Apparatus for automated identification of cell groupings on a biological specimen

• US 5,978,498 A
• Filed: 09/12/1997
• Issued: 11/02/1999
• 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): Bella, Matthew C.
• Application Number:
  US08/927,379
• Time in Patent Office: 781 Days
• Field of Search: 382/128, 382/133, 382/134, 382/224, 382/266, 382/272, 382/257
• US Class Current: 382/133
• CPC Class Codes: G06V 20/69 Microscopic objects, e.g. biological cells or cellular parts
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/155 Removing patterns interfering with the pattern to be recognised, such as ruled lines or underlines

Abstract

The detection of cellular aggregates within cytologic samples. An image analysis system with an image gathering system includes a camera, a motion controller, an illumination system and an interface obtains images of cell groupings. The image gathering system is constructed for gathering image data of a specimen mounted on a slide and is coupled to a data processing system. Image data is transferred from the image gathering system to the data processing system. The data processing system obtains objects of interest. A four step process finds cellular aggregates. The first step is acquisition of an image for analysis. The second step is extraction of image features. The third step is classification of the image to determine if any potential cellular aggregates may exist in the image. The fourth step is segmentation of objects which includes the substeps of detecting and locating potential cellular aggregates.

Claims

1. An automated cell group recognizer for locating groups of cells on a biological specimen comprising:

(a) an automated microscope for acquiring an image representation of a portion of the biological specimen wherein the automated microscope has an image representation output, wherein the image representation is taken at a single focus position;

(b) an image feature extractor coupled to the automated microscope to receive the image representation output, wherein the image feature extractor has an image feature vector output;

(c) a pixel intensity averager connected to the image representation output wherein, the pixel intensity averager has an average pixel output for pixels having an intensity greater than a first predetermined intensity;

(d) a small dark averager connected to the image representation output having a small dark averager output;

(e) a high pixel counter connected to the image representation to count a number of pixels greater than a second predetermined intensity wherein the high pixel counter has a count output; and

(f) a classifier coupled to the image feature vector output, wherein the classifier has an output indicative of a likelihood that a predetermined part of the biological specimen contains a group of cells wherein the group of cells includes a cellular aggregate, and wherein the classifier also has a likelihood output.

2. The automated cell group recognizer of claim 1 further including a segmenter connected to the image representation output, the image feature vector output, and the likelihood output where the segmenter has an output that locates cell groups on the biological specimen.

3. The automated cell group recognizer of claim 2 further including an object feature extractor connected to the group location output, feature vector output and image representation having a group feature vector output.

4. The automated cell group recognizer of claim 3 further including an object classifier connected to the group feature vector output, image feature vector output wherein the object classifier has an object classification output.

5. The automated cell group recognizer of claim 2 wherein the segmenter further comprises:

(a) a large object remover to filter objects greater than a first predetermined size from the image representation output having a first image output;

(b) a small object remover connected to the first image output to filter objects smaller than a second predetermined size having a second image output;

(c) an image thresholder connected to the second image output to locate a nuclei-like object having a third image output;

(d) an object refiner to fill small holes and remove jagged edges from the third image output having a fourth image output; and

(e) a cluster connected to the fourth image output to group nuclei that are in close proximity and eliminate those that are not in close proximity wherein the cluster has an output that locates all groups on the biological specimen.

6. The automated cell group recognizer of claim 5 where the large object remover and small object remover comprise grey scale morphological operators.

7. The automated cell group recognizer of claim 5, wherein the large object remover further comprises a field of view computer.

8. The automated cell group recognizer of claim 5 wherein the small object remover further comprises a morphological processor to perform an opening operation.

9. The automated cell group recognizer of claim 8 wherein the opening operation comprises a 9×

9 flat top operation.

10. The automated cell group recognizer of claim 1 wherein the biological specimen is a pap smear.

11. The automated cell group recognizer of claim 1 wherein the biological specimen is a gynecological specimen.

12. The automated cell group recognizer of claim 1 wherein the group of cells comprise cells that overlap.

13. The automated cell group recognizer of claim 1 wherein the portion of the biological specimen comprises at least one field of view of the automated microscope.

14. The automated cell group recognizer of claim 1 wherein the first predetermined intensity is at least 200 for an 8 bit pixel.

15. The automated cell group recognizer of claim 1 wherein the small dark averager further comprises a computer processor coupled to the image representation output, where the image representation output includes an original image, and where the computer processor computes a small dark average according to the following equation:

##EQU4## where N_Pixels is the number of pixels in an image, All_Pixels indicates that the summation covers all pixels in the image, I_org is the original image, and is the morphological dilation operator.

16. The automated cell group recognizer of claim 1 wherein the classifier further comprises an average high pixel value comparator to compare an average high pixel value against a first predetermined intensity threshold wherein the average high pixel value comparator outputs the likelihood output.

17. The automated cell group recognizer of claim 1 wherein the classifier further comprises a small dark averager/high pixel comparator connected to the image representation output wherein the small dark averager/high pixel comparator outputs the likelihood output.

18. An automated cell group recognizer for locating groups of cells on a biological specimen comprising:

(a) an automated microscope for acquiring an image representation of a portion of the biological specimen wherein the automated microscope has an image representation output, wherein the image representation is taken at a single focus position;

(b) an image feature extractor coupled to the automated microscope to receive the image representation, wherein the image feature extractor has an image feature vector output;

(c) a classifier coupled to the image feature vector output, wherein the classifier has an output indicative of a likelihood that a predetermined part of the biological specimen contains a group of cells wherein the group of cells includes a cellular aggregate, and wherein the classifier also has a likelihood output; and

(d) a computer processor coupled to the image representation output, where the image representation output includes an original image, and where the computer processor computes a small dark average according to the following equation;

where N_Pixels is the number of pixels in an image, All_Pixels indicates that the ##EQU5## summation covers all pixels in the image, I_org is the original image, and ⊕

is the morphological dilation operator.

19. An automated cell group recognizer for locating groups of cells on a biological specimen comprising:

(a) an automated microscope for acquiring an image representation of a portion of the biological specimen wherein the automated microscope has an image representation output wherein the image representation is taken at a single focus position;

(b) an image feature extractor coupled to the automated microscope to receive the image representation, wherein the image feature extractor has an image feature vector output;

(c) a classifier coupled to the image feature vector output wherein the classifier has an output indicative of a likelihood that a predetermined part of the biological specimen contains a group of cells wherein the group of cells includes a cellular aggregate, and wherein the classifier also has a likelihood output; and

(d) a small dark averager/high pixel comparator connected to the image representation output, where the image representation output includes an original image, wherein the small dark averager/high pixel comparator further comprises a processor where image representation output is used to compute a small dark average by the following equation;

##EQU6## where N_Pixels is the number of pixels in an image, All_Pixels indicates that the summation covers all pixels in the image, I_org is the original image, and ⊕

is the morphological dilation operator.

20. A method for automated cell group recognition for locating groups of cells on a biological specimen comprising the steps of:

(a) acquiring an image representation of a portion of the biological specimen, wherein the image representation is taken at a single focus position;

(b) image feature extracting an image feature vector from the image representation;

(c) pixel intensity averaging the image representation to generate an average pixel output for pixels having an intensity greater than a first predetermined intensity;

(d) small dark averaging the image representation to generate a small dark average output;

(e) high pixel counting the image representation to count a number of pixels greater than a second predetermined intensity; and

(f) classifying the image feature vector to generate a likelihood that a predetermined part of the biological specimen contains a group of cells wherein the group of cells includes a cellular aggregate.

21. The method of claim 20 further comprising the step of locating cell groups of the biological specimen based on the image representation, the image feature vector and the likelihood that a predetermined part of the biological specimen contains a group of cells.

22. An image segmenting apparatus for segmenting an image, wherein the image is taken at a single focus position, the image segmenting apparatus comprising:

(a) a large object remover to filter objects greater than a first predetermined size from the image having a first image output;

(b) a small object remover connected to the first image output to filter objects small than a second predetermined size having a second image output;

(c) an image thresholder connected to the second image output to locate a nuclei-like object having third image output;

(d) an object refiner to fill small holes and remove jagged edges from the third image output having a fourth image output; and

(e) a clusterer connected to the fourth image output for grouping nuclei that are in a cellular aggregate and for eliminating nuclei that are not in the cellular aggregate.