Method and apparatus for detection of unsuitable conditions for automated cytology scoring

• US 5,715,327 A
• Filed: 09/20/1994
• Issued: 02/03/1998
• Est. Priority Date: 09/20/1994
• Status: Expired due to Term

• Current Assignee: Tripath Imaging Incorporated (Becton, Dickinson & Co)
• Original Assignee: NeoPath, Inc. (Becton, Dickinson & Co)
• Inventors: Lee, Shih-Jong J., Wilhelm, Paul S.
• Primary Examiner(s): Couso, Jose L.
• Assistant Examiner(s): Do, Anh Hong
• Application Number:
  US08/309,117
• Time in Patent Office: 1,232 Days
• Field of Search: 382/128, 382/133, 364/413.1, 364/555, 435/7.2, 424/3, 359/397, 356/372
• US Class Current: 382/128
• CPC Class Codes: G01N 15/1475 using image analysis for extracting features of the particle
• CPC Class Codes: G01N 21/95 characterised by the material or shape of the object to be examined G01N21/89 - G01N21/91, G01N21/94 take precedence
• CPC Class Codes: G06V 20/69 Microscopic objects, e.g. biological cells or cellular parts

Abstract

A method and apparatus for determining whether a slide is suitable for processing. A suite of suitability tests are performed by an automated microscope system. The tests include magnification error flags, staining flags, main optical density flags, including detected intermediate cell nuclei, rings around detected intermediate cell nuclei, average texture measure of detected intermediate cell nuclei, average contrast to detected intermediate cell nuclei to cytoplasm, standard deviation of detected intermediate cell nuclei optical densities, detected intermediate cell ratios, average stripe area, measure of a saturated magnification of the image, measure of a grossly saturated magnification, and the percentage of images focused properly on a first try, including images never focused properly. The automated microscope quantifies the measurements in a reliable and repeatable way.

Claims

1. A method of determining whether a slide processing system has suitably processed a biological specimen slide comprising the steps of:

(a) processing the biological specimen slide with the slide processing system;

(b) measuring at least one machine processing effectiveness parameter;

(c) checking if the at least one machine processing effectiveness parameter has exceeded a limit; and

(d) accumulating scan processing error flags.

2. The method of claim 1 wherein the step of processing the biological specimen slide comprises:

(a) obtaining at least one image of the biological specimen slide;

(b) processing the at least one image of the biological specimen slide to detect at least one object of interest; and

(c) scoring the slide as either normal or abnormal based on the at least one object of interest.

3. The method of claim 1 wherein the scan processing error flags are generated by checking if the at least one machine processing effectiveness parameter is within a range.

4. A method of determining whether a slide processing system has suitably processed a biological specimen slide comprising the steps of:

(a) processing the biological specimen slide with the slide processing system;

(b) measuring at least one machine processing effectiveness parameter;

(c) checking if the at least one machine processing effectiveness parameter has exceeded a limit; and

(d) calculating at least one percentage of images that have more than at least one predetermined number of pixels saturated.

5. A method of determining whether a slide processing system has suitably processed a biological specimen slide comprising the steps of:

(a) processing the biological specimen slide with the slide processing system;

(b) measuring at least one machine processing effectiveness parameter;

(c) checking if the at least one machine processing effectiveness parameter has exceeded a limit; and

(d) calculating at least one percentage of images acquired in focus on at least one predetermined number of tries.

6. A method of determining whether a slide processing system has suitably processed a biological specimen slide comprising the steps of:

(a) processing the biological specimen slide with the slide processing system;

(b) measuring at least one machine processing effectiveness parameter;

(c) checking if the at least one machine processing effectiveness parameter has exceeded a limit; and

(d) calculating a percentage of images that were never adequately focused.

7. A method of determining whether a specimen collection result is suitable for automatic processing comprising the steps of:

(a) processing a slide to obtain at least one slide result;

(b) measuring at least one specimen collection result parameter;

(c) checking if at least one of the at least one specimen collection result parameters have exceeded a limit;

(d) identifying a number of objects detected as intermediate cells;

(e) identifying a total number of objects detected; and

(f) dividing the number of objects detected as intermediate cells by the total number of objects detected to generate an intermediate cell ratio.

8. The method of claim 7 wherein intermediate cell nuclei are detected by an intermediate cell classifier, and wherein the intermediate cell nuclei has at least one feature value, further including the step of computing a mean of the at least one feature value over all detected intermediate cell nuclei.

9. The method of claim 8 wherein a mean optical density for each detected intermediate cell nucleus is calculated and a histogram of all intermediate cell optical densities is created.

10. The method of claim 9 further including the step of computing a standard deviation of mean optical densities for all detected intermediate cell nuclei on the slide.

11. The method of claim 8 further including the step of calculating an average feature value for all detected intermediate cell nuclei on the slide and calculating a mean of an optical density in an area produced by finding a dilation residue minus a closing of a nuclear mask for each detected intermediate cell nucleus.

12. The method of claim 8 further including the step of blurring a detected intermediate cell image using a binomial filter to create a blurred image.

13. The method of claim 12 further including the step of creating a residue image that is the magnitude of the difference between an original image and the blurred image.

14. The method of claim 13 wherein an average of the magnitude difference comprises a nuclear blur average.

15. The method of claim 8 further including the step of dilating the image of each detected intermediate cell to generate a dilated image.

16. The method of claim 15 further including the step of computing a residue image by subtracting the original image from the dilated image.

17. The method of claim 16 wherein an average of the residue image in a ring inside a nucleus is subtracted from a residue average in a ring outside the nucleus comprises an edge contrast value.

18. The method of claim 7 further including the step of calculating at least one percentage of images acquired in focus on at least one predetermined number of tries.

19. The method of claim 7 further including calculating a percentage of images that were never adequately focused.

20. A method of determining whether a slide handling result is suitable for automatic processing comprising the steps of:

(a) processing the slide;

(b) measuring at least one slide handling result parameter;

(c) checking if the at least one slide handling parameter has exceeded a limit; and

(d) calculating at least one percentage of images that have more than at least one predetermined number of pixels saturated.

21. The method of claim 20 wherein the step of processing the slide comprises:

(a) obtaining at least one image of the slide;

(b) processing the at least one image of the slide to detect at least one object of interest; and

(c) scoring the slide as either normal or abnormal based on the at least one object of interest.

22. The method of claim 20 further including the step of calculating an average area segmented by a modulation pattern detector as a total patterned area divided by a number of images in which modulation areas are detected.