 |
Description  |
|
|
TECHNICAL FIELD
The technical field of this invention relates to optical methods of
chemical analysis.
BACKGROUND
Chemical analysis involving the detection and quantization of light occurs
in a large variety of situations. One application of this need is the
detection of analytes for the determination of the presence or amount of a
particular analyte. In many assays for analytes, one is concerned with
either absorption or emission (e.g., fluorescence or chemiluminescence) of
light. In many situations, one irradiates a sample with light and then
attempts to detect the effect of the sample on the transmitted or emitted
light. In the case of emitted light resulting from irradiation,
non-analyte molecules may also emit light resulting in a relatively large
background noise, which results in the introduction of substantial error
in the measurement of the effect of the sample on the light. There are
also additional systematic errors which collectively contribute to the
noise associated with the measurement.
The quality of chemical measurements involving light can be defined in
terms of the ratio of a suitable measure of the optical signal from a
sample due to the presence of analyte to the noise variation inherent
within the signal. In general, efforts to augment this signal to noise
(S/N) ratio have centered on improving the sensitivity of a measurement
apparatus so as to reduce the "detection limit" associated with a
particular analyte. The detection limit refers to the analyte
concentration within a sample above which the signal attributable to the
presence of analyte is such that a desired S/N ratio is achieved. In
practice, this detection limit is ascertained by conducting an
experimental procedure designed to elicit an optical signal related to
analyte concentration. Specifically, data relating to signal and noise
intensity is plotted in the form of a calibration curve for a range of
analyte concentrations, thereby enabling straightforward determination of
the detection limit.
The determination of concentration in unknown samples is effected by
comparing the signal obtained experimentally from the unknown with the
calibration curve. A typical unit of concentration in chemical
measurements is moles/liter [i.e., Molarity (M)], where a mole is defined
as Avogadro's number (6.0225.times.10.sup.23). Unfortunately, even the
most sensitive conventional experimental techniques have detection limits
on the order of about one femtomolar (fM), or nearly one billion analytes
per liter.
Measurements in which concentration is determined by reference to a
calibration curve may be characterized as being inherently "analog" rather
than "digital". That is, a signal correlated with analyte concentration is
initially produced by the measurement device. The calibration curve is
then consulted to obtain an approximation of the analyte concentration.
Since the calibration curve may be made continuous as a function of
concentration, the concentration derived from the calibration curve will
generally not be an integer. In contrast, measurement data in the digital
domain are often embodied in binary (i.e., two-level) signals which
unequivocally represent specific integers. Accordingly, a fundamental
difference between analog and digital modes of measurement is that the
addition of a single additional analyte to a sample analyzed using analog
means cannot be unambiguously detected. Although dramatic improvements
have been made in the accuracy of chemical measurements, such advancements
have been based on the fundamentally analog concepts of increasing signal
and reducing noise.
In molecular samples involving low levels of analyte concentration a
digital measurement methodology would afford at least two advantages: (i)
reference to a calibration curve would not be required, and (ii) the
addition of a single additional molecule to a sample could conceivably be
detected. Such a digital technique would be of utility in samples where
the analyte concentration is sufficiently low that statistical noise
accompanying each binary measurement value remains less than the
difference between successive integers. Accordingly, it is an object of
the present invention to provide an optical technique for determining low
levels of analyte concentration by means of an intrinsically digital
measurement scheme.
Relevant Literature
Whitten et al., Anal. Chem. (1991) 63:1027-1031 describe detection of
rhodamine-6G molecules by using a laser to excite fluorescence from
electrodynamically levitated microdroplets.
Stevenson et al., Applied Spectroscopy, (1992) 46(3):407-419 discuss
theoretical considerations relating to laser spectroscopic methods capable
of detecting single atoms or molecules in the laser beam.
Nguyen et al., J. Optical Society of America, (1987) 4(2):138-143 disclose
an apparatus for the detection of fluorescent species in hydrodynamically
focused flows.
Tou et al., Pattern Recognition Principles, (1974) (ISBN 0-201-07587-3),
chps. 3 and 7, discuss techniques for determining characteristic
prototypes or cluster centers from a given set of data. Methods of pattern
preprocessing and feature selection are also described.
SUMMARY OF THE INVENTION
A system and method for digitally detecting the presence of analyte
particles within a molecular sample is disclosed herein. Each analyte
particle is disposed to emit an optically detectable response upon
stimulation (e.g., illumination) in a known manner. For stimulation of
fluorescence as distinct from chemiluminescence, the digital analyte
detection system includes optical apparatus for illuminating a
multiplicity of distinct pixel regions within the sample so as to induce
each of the analyte particles included therein to emit an optical signal,
i.e., photons. The pixel regions are dimensioned such that the number of
analyte particles included within each region is sufficiently small that
the aggregate optical signal emitted by each region is less than a maximum
detection threshold proportional to variability of the optical responses.
The digital detection system further includes apparatus for measuring the
optical signal emitted from each pixel region. A data processing network
receives the measurements of the optical signals and, based on the
measurements, counts the number of analyte particles within each pixel
region so as to determine the number of analyte particles within the
sample.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a block diagrammatic representation of a preferred embodiment of
the digital molecular analyte detection system of the present invention.
FIG. 2 shows a diagrammatic elevational view of a preferred implementation
of a sensitive light detection apparatus incorporated within the inventive
detection system.
FIG. 3 depicts the manner in which a sample to be analyzed is divided into
a two-dimensional array of contiguous pixel regions.
FIG. 4 is a plan view of a first alternative implementation of the
sensitive light detection apparatus.
FIG. 5 is a plan view of a second alternative implementation of the
sensitive light detection apparatus.
FIG. 6 depicts a third alternative implementation of the sensitive light
detection apparatus.
FIG. 7 is a flow chart representative of the pixel clustering process
utilized in the digital detection scheme of the present invention.
DESCRIPTION OF THE SPECIFIC EMBODIMENTS
Apparatus and methods are provided for digitally detecting the presence of
small concentrations of molecular analyte particles within a sample. As is
discussed below, the term analyte particle refers to the combination of a
molecular analyte of interest and a suitably homogeneous label disposed to
emit light upon being stimulated. Such particles may be any entity capable
of being detected, including those as small as a DNA macromolecule or a
polysaccharide, or as a large as, for example, a latex particle. However,
with large particles other detection techniques may be equally
advantageous. The apparatus is disposed to detect low levels of emitted
light from a plurality of pixel regions within the sample, where the light
emitted by each pixel region is engendered by stimulation of analyte
particles included therein. The emitted light is digitally analyzed so as
to determine the integral number of analytes within each pixel region.
Photon emission may be achieved by irradiation of fluorophores in their
absorption range, or by chemical or physical stimulation of
chemiluminescent labels, which emit light upon irradiation or appropriate
chemical reaction. For the most part fluorescence is more convenient and
is the preferred mode. Therefore for the purpose of the description of the
subject invention, fluorescence will be described. For fluorescence, the
subject device will include means for irradiation of the sample to cause
fluorescence. The irradiation means may not be necessary for
chemiluminescence, since stimulation may be achieved by addition of a
chemical reagent. Where a chemical reagent is employed, care will be taken
in the measurement to relate the meausurement to the time from mixing.
Alternatively, the change in rate of emission may be monitored over a
predetermined time period. For the purpose of this invention, the optical
signal from chemiluminescence may be equated to the optical signal from
fluorescence.
For fluorescence measurements particularly, collimated light of a narrow
wavelength range is directed onto each pixel region within the sample. The
intensity of the emitted light will be in discrete levels in relation to
the number of analyte particles present within each pixel region. The
emitted light is then collected with a discrete element collector system
and directed to a photodetector for quantization and analysis. A discrete
element collector is an array of discrete optical elements arranged to
refract or reflect light from the sample point into a quasi-collimated
beam. An example of such a collector is a low f-number Fresnel
lens/reflector array of the type produced by 3M Corporation (Minneapolis,
Minn.).
Referring to FIG. 1, a preferred embodiment of the digital analyte
detection system 10 of the present invention includes a general purpose
computer system 100 operative to analyze measurement data accumulated by a
sensitive light detection apparatus 101. The computer system 100 includes
a central processing unit 102 that is interconnected by a system bus 104
to secondary memory 106 (e.g., a magnetic disk storage device), to primary
memory 112 (i.e., high speed, random access memory), and to one or more
user interfaces 120. Each user interface 120 typically includes a display
or monitor 121, a keyboard 122, and a mouse pointer device 123 with item
selection button 124.
Stored in primary memory 112 are a data array 126 as well as a variety of
computer programs (software) used to implement the digital detection
method of the present invention. Included among the software contained in
primary memory 112 is an instrument maintenance program 128 for
controlling the light detection apparatus 101. Also stored in primary
memory 112 are a number of data analysis programs described in detail
below. Specifically, an inventive data clustering scheme facilitating
detection of analyte particles comprises a background/noise threshold
routine 132, a data sorting routine 136, and a cluster synthesis routine
140.
The sensitive light detection apparatus 101 will typically include an
optical system substantially similar to that described in, for example,
copending U.S. patent application, Ser. No. 07/855565, filed Mar. 23,
1992, and entitled SENSITIVE LIGHT DETECTION SYSTEM, which is herein
incorporated by reference. As mentioned above, the apparatus 101 is
operative to illuminate a plurality of pixel regions within a sample of
analyte particles in order to induce fluorescence of the analyte particles
within each region. The light (optical response) emitted by each analyte
particle upon illumination at known intensity for a predetermined time is
directed to a detector operative to generate an electrical detection
signal in response thereto. The magnitude of the detection signal
corresponding to each pixel region is then stored in data array 126.
FIG. 2 shows a diagrammatic elevational view of a preferred implementation
of the sensitive light detection apparatus 101. The apparatus 101 has a
laser beam 212 which is passed through a first lens 214 which expands the
laser beam and directs the beam to second lens 216 where the beam is
recollimated. The collimated beam is then passed through an adjustable
aperture 218 to define the diameter of the beam and finally through
focusing lens 220. Focusing lens 220 is mounted on a stage, not shown,
where focusing lens 220 may be moved along the optical axis, so as to
change the beam diameter as it is incident on the sample. The laser beam
212 exits the focusing lens 220 and is reflected by a first surface mirror
222. The light then passes through a hole 224 in discrete element
collection system 226, which comprises a first discrete element collector
lens 228 for collecting the emitted light and a second discrete element
collector lens 230 for focusing the light. Usually, the focal length of
the first discrete element collector lens, which will have a structure
analogous to a Fresnel lens, will be substantially less than the focal
length of the second lens (which may or may not be a Fresnel lens),
generally being less than about 60% of the focal length of the second
lens. The collection angle for the first discrete element collector lens
will usually be at least about 90.degree. for direct emission. The emitted
focused light will then be transmitted to a second lens 232 and directed
to a filter pack 234, which serves to exclude light outside of the
wavelength range of the light emitted from the fluorophore label within
each analyte particle. The filter pack may be implemented to effect
wavelength isolation by using, for example, conventional optical filters,
monochromators or time-gating techniques. The light of the desired
wavelength is then detected by a photomultiplier tube and preamplifier 236
for transmission to electronic circuitry for analysis.
The sample 238 is supported by X-Y stage 240, which allows the incident
light to illuminate each pixel region included therein. As is shown in
FIG. 3 and is described more fully below, the sample 238 will preferably
be divided into a two-dimensional array of contiguous pixel regions 246.
The pixel regions 246 will be selected to be small enough so that
generally less than a desired number (e.g., two) of analyte particles will
be present within each pixel region. For example, the volume of sample to
be analyzed may be selected such that there exist approximately five pixel
regions for each analyte particle. Such a ratio nearly ensures that the
aggregate fluorescent energy (emitted signal) engendered by analyte
particles within each pixel region will be sufficiently small that the
statistical variation therein remains less than the optical response of a
single analyte particle. However, if less than absolute precision may be
tolerated a sample volume may be selected which results in a smaller ratio
of pixel regions to analyte particles. The number of analyte particles
within a given pixel region is determined by finding the integer nearest
the ratio of the emitted signal collected from the pixel region to one
optical response. This set of integers is stored in data array 126, and
may be summed to enable a digital determination of the number of analyte
particles within the sample 238. A stepper motor 242 (FIG. 2) is provided
for accurate movement of the sample 238 in the X and Y directions as
directed by the instrument maintenance program 128.
The X-Y stage 240 may be assembled from, for example, a Newport model 405
dual-axis translation stage. Appropriate modification would entail
accommodation of a 20 turn per inch lead screw fabricated from 1/4 inch
--20 stock stainless steel threaded rod. Use of two stepper motors
provides 3.175 microns of linear travel per step for both the X and Y
axis. Total travel for X and Y is limited to about one half inch. Limit
switches at each axis end provide a means of centering the stage. This
establishes a known reference x,y of 0,0. Image scanning of the sample can
then be performed for pixel regions having lateral dimensions on the order
of 3.175 microns.
In FIG. 4, a first alternative implementation 250 of the light detection
apparatus is shown diagrammatically with a laser source 252 fitted with
beam shutter 254. The laser beam 256 is directed through a line filter 258
and then is reflected by turning mirror 260 into the lens system
comprising an expanding lens 262, a collimating lens 264, an adjustable
aperture 265, and a focusing lens 266 which is mounted on a stepper driven
stage 268. A sample stepper stage 270 is provided to support and move the
sample per instruction of instrument maintenance program 128, holding the
sample in place with slide holder 271, which is positioned under moveable
discrete element collector system 272 (shown displaced from operational
position directly above the sample).
For the most part, the light source will be a laser, where the intensity of
the beam may-vary from a power rating of 1 .mu.W to about 100 mW. The
light wavelength may be varied widely, depending upon the absorption
characteristics of the fluorescent label included within each analyte
particle. For the most part, the light will be at a wavelength above 350
nm, usually about 400 nm, and usually below 700 nm, more usually below
about 550 nm. Desirably, each analyte particle will include a fluorescer
providing for a large Stokes shift, usually at least about 20 nm,
preferably at least about 50 nm.
For varying the beam size, one may use a movable lens, which by varying the
distance from the sample will change the beam size. By employing
appropriate stepper motors, one can provide for smaller or greater changes
in the beam with each step.
The light which is emitted from each pixel region 246 is then efficiently
collected using a discrete element collector system which provides for the
collection of the light and its transmission to a photodetector. The
discrete element collector system will usually be a multi-lens system,
where the collector lens, proximal to the sample, will generally have a
low f-number, usually less than about 2 and greater than about 0.05,
generally being in the range of about 0.075 to 1.0. The low f-number
discrete element collectors do not normally produce a high quality
collimated beam. However, by employing a second lens of about the same
diameter and a larger f-number, usually greater than 0.5, generally from
about 1 to 10, the light may be focused on the detector. The discrete
element collector lens and the second lens, will generally have a
separation of from about 0 to 50 cm, usually being in close proximity of
from about 0.1 to 5 cm. The discrete element collector lens will usually
be at least about 1 cm.sup.2 and may be as large as about 1.times.10.sup.4
cm.sup.2, usually not greater than about 500 cm.sup.2 in area.
By including a perfect spherical reflector behind the emitter to reflect
the light going away from the detector back through the source and into
the discrete element collector system, the fraction of solid angle
collected into the collimated beam emergent from the discrete element
collector system can be doubled. The spherical reflector must be placed
one sphere radius away from the source and on the opposite side of the
emitter from the discrete element collector. For pixel regions of area A,
the spherical reflector radius should be on the order of 10 times A, or
greater, in order to cause the source to remain point-like relative to the
reflector. If this is not the case, larger portions of the light will not
be reflected back into the lens.
The solid angle of light which will be collected can be calculated based on
the collection system employed.
By using the above system the solid angle fraction collected in a system
with an ideal collector can be as high as 0.90, while more realistically
with f-numbers for the discrete element collector varying from 0.5 to
0.05, estimated solid angle fractions will be in the range of about 0.26
to 0.81.
In a second alternative implementation of the sensitive light detection
system generally indicated by reference numeral 300 in FIG. 5, instead of
having the light incident to the sample being at right angles to the
sample, the incident light is at other than normal to the sample. In FIG.
5, a laser beam 301, which has been processed as described in FIG. 2, is
reflected by a reflecting mirror (not shown) so as to pass through hole
304 in discrete element collector system 306. The light strikes the sample
308 and is reflected from the sample through hole 310 and then discarded
by any convenient means. The emitted light from the sample 308 is focused
by discrete element collector system 306 and directed to lens 312 to be
processed as previously described in FIG. 2.
In a third alternative implementation of the sensitive light detection
system generally indicated by reference numeral 350 in FIG. 6, instead of
having the light incident to the sample originating above the sample and
passing through holes in the discrete element collector system, the sample
is mounted on a transparent substrate and the incident light strikes the
sample from below. In FIG. 6, a laser beam 351 is expanded through a first
lens 352, which expands the laser beam and directs it to a second lens 354
where the beam is recollimated. The collimated beam is reflected by a
first surface mirror 356 and directed to a focusing lens 358 mounted on an
adjustable stage 360 which allows the beam to be focussed at the sample
plane. The sample 362 is supported by X-Y stage 364 which allows each
pixel region of the sample to be separately illuminated by the incident
light. A stepper motor 366 is provided for accurate movement of the sample
362 in the X and Y directions per instructions issued by instrument
maintenance program 128.
Unabsorbed excitation light passes through the discrete element collector
system 368 and is absorbed by the long pass filter 370. Alternatively, the
long pass filter may be replaced by a dichroic mirror placed at a
45.degree. angle to the unabsorbed incident beam and reflecting the beam
away from the detector where the light beam can be discarded by any
convenient means.
The emitted light from the sample 362 is focussed by the discrete element
collector system 368 and directed through spatial aperture 372 and to
collector lens 374 to be processed as previously described in FIG. 2.
Analyte Particle Characteristics
The digital detection technique of the present invention can be used for
detecting a wide variety of analytes employing emission spectroscopy,
where emitted light is the detected signal. While emitted light may be as
a result of, for example, fluorescence, chemiluminescence, or
phosphorescence, the term "optical response" as employed herein is
intended to collectively refer to the light emission from a single
analyte, however induced. In addition, the term "emitted signal" as used
herein corresponds to a measurement of the optical responses detected from
a particular pixel region. The sample will preferably be a solid sample in
which detectable label not bound to an analyte may be removed through a
conventional washing procedure. In addition, the sample assay will be
uniform in order to ensure that each analyte particle elicits a
substantially identical optical response subsequent to being exposed to
equal illumination energy.
In a preferred embodiment the analyte particles within each pixel region
are measured individually based on discrete signal units providing optical
responses substantially above a background noise level. The magnitude of
each optical response is required to be large enough to allow the
particular photodetection apparatus employed to discriminate between
optical responses and ambient background noise. One or more optical
responses of a signal unit may be associated with a signal analyte
particle, but the number of units will be low and substantially identical
for each analyte particle. For the most part, the number of signal units
per analyte particle will be one. A signal unit within any given analyte
particle could be a fluorescent particle or fluorescer intercalated DNA
molecule of known fluorescence and capable of reproducible production. The
analyte included within the analyte particle could be a single molecule or
an aggregation.
The assay medium will have low concentrations of analyte, generally at
picomolar or less, frequently femtomolar or less. Assay volumes will
usually be less than about 100 .mu.l, frequently less than 10 .mu.l and
may be 1 .parallel.l or less. Each pixel region 246 (FIG. 3) will
generally be not greater than 100 um.sup.2, usually not greater than about
50 um.sup.2 and may, as mentioned above, be as low as 10 um.sup.2. The
size of each pixel region 246 will depend on the concentration of the
analyte, the rate at which it is desired to process each sample 238, the
ease with which individual optical responses may be discriminated, and the
like.
The assays will normally involve specific binding pairs, where by specific
binding pairs it is intended that a molecule has a complementary molecule,
where the binding of the members of the specific binding pair is at a
substantially higher affinity than random complex formation. Thus,
specific binding pairs may involve haptens and antigens (referred to as
"ligands") and complementary binding members, such as antibodies, enzymes,
surface membrane protein receptors, lectins, etc. (generally known as
"receptors") and nucleic acid sequences, both naturally occurring and
synthetic, either RNA or DNA, where for convenience nucleic acids will be
included within the concept of specific binding members comprising ligands
and receptors.
In carrying out the assay, there will normally be involved a conjugate of a
specific binding member and a detectable label. As already indicated, a
fluorescent label will be preferred, but other discrete labels which may
be detected include chemiluminescent labels. Methods of preparing these
conjugates are well known in the literature. Depending upon the analyte,
various protocols may be employed, which may be associated with
commercially available reagents or such reagents which may be modified.
Data Acquisition Using Sensitive Light Detection System
As noted above, a detailed discussion of the manner in which optical
responses from each pixel region within a particular sample may be
collected and quantized by the sensitive light detection apparatus is
provided in aforementioned copending U.S. patent application, Ser. No.
07/855565. Nonetheless, in order to facilitate explanation of the digital
detection technique of the present invention, preferred implementations of
the computer, electronics and software utilized by the light detection
system to acquire emitted signal data are set forth below:
Electronics
Subsequent to separation of the laser illumination from the optical
response energy using conventional optical filtering techniques, the
optical responses will typically be detected using a photomultiplier tube
(PMT). The PMT used is the Hamamatsu 1477 one inch side on tube. The tube
is constructed with a multialkali photo-cathode with a UV glass window and
has a photon to electron gain on the order of 5.3.times.10.sup.6 at an
anode to cathode potential difference of 1000 volts.
The current output of the PMT is directly coupled to the pulse generator
circuit disposed to operate in a digital mode. In digital mode operation
single electronic pulses are counted as single photon events. This mode,
also known as the photon counting mode, is only possible at low levels of
light where photons are generally spaced far enough apart to prevent dc
biasing. It is assumed that the dimensions of each pixel region are chosen
to be sufficiently small that such dc biasing, which tends to accompany
higher intensity optical response energy, does not occur.
The circuit converts the current from the tube into a voltage level,
compares the voltage value against a threshold discriminator value, and
generates a pulse to the remote counter circuit. Each pulse to the counter
is then counted as a single photon event.
A pair of operational amplifiers comprise the preamp portion of the pulse
generator circuit. The preamp portion converts current pulses around a
micro-ampere into a voltage pulse of approximately 1.0 volt. A comparator
is used as a discriminator to detect when these pulses are greater than a
preset value. When the pulses are greater than this discriminating level,
a logic transition occurs and a pulse shaper circuit converts this into a
defined 10 nsec pulse.
The computer interface consists of circuitry to convert the PC computer
instructions into logical commands and status information into computer
logic. This interface is built on a full size card for a PC slot. The
basic components on the card are the bus transceiver, address decoders,
function latches, line receiver, and counter circuits.
The computer interface uses 8 I/O port addresses, hex 0310 to 0317. Reads
from ports 0310 to 0312 provide 20 bits from the counters. When the
counters are gated by the timer circuit, with a write to 0317, pulses
received by the line receiver are counted for the defined gate period. The
gate period is programmed by writes to locations 0310 to 0312.
The 20 bits of counts provide capability to greater than 1 million.
Additional bit patterns written to ports 0313 to 0315 provide control of
laser shutter, high voltage power supply, the three axes (i.e., X, Y and
Z) stepper motors, and the operating mode of the timer circuit and the
remote pulse generator. A read from 0313 provides information regarding
the three axis limit switches.
The motor driven board accepts logic signals from the computer interface
and translates them into position commands. For incident illumination on
the order of 0.1 to 1.0 mW, the motor driven board will cause each pixel
region to be illuminated for dwell times of approximately 1 to 100 msec.
For illumination of between 10 and 100 mW, pixel dwell times typically
range from 10 to 100 .mu.sec. Logic lines for each axis provide motor
enable, direction, step size, and the step pulse. Limit switches at the X,
Y and Z axis ends are delivered to the computer interface and also wired
to protection circuitry. The protection circuitry prevents the stepping of
the motors when the limits have been reached in the event that computer
instructions fail to detect the limit.
Computer
A PC compatible computer is used to control the instrument, make the
measurements, and perform the analysis and imaging. A printer and color
plotter are attached for hard copy outputs.
The computer motherboard is equipped with an INTEL.RTM. 80286
microprocessor and 80287 math coprocessor which together comprise the CPU
102 (FIG. 1). The motherboard further includes a 1024 Kbyte memory,
keyboard and speaker connections, and 8 expansion slots.
An AST ADVANTAGE.RTM. multifunction board provides 576 Kbytes of primary
memory 112 (FIG. 1), one serial port for the plotter, and one parallel
port for the printer. An IBM.RTM. hard/floppy disk controller board and
external floppy disk driver board provide communication links to the disk
storage media 106. A VIDEO SEVEN VEGA VGA.RTM. board provides high
resolution graphic capability up to 800.times.600 pixels.
Of the three board slots remaining, one is used by a custom developed
circuit card for control and monitoring of the instrument and two slots
are available for expansion. The custom board is the computer interface
board discussed in the electronic section. It provides the means of
controlling the instrument data collection.
Software
Data collection software included within the instrument maintenance program
128 (FIG. 1) has been developed using the C programming language. It was
compiled with the MICROSOFT.RTM. C compiler version 5.1 and linked with
libraries from MICROSOFT.RTM. and VERMONT CREATIVE SOFTWARE.RTM.. It is a
window oriented package with several levels for data collection and
maintenance operations. Data files written to the 3.5 inch drive include a
binary FCD file containing header information, operating parameters, and
the data. Also written are .GRD files for 3-D image plotting of photon
count data as a function of pixel region, .DAT files containing X and Y
coordinates and photon count data, and .COR files with just coordinates.
The .DAT file can be used to provide a .GRD file that uses smoothing
factors for interpolation of missing data points. The .COR file can be
used to re-image at the same locations as a previous run.
The graphic software is written in C and linked with the GSS Graphic
libraries. This provides a visual display of the data collected and
capability of sending the display to the color plotter.
The GOLDEN SURFER.RTM. software package is a commercial product that does
the 3-D surface imaging and 2-D topographical views. There are six
programs, GRID, SURF, TOPO, VIEW, PLOTCALL, and PLOT. GRID reads data
files containing the photon counts (emitted signal data) from each pixel
region and provides the means to generate a grid image of the data. The
output from this program can be read by SURF for 3-D imaging or by TOPO
for 2-D contours. PLOTCALL and PLOT provide the means to output these
views to the color plotter. VIEW is a utility that displays the plotter
generated files on the screen and provides zoom and pan functions.
The LOTUS 123.RTM. spreadsheet software package is also used in analysis of
data. Data that is not normally collected automatically by the instrument
can be processed and displayed by the graphic functions. Other scientific
processing can also be applied to data collected by the instrument, such
as applying photon pulse pile up correlation equations to the actual d | | |
