sensors and between the attenuation ratios of the beam
splitters. For example, the power of each source 108 can be divided
by its relative intensity at a corresponding wavelength to
normalize the source powers to unit.
[0123] Escherichia coli O157:H7 EDL933, Listeria monocytogenes
F4244, Salmonella enteritidis PT21 and Staphylococcus aureus ATCC
25923 were selected as model organisms for the experiments. For the
agar plate preparation, all cultures were grown in 5 ml brain heart
infusion (BHI) (Difco, MD, USA) broth for 15 h at 37.degree. C. at
130 rpm in an incubator shaker. After incubating, the cultures were
serially diluted and surface plated on BHI agar plates (100
mm.times.15 mm) to achieve a bacterial counts of 50-100 CFU/plate.
The plates were incubated at 37.degree. C. until the size of the
colonies reached a diameter range of 900-1100 .mu.m. The diameters
of the bacterial colonies were measured using both a bright-field
microscope (Leica Microsystems, Bannockburn, Ill., USA) equipped
with CCD camera (Leica Microsystems, Leica DFC310 FX, Bannockburn,
Ill., USA) and Leica Application Suite V4.20 build 607 (Leica
Microsystems, Bannockburn, Ill., USA) using a 10.times. objective,
and the multispectral BARDOT described herein. For 1000 .mu.m
colony diameter of each genus bacteria, 10.5 h, 22.5 h, 11.5 h, and
13.5 h were used for culturing E. coli, Listeria, Salmonella, and
S. aureus, respectively. The thickness of the agar of each plate
was maintained at approximately 8 mm.
[0124] For the liquid sample preparation, a pure colony of each
genus was harvested and diluted in a single tube, and incubated for
12 hrs at 37.degree. C. Then, aliquots of the samples were
transferred to a disposable cuvette and each stock was serially
diluted 3 times at 1:10 ratio. ODs of the diluted samples at
300-800 nm were measured with a DU 800 spectrophotometer (Beckman
Coulter Inc., CA, USA). Spectral absorption curves were recorded
three times for each of 5 different samples (total 15 data sets)
for each genus, and average spectral response curves were
calculated. For quantitative comparison, the area under the curve
was calculated and this was used for normalization.
[0125] For the solid sample experiments, 5 plates of each genus
were prepared for a single-day dataset, and repeated on three
different days in order to accommodate the biological variability.
At least 20 points and colonies were measured per plate for only
the BHI agar area and the bacteria colony respectively. The
spectral OD of BHI agar is defined as in Eq. 22:
OD agar ( .lamda. ) = - log 10 ( I ( .lamda. ) agar I ( .lamda. )
input ) ( Eq . 22 ) ##EQU00012##
[0126] where I(.lamda.) refers to the intensity of light or other
output of the photodiode at wavelength .lamda.. The intensity can
be expressed, e.g., in volts or digital representations of volts.
The mean value of each OD was computed as 0.503, 0.129, 0.072 for
405 nm, 635 nm, and 904 nm respectively. Since it is difficult to
measure the actual OD of the bacterial colony without destroying
the colony structure on semi-solid agar, an indirect method can be
used to obtain the OD of the bacteria colony. OD of the colony was
computed as in Eq. 23:
OD colony ( .lamda. ) = - log 10 ( I ( .lamda. ) agar + colony I (
.lamda. ) input ) - OD agar ( .lamda. ) ( Eq . 23 )
##EQU00013##
[0127] Since the first term of the right side of the (Eq. 23) is
the combined OD of BHI agar and the colony, the OD of agar can be
subtracted to obtain the attenuation from the colony only.
[0128] An experiment was performed using Staphylococcus aureus ATCC
25923 (S. aureus). For agar plate preparation, the frozen S. aureus
stock was streaked on BHI agar plate, and grown at 37.degree. C.
incubator for 13 h. A single S. aureus colony was collected with a
sterilized loop, and grown in 5 ml brain heart infusion (BHI)
(Difco, MD, USA) broth for 15 h at 37.degree. C. at 130 rpm in an
incubator shaker to maintain a purity of the culture. After
incubating, the cultures were serially diluted and surface plated
on BHI agar plate (100 mm.times.15 mm) to achieve bacterial counts
of 50-100 CFU/plate. The plates were incubated at 37.degree. C.
until the size of the colonies reached a diameter range of
800.about.1100 .mu.m. 13-15 h of incubation time was necessary to
achieve this colony diameter. The diameters of the bacterial
colonies were measured using both a bright-field microscope (Leica
Microsystems, Bannockburn, Ill., USA) equipped with CCD camera
(Leica Microsystems, Leica DFC310 FX, Bannockburn, Ill., USA) and
Leica Application Suite V4.20 build 607 (Leica Microsystems,
Bannockburn, Ill., USA) with a 10.times. objective, and ICMA
(Integrated Colony Morphology Analyzer, Purdue University, IN,
USA). The agar thickness of each plate was kept at 8 mm to maintain
similar conditions between duplications.
[0129] FIGS. 16A-B and FIGS. 17A-D compare the spectroscopic OD
measurements for liquid and solid samples. This information can be
used for the selection of the best discriminative wavelength
region. For example, S. aureus showed almost 1/3 higher OD than L.
mono at 400 nm wavelength. For the four tested genera, the selected
wavelengths of 405 nm, 635 nm and 904 nm provide measurable
differences in spectral absorption at those wavelengths. OD was
also measured for solid samples. All genera showed monotonically
decreasing OD trends as the wavelength increased except for S.
aureus. By nature, liquid samples are more homogenously spread out
through the whole volume, so only the individual cell shape or
other particulates can be argued as the contributing factors for
the observed difference. However, in solid samples, growing
microbial films (bacterial colonies) have more characteristic
information beside the individual cell shape. For example, E. coli,
Listeria and Bacillus cells are all rod-shaped, but their colony
characteristics show dramatic differences. In addition, nutrition,
agar hardness, and environmental factors can change the morphology
of the solid colony. Therefore, an instrument such as multispectral
BARDOT that can capture the multimodal characteristics of a colony
can better provide differentiable traits from a given bacterial
sample.
[0130] FIG. 16A shows spectral absorption (OD) measurements of E.
coli, L. mono, S. enteritidis, and S. aureus in liquid BHI stock
(300-800 nm range) at a 1:100 dilution. All the genera showed peak
OD values near 400 nm, and ODs gradually decreased with increasing
wavelength. S. aureus had the highest OD value, while L. mono had
the lowest OD value among the tested genera. Different genera of
bacteria showed different spectral OD for incident wavelength, and
wavelengths were selected to provide effective OD
discrimination.
[0131] FIG. 16B shows the relative OD values of the interrogated
genera with respect to L. mono. Shown are spectral absorption
measurements representing relative spectral optical density (OD) of
common pathogens, Escherichia coli O157:H7 EDL933, Salmonella
enteritidis PT21, and Staphylococcus aureus ATCC25923, to Listeria
monocytogenes F4244. The vertical lines stand for example
wavelengths of the employed laser diode (LD). The 405, 635, and 904
nm laser lines were selected on the basis of line separation which
provided selected levels of OD difference between the interrogated
genera. In various examples, the OD information can be used as a
simple classification method for bacteria genera.
[0132] In various aspects, three pellicle beam splitters mounted in
optical cages can be adopted to avoid a ghost effect from the use
of thick plate beam splitters, and to provide improved alignment of
the light source and the CMOS camera. In addition, two Si
photodiodes (PD) can be included such that laser intensity can be
monitored before and after the laser passes the bacterial colony.
The ratio of the voltage readings or other data from those
photodiodes can provide the spectral OD of the whole colony. Each
diode laser can illuminate a bacterial colony sequentially to
capture spectral forward scattering images and OD, and overall
measurement time for each colony can be 3-4 sec.
[0133] FIG. 17A shows the measurement points, where input laser
intensity was measured, e.g., at PD 114 (FIG. 1A), while the media
with one or more bacterial colonies were measured at PD 216 of FIG.
2.
[0134] FIG. 17B illustrates that the spectral absorption from bare
agar areas displays similar characteristics to the liquid samples.
When the wavelength increases, the OD of the BHI agar
decreases.
[0135] FIG. 17C illustrates that the net OD ((agar+colony)-agar)
from bacterial colonies generally shows a decrease as the
wavelength increases. S. aureus shows a peak value of 0.38 for net
OD from 635 nm, while S. entertidis shows a value of 0.22 (FIG.
17C). The net OD at 904 nm for all genera was the lowest among the
examined wavelengths. When the samples were dissolved in a liquid
format, Listeria showed the lowest OD. Compared to the liquid
sample result, the OD of L. mono on BHI agar is not the lowest OD
through the interrogated wavelengths. S. enteritidis shows the
lowest OD at 635 nm and 904 nm. Since single-wavelength OD values
provide limited differentiability among genera, three-wavelength
combinations of the OD can be utilized as a first step
classification method for the tested genera.
[0136] FIG. 17D illustrates spectral OD differences calculated to
enhance the differentiation of the spectral OD in various aspects.
The X-axis represents three combinations of the OD difference
(#1=OD.sub.405-OD.sub.635, #2=OD.sub.405-OD.sub.904, and
#3=OD.sub.635-OD.sub.904). Using this method, differences between
the genera are visually enhanced, and it becomes easy to recognize
their OD trend across wavelengths. For instance, S. aureus has a
negative OD difference at #1. Other genera show 0 to 0.07 OD
difference at #1, which means those genera have similar ODs at 405
nm and 635 nm. At #2, the OD difference between the 405 nm and 904
nm regions, L. mono has the highest OD differences, while S. aureus
has the smallest OD difference. E. coli and S. enteritidis have
similar OD difference trends at both #2 and #3 combinations.
[0137] FIG. 18A-18D show prediction results of multispectral
forward scattering pattern for a bacterial colony based on spectral
ELS modeling. FIGS. 18A-18C are shown with black and white reversed
for clarity of presentation. S. aureus is selected as a target
micro-organism since it has a bell-curve with tailing edge colony
shape and it generates a concentric circle shaped forward
scattering pattern. For the prediction, the colony diameter is set
at 1000 .mu.m, and the aspect ratio (colony center height to
diameter ratio) is set as 1:6.25. The wavelength term is found in
both the amplitude and the phase components of the modeling (Eq.
16). As the predicted model shows, the pattern size and the number
of rings decrease, while ring width and ring gap increase with
longer wavelengths of incident light.
[0138] Table 2 shows the computed result for the maximum
diffraction angle and the number of diffraction rings from Eq. 20
and 21, and shows a good match with the result from the
modeling.
TABLE-US-00002 TABLE 2 Half of maximum Number of .lamda. (nm)
N.sub.bac diffraction angle (rad) diffraction rings (ea) 405 1.4834
0.0483 94 635 1.4684 0.0468 60 405 1.4623 0.0462 40
[0139] FIGS. 18A-18C show simulated scattering patterns at 405 nm,
635 nm, and 904 nm, respectively. FIG. 18D shows a cross-sectional
view near the boundary region of the predicted patterns.
[0140] FIG. 19 shows the spectral forward scattering patterns of
the four example genera of bacteria on semi-solid BHI agar as
measured at three wavelengths by an MS-BARDOT instrument such as
described above with reference to FIGS. 1A-5A. The pellicle beam
splitter 214 (FIG. 2) permits capturing the forward scattering
patterns and the OD of the bacteria colony simultaneously. Visual
inspection of the spectral scatter patterns shows qualitative
differences in the patterns. Observing the difference along a
column of FIG. 19 shows that different bacteria show different
prominent patterns. For E. coli and S. enteritis samples, the
405-nm patterns show fine structures of spokes, speckles, and
rings, unlike the patterns collected at the other wavelengths. For
L. mono samples, the 405-nm pattern has the largest diameter
pattern, and the central portion of the pattern indicates higher
signal intensity when compared with 635 and 904 nm.
[0141] Bacterial colonies have two major regions: edge regions that
are generally less dense, have greater water content, and wherein
division of cells occurs, and the center part. The pattern
information can provide some understanding on how the bacteria are
spreading at the edge and how cells are accumulating at the center
part. One organism that uniquely stands out is S. aureus. The
patterns show very weak rings at the edge, with little detail
information that can be extracted (except for 635 nm).
[0142] FIG. 20A shows the image of FIG. 19 (spectral forward
scattering patterns of S. aureus for 405, 635, and 904 nm)
transformed into rectangular coordinates, where X and Y axes
represent angle and radius, respectively. This polar pattern
representation reveals a clear ripple structure in the 635-nm
patterns, while the 405-nm and 904-nm patterns show low intensity
outside of the central bright spots. The average intensity across
the whole circular regions shows the ripple structure for 635 nm
and 904 nm, e.g., for radius values between 75 and 120, though the
latter wavelength shows 50% lower average intensity. For 405 nm,
average intensity is comparable to the 635 nm patterns, but very
small speckle patterns are observed without any ring structures
(FIG. 20B).
[0143] FIG. 21 is a graph showing classification accuracy for
various numbers of features.
[0144] Multispectral forward scattering pattern and OD based
bacterial phenotyping techniques according to various examples
herein can measure three different wavelengths (405, 635, and 904
nm) of both forward scattering patterns and ODs for a target
bacterial colony simultaneously. Utilizing stackable pellicle beam
splitters structure, some examples reduce unexpected optical side
effects such as ghost effects. Some examples can be readily
expanded to include light sources of additional wavelengths. Using
pseudo-Zernike (GPZ) polynomials/moment, the results of the four
different bacterial genera were analyzed and classified.
[0145] The spectral-scatter patterns were analyzed as described
above using GPZ moments as features. In some examples, since three
separate laser wavelengths are used, the number of extracted
features per colony is three times larger than in a
single-wavelength system.
[0146] The feature extraction/recognition of scatter patterns was
performed using pseudo-Zernike (GPZ) polynomials/moments. The GPZ
polynomials are formally defined as in Eq. 24:
k p .lamda. .alpha. ( z ) = z p + .lamda. 2 ( z * ) p + .lamda. 2 (
.alpha. + 1 ) p - .lamda. ( p - .lamda. ) ! 2 F 1 ( - p + .lamda. ,
- p - .lamda. - 1 ; .alpha. + 1 ; 1 - 1 ( zz * ) 1 / 2 ) ( Eq . 24
) ##EQU00014##
[0147] where * denotes the complex conjugate and z=re.sup.j.theta..
The parameter a is user-selectable, and scales the polynomial
values. The repetition X is set to be between 0 and p.
[0148] The polynomial is defined in polar coordinates as in Eq.
25:
k.sub.p.lamda..sup..alpha.(r,.theta.)=k.sub.p.lamda..sup..alpha.(re.sup.-
j.theta.)=R.sub.p.lamda..sup..alpha.(r)e.sup.j.lamda..theta., (Eq.
25)
[0149] Where the real-values radial polynomial
R.sub.p.lamda..sup..alpha.(r) is given by Eq. 26:
R p .lamda. .alpha. ( r ) = ( p + .lamda. + 1 ) ! ( .alpha. + 1 ) p
+ .lamda. + 1 s = 0 p - .lamda. ( - 1 ) s ( .alpha. + 1 ) 2 p + 1 -
s s ! ( p - .lamda. - s ) ! ( p + .lamda. + 1 - s ) ! r p - s ( Eq
. 26 ) ##EQU00015##
[0150] The radial polynomial R.sub.p.lamda..sup..alpha.(r) is
computed using the recurrence relation given in Eq. 27:
R.sub.p.lamda..sup..alpha.(r)=(M.sub.1r+M2)R.sub.p-1,.lamda..sup..alpha.-
(r)+M.sub.3R.sub.p-2,.lamda..sup..alpha.(r), (Eq. 27)
with the following coefficients in Eq. 28:
M 1 = ( 2 p + 1 + .alpha. ) ( 2 p + .alpha. ) ( p + .lamda. + 1 +
.alpha. ) ( p - .lamda. ) M 2 = ( p + .lamda. + 1 ) ( .alpha. + 2 p
) ( p + .lamda. + .alpha. + 1 ) + M 1 ( p + .lamda. ) ( p - .lamda.
- 1 ) 2 p - 1 + .alpha. M 3 = ( p + .lamda. ) ( p + .lamda. + 1 ) (
2 p - 2 + .alpha. ) ( 2 p - 1 + .alpha. ) 2 ( p + .lamda. + .alpha.
+ 1 ) ( p + .lamda. + .alpha. ) ++ M 2 ( p + .lamda. ) ( 2 p - 2 +
.alpha. ) p + .lamda. + .alpha. - M 1 ( p + .lamda. ) ( p + .lamda.
- 1 ) ( p - .lamda. - 2 ) 2 ( p + .lamda. + .alpha. ) ( Eq . 28 )
##EQU00016##
and in Eq. 29:
[0151] R.sub..lamda..lamda..sup..alpha.(r)=r.sup..lamda.
R.sub..lamda.+1,.lamda..sup..alpha.(r)=[(.alpha.+3+2.lamda.)r-2(.lamda.+-
1)]R.sub..lamda..lamda..sup..alpha.(r) (Eq. 29)
[0152] Various systems based on monochromatic elastic light scatter
produce features which can lead to high classification accuracies.
Therefore, performance of various examples can be evaluated using
sensitivity and specificity of the tested systems. Consequently,
robust increases in classification success in the range of 1-2% can
be provided.
[0153] In various aspects, feature selection can be based on a
random forest algorithm (RF), in which for every run the RF selects
a random features subset and generates a classification tree. The
importance of the analyzed features can be determined by the
accuracy of these trees. In various aspects, the improvement of
classification can be related to the increased feature numbers, and
the range of 10-20 features can be used (See FIG. 21). In an
experiment, 15 was selected as the number of features. After
determining the subset of best features in single wavelength and
multiple wavelength settings using random forests, classification
was performed using standard SVM implementation with a linear
kernel. The performance of the classifiers was determined by
5.times.2 crossvalidation. The entire crossvalidation procedure was
repeated 10 times with different seeds for a random number
generator. The final results are reported in Tables 3-6. Tables 3-6
show confusion matrices (as percentages) with accompanying standard
deviations, for 405 nm (A), 635 nm (B), 904 nm (C), and a mix of
features from all three wavelengths (D), respectively.
TABLE-US-00003 TABLE 3 (A) E. coli Listeria Salmonella Staph. E.
coli 93.75 (0.38) 0.43 (0.49) 1.25 (0.14) 2.5 (0.17) Listeria 3.64
(0.256) 99.16 (0.58) 0.23 (0.22) 0.13 (0.168) Salmonella 2.57
(0.302) 0.43 (0.27) 96.12 (0.3) 0.66 (0) Staph. 0.00 (0) 0 (0) 2.39
(0.46) 96.73 (0)
TABLE-US-00004 TABLE 4 (B) E. coli Listeria Salmonella Staph. E.
98.85 (0.233) 0 (0) 0.66 (0) 0 (0) coli Liste- 1.14 (0.23) 99.35
(0) 0.65 (0) 0 (0) ria Salmo- 0.00 (0) 0.66 (0) .sup. 98.59 (0.22)
0 (0) nella Staph. 0.00 (0) 0 (0) 0.1 (0.22) 100 (0)
TABLE-US-00005 TABLE 5 (C) E. coli Listeria Salmonella Staph. E.
98.32 (0.423) 1.12 (0.32) 0.13 (0.23) 0.33 (0.219) coli Liste- 1.27
(0.284) 98.9 (0.31) 1.82 (0.38) 0 (0) ria Salmo- 0.39 (0.34) 0 (0)
95.76 (0.68) 0.07 (0.139) nella Staph. 0.00 (0) 0 (0) 2.25 (0.36)
99.61 (0.207)
TABLE-US-00006 TABLE 6 (D) E. coli Listeria Salmonella Staph. E.
99.93 (0.139) 0 (0) 0.1 (0.16) 0.59 (0.139) coli Liste- 0.00 (0)
100 (0) 1.33 (0.42) 0 (0) ria Salmo- 0.07 (0.139) 0 (0) 98.45
(0.47) 0 (0) nella Staph. 0.00 (0) 0 (0) 0.1 (0.16) 99.41
(0.138)
[0154] FIGS. 22A-22D are three-dimensional linear-discriminant
analysis plots showing feature-space distances between colonies of
different types of micro-organisms. The linear discriminant
analysis plots show the structure of the data point clouds in 3-D,
and illustrate differences between classifiers determined from
different features sets (FIGS. 22A-22D).
[0155] A variety of embodiments of MS-BARDOT instrument provide a
stepping motor and right angle gold mirror to physically move the
three lasers sequentially over a distance, which can be e.g. 10 mm,
over a period of seconds, during which the patterns at the
different wavelengths can be recorded. One benefit of this set of
embodiments is it maintains the optics-free design of single
wavelength BARDOT, which reduces stray scattered light that might
affect the acquired scatter patterns.
[0156] A variety of embodiments of MS-BARDOT instrument provide a
laser source module which incorporates multiple, e.g., three,
incident laser wavelengths, a photodiode, and one or more, e.g.,
two, pellicle beam splitters. Some of these embodiments provide
multiple wavelength laser sources in a compact system or permit
acquisition of multiple wavelength images in a rapid manner (e.g.,
.about.3.5 sec per colony). Various examples include an additional
photodiode configured to acquire absorption data during
irradiation, permitting monitoring the input intensity. In a
variety of aspects, a pellicle beam splitter can reduce or remove
multiple-reflection images from prior two-beam splitters.
[0157] Due the spectral nature of the new modeling approach, all
the derived formulas include the wavelength term. In addition, the
spectral dependency of the refractive index plays an important role
in calculating the two major characteristics of the scatter
patterns ( ). As discussed above with reference to FIG. 10, both
the number of rings and the half diffraction angle show an
inversely proportional relationship to the incident wavelength and
correlated with spectrally-varying refractive index. In some
examples, due to the biological growth nature of the bacterial
colony, the colony forms a dense core area and a rim area where
cells are constantly dividing and expanding the boundary. The
diffraction patterns can be correlated with the biological
structure in the rim area, e.g., exhibiting a lower outermost-ring
intensity than the computed simulation for a Gaussian shape.
[0158] Multispectral forward scattering can provide valuable
information regarding the bacterial colony. A benefit of a variety
of aspects is that optical absorption data (e.g., optical density,
or OD) can be incorporated into a Zernike or other spatial scatter
pattern analysis. This can permit interrogating different kinds of
bacterial colonies, since some pathogenic and non-pathogenic
bacteria have different extracellular material such as
capsular-polysaccharides. Combining spatial scattering patterns and
optical absorption can provide improved resolution and
classification in different phylogenic bacteria. A further benefit
is that understanding of the multispectral system allows expansion
of it to a hyperspectral forward scatterometer which can be
designed with acousto-optic tunable filters (AOTF) and
super-continuum lasers.
[0159] In a variety of aspects, cage-mounted pellicle beam
splitters can reduce ghosting. The cage system itself provides
proper alignment of the incident light. Other optical mounting
systems can also be used. Various aspects include three lasers, a
translation stage, and a CMOS camera control which includes one
IEEE1394 port, seven digital input/outputs (I/O), and two
analog-to-digital converters (ADC). Calibration at each wavelength
can be performed to accommodate different reflectance/transmission
ratios from the pellicle beam splitter and spectral quantum
efficiencies from the CMOS camera. The incoming spectral intensity
can be measured and compensated for each wavelength such that
approximately the same intensity is perceived by the CMOS camera
(FIG. 2).
[0160] Throughout this description, some aspects are described in
terms that would ordinarily be implemented as software programs.
Those skilled in the art will readily recognize that the equivalent
of such software can also be constructed in hardware, firmware, or
micro-code. Because data-manipulation algorithms and systems are
well known, the present description is directed in particular to
algorithms and systems forming part of, or cooperating more
directly with, systems and methods described herein.
[0161] FIG. 23 shows a flowchart illustrating an exemplary method
for, e.g., training or employing a computational model to identify
types of micro-organisms in colonies. Also shown are data produced
by some of the blocks. The blocks can be performed in any order
except when otherwise specified, or when data from an earlier block
is used in a later block. In at least one example, processing
begins with block 2305, block 2310, or block 2320. For clarity of
explanation, reference is herein made to various components shown
in FIG. 1A, 1B, 2, 3A-3E, 4A, 4B, 5A, 14A, or 14B that can carry
out or participate in the steps of the exemplary method. It should
be noted, however, that other components can be used; that is,
exemplary method(s) shown in FIG. 2 are not limited to being
carried out by the identified components.
[0162] In some examples, at block 2305, images are captured of
colonies of micro-organisms, e.g., under irradiation of one or more
wavelength(s). For example, the images can be captured using an
imager during irradiation of corresponding ones of the colonies
with corresponding ones of the wavelengths. In some examples, the
images can be captured using a multispectral transmissive system
such as that described above with reference to FIG. 1A, 1B, 2, 14A,
or 14B; a multispectral reflective or transmissive/reflective
system such as that described above with reference to FIG. 3A-3E,
4A, 4B, or 5A; or a reflective system such as that described above
with respect to transmissive/reflective systems, omitting or not
using the transmissive portion.
[0163] At block 2310, feature values are determined based at least
in part on images, e.g., training images, of colonies of
micro-organisms under irradiation of different wavelengths. The
images can be, e.g., forward or reverse scatter images such as
discussed above with reference to, e.g., FIGS. 5B-5I. The feature
values can be determined, e.g., using Zernike polynomials such as
discussed above with reference to FIG. 21. For example, at least
some of the feature values can include one or more Zernike or
pseudo-Zernike moments for individual ones of the images.
[0164] In some examples, at block 2310, a first feature value of
the at least some of the determined feature values is determined
based at least in part on a first one of the images corresponding
to irradiation of a first one of the wavelengths. A second feature
value of the at least some of the determined feature values is
determined based at least in part on a second one of the images
corresponding to irradiation of a second, different one of the
wavelengths. For example, the images can include a 405 nm image of
a colony and a 635 nm image of the same colony. One of the feature
values can be, e.g., the Z.sub.2.sup.0 Zernike moment of the 405 nm
image, and another one of the feature values can be, e.g., the
Z.sub.2.sup.0 Zernike moment of the 635 nm image.
[0165] At block 2315, at least some of the determined feature
values are clustered based at least in part on
colony-identification values of the images. The
colony-identification values can be, e.g., values representing the
genus, species, sub-species (e.g., serovar), or other type of
micro-organism. Each image can be associated with such a value. For
example, the three images in the left-hand column of FIG. 19 can
share a colony-identification value corresponding to E. coli
O157:H7. The clustering can be done using, e.g., support vector
machines (SVMs), hierarchical or centroid-based clustering
algorithms, distribution- or density-based clusters, or other
clustering algorithms. In some examples, the clustering comprises
training a classification model, e.g., an SVM, using the at least
some of the determined feature values as training data and the
colony-identification values as class data. Clusters of feature
values in feature space can be associated with
colony-identification values.
[0166] At block 2320, e.g., after block 2315, test feature values
can be determined based at least in part on images of a test
micro-organism colony under irradiation of different wavelengths.
The images can be scatter images. The feature values can be
determined, e.g., as discussed above with reference to block
2310.
[0167] At block 2325, a test colony-identification value of the
test micro-organism colony can be determined by applying the test
feature values to the trained classification model. For example,
the test colony-identification value can be selected as the
colony-identification value associated with the cluster of the
trained classification model to which the test feature values
belong.
[0168] In some examples, block 2315 includes blocks 2375, 2380,
2385, 2390, or 2395.
[0169] At block 2375, multiple subsets of the determined feature
values are selected. For example, the subsets can be selected
randomly. In some examples, a random forest algorithm is used, as
discussed above with reference to FIG. 21.
[0170] At block 2380, candidate classification models can be
trained for respective ones of the subsets. For example, clustering
can be performed separately based on each subset. The training can
be done, e.g., as described above with reference to block 2315.
[0171] At block 2385, accuracy values are determined for respective
ones of the trained candidate classification models. For example,
feature values of evaluation images not included in the training
can be applied to the models.
[0172] At block 2390, at least some of the determined feature
values are selected based at least in part on the determined
accuracy values. This permits determining, for a specific training
set of images, which combination of features permits most
effectively distinguishing micro-organism types from each other or
identifying micro-organism types.
[0173] At block 2395, the clustering can be performed using the
determined feature values. For example, an SVM can be trained using
the determined feature values, as discussed above with reference to
FIG. 21.
[0174] In view of the foregoing, various aspects provide
measurement of bacterial colonies and analysis of measured data. A
technical effect of some examples is to determine the type of
bacteria growing in a measured bacterial colony. A further
technical effect of some examples is to control operation of, e.g.,
an X-Y stage or a laser source to successively irradiate one or
more colonies with light of one or more wavelengths.
[0175] FIG. 24 is a high-level diagram showing the components of an
exemplary data-processing system 2401 for analyzing data and
performing other analyses described herein, and related components.
The system 2401 includes a processor 2486, a peripheral system
2420, a user interface system 2430, and a data storage system 2440.
The peripheral system 2420, the user interface system 2430 and the
data storage system 2440 are communicatively connected to the
processor 2486. Processor 2486 can be communicatively connected to
network 2450 (shown in phantom), e.g., the Internet or a leased
line, as discussed below. Sequence controller 104 or other labeled
components shown in FIG. 1B, X-Y translation stage 218 (FIG. 2), or
labeled components of FIG. 3A or 5A, can each include one or more
of systems 2486, 2420, 2430, 2440, and can each connect to one or
more network(s) 2450. Processor 2486, and other processing devices
described herein, can each include one or more microprocessors,
microcontrollers, field-programmable gate arrays (FPGAs),
application-specific integrated circuits (ASICs), programmable
logic devices (PLDs), programmable logic arrays (PLAs),
programmable array logic devices (PALs), or digital signal
processors (DSPs).
[0176] Processor 2486 can implement processes of various aspects
described herein, e.g., with reference to FIG. 23. Processor 2486
and related components can, e.g., carry out processes for operating
imaging systems to capture images of colonies or processes for
analyzing image data to train computational models or identify
bacteria.
[0177] Processor 2486 can be or include one or more device(s) for
automatically operating on data, e.g., a central processing unit
(CPU), microcontroller (MCU), desktop computer, laptop computer,
mainframe computer, personal digital assistant, digital camera,
cellular phone, smartphone, or any other device for processing
data, managing data, or handling data, whether implemented with
electrical, magnetic, optical, biological components, or
otherwise.
[0178] The phrase "communicatively connected" includes any type of
connection, wired or wireless, for communicating data between
devices or processors. These devices or processors can be located
in physical proximity or not. For example, subsystems such as
peripheral system 2420, user interface system 2430, and data
storage system 2440 are shown separately from the data processing
system 2486 but can be stored completely or partially within the
data processing system 2486.
[0179] The peripheral system 2420 can include or be communicatively
connected with one or more devices configured or otherwise adapted
to provide digital content records to the processor 2486 or to take
action in response to processor 186. For example, the peripheral
system 2420 can include digital still cameras, digital video
cameras, cellular phones, or other data processors. The processor
2486, upon receipt of digital content records from a device in the
peripheral system 2420, can store such digital content records in
the data storage system 2440. In the illustrated example,
peripheral system 2420 is communicatively connected to control
laser(s) or a stage (e.g., a stage holding an agar plate with a
colony growing thereon), and to receive information from imager(s)
or photodiode(s) collecting light above or below (on a forward or
reverse side of) the colony.
[0180] The user interface system 2430 can convey information in
either direction, or in both directions, between a user 2438 and
the processor 2486 or other components of system 2401. The user
interface system 2430 can include a mouse, a keyboard, another
computer (connected, e.g., via a network or a null-modem cable), or
any device or combination of devices from which data is input to
the processor 2486. The user interface system 2430 also can include
a display device, a processor-accessible memory, or any device or
combination of devices to which data is output by the processor
2486. The user interface system 2430 and the data storage system
2440 can share a processor-accessible memory.
[0181] In various aspects, processor 2486 includes or is connected
to communication interface 2415 that is coupled via network link
2416 (shown in phantom) to network 2450. For example, communication
interface 2415 can include an integrated services digital network
(ISDN) terminal adapter or a modem to communicate data via a
telephone line; a network interface to communicate data via a
local-area network (LAN), e.g., an Ethernet LAN, or wide-area
network (WAN); or a radio to communicate data via a wireless link,
e.g., WIFI or GSM. Communication interface 2415 sends and receives
electrical, electromagnetic or optical signals that carry digital
or analog data streams representing various types of information
across network link 2416 to network 2450. Network link 2416 can be
connected to network 2450 via a switch, gateway, hub, router, or
other networking device.
[0182] In various aspects, system 2401 can communicate, e.g., via
network 2450, with a data processing system 2402, which can include
the same types of components as system 2401 but is not required to
be identical thereto. Systems 2401, 2402 are communicatively
connected via the network 2450. Each system 2401, 2402 executes
computer program instructions to, e.g., operate measurement
instruments or analyze data. In an example, system 2401
operates
[0183] Processor 2486 can send messages and receive data, including
program code, through network 2450, network link 2416 and
communication interface 2415. For example, a server can store
requested code for an application program (e.g., a JAVA applet) on
a tangible non-volatile computer-readable storage medium to which
it is connected. The server can retrieve the code from the medium
and transmit it through network 2450 to communication interface
2415. The received code can be executed by processor 2486 as it is
received, or stored in data storage system 2440 for later
execution.
[0184] Data storage system 2440 can include or be communicatively
connected with one or more processor-accessible memories configured
or otherwise adapted to store information. The memories can be,
e.g., within a chassis or as parts of a distributed system. The
phrase "processor-accessible memory" is intended to include any
data storage device to or from which processor 2486 can transfer
data (using appropriate components of peripheral system 2420),
whether volatile or nonvolatile; removable or fixed; electronic,
magnetic, optical, chemical, mechanical, or otherwise. Exemplary
processor-accessible memories include but are not limited to:
registers, floppy disks, hard disks, tapes, bar codes, Compact
Discs, DVDs, read-only memories (ROM), erasable programmable
read-only memories (EPROM, EEPROM, or Flash), and random-access
memories (RAMs). One of the processor-accessible memories in the
data storage system 2440 can be a tangible non-transitory
computer-readable storage medium, i.e., a non-transitory device or
article of manufacture that participates in storing instructions
that can be provided to processor 2486 for execution.
[0185] In an example, data storage system 2440 includes code memory
2441, e.g., a RAM, and disk 2443, e.g., a tangible
computer-readable rotational storage device or medium such as a
hard drive. Computer program instructions are read into code memory
2441 from disk 2443. Processor 2486 then executes one or more
sequences of the computer program instructions loaded into code
memory 2441, as a result performing process steps described herein.
In this way, processor 2486 carries out a computer implemented
process. For example, steps of methods described herein, blocks of
the flowchart illustrations or block diagrams herein, and
combinations of those, can be implemented by computer program
instructions. Code memory 2441 can also store data, or can store
only code.
[0186] Various aspects described herein may be embodied as systems
or methods. Accordingly, various aspects herein may take the form
of an entirely hardware aspect, an entirely software aspect
(including firmware, resident software, micro-code, etc.), or an
aspect combining software and hardware aspects These aspects can
all generally be referred to herein as a "service," "circuit,"
"circuitry," "module," or "system."
[0187] Furthermore, various aspects herein may be embodied as
computer program products including computer readable program code
("program code") stored on a computer readable medium, e.g., a
tangible non-transitory computer storage medium or a communication
medium. A computer storage medium can include tangible storage
units such as volatile memory, nonvolatile memory, or other
persistent or auxiliary computer storage media, removable and
non-removable computer storage media implemented in any method or
technology for storage of information such as computer-readable
instructions, data structures, program modules, or other data. A
computer storage medium can be manufactured as is conventional for
such articles, e.g., by pressing a CD-ROM or electronically writing
data into a Flash memory. In contrast to computer storage media,
communication media may embody computer-readable instructions, data
structures, program modules, or other data in a modulated data
signal, such as a carrier wave or other transmission mechanism. As
defined herein, computer storage media do not include communication
media. That is, computer storage media do not include
communications media consisting solely of a modulated data signal,
a carrier wave, or a propagated signal, per se.
[0188] The program code includes computer program instructions that
can be loaded into processor 2486 (and possibly also other
processors), and that, when loaded into processor 2486, cause
functions, acts, or operational steps of various aspects herein to
be performed by processor 2486 (or other processor). Computer
program code for carrying out operations for various aspects
described herein may be written in any combination of one or more
programming language(s), and can be loaded from disk 2443 into code
memory 2441 for execution. The program code may execute, e.g.,
entirely on processor 2486, partly on processor 2486 and partly on
a remote computer connected to network 2450, or entirely on the
remote computer.
Example Clauses
[0189] Throughout these example clauses, parenthetical remarks are
examples and are not limiting. Examples given in the parenthetical
remarks of specific example clauses can also apply to the same
terms appearing elsewhere in these example clauses.
[0190] A: A system for the identification of micro-organisms, the
system comprising: an irradiation unit (e.g., including sources
108A, 108B, and 108C, and beamsplitters 106A, 106B, all FIG. 1A)
adapted to sequentially provide coherent electromagnetic radiation
of multiple wavelengths along a common optical path (horizontal
axis in FIGS. 14A, 14B); a holder (fork structure shown in FIG. 2
under petri dish 110 and shown in FIG. 3A under the bacteria agar
plate) adapted to retain a substrate having a surface adapted for
growth of a micro-organism colony (petri dish 110, FIGS. 1A, 2); a
beamsplitter (304) adapted to direct the coherent electromagnetic
radiation from the common optical path towards the retained
substrate; and an imager (backward scattering pattern grabber 306,
FIGS. 3A and 3E) arranged opposite the beamsplitter from the
retained substrate and adapted to obtain images of
backward-scattered light patterns from the micro-organism colony
("bacterial colony," FIG. 1A) irradiated by the respective
wavelengths of the directed coherent electromagnetic radiation.
[0191] B: The system according to paragraph A, further comprising:
a stage ("2-axis lateral stage," FIG. 1B) adapted to translate the
retained substrate or the beamsplitter with respect to each other
so that the directed coherent electromagnetic radiation irradiates
the micro-organism colony. (For example, the substrate can move and
the beamsplitter remain stationary, the beamsplitter can move and
the substrate remain stationary, or both the substrate and the
beamsplitter can move.)
[0192] C: The system according to paragraph B, further comprising a
controller (104) configured to: operate the stage and the
irradiation unit to irradiate a first colony of a plurality of
micro-organism colonies on the retained substrate; operate the
imager to obtain a first image (e.g., FIG. 5B) and a second image
(e.g., FIG. 5D) of backward-scattered light patterns from the first
colony, the first image corresponding to a first wavelength and the
second image corresponding to a second, different wavelength;
subsequently (e.g., subsequent to the obtaining of the first image
and the second image), operate the stage and the irradiation unit
to irradiate a second colony of the plurality of micro-organism
colonies on the retained substrate; and operate the imager to
obtain a third image (e.g., FIG. 5E) and a fourth image (e.g., FIG.
5F) of backward-scattered light patterns from the second colony,
the third image corresponding to a third wavelength and the fourth
image corresponding to a fourth wavelength different from the third
wavelength.
[0193] D: The system according to any of paragraphs A-C, wherein
the irradiation unit comprises: multiple sources (108A, 108B, 108C)
for the respective wavelengths of the coherent electromagnetic
radiation; and one or more source beamsplitters (106A, 106B)
configured to direct the coherent electromagnetic radiation from
the sources to the common optical path.
[0194] E: The system according to paragraph D, wherein the sources
comprise respective lasers (e.g., laser diodes as described above,
or gas, dye, or solid lasers).
[0195] F: The system according to paragraph D or E, wherein the
source beamsplitters comprise respective pellicle
beamsplitters.
[0196] G: The system according to any of paragraphs D-F, wherein
the source beamsplitters comprise R45:T55 beamsplitters or
beamsplitters of other R:T ratios.
[0197] H: The system according to any of paragraphs D-G, wherein
the source beamsplitters comprise cage mounts or other optical
mounts.
[0198] I: The system according to any of paragraphs D-H, wherein
the one or more source beamsplitters consist of a number of
beamsplitters equal to the number of sources minus one.
[0199] J: The system according to any of paragraphs A-I, wherein
the wavelengths comprise one or more of 405 nm, 635 nm, or 904
nm.
[0200] K: The system according to any of paragraphs A-J, wherein
the irradiation unit further comprises a sensor configured to
detect a level value (e.g., intensity, power, radiance, irradiance,
or any other radiometric or photometric quantity indicative of
coherent electromagnetic radiation level detected by the sensor) of
the coherent electromagnetic radiation.
[0201] L: The system according to paragraph K, wherein: the level
value corresponds to a selected one of the wavelengths; and the
system further comprises a controller (124) responsive to the level
value and a selected set point (based, e.g., on sensor response) to
adjust an output level (e.g., drive power, voltage, or current,
actual watts or lumens out, or other quantities indicative or
determinative of coherent electromagnetic radiation level emitted
by the source(s)) of the coherent electromagnetic radiation of the
selected one of the wavelengths.
[0202] M: The system according to paragraph K or L, wherein the
controller is further configured to: determine respective level
values of the multiple wavelengths using the sensor; and adjust
respective output levels of the coherent electromagnetic radiation
of the respective ones of the wavelengths based at least in part on
the respective level values and a selected set point.
[0203] N: The system according to any of paragraphs K-M, wherein
the sensor is arranged substantially upstream of the beamsplitter
(e.g., closer to the source(s) than the beamsplitter) along the
common optical path (e.g., as part of laser source 302, FIG.
3A).
[0204] O: The system according to any of paragraphs K-N, wherein
the sensor is arranged optically between the beamsplitter and the
retained substrate (e.g., between 304 and the agar plate, FIG. 3A;
this can be done using a beamsplitter such as the lower
beamsplitter 106, FIG. 2).
[0205] P: The system according to any of paragraphs K-O, wherein:
the sensor is arranged optically upstream of the retained substrate
(e.g., PD #1 114, FIG. 2); and the system further comprises a
second sensor (e.g., PD #2 216, FIG. 2) arranged optically
downstream of the retained substrate and configured to detect a
second level value of the coherent electromagnetic radiation.
[0206] Q: The system according to paragraph P, further comprising a
computation unit (124 or 104, FIG. 1B) configured to determine an
optical density of the micro-organism colony irradiated by the
directed coherent electromagnetic radiation based at least in part
on the level value and the second level value.
[0207] R: The system according to any of paragraphs A-Q, further
comprising: a second imager (forward scattering pattern grabber,
FIG. 3E) arranged opposite the retained substrate from the
beamsplitter and adapted to obtain images of forward-scattered
light patterns from the micro-organism colony irradiated by the
respective wavelengths of the directed coherent electromagnetic
radiation.
[0208] S: The system according to paragraph R, further comprising:
a first sensor (114) arranged optically upstream of the retained
substrate and configured to detect a first level value of the
coherent electromagnetic radiation; and a second sensor (216)
arranged optically downstream of the retained substrate and
configured to detect a second level value of the coherent
electromagnetic radiation.
[0209] T: The system according to paragraph S, further comprising a
second beamsplitter (214) arranged between the retained substrate
and the second imager and configured to direct at least some
electromagnetic radiation passing through the retained substrate to
the second sensor.
[0210] U: The system according to any of paragraphs A-T, wherein
the second beamsplitter comprises a pellicle beamsplitter.
[0211] V: The system according to any of paragraphs A-U, wherein
the second beamsplitter comprises a plate beamsplitter coated with
a wideband antireflective coating.
[0212] W: The system according to any of paragraphs A-V, wherein
the second beamsplitter comprises an R45:T55 beamsplitter.
[0213] X: The system according to any of paragraphs A-W, wherein
the second beamsplitter comprises a cage mount.
[0214] Y: The system according to any of paragraphs A-X, wherein
the beamsplitter comprises a pellicle beamsplitter.
[0215] Z: The system according to any of paragraphs A-Y, wherein
the beamsplitter comprises a plate beamsplitter coated with a
wideband antireflective coating.
[0216] AA: The system according to any of paragraphs A-Z, wherein
the beamsplitter comprises an R45:T55 beamsplitter.
[0217] AB: The system according to any of paragraphs A-AA, wherein
the beamsplitter comprises a cage mount.
[0218] AC: The system according to any of paragraphs A-AB, wherein
the optical path between the retained substrate and the imager
consists of one or more non-focusing optical elements (e.g.,
beamsplitters such as pellicle beamsplitters or polarizing
beamsplitters, mirrors, prism-based reflectors, or other elements
not having a focal distance or otherwise configured to direct light
without focusing the light).
[0219] AD: A system for the identification of micro-organisms, the
system comprising: an irradiation unit adapted to provide coherent
electromagnetic radiation of a selected wavelength along an optical
path; a holder adapted to retain a substrate having a surface
adapted for growth of a micro-organism colony; a beamsplitter
adapted to direct the coherent electromagnetic radiation from the
optical path towards the retained substrate; and an imager arranged
opposite the beamsplitter from the retained substrate and adapted
to obtain an image of a backward-scattered light pattern from the
micro-organism colony irradiated by the directed coherent
electromagnetic radiation.
[0220] AE: The system according to paragraph AD, further
comprising: a stage adapted to translate the retained substrate or
the beamsplitter with respect to each other so that the directed
coherent electromagnetic radiation irradiates the micro-organism
colony.
[0221] AF: The system according to paragraph AE, further comprising
a controller configured to: operate the stage and the irradiation
unit to successively irradiate ones of a plurality of
micro-organism colonies on the retained substrate; and operate the
imager to obtain a plurality of images of backward-scattered light
patterns from the successively-irradiated micro-organism colonies,
the plurality of images including at least first and second images
of a first colony at respective, different wavelengths, and third
and fourth images of a second, different colony at respective,
different wavelengths.
[0222] AG: The system according to any of paragraphs AD-AF, wherein
the irradiation unit further comprises a sensor configured to
detect a level value of the coherent electromagnetic radiation.
[0223] AH: The system according to paragraph AG, further comprising
a controller responsive to the level value and a selected set point
to adjust an output level of the coherent electromagnetic
radiation.
[0224] AI: The system according to any of paragraphs AD-AH, wherein
the beamsplitter comprises a pellicle beamsplitter.
[0225] AJ: The system according to any of paragraphs AD-AI, wherein
the beamsplitter comprises a plate beamsplitter coated with a
wideband antireflective coating.
[0226] AK: The system according to any of paragraphs AD-AJ, wherein
the beamsplitter comprises an R45:T55 beamsplitter.
[0227] AL: The system according to any of paragraphs AD-AK, wherein
the beamsplitter comprises a cage mount.
[0228] AM: The system according to any of paragraphs AD-AL, wherein
the optical path between the retained substrate and the imager
consists of one or more non-focusing optical elements.
[0229] AN: A system for the identification of micro-organisms, the
system comprising: an irradiation unit adapted to sequentially
provide coherent electromagnetic radiation of multiple wavelengths
along a common optical path; a holder adapted to retain a substrate
having a surface adapted for growth of a micro-organism colony in
operative arrangement to receive the coherent electromagnetic
radiation along the common optical path; and an imager arranged
optically downstream of the retained substrate and adapted to
obtain images of forward-scattered light patterns from the
micro-organism colony irradiated by the respective wavelengths of
the directed coherent electromagnetic radiation.
[0230] AO: The system according to paragraph AN, further
comprising: a stage adapted to translate the retained substrate or
irradiation unit with respect to each other so that the directed
coherent electromagnetic radiation irradiates the micro-organism
colony.
[0231] AP: The system according to paragraph AO, further comprising
a controller configured to: operate the stage and the irradiation
unit to successively irradiate ones of a plurality of
micro-organism colonies on the retained substrate; and operate the
imager to obtain a plurality of images of backward-scattered light
patterns from the successively-irradiated micro-organism colonies,
the plurality of images including at least first and second images
of a first colony at respective, different wavelengths, and third
and fourth images of a second, different colony at respective,
different wavelengths.
[0232] AQ: The system according to any of paragraphs AN-AP, wherein
the irradiation unit comprises: multiple sources for the respective
wavelengths of the coherent electromagnetic radiation; and one or
more source beamsplitters configured to direct the coherent
electromagnetic radiation from the sources to the common optical
path.
[0233] AR: The system according to paragraph AQ, wherein the
sources comprise respective lasers.
[0234] AS: The system according to paragraph AQ or AR, wherein the
source beamsplitters comprise respective pellicle
beamsplitters.
[0235] AT: The system according to any of paragraphs AQ-AS, wherein
the source beamsplitters comprise R45:T55 beamsplitters.
[0236] AU: The system according to any of paragraphs AQ-AT, wherein
the source beamsplitters comprise cage mounts.
[0237] AV: The system according to any of paragraphs AQ-AU, wherein
the one or more source beamsplitters consist of a number of
beamsplitters equal to the number of sources minus one.
[0238] AW: The system according to any of paragraphs AN-AV, wherein
the wavelengths comprise one or more of 405 nm, 635 nm, or 904
nm.
[0239] AX: The system according to any of paragraphs AN-AW, wherein
the irradiation unit further comprises a sensor configured to
detect a level value of the coherent electromagnetic radiation.
[0240] AY: The system according to paragraph AX, wherein: the level
value corresponds to a selected one of the wavelengths; and the
system further comprises a controller responsive to the level value
and a selected set point to adjust an output level of the coherent
electromagnetic radiation of the selected one of the
wavelengths.
[0241] AZ: The system according to paragraph AX or AY, wherein the
controller is further configured to: determine respective level
values of the multiple wavelengths using the sensor; and adjust
respective output levels of the coherent electromagnetic radiation
of the respective ones of the wavelengths based at least in part on
the respective level values and a selected set point.
[0242] BA: The system according to any of paragraphs AX-AZ, wherein
the sensor is arranged substantially upstream of the retained
substrate along the common optical path.
[0243] BB: The system according to paragraph BA, further comprising
a second sensor arranged optically downstream of the retained
substrate and configured to detect a second level value of the
coherent electromagnetic radiation.
[0244] BC: The system according to paragraph BB, further comprising
a computation unit configured to determine an optical density of
the micro-organism colony irradiated by the directed coherent
electromagnetic radiation based at least in part on the level value
and the second level value.
[0245] BD: The system according to paragraph BB or BC, further
comprising a beamsplitter arranged between the retained substrate
and the second imager and configured to direct at least some
electromagnetic radiation passing through the retained substrate to
the second sensor.
[0246] BE: The system according to paragraph BD, wherein the
beamsplitter comprises a pellicle beamsplitter.
[0247] BF: The system according to paragraph BD or BE, wherein the
beamsplitter comprises a plate beamsplitter coated with a wideband
antireflective coating.
[0248] BG: The system according to any of paragraphs BD-BF, wherein
the beamsplitter comprises an R45:T55 beamsplitter.
[0249] BH: The system according to any of paragraphs BD-BG, wherein
the beamsplitter comprises a cage mount.
[0250] BI: A method comprising: determining feature values based at
least in part on images of colonies of micro-organisms under
irradiation (e.g., visible-light or otherwise, e.g., 300 nm-800 nm,
or ultraviolet to near-infrared) of different wavelengths (e.g.,
one wavelength per image); and clustering at least some of the
determined feature values based at least in part on
colony-identification values of the images.
[0251] BJ: The method according to paragraph BI, wherein the
clustering comprises training a classification model using the at
least some of the determined feature values as training data and
the colony-identification values as class data.
[0252] BK: The method according to paragraph BJ, wherein the
classification model includes a support vector machine.
[0253] BL: The method according to paragraph BJ or BK, further
comprising: determining test feature values based at least in part
on images of a test micro-organism colony (e.g., images not
included in the images used for the clustering) under irradiation
of different wavelengths; and determining a test
colony-identification value of the test micro-organism colony by
applying the test feature values to the trained classification
model.
[0254] BM: The method according to any of paragraphs BI-BL, wherein
the clustering comprises: selecting multiple subsets of the
determined feature values; training candidate classification models
for respective ones of the subsets; determining accuracy values for
respective ones of the trained candidate classification models; and
selecting the at least some of the determined feature values based
at least in part on the determined accuracy values (e.g., random
forest selection as described above).
[0255] BN: The method according to any of paragraphs BI-BM, wherein
the determining comprises determining, as at least some of the
feature values, one or more Zernike or pseudo-Zernike moments for
individual ones of the images.
[0256] BO: The method according to any of paragraphs BI-BN, further
comprising: determining a first feature value of the at least some
of the determined feature values based at least in part on a first
one of the images corresponding to irradiation of a first one of
the wavelengths; and determining a second feature value of the at
least some of the determined feature values based at least in part
on a second one of the images corresponding to irradiation of a
second, different one of the wavelengths (e.g., using plural
features determined from images captured at respective, different
wavelengths).
[0257] BP: The method according to any of paragraphs BI-BO, further
comprising: capturing the images using an imager during irradiation
of corresponding ones of the colonies with corresponding ones of
the wavelengths.
[0258] BQ: The method according to any of paragraphs BI-BP, further
comprising: capturing the images using a system as recited in any
of paragraphs A-AC (e.g., a multispectral reflective or
reflective/transmissive imaging system).
[0259] BR: The method according to any of paragraphs BI-BP, further
comprising: capturing the images using a system as recited in any
of paragraphs AD-AM (e.g., a single-wavelength reflective imaging
system).
[0260] BS: The method according to any of paragraphs BI-BP, further
comprising: capturing the images using a system as recited in any
of paragraphs AN-BH (e.g., a multispectral transmissive imaging
system).
[0261] BT: A computer-readable medium, e.g., a computer storage
medium, having thereon computer-executable instructions, the
computer-executable instructions upon execution configuring a
computer to perform operations as any of any of paragraphs BH-BS
recite.
[0262] BU: A device comprising: a processor; and a
computer-readable medium, e.g., a computer storage medium, having
thereon computer-executable instructions, the computer-executable
instructions upon execution by the processor configuring the device
to perform operations as any of paragraphs BH-BS recite.
[0263] BV: A device comprising: a processor; and a
computer-readable medium, e.g., a computer storage medium, having
thereon computer-executable instructions, the computer-executable
instructions executable by the processor to cause the processor to
perform operations as any of any of paragraphs BH-BS recite.
[0264] BW: A system comprising: means for processing; and means for
storing having thereon computer-executable instructions, the
computer-executable instructions including means to configure the
system to carry out a method as any of any of paragraphs BH-BS
recite.
CONCLUSION
[0265] The invention is inclusive of combinations of the aspects
described herein. References to "a particular aspect" (or
"embodiment" or "version") and the like refer to features that are
present in at least one aspect of the invention. Separate
references to "an aspect" (or "embodiment") or "particular aspects"
or the like do not necessarily refer to the same aspect or aspects;
however, such aspects are not mutually exclusive, unless so
indicated or as are readily apparent to one of skill in the art.
The use of singular or plural in referring to "method" or "methods"
and the like is not limiting. The word "or" is used herein in a
non-exclusive sense, unless otherwise explicitly noted.
[0266] The invention has been described in detail with particular
reference to certain preferred aspects thereof, but it will be
understood that variations, combinations, and modifications can be
effected by a person of ordinary skill in the art within the spirit
and scope of the invention.
* * * * *