WO1997043638A1 - Method and apparatus for verifying uniform flow of a fluid sample through a flow cell - Google Patents

Method and apparatus for verifying uniform flow of a fluid sample through a flow cell Download PDF

Info

Publication number
WO1997043638A1
WO1997043638A1 PCT/US1997/008211 US9708211W WO9743638A1 WO 1997043638 A1 WO1997043638 A1 WO 1997043638A1 US 9708211 W US9708211 W US 9708211W WO 9743638 A1 WO9743638 A1 WO 9743638A1
Authority
WO
WIPO (PCT)
Prior art keywords
flow
particles
examination area
fluid sample
degree
Prior art date
Application number
PCT/US1997/008211
Other languages
French (fr)
Inventor
Harvey L. Kasdan
Jaehyl Ko
Original Assignee
International Remote Imaging Systems, Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by International Remote Imaging Systems, Inc. filed Critical International Remote Imaging Systems, Inc.
Priority to AU31261/97A priority Critical patent/AU3126197A/en
Publication of WO1997043638A1 publication Critical patent/WO1997043638A1/en

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N15/00Investigating characteristics of particles; Investigating permeability, pore-volume, or surface-area of porous materials
    • G01N15/10Investigating individual particles
    • G01N15/14Electro-optical investigation, e.g. flow cytometers
    • G01N15/1468Electro-optical investigation, e.g. flow cytometers with spatial resolution of the texture or inner structure of the particle
    • G01N15/147Electro-optical investigation, e.g. flow cytometers with spatial resolution of the texture or inner structure of the particle the analysis being performed on a sample stream

Definitions

  • the present invention relates to fluid samples in flow cells and distributed on slides and, in particular, to a method for verifying that a fluid sample is flowing uniformly through the flow cell and is uniformly distributed on the slide.
  • FIG. 1 A conventional flow cell and particle analyzer is shown, in plan view, in Fig. 1 , and a particle analyzer is shown in block schematic form in Fig. 2.
  • the flow cell and particle analyzer automate the job of counting fluid particles suspended in a biological sample (for example, blood cells suspended in a serum sample or particles suspended in a urine sample). More generally, the flow cell and particle analyzer may be used to automate the job of counting any "particles of interest" that are suspended in a fluid sample.
  • a biological sample for example, blood cells suspended in a serum sample or particles suspended in a urine sample.
  • the flow cell includes a body 10 containing a flow chamber having an inlet 12 through which the fluid sample is provided.
  • the flow chamber also has an outlet 14.
  • a passageway 16 extends between the inlet 12 and the outlet 14 such that the fluid sample provided at the inlet
  • sheath fluid e.g., saline solution
  • sheath fluid e.g., saline solution
  • a strobe light 32 Fig. 2 illuminates the examination area 180 from below and a microscope 30 is focused on the examination area
  • a CCD camera 34 receives the output of the microscope 30 to form a series of still frame images.
  • a frame grabber 40 "grabs" the still frame images and stores them in a memory.
  • a CPU 56 combines selected components of the still frame images to generate composite images which represent the particles in multiple images. Thus, individual particles (for example, different types of blood cells) can be analyzed and sorted.
  • the fluid sample having the suspended particles of interest must flow uniformly through the flow cell examination area 18, without interruption. Even if interruptions occur relatively infrequently, when they do occur they will cause the determined particle counts to be inaccurate.
  • a particle analyzer verifies the uniformity of fluid flow through a flow cell examination area without reference to particles of interest (that may or may not be suspended therein). First, the particle analyzer inspects still frame images generated by a frame grabber of the flow cell and the light level in the examination area is determined. Then, a heavy stain is injected into the flow cell. Thereafter, the particle analyzer inspects further still frame images generated by the frame grabber of the flow cell and determines from the further images whether there is a uniformly decreased intensity in light level in the examination area (corresponding to the heavy stain flowing through the examination area). If so, then the flow of fluid through the flow cell examination area is indicated as being uniform.
  • a particle analyzer executes statistical tests on counts of particles of interest in the fluid sample under analysis and uses these counts to determine whether the fluid sample is flowing uniformly past an examination area of the fluid particle analyzer. In particular, statistical parameters of the particle counts are calculated. Then, it is determined whether the calculated statistical parameters are within an allowable range for the parameters. Based on this determination, an indication is provided of whether the fluid sample is flowing uniformly past the examination area.
  • Fig. 1 is a plan view of a flow cell for moving a fluid sample past an examination area.
  • Fig. 2 is a schematic block diagram of a particle analyzer for use with the flow cell of Fig. 1 in practicing the present invention.
  • Fig. 3 is a flowchart which illustrates a process in accordance with a first embodiment of the invention for determining whether a fluid sample is flowing uniformly past an examination area of the apparatus of Fig. 1.
  • Fig. 4 is a top view of the examination area of the flow cell of Fig. 1.
  • Fig. 5 is a flowchart which illustrates a process in accordance with a second embodiment of the invention for determining whether fluid is flowing uniformly past the examination area of the apparatus of Fig. 1.
  • Fig. 6 is a bar graph which illustrates how the determiriing step 512 may be carried out in the process illustrated in Fig. 5.
  • Fig. 7 is a flowchart which illustrates a process in accordance with a third embodiment of the invention for determining whether fluid is flowing smoothly past the examination area of the apparatus of Fig. 1.
  • Fig. 8 is a bar graph which illustrates how the determination step may be carried out in the process illustrated in Fig. 7.
  • Fig. 9 is a flowchart which illustrates a process in accordance with a fourth embodiment of the invention for determining whether fluid is flowing smoothly past the examination area of the apparatus of Fig. 1.
  • Fig. 10 is a illustrates the data flow of the process shown in the flowchart of Fig. 9.
  • Figs. 11-16 are examples of membership functions for computing a degree of support for various measurement values in step 904 of the flowchart of Fig. 9.
  • Fig. 17 is an example of a membership function for computing a defuzzified value, FD(Center of Maxes), in step 908 of the flowchart of
  • FIG. 3 A first embodiment of the present invention is now described with reference to the flow chart in Fig. 3.
  • the steps illustrated in the flowchart of Fig. 3 are executed by the processor 36 of the flow analyzer shown in Fig. 2.
  • the processor determines an average light level intensity of the examination area (or of a portion of the examination area).
  • the average light intensity is measured by averaging the intensity of all portions, of more than one still frame image, that correspond to the examination area.
  • a heavy stain is caused to be injected into the flow cell.
  • the heavy stain may be provided at the inlet 12 of the Fig.
  • the heavy stain may be provided concurrently with a fluid sample under analysis, provided that the stain is not so opaque that it interferes with the ability of the particle analyzer to distinguish particles in the fluid sample under analysis.
  • the heavy stain may be provided exclusive of a fluid sample under analysis (e.g., at the end of a sample run).
  • the particle analyzer After injecting the stain into the flow cell, at step 303, the particle analyzer looks for a period of uniformly decreased light level intensity in the still frame images, as measured in the direction perpendicular to the fluid flow. At step 304, if the uniformly decreased light level intensity was expected to be seen within a first time period since injection of the stain into the flow cell, and the first time period has passed, then, at step 306, it is reported that fluid is not flowing uniformly past the examination area 18 of the flow cell.
  • the examination area 18 is logically divided into sub-areas 18a through 18f, perpendicular to the direction of fluid flow, as shown in Fig. 4.
  • the direction of fluid flow is shown in Fig. 4 by an arrow A.
  • Fig. 5 illustrates the flow of the basic process executed by the processor 36 in accordance with the second embodiment of the invention.
  • the processor 36 initializes an index variable "i".
  • the processor 36 determines a count of the number of particles "of interest” in each sub-area 18a through 18f of the examination area 18 and accumulates this count for all grabbed frames thus far processed.
  • the particles that are "of interest” are the particles that the particle analyzer is designed to detect and analyze.
  • the particles "of interest” may be red blood cells.
  • the "particles of interest” may be calibrator particles which have been added to a fluid sample for volume calibration of the particles that the particle analyzer is designed to detect and analyze.
  • the processor 36 increments the index variable "i" at step 506. Once a predetermined number of frames have been processed, as indicated by "i" being equal to "n” at step 508, the processor 36 proceeds (at step 510) to calculate a statistical distribution of the number of particles counted in the sub-areas 18a through 18f. Then, at step 512, the processor 36 determines from the calculated statistical distribution the fluid flow uniformity within the examination area 18.
  • Fig. 6 is a bar graph whose bars 20a through 20f represent the accumulated counts of the number of the particles in the sub-areas 18a through 18f.
  • the curve 22 in Fig. 6 represents an expected distribution of particles in the sub-areas 18a through 18f if fluid is flowing uniformly past the examination area 18. Generally, a uniform distribution (or some other expected distribution determined by fluid dynamics) of particles is expected.
  • the processor 36 determines that the fluid is flowing uniformly past the examination area 18. On the other hand, if the distribution of accumulated counts is outside the allowable range, then the processor 36 determines that fluid is not flowing uniformly past the examination area 18.
  • counts of particles of interest are accumulated during specified time periods, and the processor 36 calculates statistical parameters of the accumulated counts.
  • FIG. 7 illustrates the flow of the basic process executed by the processor 36 in accordance with the third embodiment of the invention.
  • the processor 36 initializes an index variable "i".
  • the processor 36 sets a count variable, indexed by "i", to the number of sample particles counted in those frames (step 704).
  • the processor 36 increments the index variable "i" at step 706.
  • the processor 36 proceeds (at step 710) to calculate a statistical distribution of the number of sample particles counted in the groups of "1" frames. Then, at step 712, the processor 36 determines from the calculated statistical distribution if fluid has flown uniformly past the examination area 18 during the sample run.
  • the height of each bar m,, m 2 , . . ., m n indicates the number of sample particles counted for the corresponding group of "1" frames.
  • the dotted line labelled m ave indicates an overall mean number of sample particles, calculated by the processor 36 at step 610, for each group of "1" frames over the "n” groups.
  • the processor 36 determines, for each group of "1" frames, how far the overall mean number of particles, m ave , per group of "1" frames differs from the actual count, m j , of sample particles in each group of "1" frames. This difference is shown graphically in Fig. 8 by a hatch pattern. Then, at step 712, the processor determines if the differences are within an allowable range. If the differences are within the allowable range, the processor 36 indicates that the fluid sample is flowing uniformly past the examination area 18.
  • the processor 36 determines if "x" is within an allowable range.
  • the processor 36 calculates a parameter
  • y indicates the sum of the deviations for the individual groups. Then, at step 712, the processor 36 determines if the parameter y is within some allowable range. The parameter y being within the allowable range indicates, within some degree of statistical certainty, that the fluid sample is flowing uniformly past the examination area 18.
  • the examination area 18 is logically divided by the processor 36 into the sub-examination areas 18a to 18f (Fig. 4).
  • the processor 36 executes the process shown in steps 702-712 of the Fig. 7 flowchart for each sub-examination area 18a to 18f.
  • the processor 36 calculates the statistical parameter x, as shown above in equation (1), for each of the areas 18a to 18f. Then, at step 712, the processor 36 determines whether the x calculated for each area 18a to 18f is within an allowable range.
  • the processor 35 calculates the statistical parameters x and y, shown above in equations (2) and (3), for each of the areas 18a to 18f. Then, at step 712, the processor 36 determines whether the y calculated for each area 18a to 18f is within an allowable range.
  • the processor 36 calculates the statistical parameters from counts of particles in only one or more selected ones of the areas 18a to 18f.
  • the processor 36 determines whether the calculated statistical parameters, for the selected areas, are within an allowable range. It is noted that this embodiment may not be as reliable as embodiments which utilize the entire examination area 18.
  • values are obtained of several measurements, where each measurement is at least partially indicative of the flow of the fluid sample through the flow cell. Then, these partial indications are combined to generate a "crisp" indication of the fluid flow.
  • the combining is preferably implemented by the CPU 56 (Fig. 2) executing special purpose software or firmware.
  • One relatively simple way to combine the partial indications is to use a linear discriminant function, in which the partial indications are each multiplied by a weighting factor (indicative of the contribution of the measurement) to generate a weighted partial indication.
  • the weighted partial indications are summed together and compared to a threshold to produce the "crisp" indication of the fluid flow.
  • fuzzy logic Another way to combine the partial indications is to use "fuzzy logic". Such a fuzzy logic method is advantageous for use with an automated particle analyzer, which is normally unattended. The method preferably implemented by the CPU 56 (Fig. 2) executing special purpose software.
  • Fig. 9 is a flowchart illustrating an embodiment of the fuzzy logic detection method.
  • the fuzzy logic detection method obtains values of the several measurements (measurement values) that are at least partially indicative of the flow of the fluid sample through the examination area of the flow cell.
  • a degree of support for each measurement value is determined from a membership function of the measurement.
  • values indicating the degree of support are provided to inference machines or rule functions , which calculate "fuzzy" degree of flow disruption values.
  • calculated degree of flow disruption values are defuzzified to a "crisp" number to determine whether a flow disruption has occurred.
  • linguistic variable or measurement a linguistic description, parameter or metric related to an event or control input (e.g. temperature) terms ranked linguistic descriptions of a measurement.
  • temperature ⁇ low, normal, high ⁇ degree of support degree to which a measurement is a member of the term; normalized to a number between 0 and 1 membership function a function that yields its degree of support given a measurement value inference machine rules of the fuzzy logic defuzzification a process of converting "fuzzy" values to a "crisp” and real value
  • fuzzy logic embodiment is now discussed in more detail with reference to the data flow diagram shown in Fig. 10.
  • a degree of support value for each measurement value is computed from the membership function for the measurement (step 904).
  • a number of sample runs are carried out and measurement values determined.
  • an operator observes the runs and determines for each run which of the one or more degrees of flow occurred.
  • the determinations of which degree of flow occurred are correlated to the measurement values to determine the membership function for each measurement. That is, a membership function for a particular measurement provides a numeric indicator of association of each of one or more linguistic descriptions of flow occurrence (e.g., no clog, partial clog and full clog) with a particular measurement value.
  • the computed degree-of-support values for each degree-of-flow are provided to an inference machine for that degree of flow.
  • the inference machine calculates degree-of-flow disruption "fuzzy" values (step 906).
  • the inference machine may calculate "fuzzy" values CLOG[low], CLOGjnorm] and CLOG[high] for each degree of flow where each represents a numeric indicator of an association (low association, normal association, and high association, respectively) that the particular degree of flow is the actual degree of flow.
  • the "fuzzy values” CLOGflow], CLOG[norm] and CLOG[high] are "defuzzified” into a crisp prediction of whether the fluid sample is flowing through the flow cell with that degree of flow (step 908).
  • approximately 1000 observed samples may be classified into three groups: no-clog, some-clog and full-clog. The mean and standard deviation of the measurements for each group of samples are then calculated.
  • Temporal Flow Variance The TFV measures the variance of the flow rate from frame to frame, or from group of frames to group of frames. For example, in one embodiment, the frames are grouped into about 10 bins and the number of analytes in each bin are counted. These counts/bins are then used to calculate Tnorm, a Poisson distribution based parameter. If there are a sufficient number of analytes in each frame, then the TFV can be calculated on an individual frame basis. Example set measurement functions for TFV are shown in Fig. 11. Spatial Flow Variance (SFV) Measurement The SFV provides an indication of a variance of distribution of particles of interest in a direction perpendicular to the direction of fluid flow in the flow cell.
  • SFV Spatial Flow Variance
  • the examination area is equally divided into a number of sub-areas (18a through 18f in Fig. 4) and the particles of interest (analytes) in each sub-area are accumulated. Then, these accumulated counts are used to calculate Tnorm, a Poisson distribution based parameter. As with TFV discussed above, if there are a sufficient number of analytes in each sub-area of each frame, then the SFV can be calculated on an individual frame basis. Example set measurement functions for SFV are shown in Fig. 12.
  • SDV Spatial Delta(max-min) Variance
  • the SDV also provides an indication of a variance of distribution of particles along a direction perpendicular to the direction of fluid flow in the flow cell.
  • the SDV is more sensitive to sharp peaks and valleys in this distribution than is the SFV.
  • delta (maximum - minimum) / average.
  • Concentration Ratio of LPF to HPF Measurement The CR measures the difference in concentration of particles of interest between the low power field and high power field. The concentration of the small particles detected in low power field, designated as size group 4
  • the CMS measures the temporal shift of analytes along the direction perpendicular to fluid flow in the flow cell.
  • the "center of mass" in the direction perpendicular to fluid flow for each sub-area 18a through 18f is calculated. Theoretically, the CMS measurement should be about 127.5 if fluid is flowing perfectly uniformly through the flow cell.
  • the calculated center of mass/frames is then used to calculate Tnorm, a Poisson distribution based parameter.
  • Example set measurement functions for CMS are shown in Fig. 16.
  • Step 1 Apply Rules (Inference Rules)
  • the measurements used in determining Partial Clog are LPC, CMS, CR, TFV, SFV and SDV.
  • Step 2 Calculate Final Score (Defuzzification). Defuzzification is done the same way as the Full Clog fuzzy logic.
  • Step 1 Find Degree of Support from Each Measurement
  • the 3 sets are GUNKY, LEARNING and TEST.
  • the GUNKY set represents samples that are likely to cause clogs in the flow cell.
  • Amorphous specimens, specimens with lots of crystals and specimens with lots of bacteria are examples of the samples which comprise the GUNKY set.
  • the GUNKY set does not represent the population.
  • the LEARNING set is a better representation of the population and more realistic samples.
  • the LEARNING set was used in fine-tuning the fuzzy logic. After fine-tuning the fuzzy logic, it was tested on TEST set.
  • the data are collected from two different flow cell systems. When these samples were run, the operator visually observed the processing of each specimen and classified them into no clog, partial clog and full clog.
  • the Fuzzy Logic was applied to these data sets.
  • the Fuzzy Logic was applied to these data sets.
  • the Fuzzy Logic was applied to these data sets.
  • the Fuzzy Logic was applied to these data sets.

Abstract

Particles in a fluid sample are counted as the fluid flows through a plurality of examination sub-areas (18a-18f) in a particle analyzer. Parameters (510) based upon the particle count in each sub-area (18a-18f) are calculated. It is determined whether the calculated parameters (510) are within an allowable range for the parameters (510). Based on this determination, an indication (512) is provided of the uniformity of the fluid sample in the examination sub-areas (18a-18f).

Description

METHOD AND APPARATUS FOR VERIFYING UNIFORM FLOW OF A FLUID SAMPLE
THROUGH A FLOW CELL
Technical Field The present invention relates to fluid samples in flow cells and distributed on slides and, in particular, to a method for verifying that a fluid sample is flowing uniformly through the flow cell and is uniformly distributed on the slide.
Background A conventional flow cell and particle analyzer is shown, in plan view, in Fig. 1 , and a particle analyzer is shown in block schematic form in Fig. 2.
The flow cell and particle analyzer automate the job of counting fluid particles suspended in a biological sample (for example, blood cells suspended in a serum sample or particles suspended in a urine sample). More generally, the flow cell and particle analyzer may be used to automate the job of counting any "particles of interest" that are suspended in a fluid sample.
Referring now to Fig. 1, the flow cell includes a body 10 containing a flow chamber having an inlet 12 through which the fluid sample is provided.
The flow chamber also has an outlet 14. A passageway 16 extends between the inlet 12 and the outlet 14 such that the fluid sample provided at the inlet
12 flows past an examination area 18 (shown by a dashed line in Fig. 1) and is discharged out of the passageway 16 via the outlet 14.
During a sampling run, the fluid sample is surrounded on both sides by sheath fluid (e.g., saline solution) introduced into the passageway 16 from inlets 13a and 13b. A strobe light 32 (Fig. 2) illuminates the examination area 180 from below and a microscope 30 is focused on the examination area
18 from above. A CCD camera 34 receives the output of the microscope 30 to form a series of still frame images. In particular, a frame grabber 40 "grabs" the still frame images and stores them in a memory. A CPU 56 combines selected components of the still frame images to generate composite images which represent the particles in multiple images. Thus, individual particles (for example, different types of blood cells) can be analyzed and sorted.
In order for the particle analyzer to perform reliably, the fluid sample having the suspended particles of interest must flow uniformly through the flow cell examination area 18, without interruption. Even if interruptions occur relatively infrequently, when they do occur they will cause the determined particle counts to be inaccurate.
Summary
In accordance with one embodiment of the invention, a particle analyzer verifies the uniformity of fluid flow through a flow cell examination area without reference to particles of interest (that may or may not be suspended therein). First, the particle analyzer inspects still frame images generated by a frame grabber of the flow cell and the light level in the examination area is determined. Then, a heavy stain is injected into the flow cell. Thereafter, the particle analyzer inspects further still frame images generated by the frame grabber of the flow cell and determines from the further images whether there is a uniformly decreased intensity in light level in the examination area (corresponding to the heavy stain flowing through the examination area). If so, then the flow of fluid through the flow cell examination area is indicated as being uniform. Otherwise, the flow of fluid through the flow cell examination area is indicated as being non-uniform. In accordance with another embodiment of the invention, a particle analyzer executes statistical tests on counts of particles of interest in the fluid sample under analysis and uses these counts to determine whether the fluid sample is flowing uniformly past an examination area of the fluid particle analyzer. In particular, statistical parameters of the particle counts are calculated. Then, it is determined whether the calculated statistical parameters are within an allowable range for the parameters. Based on this determination, an indication is provided of whether the fluid sample is flowing uniformly past the examination area. A better understanding of the features and advantages of the invention will be obtained by reference to the following detailed description and accompanying drawings which set forth an illustrative embodiment in which the principles of the invention are utilized.
Brief Description ofthe Figures Fig. 1 is a plan view of a flow cell for moving a fluid sample past an examination area.
Fig. 2 is a schematic block diagram of a particle analyzer for use with the flow cell of Fig. 1 in practicing the present invention.
Fig. 3 is a flowchart which illustrates a process in accordance with a first embodiment of the invention for determining whether a fluid sample is flowing uniformly past an examination area of the apparatus of Fig. 1.
Fig. 4 is a top view of the examination area of the flow cell of Fig. 1. Fig. 5 is a flowchart which illustrates a process in accordance with a second embodiment of the invention for determining whether fluid is flowing uniformly past the examination area of the apparatus of Fig. 1.
Fig. 6 is a bar graph which illustrates how the determiriing step 512 may be carried out in the process illustrated in Fig. 5.
Fig. 7 is a flowchart which illustrates a process in accordance with a third embodiment of the invention for determining whether fluid is flowing smoothly past the examination area of the apparatus of Fig. 1.
Fig. 8 is a bar graph which illustrates how the determination step may be carried out in the process illustrated in Fig. 7. Fig. 9 is a flowchart which illustrates a process in accordance with a fourth embodiment of the invention for determining whether fluid is flowing smoothly past the examination area of the apparatus of Fig. 1.
Fig. 10 is a illustrates the data flow of the process shown in the flowchart of Fig. 9.
Figs. 11-16 are examples of membership functions for computing a degree of support for various measurement values in step 904 of the flowchart of Fig. 9.
Fig. 17 is an example of a membership function for computing a defuzzified value, FD(Center of Maxes), in step 908 of the flowchart of
Fig. 9.
Detailed Description
A first embodiment of the present invention is now described with reference to the flow chart in Fig. 3. The steps illustrated in the flowchart of Fig. 3 are executed by the processor 36 of the flow analyzer shown in Fig. 2.
At step 301, when it is desired to determine whether fluid is flowing uniformly past an examination area of a flow cell (such as the examination area 18 of the flow cell shown in Fig. 1), the processor determines an average light level intensity of the examination area (or of a portion of the examination area). Preferably, the average light intensity is measured by averaging the intensity of all portions, of more than one still frame image, that correspond to the examination area. However, while not as reliable, it is within the scope of the invention to determine the light level intensity by considering only a representative portion of each still frame image (or even of only one still frame image). Then, at step 302, a heavy stain is caused to be injected into the flow cell. For example, the heavy stain may be provided at the inlet 12 of the Fig. 1 flow cell. The heavy stain may be provided concurrently with a fluid sample under analysis, provided that the stain is not so opaque that it interferes with the ability of the particle analyzer to distinguish particles in the fluid sample under analysis. Alternately, the heavy stain may be provided exclusive of a fluid sample under analysis (e.g., at the end of a sample run).
After injecting the stain into the flow cell, at step 303, the particle analyzer looks for a period of uniformly decreased light level intensity in the still frame images, as measured in the direction perpendicular to the fluid flow. At step 304, if the uniformly decreased light level intensity was expected to be seen within a first time period since injection of the stain into the flow cell, and the first time period has passed, then, at step 306, it is reported that fluid is not flowing uniformly past the examination area 18 of the flow cell.
Otherwise, if the period of uniformly decreased light level intensity is seen within the first time period, then, at step 312, it is reported that the fluid is flowing uniformly past the examination area 18 of the flow cell. In accordance with a second embodiment of the invention, the examination area 18 is logically divided into sub-areas 18a through 18f, perpendicular to the direction of fluid flow, as shown in Fig. 4. The direction of fluid flow is shown in Fig. 4 by an arrow A. Fig. 5 illustrates the flow of the basic process executed by the processor 36 in accordance with the second embodiment of the invention. At step 502, the processor 36 initializes an index variable "i". Then, at each execution of step 504, for the current frame grabbed by the frame grabber 40, the processor 36 determines a count of the number of particles "of interest" in each sub-area 18a through 18f of the examination area 18 and accumulates this count for all grabbed frames thus far processed. (The particles that are "of interest" are the particles that the particle analyzer is designed to detect and analyze.) For example, if the particle analyzer is a blood cell analyzer, the particles "of interest" may be red blood cells. Alternately, the "particles of interest" may be calibrator particles which have been added to a fluid sample for volume calibration of the particles that the particle analyzer is designed to detect and analyze.
The processor 36 increments the index variable "i" at step 506. Once a predetermined number of frames have been processed, as indicated by "i" being equal to "n" at step 508, the processor 36 proceeds (at step 510) to calculate a statistical distribution of the number of particles counted in the sub-areas 18a through 18f. Then, at step 512, the processor 36 determines from the calculated statistical distribution the fluid flow uniformity within the examination area 18. Fig. 6 is a bar graph whose bars 20a through 20f represent the accumulated counts of the number of the particles in the sub-areas 18a through 18f. The curve 22 in Fig. 6 represents an expected distribution of particles in the sub-areas 18a through 18f if fluid is flowing uniformly past the examination area 18. Generally, a uniform distribution (or some other expected distribution determined by fluid dynamics) of particles is expected.
If the distribution of accumulated counts as represented by bars 20a through 20f in Fig. 6 is within an allowable range of the expected distribution as represented by curve 22, then the processor 36 determines that the fluid is flowing uniformly past the examination area 18. On the other hand, if the distribution of accumulated counts is outside the allowable range, then the processor 36 determines that fluid is not flowing uniformly past the examination area 18.
In accordance with a third embodiment of the invention, counts of particles of interest are accumulated during specified time periods, and the processor 36 calculates statistical parameters of the accumulated counts.
Thus, it can be seen that the third embodiment of the invention is "time- oriented", in contrast to the second embodiment of the invention which is "space-oriented". Fig. 7 illustrates the flow of the basic process executed by the processor 36 in accordance with the third embodiment of the invention. At step 702, the processor 36 initializes an index variable "i". Then, for each group of "1" frames (e.g., 50 frames, where each frame represents sampling at different times) grabbed by the frame grabber 40, the processor 36 sets a count variable, indexed by "i", to the number of sample particles counted in those frames (step 704). The processor 36 increments the index variable "i" at step 706. Once the sample particles have been counted in each of "n" groups of "1" frames (as determined at step 708), the processor 36 proceeds (at step 710) to calculate a statistical distribution of the number of sample particles counted in the groups of "1" frames. Then, at step 712, the processor 36 determines from the calculated statistical distribution if fluid has flown uniformly past the examination area 18 during the sample run.
A first aspect of the third embodiment of the invention is now discussed with reference to the bargraph shown in Fig. 8. Fig. 8 shows a bargraph of an example of sample particle counts for each of "n" (in die example of Fig. 8, n = 1 1) groups, Gl 5 G2, . . . , Gn of "1" frames (e.g., 50 frames). The height of each bar m,, m2, . . ., mn indicates the number of sample particles counted for the corresponding group of "1" frames. The dotted line labelled mave indicates an overall mean number of sample particles, calculated by the processor 36 at step 610, for each group of "1" frames over the "n" groups. Also at step 710, the processor 36 determines, for each group of "1" frames, how far the overall mean number of particles, mave, per group of "1" frames differs from the actual count, mj, of sample particles in each group of "1" frames. This difference is shown graphically in Fig. 8 by a hatch pattern. Then, at step 712, the processor determines if the differences are within an allowable range. If the differences are within the allowable range, the processor 36 indicates that the fluid sample is flowing uniformly past the examination area 18.
In accordance with a second aspect of the third embodiment of the invention, the processor 36 calculates, at step 710, a parameter x = ^ i *Mi > i = 1 TO n (1)
assuming that x defines a weighted sum of Poisson random variables. Then, at step 712, the processor 36 determines if "x" is within an allowable range.
In accordance with a third aspect of the third embodiment of the invention, at step 710, the processor 36 calculates a parameter
Figure imgf000010_0001
where x is the average number of particles counted in each group of "1" frames". The processor 36 then calculates a parameter
Figure imgf000010_0002
Since | m{ - x | indicates how much the particle count for an individual group of "1" frames deviates from the overall average, y indicates the sum of the deviations for the individual groups. Then, at step 712, the processor 36 determines if the parameter y is within some allowable range. The parameter y being within the allowable range indicates, within some degree of statistical certainty, that the fluid sample is flowing uniformly past the examination area 18.
In accordance with still further aspects according to the third embodiment of the invention, the examination area 18 is logically divided by the processor 36 into the sub-examination areas 18a to 18f (Fig. 4). The processor 36 executes the process shown in steps 702-712 of the Fig. 7 flowchart for each sub-examination area 18a to 18f.
For example, in accordance with a fourth aspect of the third embodiment, at step 710, the processor 36 calculates the statistical parameter x, as shown above in equation (1), for each of the areas 18a to 18f. Then, at step 712, the processor 36 determines whether the x calculated for each area 18a to 18f is within an allowable range. In accordance with a fifth aspect of the third embodiment, at step 710, the processor 35 calculates the statistical parameters x and y, shown above in equations (2) and (3), for each of the areas 18a to 18f. Then, at step 712, the processor 36 determines whether the y calculated for each area 18a to 18f is within an allowable range. Alternately, and particularly for situations where processing power is at a premium, at step 710, the processor 36 calculates the statistical parameters from counts of particles in only one or more selected ones of the areas 18a to 18f. At step 712, the processor 36 determines whether the calculated statistical parameters, for the selected areas, are within an allowable range. It is noted that this embodiment may not be as reliable as embodiments which utilize the entire examination area 18.
In accordance with a further embodiment of the invention, values are obtained of several measurements, where each measurement is at least partially indicative of the flow of the fluid sample through the flow cell. Then, these partial indications are combined to generate a "crisp" indication of the fluid flow. The combining is preferably implemented by the CPU 56 (Fig. 2) executing special purpose software or firmware.
One relatively simple way to combine the partial indications is to use a linear discriminant function, in which the partial indications are each multiplied by a weighting factor (indicative of the contribution of the measurement) to generate a weighted partial indication. The weighted partial indications are summed together and compared to a threshold to produce the "crisp" indication of the fluid flow.
Another way to combine the partial indications is to use "fuzzy logic". Such a fuzzy logic method is advantageous for use with an automated particle analyzer, which is normally unattended. The method preferably implemented by the CPU 56 (Fig. 2) executing special purpose software.
Fig. 9 is a flowchart illustrating an embodiment of the fuzzy logic detection method. At step 902, the fuzzy logic detection method obtains values of the several measurements (measurement values) that are at least partially indicative of the flow of the fluid sample through the examination area of the flow cell. At step 904, for each of degrees of flow 1 , 2 and 3 (e.g., clog, partial clog and no clog), a degree of support for each measurement value is determined from a membership function of the measurement. At step 906, values indicating the degree of support are provided to inference machines or rule functions , which calculate "fuzzy" degree of flow disruption values. Then, at step 908, calculated degree of flow disruption values are defuzzified to a "crisp" number to determine whether a flow disruption has occurred. The following terms, defined below, are employed in describing the fuzzy logic detection method in accordance with an embodiment of the invention. linguistic variable or measurement a linguistic description, parameter or metric related to an event or control input (e.g. temperature) terms ranked linguistic descriptions of a measurement. E.g., temperature {low, normal, high} degree of support degree to which a measurement is a member of the term; normalized to a number between 0 and 1 membership function a function that yields its degree of support given a measurement value inference machine rules of the fuzzy logic defuzzification a process of converting "fuzzy" values to a "crisp" and real value The fuzzy logic embodiment is now discussed in more detail with reference to the data flow diagram shown in Fig. 10. First, for each of one or more degrees of flow (e.g., "no clog", "partial clog" and "full clog" - shown as degree of flow 1, 2 and 3 in Fig. 10) of the fluid in the flow cell, a degree of support value for each measurement value is computed from the membership function for the measurement (step 904). To determine the membership function for a measurement, a number of sample runs are carried out and measurement values determined. Meanwhile, an operator observes the runs and determines for each run which of the one or more degrees of flow occurred. Then, the determinations of which degree of flow occurred are correlated to the measurement values to determine the membership function for each measurement. That is, a membership function for a particular measurement provides a numeric indicator of association of each of one or more linguistic descriptions of flow occurrence (e.g., no clog, partial clog and full clog) with a particular measurement value.
Next, the computed degree-of-support values for each degree-of-flow are provided to an inference machine for that degree of flow. The inference machine calculates degree-of-flow disruption "fuzzy" values (step 906). For example, the inference machine may calculate "fuzzy" values CLOG[low], CLOGjnorm] and CLOG[high] for each degree of flow where each represents a numeric indicator of an association (low association, normal association, and high association, respectively) that the particular degree of flow is the actual degree of flow.
Next, for each degree of flow, the "fuzzy values" CLOGflow], CLOG[norm] and CLOG[high] are "defuzzified" into a crisp prediction of whether the fluid sample is flowing through the flow cell with that degree of flow (step 908).
An example of the parameters, the terms of measurements and the graphical representation of the membership functions are now described. The "critical" points of each term are determined statistically, in advance, using the mean and standard deviation, and it is the critical points that determine the membership function for the parameter.
For example, approximately 1000 observed samples may be classified into three groups: no-clog, some-clog and full-clog. The mean and standard deviation of the measurements for each group of samples are then calculated.
Then, either one or two standard deviations are added to determine a range of each term.
Stain Measurement (ST)
When it is desired to determine whether the flow cell is clogged (e.g., at the end of a particular specimen analysis run), additional stain is injected into the flow cell. Then, the light intensity of subsequently grabbed still frame images is measured. If the flow cell is completely clogged, there will be no discernable difference in light reading before and after the additional stain is injected into the flow cell. Thus, the measured intensity is then translated into a numerical measure of association of whether the flow cell is clogged (true indicating a greater association of the numeric value with the condition that the flow cell is clogged and false indicating a greater association with the condition that the flow cell is not clogged).
Temporal Flow Variance (TFV) The TFV measures the variance of the flow rate from frame to frame, or from group of frames to group of frames. For example, in one embodiment, the frames are grouped into about 10 bins and the number of analytes in each bin are counted. These counts/bins are then used to calculate Tnorm, a Poisson distribution based parameter. If there are a sufficient number of analytes in each frame, then the TFV can be calculated on an individual frame basis. Example set measurement functions for TFV are shown in Fig. 11. Spatial Flow Variance (SFV) Measurement The SFV provides an indication of a variance of distribution of particles of interest in a direction perpendicular to the direction of fluid flow in the flow cell. As described above with respect to Fig. 4, the examination area is equally divided into a number of sub-areas (18a through 18f in Fig. 4) and the particles of interest (analytes) in each sub-area are accumulated. Then, these accumulated counts are used to calculate Tnorm, a Poisson distribution based parameter. As with TFV discussed above, if there are a sufficient number of analytes in each sub-area of each frame, then the SFV can be calculated on an individual frame basis. Example set measurement functions for SFV are shown in Fig. 12.
Spatial Delta(max-min) Variance (SDV) Measurement The SDV also provides an indication of a variance of distribution of particles along a direction perpendicular to the direction of fluid flow in the flow cell. The SDV, however, is more sensitive to sharp peaks and valleys in this distribution than is the SFV. The accumulated counts for each sub-area are used to calculate delta = (maximum - minimum) / average. As with TFV and SFV discussed above, if there are a sufficient number of analytes in each sub-area of each frame, then the SDV can be calculated on an individual frame basis. Example set measurement functions for SDV are shown in
Fig. 13.
Concentration Ratio of LPF to HPF (CR) Measurement The CR measures the difference in concentration of particles of interest between the low power field and high power field. The concentration of the small particles detected in low power field, designated as size group 4
(LPF4) is divided by the concentration sum of particles in high power size groups 1, 2, 3, and 4, thus excluding the smallest high power group, designated as high power field 5. Theoretically, this value should be about 16. The CR gives an indication of whether a clog occurred between frames grabbed in low power mode of the frame grabber 40 and frames grabbed in high power mode of the frame grabber 40. Example set membership functions for CR are shown in Fig. 14.
LPF4 (<EPI class) Concentration (LPC) Measurement
It was empirically determined and reasonable that a high concentration of small particles (e.g., LPF4 class analytes) is associated with a higher probability of the flow cell becoming clogged. Example set measurement functions for LPC are shown in Fig. 15.
Center of Mass Shift Variance (CMS) Measurement
The CMS measures the temporal shift of analytes along the direction perpendicular to fluid flow in the flow cell. The "center of mass" in the direction perpendicular to fluid flow for each sub-area 18a through 18f is calculated. Theoretically, the CMS measurement should be about 127.5 if fluid is flowing perfectly uniformly through the flow cell. The calculated center of mass/frames is then used to calculate Tnorm, a Poisson distribution based parameter.
Center of Mass = (X[l] + X[2] + ....X[n])./ n, where n is equal to the number of analytes and X[i] is equal to x-position of i-th analyte. Again, if there are a sufficient number of analytes in each sub- area of each frame, then the CM can be calculated on an individual frame basis.
Example set measurement functions for CMS are shown in Fig. 16.
An embodiment of the inference machines is now described. Full Clog Inference Machine
The measurements provided to the Full Clog Inference Machines are ST, TFV, SFV and SDV. Step 1 : Apply Rules (Inference Rules)
CLOGflow] = (ST[false] + TFVflow] + SFV[low] + SDV[low]) / 4 CLOG[norm] = min (TFV[norm], SFV[norm], SDV[norm]) CLOG[high] = (ST[true] + TFV[high] + SFVfhigh] + SDV[high]) / 4 Step 2: Calculate Final Score (Deffuzification) See Fig. 17 for an example of a membership function for the defuzzified value, FD(Center of Maxes) FD(low) = - CLOG[low] * 100 FD(high) = CLOG[high] * 100 if (CLOG[low] < CLOGfhigh]) FD(norm) = - (CLOG[norm] * 50) + 50 else
FD(norm) = (CLOG[norm] * 50) - 50 Weighted Sum =
(FD(low)* CLOG[low]) + (FD(norm) * CLOGfnorm]) + (FD(high) * CLOG[high])
FD(Center of Maxes) =
Weighted Sum / (CLOGflow] + CLOGfnorm] + CLOGfhigh]) The range of FD(Center of Maxes) is from -100 to 100.
if (FD(Center of Maxes) > 20) > Full Clog
Partial Clog Inference Machine
The measurements used in determining Partial Clog are LPC, CMS, CR, TFV, SFV and SDV.
Step 1 : Apply Rules (Inference Rules) Shift Variance (low) = (LPCflow] + CMSflow]) / 2
Flow Variance (low) = max (CRfnorm], TFVflow], SFVflow], SDVflow]) CLOGflow] = (Shift Variance(low) + Flow Variance(low)) / 2 Shift Variance(norm) = (LPCfnorm] + CMSfnorm]) / 2 Flow Variance (norm) - max ( TFVfnorm], SFVfnorm], SDVfnorm]) CLOGfnorm] = (Shift Variance(norm) + Flow Variance(norm)) / 2 Shift Variance(high) = (LPCfhigh] + CMSfhigh]) / 2
Flow Variance(high) = max (max(CRflow], CRfhigh]), TFVfhigh], SFVfhigh], SDVfhigh]) CLOGfhigh] =
(Shift Variance(high) + Flow Variance(high)) / 2
Step 2: Calculate Final Score (Defuzzification). Defuzzification is done the same way as the Full Clog fuzzy logic.
if (Center of Max >= 25) > Partial Clog
Example
CR = 5.02, ST = false, LPC = 69.5, CMS = 7.6, TFV = 3.8, SFV = 8.02, SDV = 1.94
Step 1 : Find Degree of Support from Each Measurement
STffalse] = 0.7 STftrue] = 0.0
CRflow] = 0.0 CRfnorm] = 0.155 CRfhigh] = 0.0
LPCflow] = 0.0 LPCfnorm] = 0.0 LPCfhigh] = 1.0 CMSflow] = 0.0 CMSfnorm] = 0.0 CMSfhigh] = 0.326
TFVflow] = 0.0 TFVfnorm] = 0.823 TFVfhigh] = 0.0
SFVflow] = 0.0 SFVfnorm] = 0.0 SFVfhigh] = 0.252
SDVflow] = 0.0 SDVfnorm] = 0.641 SDVfhigh] = 0.0 Step 2: Feed into Full Clog Inference Machine
CLOGflow] = 0.175 CLOGfnorm] = 0.0 CLOGfhigh] = 0.063
FD(Center of Maxes) = -11 < 20, therefore no full clog is inferred.
Step 3: Feed into Partial Clog Inference Machine CLOGflow] = 0.0775 CLOGfnorm] = 0.411 CLOGfhigh] = 0.458
FD(Center of Maxes) = 34 > 25, therefore partial clog is inferred.
An example is now described. A. Test Setup
Three sets of data were tested. The 3 sets are GUNKY, LEARNING and TEST. As the name implies the GUNKY set represents samples that are likely to cause clogs in the flow cell. Amorphous specimens, specimens with lots of crystals and specimens with lots of bacteria are examples of the samples which comprise the GUNKY set. The GUNKY set does not represent the population. The LEARNING set, however, is a better representation of the population and more realistic samples. The LEARNING set was used in fine-tuning the fuzzy logic. After fine-tuning the fuzzy logic, it was tested on TEST set. The data are collected from two different flow cell systems. When these samples were run, the operator visually observed the processing of each specimen and classified them into no clog, partial clog and full clog. The Fuzzy Logic was applied to these data sets. The Fuzzy
Logic classification was compared to the Operator's classification.
Test Results
The following terms, defined below, are employed in describing the test and results. True Positive (TP)
Operator reported Clog and Fuzzy Logic yielded Clog True Negative (TN) Operator reported No Clog and Fuzzy Logic yielded No Clog False Positive (FP)
Operator reported No Clog but Fuzzy Logic yielded Clog False Negative (FN)
Operator reported Clog but Fuzzy Logic yielded No Clog Sensitivity
TP / (TP + FN) Specificity
TN / (FP + TN) Positive Predictive Value (PPV)
TP/ (TP + FP) Negative Predictive Value (NPV)
TN / (TN + FN) Efficiency
(TP + TN) / (TP + FP + FN + TN)
The test results for the Fuzzy Logic classifications are shown in tables below:
Table 1 LEARNING Set
Figure imgf000020_0001
Table 2 LEARNING Set With Positive Full and Partial Clogs Added
Figure imgf000021_0001
Table 3 TEST Set
Figure imgf000021_0002
Table 4 TEST Set With Positive Full and Partial Clogs Added
Figure imgf000021_0003
Table 5 GUNKY Set
Figure imgf000021_0004
Table 6 GUNKY Set With Positive Full and Partial Clogs Added
Figure imgf000022_0001
Table 7 All Samples
Figure imgf000022_0002
Table 8 All Samples With Positive Full and Partial Clogs Added
Figure imgf000022_0003
Table 9 Fuzzy Logic Performance
Figure imgf000023_0001
It should be understood that various embodiments of the invention described hereinabove are applicable beyond the area of fluid flow through a flow cell. For example, it is important that particles contained within a fluid on a microscope slide (e.g., presented as a wet mount or as a dry smear), to be examined microscopically, are uniformly distributed throughout an examination area. For example, blood cells in a blood smear should be uniformly distributed over the examination area of the microscope slide. In such an embodiment, fields of the examination area are scanned (and examined in temporal succession) as disclosed, for example, in U.S. Patent No. 5,625,709 (incoφorated herein by reference in its entirety), the particles of these scanned fields are counted, and the counts subjected to the statistical analysis set forth above. In addition, in enhancements of this embodiment, it is determined which particular sub-areas of the examination area have a uniform distribution of particles, and the results of the particle analysis are then based only on these particular sub-areas.
It should be further understood that, while in the described embodiments many of the method steps for determining flow quality are carried out by the CPU 56 executing special purpose firmware or software, it is within the scope of the invention for some or all of the method steps to be performed by hardwired circuitry.

Claims

WHAT IS CLAIMED IS:
1. A method of determining flow quality of a fluid sample through an examination area of a flow chamber in a particle analyzer, the fluid sample including particles suspended therein, the method comprising: counting the particles by the particle analyzer as the fluid sample flows through the examination area of the flow chamber to generate a plurality of particle counts; calculating at least one parameter based upon the particle counts; determining whether the at least one calculated parameter is within an allowable range for the parameter to generate an allowable range determination; determining the flow quality of the fluid sample through the examination area in response to the allowable range determination.
2. The method of Claim 1, wherein the counting step includes: imaging substantially still frame images of the fluid as the fluid passes through the examination area; and for each of a plurality of groups of the still frame images, counting the particles from the group of still frame images to generate a particle count, wherein the particle counts generated for the plurality of groups of the still frame images are the plurality of particle counts of which the parameters are calculated in the calculating step.
3. The method of Claim 1, wherein the particles are particles that the particle analyzer was designed to detect and analyze.
4. The method of Claim 1, wherein the particles are calibrator particles added to the fluid sample.
5. The method of Claim 1, wherein the at least one parameter includes a weighted sum of the particle counts.
6. A particle analyzer for analyzing a fluid sample including particles suspended therein, comprising: a flow chamber including an examination area; an imager that images substantially still frame images of the fluid in the examination area as the fluid passes through the examination area; a counter that counts the particles in still frame images to generate a plurality of particle counts; a calculator that calculates parameters based upon the particle counts; a first determiner that determines whether the at least one calculated parameter is within an allowable range for the parameter to generate an allowable range determination; a second determiner that determines the flow quality of the fluid sample through the examination area in response to the allowable range determination.
7. The apparatus of Claim 6, wherein the particles are particles that the particle analyzer was designed to detect and analyze.
8. The apparatus of Claim 6, wherein the particles are calibrator particles added to the fluid sample.
9. The apparatus of Claim 6, wherein the at least one parameter includes a weighted sum of the particle counts.
10. A method of determining flow quality of a fluid sample through an examination area of a flow chamber, the fluid sample including particles suspended therein and the examination area including one or more sub-areas each through which only a portion of the fluid sample flows, the method comprising: for each sub-area counting the particles as the fluid flows through the sub-area to generate particle counts for the sub-area; calculating at least one parameter based upon the particle counts for the sub-area; and determining whether the at least one calculated parameter for the sub-area is within an allowable range for the parameter to generate at least one allowable range determination; and determining the flow quality of the fluid sample in response to the allowable range determination.
11. The method of Claim 10, and including: imaging one or more substantially still frame images of the fluid in the examination area as the fluid passes through the examination area, wherein the step of counting the particles for each sub-area includes counting the particles from a portion of each still frame image, the portion corresponding to that sub-area, to generate the particle count for that sub- area.
12. The method of Claim 10, wherein the particles are particles that the particle analyzer was designed to detect and analyze.
13. The method of Claim 10, wherein the particles are calibrator particles added to the fluid sample.
14. The method of Claim 10, wherein the at least one parameter includes a weighted sum of the particle counts.
15. A particle analyzer for analyzing a fluid sample including particles suspended therein, comprising: a flow chamber including an examination area, the examination area including one or more sub-areas each through which only a portion of the fluid sample flows; an imager that images substantially still frame images of the fluid in the examination area as the fluid passes through the examination area; a counter that, for each sub-area, counts the particles in a corresponding portion of the still frame images to generate a plurality of particle counts for the sub-area; a calculator that calculates, for each sub-area, parameters based upon the particle counts for the sub-area; a first determiner that determines whether the at least one calculated parameter for each sub-area is within an allowable range for the parameter, to generate an allowable range determination; a second determiner that determines the flow quality of the fluid sample through the examination area in response to the allowable range determination.
16. The apparatus of Claim 15, wherein the particles are particles that the particle analyzer was designed to detect and analyze.
17. The apparatus of Claim 15, wherein the particles are calibrator particles added to the fluid sample.
18. The apparatus of Claim 15, wherein the at least one parameter includes a weighted sum of the particle counts.
19. An apparatus for analysis of particles suspended in a fluid, comprising: a flow chamber through which the fluid flows, the flow chamber including an examination area; a particle counter for counting the particles as the fluid flows past the examination area, to generate a count; a calculator for calculating at least one parameter based upon the particle count; and a determiner that determines whether the at least one calculated parameter is within an allowable range for the parameter to determine that the flow quality of the fluid sample flowing through the examination area is acceptable.
20. The apparatus of Claim 19, wherein the particle counter includes: image pick-up means for picking up substantially still frame images of the fluid as the fluid passes through the examination area; and means for counting the particles from the still frame images to generate the particle count.
21. The apparatus of Claim 20, wherein the examination area including a plurality of sub-areas through which only a portion of the fluid sample flows, wherein the counting means for counting includes means for counting, for each sub-area, the particles from portions of the still frame images that correspond to that sub-area, to generate the particle count for that sub-area, the calculating means includes means, for each sub-area, for calculating at least one parameter based upon he particle count for that sub- area, and the determining means includes means for determining whether the at least one calculated parameter for each sub-area is within an allowable range for the parameter.
22. The apparatus of Claim 19, wherein the at least one parameter includes a weighted sum of the particle counts.
23. A method of determining flow quality of a fluid sample through an examination area of a flow chamber, the fluid sample including particles suspended therein and the examination area including a plurality of sub-areas through which only a portion of the fluid sample flows, the method comprising: for each sub-area, counting the particles as the fluid flows through the sub-area over time to generate an accumulated particle count for each sub- area; calculating at least one parameter based upon the accumulated particle counts; determining whether the at least one calculated parameter is within an allowable range for the parameter to generate an allowable range determination; and determining the flow quality of the fluid sample in response to the allowable range determination.
24. The method of Claim 23, and including: imaging one or more substantially still frame images of the fluid in the examination area as the fluid passes through the examination area, wherein the step of counting the particles for each sub-area includes counting the particles from a portion of the still frame images, the portion corresponding to that sub-area, to generate the particle count for that sub-area over time.
25. The method of Claim 23, wherein the particles are particles that the particle analyzer was designed to detect and analyze.
26. The method of Claim 23, wherein the particles are calibrator particles added to the fluid sample.
27. The method of Claim 23, wherein the at least one parameter includes a weighted sum of the particle counts.
28. A particle analyzer for analyzing a fluid sample including particles suspended therein, comprising: a flow chamber including an examination area, the examination area including one or more sub-areas each through which only a portion of the fluid sample flows; an imager that images substantially still frame images of the fluid in the examination area as the fluid passes through the examination area; a counter that, for each sub-area, counts the particles in a corresponding portion of the still frame images as the fluid flows through the sub-area over time to generate an accumulated particle count for each sub- area; a calculator that calculates parameters based upon the accumulated particle counts; a first determiner that determines whether the at least one calculated parameter is within an allowable range for the parameter, to generate an allowable range determination; and a second determiner that determines the flow quality of the fluid sample in response to the allowable range determination.
29. The apparatus of Claim 28, wherein the particles are particles that the particle analyzer was designed to detect and analyze.
30. The apparatus of Claim 29, wherein the particles are calibrator particles added to the fluid sample.
31. The apparatus of Claim 29, wherein the at least one parameter includes a weighted sum of the particle counts.
32. A method of determining flow quality of a fluid sample, in a particle analyzer, through an examination area of a flow chamber, comprising: providing a heavy stain as at least a portion of the fluid sample; determining, by the particle analyzer, if there is a period of time during which a stained portion of the fluid sample passes through the examination area after the providing step and is uniformly distributed within a first portion of the examination area, to generate a determination signal; and determining the flow quality of the fluid sample in response to the determination signal.
33. The method of Claim 32, wherein the first portion of the examination area is a portion of the examination area which spans the examination area in a direction perpendicular to a direction of flow of the fluid sample through the examination area.
34. The method of Claim 32, wherein the providing step is executed by the particle analyzer concurrently with the particle analyzer analyzing particles in a fluid sample.
35. The method of Claim 32, wherein the providing step is executed by the particle analyzer at a time when the particle analyzer is not analyzing particles in a fluid sample.
36. The method of Claim 32, wherein the particle analyzer includes a light source illuminating the examination area from a first side and an imaging source imaging the examination area from a second side, opposite the first side, to generate a plurality of images, and wherein, in the determining step, the particle analyzer determines from the images if there is a period of time during which a stained portion of the fluid sample passes through the examination area after the providing step and is uniformly distributed within the first portion of the examination area.
37. The method of Claim 36, wherein the first portion of the examination area is a portion of the examination area which spans the examination area in a direction perpendicular to a direction of flow of the fluid sample through the examination area.
38. The method of Claim 32, wherein the at least one parameter includes a weighted sum of the particle counts.
39. An apparatus for analysis of particles suspended in a fluid: a flow chamber through which the fluid flows, the flow chamber including an examination area; a particle counter for counting the particles as the fluid flows past the examination area, to generate a count; means for providing a heavy stain as at least a portion of the fluid sample; means for determining if there is a period of time during which a stained portion of the fluid sample passes through the examination area, after the heavy stain is provided, and is uniformly distributed within a first portion of the examination area, to generate a determination signal; and means for determining flow quality of the fluid sample in response to the determination signal.
40. The apparatus of Claim 39, wherein the first portion of the examination area is a portion of the examination area which spans the examination area in a direction perpendicular to a direction of flow of the fluid sample through the examination area.
41. The apparatus of Claim 39, wherein the means for providing provides the heavy stain concurrently with the particle analyzer analyzing particles in a fluid sample.
42. The apparatus of Claim 39, wherein the means for providing provides the heavy stain at a time when the particle analyzer is not analyzing particles in a fluid sample.
43. The apparatus of Claim 39, wherein the particle analyzer includes a light source illuminating the examination area from a first side and an imaging source imaging the examination area from a second side, opposite the first side, to generate a plurality of images, and wherein the means for determining determines from the images if there is a period of time during which a stained portion of the fluid sample passes through the examination area after the providing step and is uniformly distributed within the first portion of the examination area.
44. The apparatus of Claim 43, wherein the first portion of the examination area is a portion of the examination area which spans the examination area in a direction perpendicular to a direction of flow of the fluid sample through the examination area.
45. The apparatus of Claim 39, wherein the at least one parameter includes a weighted sum of the particle counts.
46. A method of determining flow quality of a fluid sample, in a particle analyzer, through an examination area of a flow chamber, comprising: measuring the fluid sample flowing through the examination area of the flow chamber to determine a plurality of measurement values each at least partially indicative of a degree of flow of the fluid sample through the examination area; combining the measurement values to generate a combined value; and determining the flow quality of the fluid sample responsive to the combined value.
47. The method of claim 46, wherein the measuring step includes for each of a plurality of degrees of flow, determining degree of support values for each measurement value by applying a membership function to the measurement value, the membership function applied corresponding to the measurement and the degree of flow, and the degree of support values for a particular degree of flow and for a particular measurement value corresponding to an association of a linguistic description of the particular degree of flow of the fluid sample with the particular measurement value; and wherein the combining step includes for each of the plurality of degrees of flow, determining degree of flow disruption prediction values by applying an inference rule to one or more of the degree of support values determined for the degree of flow, the inference rule corresponding to the degree of flow, and the degree of flow disruption value for each degree of flow corresponding to the association of that particular degree of flow based collectively on the measurement values from which the degree of support values determined for the that particular degree of flow; and for each degree of flow, determining values of a final prediction for the degree of flow by combining the determined degree of flow disruption prediction values for that degree of flow, wherein the final prediction values are the combined value.
48. The method of claim 47, wherein the step of determining values of a final prediction for each degree of flow includes: determining a weighted sum of the determined degree of flow disruption prediction values for the that degree of flow; and determining a center of maxes of the determined degree of flow disruption prediction values in response to a weighted sum of the determined degree of flow prediction values and to a sum of the determined degree of flow prediction values.
49. The method of claim 48, wherein the step of determining a center of maxes includes dividing the weighted sum of the determined degree of flow prediction values by the sum of the determined degree of flow prediction values.
50. The method of claim 47, wherein the step of determining the flow quality of the fluid sample further includes comparing the final prediction values to predetermined ranges of final prediction values.
51. The method of claim 47, wherein the plurality of measurements includes two or more of: stain measurement; temporal flow variance; spatial flow variance; spatial delta variance; a comparison of two or more independent measurements of concentration of particles in respective independent size ranges; concentration of particles having a small size; and center of mass shift variance.
52. The method of claim 46, wherein the step of combining the measurement values to generate a combined value includes: multiplying functions of each measurement value by a weighting factor to generate a weighted partial indication; and summing the weighted partial indications to generate the combined value.
53. The method of claim 46, wherein the plurality of measurements includes two or more of: stain measurement; temporal flow variance; spatial flow variance; spatial delta variance; a comparison of two or more independent measurements of concentration of particles in respective independent size ranges; concentration of particles having a small size; and center of mass shift variance.
54. A particle analyzer for analyzing a fluid sample including particles suspended therein, comprising: a flow chamber including an examination area; measuring means for measuring the fluid sample flowing through the examination area of the flow chamber to determine a plurality of measurement values each at least partially indicative of a degree of flow of the fluid sample through the examination area; combining means for combining the measurement values to generate a combined value; and determining means for determining the flow quality of the fluid sample responsive to the combined value.
55. The particle analyzer of claim 54, wherein the determining means is first determining means, and wherein the measuring means includes second determining means for, for each of a plurality of degrees of flow, determining degree of support values for each measurement value by applying a membership function to the measurement value, the membership function applied corresponding to the measurement and the degree of flow, and the degree of support values for a particular degree of flow and for a particular measurement value corresponding to an association of a linguistic description of the particular degree of flow of the fluid sample with the particular measurement value; and wherein the combining means includes third determining means for, for each of the plurality of degrees of flow, determining degree of flow disruption prediction values by applying an inference rule to one or more of the degree of support values determined for the degree of flow, the inference rule corresponding to the degree of flow, and the degree of flow disruption value for each degree of flow corresponding to the association of that particular degree of flow based collectively on the measurement values from which the degree of support values determined for the that particular degree of flow; and fourth determining means for, for each degree of flow, determining values of a final prediction for the degree of flow by combining the determined degree of flow disruption prediction values for that degree of flow, wherein the final prediction values are the combined value.
56. The particle analyzer of claim 55, wherein the fourth determining means includes: fifth determining means for determining a weighted sum of the determined degree of flow disruption prediction values for the that degree of flow; and sixth determining means for determining a center of maxes of the determined degree of flow disruption prediction values in response to a weighted sum of the determined degree of flow prediction values and to a sum of the determined degree of flow prediction values.
57. The method of claim 56, wherein the sixth determining means includes dividing means for dividing the weighted sum of the determined degree of flow prediction values by the sum of the determined degree of flow prediction values.
58. The method of claim 55, wherein the first determining means includes comparing means for comparing the final prediction values to predetermined ranges of final prediction values.
59. The method of claim 55, wherein the plurality of measurements includes two or more of: stain measurement; temporal flow variance; spatial flow variance; spatial delta variance; a comparison of two or more independent measurements of concentration of particles in respective independent size ranges; concentration of particles having a small size; and center of mass shift variance.
60. The method of claim 54, wherein the combining means includes: multiplying means for multiplying functions of each measurement value by a weighting factor to generate a weighted partial indication; and summing means for summing the weighted partial indications to generate the combined value.
61. The method of claim 54, wherein the plurality of measurements includes two or more of: stain measurement; temporal flow variance; spatial flow variance; spatial delta variance; a comparison of two or more independent measurements of concentration of particles in respective independent size ranges; concentration of particles having a small size; and center of mass shift variance.
62. A method of analyzing particles in a fluid sample in a field of view, wherein the field of view includes a plurality of sub-fields, the method comprising: forming an electrical image of the field of view in a device; reading out the electrical image from the device to produce an electrical waveform; for each sub-field, counting the particles in the field of view based upon a portion of the electrical waveform that corresponds to the sub-field, to determine a particle count for the sub-field; calculating at least one parameter based upon the particle counts; determining whether the at least one calculated parameter is within an allowable range for the parameter, to generate a determination signal; determining the quality of distribution of the sample in response to the determination signal, to generate a quality determination signal.
63. A method as in Claim 62, and further comprising: performing a particle analysis using only those sub-fields indicated by the quality determination signal to be sub-fields in which particles are distributed with an adequate quality.
PCT/US1997/008211 1996-05-15 1997-05-12 Method and apparatus for verifying uniform flow of a fluid sample through a flow cell WO1997043638A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AU31261/97A AU3126197A (en) 1996-05-15 1997-05-12 Method and apparatus for verifying uniform flow of a fluid sample through a flow cell

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US1774796P 1996-05-15 1996-05-15
US60/017,747 1996-05-15
US08/852,519 1997-05-07
US08/852,519 US6184978B1 (en) 1996-05-15 1997-05-07 Method and apparatus for verifying uniform flow of a fluid sample through a flow cell and distribution on a slide

Publications (1)

Publication Number Publication Date
WO1997043638A1 true WO1997043638A1 (en) 1997-11-20

Family

ID=26690259

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US1997/008211 WO1997043638A1 (en) 1996-05-15 1997-05-12 Method and apparatus for verifying uniform flow of a fluid sample through a flow cell

Country Status (3)

Country Link
US (3) US6184978B1 (en)
AU (1) AU3126197A (en)
WO (1) WO1997043638A1 (en)

Families Citing this family (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6184978B1 (en) * 1996-05-15 2001-02-06 International Remote Imaging Systems, Inc. Method and apparatus for verifying uniform flow of a fluid sample through a flow cell and distribution on a slide
WO1999061896A1 (en) * 1998-05-25 1999-12-02 Herbert Peter Jennissen Flow-through shear analyzer for biologically active molecules in liquid layers on surfaces
JP2000206032A (en) * 1999-01-06 2000-07-28 Bayer Corp Counting method for variable-ratio-volume particles and method for decision of characteristic of sample
GB9902549D0 (en) * 1999-02-05 1999-03-24 Jorin Limited Apparatus for and method of monitoring particulate material in a fluid
US20070240496A1 (en) * 1999-02-20 2007-10-18 Bayer Healthcare Llc Variable Rate Particle Counter and Method of Use
EP1377811B1 (en) * 2001-04-03 2008-07-16 Micronics, Inc. Split focusing cytometer
US6873725B2 (en) * 2002-09-09 2005-03-29 Coulter International Corp. Simultaneous measurement and display of 3-D size distributions of particulate materials in suspensions
US7150607B2 (en) * 2002-11-18 2006-12-19 International Remote Imaging Systems, Inc. Uniform flow displacement pump
US6825926B2 (en) * 2002-11-19 2004-11-30 International Remote Imaging Systems, Inc. Flow cell for urinalysis diagnostic system and method of making same
CN103977848B (en) 2007-04-06 2016-08-24 加利福尼亚技术学院 Microfluidic device
US8885035B2 (en) * 2009-06-16 2014-11-11 Lester F. Ludwig Electronic imaging flow-microscope for environmental remote sensing, bioreactor process monitoring, and optical microscopic tomography
US9594239B1 (en) 2009-06-16 2017-03-14 Lester F. Ludwig Optical tomography for microscopy, cell cytometry, microplate array instrumentation, crystallography, and other applications
NO336049B1 (en) 2010-10-22 2015-04-27 Seabox As Technical system, method and application for online measurement and monitoring of particle contents of an injection water stream in a subsea pipeline
EP2932266A4 (en) 2012-12-17 2016-11-30 Leukodx Ltd Systems and methods for determining a chemical state
US10610861B2 (en) 2012-12-17 2020-04-07 Accellix Ltd. Systems, compositions and methods for detecting a biological condition
US20140170678A1 (en) 2012-12-17 2014-06-19 Leukodx Ltd. Kits, compositions and methods for detecting a biological condition
US9857361B2 (en) 2013-03-15 2018-01-02 Iris International, Inc. Flowcell, sheath fluid, and autofocus systems and methods for particle analysis in urine samples
EP2972208B1 (en) 2013-03-15 2018-06-13 Iris International, Inc. Method and composition for staining and sample processing
US9316635B2 (en) 2013-03-15 2016-04-19 Iris International, Inc. Sheath fluid systems and methods for particle analysis in blood samples
AU2014409119B2 (en) 2014-10-17 2020-07-16 Iris International, Inc. Systems and methods for imaging fluid samples
CN111868834A (en) 2018-03-30 2020-10-30 Idexx实验室公司 Quality control for point-of-care diagnostic systems
CN114502941A (en) 2019-10-11 2022-05-13 贝克曼考尔特公司 Methods and compositions for dyeing and sample treatment

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4338024A (en) * 1980-05-02 1982-07-06 International Remote Imaging Systems, Inc. Flow analyzer and system for analysis of fluids with particles
US4519087A (en) * 1983-06-20 1985-05-21 International Remote Imaging Systems, Inc. Method of determining the diagnostic significance of the content of a volume of biological sample containing particles
US4612614A (en) * 1980-09-12 1986-09-16 International Remote Imaging Systems, Inc. Method of analyzing particles in a fluid sample
US4696183A (en) * 1985-03-01 1987-09-29 Hitachi, Ltd. Method and apparatus of flow analysis
US5625709A (en) * 1994-12-23 1997-04-29 International Remote Imaging Systems, Inc. Method and apparatus for identifying characteristics of an object in a field of view

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4804267A (en) * 1986-07-10 1989-02-14 Scientific Imaging Instruments, Inc. System for microscopically analyzing fluids
US4393466A (en) * 1980-09-12 1983-07-12 International Remote Imaging Systems Method of analyzing particles in a dilute fluid sample
CA1223053A (en) * 1983-10-17 1987-06-16 Her Majesty The Queen In Right Of Canada As Represented By The Minister Of National Defence Of Her Majesty's Canadian Government Ferromagnetic wear detector
JPS60213850A (en) * 1984-04-09 1985-10-26 Hitachi Ltd Particle analyzer
DE69527465T2 (en) * 1994-04-21 2003-05-08 Hitachi Ltd Monitoring procedure of a coloring solution for particle analysis and calibration procedure of particle analysis
JP3411112B2 (en) * 1994-11-04 2003-05-26 シスメックス株式会社 Particle image analyzer
US6184978B1 (en) * 1996-05-15 2001-02-06 International Remote Imaging Systems, Inc. Method and apparatus for verifying uniform flow of a fluid sample through a flow cell and distribution on a slide

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4338024A (en) * 1980-05-02 1982-07-06 International Remote Imaging Systems, Inc. Flow analyzer and system for analysis of fluids with particles
US4612614A (en) * 1980-09-12 1986-09-16 International Remote Imaging Systems, Inc. Method of analyzing particles in a fluid sample
US4519087A (en) * 1983-06-20 1985-05-21 International Remote Imaging Systems, Inc. Method of determining the diagnostic significance of the content of a volume of biological sample containing particles
US4696183A (en) * 1985-03-01 1987-09-29 Hitachi, Ltd. Method and apparatus of flow analysis
US5625709A (en) * 1994-12-23 1997-04-29 International Remote Imaging Systems, Inc. Method and apparatus for identifying characteristics of an object in a field of view

Also Published As

Publication number Publication date
AU3126197A (en) 1997-12-05
US6590646B2 (en) 2003-07-08
US6184978B1 (en) 2001-02-06
US20030002037A1 (en) 2003-01-02
US6424415B1 (en) 2002-07-23

Similar Documents

Publication Publication Date Title
US6184978B1 (en) Method and apparatus for verifying uniform flow of a fluid sample through a flow cell and distribution on a slide
US4453266A (en) Method and apparatus for measuring mean cell volume of red blood cells
US4596464A (en) Screening method for red cell abnormality
US5123055A (en) Method and an apparatus for differentiating a sample of biological cells
JP3586695B2 (en) Method and apparatus for continuously monitoring and predicting slide and specimen preparation for biological specimens
CA1135979A (en) Automated method and apparatus for classification of cells with application to the diagnosis of anemia
US20080172185A1 (en) Automatic classifying method, device and system for flow cytometry
JPH0785168A (en) Display device for clinical examination result
US11340242B2 (en) Quality control method, quality control system, management apparatus, analyzer, and quality control abnormality determination method
EP2287620B1 (en) Automatic analyzer comprising means for assisting judgment of abnormality of reaction process data
JPH05149863A (en) Particle counting method
EP0529666A2 (en) Cell analyzing apparatus
JPH06201569A (en) Apparatus and method for particle analysis
EP1849125B1 (en) Method and apparatus for analyzing body fluids
WO1997043620A1 (en) Selectively emphasizing particles of interest from a fluid sample for analysis
CN111274949B (en) Blood disease white blood cell scatter diagram similarity analysis method based on structural analysis
CN112102903A (en) Quality control system based on clinical laboratory testing result
US5735274A (en) Apparatus and method for analyzing particles including shifting a judgement region
EP0375201B1 (en) A method of separating the three major types of blood cells from a white blood cell histogram
JPH04151541A (en) Cell analyzing apparatus
EP0138591A2 (en) Screening method for red blood cell abnormality
Parasuraman et al. The use of signal detection theory in research on human-computer interaction
CN111812070B (en) Nuclear left shift and value range determining method and device and cell analyzer
US11933796B2 (en) Quality control method, quality control system, management apparatus, analyzer, and quality control abnormality determination method
Abate et al. Tomography system to acquire 3D images of cells in laminar flow: software architecture

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AU CA JP

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH DE DK ES FI FR GB GR IE IT LU MC NL PT SE

DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application
NENP Non-entry into the national phase

Ref country code: JP

Ref document number: 97541114

Format of ref document f/p: F

122 Ep: pct application non-entry in european phase
NENP Non-entry into the national phase

Ref country code: CA