Calibration in an automatic chemical testing apparatus

• US 4,043,756 A
• Filed: 12/29/1976
• Issued: 08/23/1977
• Est. Priority Date: 12/29/1976
• Status: Expired due to Term

• Current Assignee: Boehringer Mannheim Diagnostics Corp. (Roche Holding AG)
• Original Assignee: Boehringer Mannheim Diagnostics Corp. (Roche Holding AG)
• Inventors: Sommervold, David E.
• Primary Examiner(s): Serwin, R.E.
• Application Number:
  US05/758,055
• Time in Patent Office: 237 Days
• Field of Search: 23/230 R, 23/230 B, 23/253 R, 23/259, 356/36, 356/39, 235/151.35
• US Class Current: 436/43
• CPC Class Codes: G01N 2035/00702 Curve-fitting; Parameter matching; Calibration constants
• CPC Class Codes: G01N 35/00693 Calibration
• CPC Class Codes: Y10T 436/11 Automated chemical analysis
• CPC Class Codes: Y10T 436/113332 with conveyance of sample along a test line in a container or rack

Abstract

In an automatic chemical testing apparatus calibration values obtained from readout signals of analysis means must be stored for use in prior art calculating means for processing of subsequent readout signals of samples containing unknown quantities of substances in a medium. Improved means are provided for selecting the calibration values, namely a blank value and a reference value. A plurality of pairs of blank and reference samples are processed. The blank samples have known zero or baseline levels or concentrations of substances, and the reference samples have known concentration levels of each substance. These known values are stored for comparison to processed values obtained by processing the readout signals obtained in response to analysis of the blank and reference samples. The validity of each such readout signal as a calibration value is determined by the successfulness of the comparison. The most current successful blank and reference signals are selected as the calibration values. For verifying validity, the selected calibration values are utilized for processing signals obtained from a calibration control sample for which known substance concentration values are stored. A blank, reference and calibration control sample are processed as above at a later time for updating the calibration values during continued operation of the apparatus.

Claims

1. In an automatic chemical testing apparatus comprising readout means for producing readout signals indicative of optical density of samples wherein the wavelength at which optical density of a sample is measured is selected in correspondence with a particular chemistry being performed and the optical density at that wavelength is a function of the concentration of a particular substance therein for the particular chemistry, processing means for calculating a value of concentration units corresponding to the optical density indicated by the readout signal, wherein said processing means utilizes a predetermined curve function and wherein the predetermined curve function must be calibrated by the measurement of a blank sample and a reference sample each having a known concentration level of substances therein to provide calibration values indicative of a baseline and of a reference level for values of readout signals, and means for providing readout signals for a plurality of substantially identical blank samples and for a pluralty of substantially identical reference samples, the improvement of means for providing calibration values to said processing means comprising:

means for storing a first blank readout signal having a first blank value, means for storing a second blank readout signal having a second blank value, means supplying said second blank value to said processing means for obtaining a first blank concentration value based on the use of said second blank value as a calibration value, comparison means for measuring blank error between the first blank concentration value to a known stored blank concentration value for said blank sample, means responsive to said comparison means for supplying said second blank value to said processing means if the blank error is below a preselected level for providing said second blank value as a calibration value for processing of unknown samples.

2. The improvement according to claim 1 further comprising means for storinga first reference readout signal having a first reference value, means for storing a second reference readout signal having a second reference value, means supplying said second reference value to said processing means for obtaining a first reference concentration value based on the use of said second reference value as a calibration value, comparison means for measuring reference error between the first reference concentration value to a known stored reference concentration value for said reference sample, means responsive to said comparision means for supplying said second reference value to said processing means if the reference error is below a preselected level for providing said second reference value as a calibration value for processing of unknown samples.

3. The improvement according to claim 2 further comprising means responsive to said comparison means for providing an error flag output to utilization means when the blank or reference error exceeds the respective preselected level.

4. The improvement according to claim 3 for providing said first blank value as a calibration value when said blank error is above the respective preselected level and for providing said first reference value as a calibration value when the reference error is above the respective preselected level.

5. The improvement according to claim 4 further comprising means for processing a calibration control readout signal obtained in response to measurement of a calibration control sample in said processing means and utilizing the provided calibration values in said processing means to provide a calibration control concentration value, wherein said comparison means measures the calibration control concentration error between said calibration control value and a preset known calibration control value, and error sensing means responsive to said comparison means for indicating validity of said provided calibration values when said calibration control concentration error is below a preset level.

6. The improvement according to claim 5 further comprising means for providing an error flag to utilization means when the calibration control concentration error is above the preset level.

7. The improvement according to claim 6 further comprising means for storing a recalibration blank value and a recalibration reference value obtained from readout signals for a further blank and a further reference sample measured after processing of an interval of a number of unknown samples, means providing said recalibration blank and reference values to said processing means to provide respective first and second recalibration concentration values, means for comparing said first and second recalibration concentration values to said blank and reference concentration values, and means for replacing said calibration values with said further blank and further reference values when the respective comparisons indicate concentration value drift below a preset limit.

8. The improvement according to claim 7 further comprising means for processing a calibration control readout signal in response to a calibration control sample after processing said further blank and reference samples and comparing the concentration value obtained therefrom to a stored known recalibration concentration value for verifying validity of said further blank and reference values as calibration values.

9. The improvement according to claim 8 further comprising means for retaining said calibration values when the respective comparisons indicate concentration value drift above the preset limit.

10. The improvement according to claim 8 further comprising means for comparing a concentration value for each unknown sample to a stored range of values and means for providing said concentration value for each unknown sample to a stored range of values and means for providing said concentration value to utilization means when said concentration value lies within said range and means for providing a readout value indicative of optical density for the unknown sample and an off scale flag to said utilization means when the concentration value is not in the range.

11. In an automatic chemical testing apparatus comprising readout means for producing in each of a plurality of channels a readout signal indicative of optical density of an aliquot of a sample wherein the wavelength at which optical density of a sample aliquot is measured is selected in correspondence with a particular chemistry being performed and the optical density at that wavelength is a function of the concentration of a particular substance therein for the particular chemistry, processing means for calculating a value of concentration units corresponding to the optical density indicated by each of the readout signals, wherein said processing means utilizes a predetermined curve function for each channel and wherein the predetermined curve function must be calibrated by the messurement of a blank sample and a reference sample each having a known concentration level of substances therein to provide calibration values indicative of a baseline and of a reference level for values of readout signals, and means for providing readout signals for a plurality of substantially identical blank samples and for a plurality of substantially identical reference samples, the improvement of means for providing calibration values for each channel to said processing means comprising:

means for storing a first blank readpout signal having a first blank value, means for storing a second blank readout signal having a second blank value, means supplying said second blank value to said processing means for obtaining a first blank concentration value based on the use of said second blank value as a calibration value, comparison means for measuring blank error between the first blank concentration value to a known stored blank concentration value for said blank sample, means for storing said second blank value, means for providing a third blank value to said processing means for obtaining a second blank concentration value, means for obtaining a second blank concentration value, said comparison means further being connected for measuring blank error between said second blank concentration value, and means for providing a successful blank value to said processing means, as a calibration value, where a successful value is one the use of which results in a blank error below a preselected level.

12. The improvement according to claim 11 further comprising means for storing a first reference readout signal having a first reference value, means for storing a second reference readout signal having a second reference value, means supplying said second reference value to said processing means for obtaining a first reference concentration value based on the use of said second reference value as a calibration value, comparison means for measuring reference error between the first reference concentration value to a known stored reference concentration value for said reference sample, means for storing said second reference value, means for providing a third reference value to said processing means for obtaining a second reference concentration value, means for obtaining a second reference concentration value, said comparison means further being connected for measuring reference error between said second reference concentration value, and means for providing a successful reference value to said processing means, as a calibration value, where a successful value is one the use of which results in a reference error below a preselected level.

13. The improvement according to claim 12 wherein said means for providing a successful blank value or reference value comprises means for providing the most recently obtained successful value.

14. The improvement according to claim 13 further comprising means for providing an error flag to utilization means when no blank value after said first blank value is a successful value and for providing the first blank value to said processing means as a calibration value.

15. The improvement according to claim 14 further comprising means for providing an error flag to utilization means when no reference value after said first reference value is a successful value and for providing the first reference value to said processing means as a calibration value.

16. The improvement according to claim 15 further comprising means for obtaining a further blank value and a further reference value and for producing a further blank concentration value and reference concentration value after an interval of processing of unknown samples, means for producing further blank error and reference error signals, and for replacing calibration values with the values of said further blank and reference signals if they are successful and for providing an error flag to utilization means if they are not.

17. In an automatic chemical testing method comprising in each of a plurality of channels a readout signal indicative of optical density of an aliquot of a sample wherein the wavelength at which optical density of a sample aliquot is measured is selected in correspondence with a particular chemistry being performed and the optical density at that wavelength is a function of the concentration of a particular substance therein for the particular chemistry, calculating in processing means a value of concentration units corresponding to the optical density indicated by each of the readout signals, wherein said processing means utilizes a predetermined curve function for each channel and wherein the predetermined curve function must be calibrated by the measurement of a blank sample and a reference sample each having a known concentration level of substance therein to provide calibration values indicative of a baseline and of a reference level for values of readout signals, and providing readout signals for a plurality of substantially identical blank samples and for a plurality of substantially identical reference samples, the improvement of a method for providing calibration values for each channel to said processing means comprising:

obtaining a blank readout signal for each blank sample having a blank value, supplying each said blank value to said processing means for obtaining a respective blank concentration value based on the use of the said blank value as a calibration value, comparing selected ones of said blank concentration values for measuring blank error between blank concentration values, comparing at least one of said blank concentration values to a known stored blank concentration value, and supplying a successful blank value to said processing means as a calibration value, where a successful value is one the use of which results in a blank error below a preselected level.

18. The improved method according to claim 17 further comprising the steps of obtaining a reference readout signal for each reference sample having a reference value, supplying each said reference value to said processing means for obtaining a respective reference concentration value based on the use of the said reference value as a calibration value, comparing selected ones of said reference concentration values for measuring reference error between reference concentration values, comparing at least one of said reference concentration values to a known stored reference concentration value for said reference samples to obtain a further reference error value, and supplying a successful reference value to said processing means as a calibration value, where a successful value is one the use of which results in a reference error below a preselected level.

19. The improved method according to claim 18 wherein said step of providing a successful blank value or reference value comprises providing the most recently obtained successful value.

20. The improved method according to claim 19 further comprising providing an error flag to utilization means when no blank value after said first blank value is a successful value and providing the first blank value to said processing means as a calibration value.

21. The improved method according to claim 20 further comprising providing an error flag to utilization means when no reference value after said first reference value is a successful value and providing the first reference value to said processing means as a calibration value.

22. The improved method according to claim 21 further comprising obtaining a further blank value and a further reference value and for producing a further blank concentration value and reference concentration value after an interval of processing of unknown samples, producing a further blank error signal and a further reference error signal, and replacing calibration values with the values of said further blank and reference signals if they are successful and providing an error flag to utilization means if they are not.