Methods and apparatus for measuring multiple optical properties of biological specimens

• US 5,072,382 A
• Filed: 10/02/1989
• Issued: 12/10/1991
• Est. Priority Date: 10/02/1989
• Status: Expired due to Term

• Current Assignee: Compucyte Corporation (Thorlabs Incorporated)
• Original Assignee: Louis A. Kamentsky
• Inventors: Kamentsky, Louis A.
• Primary Examiner(s): Lall, Parshotam S.
• Assistant Examiner(s): MELNICK, STEVEN A
• Application Number:
  US07/416,031
• Time in Patent Office: 799 Days
• Field of Search: 382/6, 382/50, 382/52, 364/526, 364/413.1, 364/413.08, 364/506, 364/525, 364/556, 364/413.07, 364/555, 364/497, 364/413.13, 250/350.1, 250/360.1, 356/408, 356/432, 358/106, 358/107, 358/903
• US Class Current: 382/133
• CPC Class Codes: G01N 15/147 the analysis being performed on a sample stream
• CPC Class Codes: G06V 20/69 Microscopic objects, e.g. biological cells or cellular parts

Abstract

The invention relates to an apparatus and methods for generating multiparameter optical data that characterize a population of cells. The invention includes the steps of scanning the cell population with a beam to produce sets of digital data samples, each sample set of digital data representing multiparameter optical interactions from a specific location within the cell population; storing the digital data, e.g., in a computer memory; locating a cell within the population, e.g., by comparing the digital data to a preselected threshold value; defining a neighborhood around the digital data representing the located cell; estimating a background level for the neighborhood based upon digital data corresponding to locations outside the neighborhood; and correcting each of the samples corresponding to the neighborhood with the estimated neighborhood background level to generate the optical data. The invention further relates to specific methods of background correction and data calibration as well as specific sampling features to enable precise estimates of multiple cellular constituents or other cell properties at high rates of speed.

Claims

1. A method for generating cell constituent data that characterizes a population of cells, the method comprising:

(a) scanning the cell population with a laser beam to produce digital data samples, different said samples representing different locations within the cell population;

(b) storing said digital data;

(c) locating a cell within said population;

(d) defining a neighborhood of digital data samples that includes the digital data samples corresponding to the located cell;

(e) estimating a background level for each digital data sample of the neighborhood based upon stored digital data from corresponding scan locations in prior or subsequent neighborhoods;

(f) correcting each of said digital data samples corresponding to the neighborhood with the estimated background level; and

(g) summing said corrected digital data samples in the neighborhood to generate the cell constituent data.

2. The method of claim 1, wherein said cell locating step comprises comparing said stored digital data to a preselected threshold.

3. The method of claim 1, wherein said sample correcting step comprises subtracting the corresponding background level from each of said stored digital data samples in a neighborhood.

4. The method of claim 1, wherein said digital data samples represent a plurality of optical properties of each cell in the population.

5. The method of claim 1, wherein the size of said neighborhood is set to statistically include a single cell.

6. The method of claim 1, wherein the scan locations of step (e) comprise the same positions of scans in said prior or subsequent neighborhoods.

7. The method of claim 1 wherein said cell population is on a movable surface that moves with a stepping motion, said stepping motion is synchronized to a scanner, and digital location data is generated from which the step and scan location corresponding to a cell can be derived.

8. The method of claim 7, wherein said cells on said movable surface can be visually observed.

9. The method of claim 1, wherein said neighborhood background level is determined by finding minimum digital data sample values in memory locations corresponding to actual locations adjacent the neighborhood for each sample in said neighborhood.

10. The method of claim 1, wherein said neighborhood background levels are determined by averaging a plurality of minimum digital data sample values in memory outside said neighborhood.

11. The method of claim 1, wherein the time corresponding approximately to the time the digital data was acquired is recorded and stored as synchronous digital time data, whereby each sample has a corresponding digital time point.

12. The method of claim 11, wherein said cell population is on a movable surface, said scanning proceeds in synchrony with a stepping motion of said surface, and digital location data is generated from which the location corresponding to a sample can be derived, whereby each sample has a corresponding digital time point and location.

13. The method of claim 12, further comprising the step of rescanning the population of cells, generating new digital data, location data and time data, and comparing said new data with the corresponding data from a previous scan to determine changes in said cell population.

14. The method of claim 1, wherein said locating step comprises detecting a digital data sample corresponding to the edge of the cell first contacted by said beam.

15. The method of claim 14, further comprising the steps of initially centering said neighborhood around said digital data sample corresponding to the edge of the cell first contacted by said beam, determining the maximum digital data sample within said neighborhood, and then recentering the neighborhood with respect to said maximum sample before estimating a background level for the neighborhood.

16. An apparatus for generating cell constituent data that characterizes a population of cells comprising:

(a) a scanner for scanning the cell population with a laser beam and a motorized stage to move the cell population at a right angle to the laser scan to produce digital data samples from light emitted from the cells upon laser excitation, different said samples representing different locations along a scan line within the cell population;

(b) a memory to store said digital data;

(c) means for locating digital data samples in memory corresponding to a cell within said population;

(d) means for defining a neighborhood around the located cell;

(e) means for estimating a background level for each digital data sample of the neighborhood based upon stored digital data corresponding to scan locations in prior or subsequent neighborhoods; and

(f) means for correcting each of said digital data samples corresponding to the neighborhood with the estimated neighborhood background level to generate the cell constituent data.

17. A method for characterizing a population of cells, the method comprising:

(a) scanning the cell population on a surface along a scan line with a laser beam with a spot size comparable in size to the cells to be scanned to produce analog signals from a sensor that detects light emitted by the cells and sampling said analog signals to produce digital data samples, said samples being taken at a sampling rate set such that the distance travelled per time as the spot is scanned between consecutive sampled locations approximates said spot size;

(b) displacing the scan line of said laser with respect to said surface such that different said samples represent different locations within the cell population; and

(c) processing the digital data to generate cell constituent data that characterizes the cells in the population.

18. The method of claim 17, wherein said sensor analog signals are derived from fluorescence, forward angle light scatter, extinction, or wide angle light scatter emissions.

19. The method of claim 17, wherein said spot size is 3 to 50 microns.

20. The method of claim 17, wherein said processing step comprises correcting said digital data samples with a background level to generate said cell constituent data.

21. The method of claim 20, wherein said processing step further comprises calibrating said optical data samples with a velocity normalization factor proportional to the velocity at which the light beam is scanned over each location to generate cell constituent data calibrated for scanning velocity variations.

22. The method of claim 21, wherein said calibrating comprises multiplying said digital data by said velocity normalization factor.

23. The method of claim 17, wherein the wavelength distribution of the beam is controlled by the data processor.

24. The method of claim 17 for characterizing a population of cells further comprising the steps of:

(a) locating a cell within said population;

(b) defining a neighborhood of digital data samples that includes the digital data samples corresponding to the located cell;

(c) summing all samples of digital data in a neighborhood to generate a cell constituent neighborhood value;

(d) comparing each digital sample in a neighborhood with a preselected morphology threshold to generate a binary neighborhood pattern of samples either above or below said threshold;

(e) storing said cell constituent neighborhood value and binary neighborhood pattern in memory; and

(f) generating cell characterization properties corresponding to said cell constituent neighborhood value and morphological properties corresponding to said binary neighborhood pattern for each cell.

25. The method of claim 24, wherein said cell locating step comprises comparing said stored digital data to a preselected threshold.

26. The method of claim 24 wherein the generation of said morphological properties comprises:

applying boolean functions to said binary neighborhood patterns to generate morphological properties of the cell corresponding to said neighborhood.

27. The method of claim 24, wherein said digital data samples represent a plurality of morphological properties of each cell in the population.

28. The method of claim 17 for characterizing a population of cells, the method further comprising:

(a) scanning a region of said surface without cells to generate intensity calibration data;

(b) generating an intensity normalization factor from said intensity calibration data and storing said intensity normalization factor in memory;

(c) locating a cell within said population;

(d) defining a neighborhood of digital data samples that includes the digital data samples corresponding to the located cell; and

(e) calibrating the digital data in said neighborhood with said intensity normalization factor prior to generating the cell constituent data.

29. The method of claim 28, wherein said cell constituent data calibrating step comprises multiplying said optical data by said intensity normalization factor.

30. The method of claim 28, wherein said surface without cells is a calibration slide comprising a uniform fluroescent dye surface.

31. The method of claim 28, wherein said surface without cell is a dye-filled cuvette.

32. The method of claim 28, wherein said surface without cells is a calibration slide comprising a uniform scattering surface.

33. The method of claim 28, wherein the derivation of said cell constituent data contained in said neighborhood comprises the steps of:

estimating background levels for the individual neighborhoods based upon stored digital data corresponding to locations outside said neighborhood; and

correcting each of said digital data samples in each neighborhood with the corresponding estimated neighborhood background level to generate said optical data.

34. The method of claim 28, wherein said cell locating step comprises comparing said stored digital data to a preselected threshold.

35. An apparatus for measuring and displaying cellular properties of a population of cells comprising:

a laser source for generating a beam of light to illuminate the cells to be measured with a beam spot comparable in size to the cells to be scanned, an optical signal being emitted as a result of said spot illuminating said cells;

a surface upon which the cells are located;

an optical path for directing said spot from said beam source to the cells on said surface;

a scanner interposed in said optical path between said beam source and the cells for scanning aid spot across said surface along a scan line;

a sensor for measuring said optical signal;

an analog-to-digital converter arranged to sample said optical signal at a specified rate and to produce digital data;

whereby said sampling rate is set such that the distance traveled per time as the spot is scanned between consecutive sampled locations approximates said spot size;

a means for displacing the scan line of said laser with respect to said surface such that said laser spot scans continuously over a plurality of cells on said surface;

a data processor for deriving cell constituent data from said digital data corresponding to individual cells, and processing said cell constituent data to display cellular properties of the cell population; and

a memory for storing said digital data produced by said converter.

36. The apparatus of claim 35, wherein said spot size is 3 to 50 microns.

37. The apparatus of claim 35, wherein said scanner is a resonant galvanometer scanner.

38. The apparatus of claim 35 wherein the wavelength distribution of said beam is controlled by the data processor.