Count accuracy control means for a blood analyses system

• US 4,706,207 A
• Filed: 06/24/1985
• Issued: 11/10/1987
• Est. Priority Date: 06/24/1985
• Status: Expired due to Fees

• Current Assignee: Nova Celltrak Incorporated
• Original Assignee: Nova Celltrak Incorporated
• Inventors: Angel, Henry R., Hennessy, James W., Carlson, Richard A.
• Primary Examiner(s): NOT, DEFINED
• Application Number:
  US06/748,003
• Time in Patent Office: 869 Days
• Field of Search: 364/414, 364/415, 364/416, 364/555, 356/39, 356/40, 356/414
• US Class Current: 702/21
• CPC Class Codes: G01N 15/12 by observing changes in resistance or impedance across apertures when traversed by individual particles, e.g. by using the Coulter principle
• CPC Class Codes: G01N 2035/00306 Housings, cabinets, control panels (details)
• CPC Class Codes: G01N 2035/00316 Detecting door closure
• CPC Class Codes: G01N 35/1097 characterised by the valves valves in general F16K

Abstract

An apparatus for making blood counts is provided with computer and instrument control to optimize performance of the apparatus. Means for counting blood cells such as platelets, white blood and red blood cells enables a plot of the number of particles counted in each of the adjacent bands or channels inclusive within a selected range of particle size so that a plot of count against particle size may be made. The computer enables analyses of plots thus obtained by system algorithms in order to correct errors of various types which may occur in the plot. For example, by calculating the logarithms of the counts of particles and of mean particle size for each particle size channel, a further plot of log count against log of blood particle size of generally symmetrical bell shape may be generated. The computer further calculates change of logarithm (Δ log) of counts in adjacent channels. A straight line of fixed slope is then plotted representing the change of logarithm counts against the logarithm of blood particle size, giving more weight to those points near the peak. The calculation steps are reversed using the points on the straight line for ultimately calculating counts in the various channels from Δ log and taking the anti-log of the log particle size to regenerate a new curve showing more accurately distribution of the selected blood particles against particle size.

Claims

1. The method of determining curve shape in a particular selected blood particle log normal size distribution curve comprising:

analyzing a blood sample using a cell counter to count blood particles within each adjacent channel of blood particle size over at least a predetermined range of particle size,

calculating the logarithms of the count of blood particles and the particle size for each particle size channel, enabling plotting of a log count vs. log of blood particle size curve of generally symmetrical bell shape,

calculating change of logarithm (Δ

log) of counts in adjacent channels,

generating a straight line fixed slope plot representing the change of logarithm of counts vs. the log of blood particle size, giving more weight to those points near the peak of the log count vs. log of blood particle size curve in determining said fixed slope line, and

reversing the calculation step using the points on the straight line plot and calculating the counts in each channel from Δ

log and taking the anti-log of log particle size to regenerate ultimately a new curve showing a redistribution of number of the selected blood particles against particle size.

2. The method of claim 1 in which an average is derived by mathematical manipulation of the regenerated new curve.

3. The method of claim 2 applied to mean platelet volume (MPV) including the additional step of finding from the ultimately regenerated curve the mean platelet volume which is mathematically the average platelet volume.

4. The method of claim 1 in which the particular selected blood particle is platelet.

5. The method of determining curve shape according to claim 1 for calculating mean cell volume (MCV) wherein;

said analyzing step counts red blood cell particles within each adjacent channel, and

using the new curve to compute MCV.

6. The method of claim 5 in which the MCV count is computed by integrating or summing up all counts under the new curve.

7. The method of claim 1 in which a mean corpuscular volume count is computed by taking the average of all counts under the regenerated new curve.

8. The method of determining curve shape for red blood count (RBC) in a blood particle normal size distribution curve comprising:

calculating the logarithm and change of logarithm (Δ

log) of counts in adjacent channels and generating a straight line plot of Δ

log against the log of blood particle size in the channels, giving more weight to those points in channels near the point where the counts in said channels are at a maximum in determining line slope, and

reversing the calculating steps using the points on the straight line plot and calculating the counts in the channels from Δ

log and taking the anti-log of log particle size in the channels to regenerate ultimately a normal curve showing a redistribution of number of particles against particle size.

9. The method of approximating curve shape in a blood particle log normal size distribution curve comprising:

analyzing a blood sample using a cell counter to count blood particles within each adjacent channel of blood particle size over at least a predetermined range of particle size,

calculating the logarithm of the count of blood particles for each particle size channel, enabling plotting a log count vs. log particle size distribution curve of generally symmetrical bell shape,

calculating change of logarithm (Δ

log) of counts in adjacent channels and the logarithm of blood cell input amplitudes,

using the least squares fit formula to generate a straight line representing change of logarithm (Δ

log) of counts in adjacent channels vs. the log of cell particle size in said channels, the straightness of which line permits examining the log curve for its uniformity, and

determining if the curve does measure up to predetermined minimum requirements, and if not, using measured data to generate the cell count for each cell size between two predetermined threshold limits.

10. The method of claim 9 done by computer, wherein counts are first subjected to a logarithm translation by a log amplifier and standards are stored for comparison together with predetermined tolerances for acceptance and rejection.

11. In a blood cell counting apparatus employing a computer, the method of assuring more accurate results comprising:

analyzing a blood sample using a cell counter to count blood particles within each adjacent channel of blood particle size over at least a predetermined range of particle size,

computing a projected count for each particle count channel,

making a periodic check of intermediate counts at predetermined intervals during each count to generate periodic results,

comparing the periodic results with said projected counts and

if any of the intermediate results deviate from the projected count, rerunning the test.

12. In the method of claim 11, employing the following intermediate steps:

following termination of counting, using a program to probe the counting system to determine possible cause of the deviation, and

taking corrective action before running the test.

13. The method of claim 11 in which before rerunning the test, the computer makes certain tests of the computer and various instrument locations using built in test capabilities and if the tests indicate correctable problems, connects those problems, but if the tests indicate uncorrectable problems, indicating an error signal.

14. The method of claim 13 in which a table of possible error signals is matched with the error signal and an indicator is selected for activation which identifies the source of the error.