Automated cytological specimen classification system and method

• US 6,327,377 B1
• Filed: 04/13/1998
• Issued: 12/04/2001
• Est. Priority Date: 04/08/1988
• Status: Expired due to Fees

• Current Assignee: Tripath Imaging Incorporated (Becton, Dickinson & Co)
• Original Assignee: Autocyte North Carolina LLC
• Inventors: Hall, Thomas L., Rutenberg, Mark R.
• Primary Examiner(s): Patel, Jayanti K.
• Application Number:
  US09/059,208
• Time in Patent Office: 1,331 Days
• Field of Search: 382/128-134, 382/224, 382/227, 382/201, 382/293, 382/181, 382/291, 382/155-160, 600/562, 422/68.1, 436/43, 436/63, 250/461.2, 128/920
• US Class Current: 382/133
• CPC Class Codes: G01N 1/30 Staining; Impregnating Fixation; Dehydration; Multistep processes for preparing samples of tissue, cell or nucleic acid material and the like for analysis
• CPC Class Codes: G01N 1/44 Sample treatment involving radiation, e.g. heat
• CPC Class Codes: G01N 15/1475 using image analysis for extracting features of the particle
• CPC Class Codes: G01N 2203/0647 Image analysis
• CPC Class Codes: G06N 3/04 Architecture, e.g. interconnection topology
• CPC Class Codes: G06V 20/69 Microscopic objects, e.g. biological cells or cellular parts
• CPC Class Codes: G06V 20/693 Acquisition
• CPC Class Codes: Y10S 128/92 Computer assisted medical diagnostics
• CPC Class Codes: Y10T 436/11 Automated chemical analysis

Abstract

An automated screening system and method for cytological specimen classification in which a neural network is utilized in performance of the classification function. Also included is an automated microscope and associated image processing circuitry.

Claims

1. A semi-automated method of classifying a specimen, comprising the steps of:

a) obtaining the specimen; and

b) classifying the specimen to determine the likelihood that individual objects in the specimen have attributes justifying further evaluation, said classifying including

i) ranking individual objects in the specimen in an order according to the likelihood that an object has attributes justifying further evaluation, and

ii) identifying location coordinates of one or more of the objects for review or further classification by a human.

2. The method of claim 1, wherein the step of identifying location coordinates comprises selecting coordinates of one or more objects according to said order.

3. The method of claim 2, wherein one or more objects associated with the selected coordinates is less than the number of ranked objects.

4. The method of claim 2, wherein the step of selecting coordinates further comprises the step of presenting images of objects corresponding to the selected coordinates.

5. The method of claim 1, wherein said ranking step is performed by a neurocomputer.

6. The method of claim 1, wherein the specimen in a cervical smear.

7. The method of claim 1, wherein said ranking step comprises assigning a value to individual objects in the specimen, said value being on a scale between a first output value associated with a first condition and a second output value associated with a second condition.

8. The method according to claim 7, wherein the first condition is benign and the second condition is non-benign.

9. The method according to claim 1, wherein said classifying step further comprises selecting individual objects to be ranked by analyzing at least one morphological feature of each object.

10. The method according to claim 9, wherein the analyzed morphological feature is selected from the group consisting of cell nucleus size and cell nucleus optical density.

11. A method for providing interactive review of objects in a specimen indicative of the highest likelihood of abnormality in the specimen, comprising the steps of:

a) obtaining the specimen; and

b) classifying the specimen to determine the likelihood that individual objects in the specimen have attributes of cell abnormality justifying further evaluation, said classifying including

i) assigning individual objects in the specimen a value according to the likelihood that an object has attributes of cell abnormality, and

ii) selecting location coordinates of one or more of the objects to provide for viewing and further classification by a human.

12. The method of claim 11, wherein said assigning step is performed by a neurocomputer.

13. The method of claim 11, wherein the specimen is a cervical smear.

14. The method according to claim 11, wherein said assigning step comprises assigning a value on a scale between a first output value associated with a first condition and a second output value associated with a second condition.

15. The method according to claim 14, wherein the first condition is benign and the second condition is non-benign.

16. A method of providing location-guided screening of a specimen for objects in the specimen having a likelihood of cell abnormality, comprising the steps of:

a) obtaining the specimen; and

b) classifying the specimen to determine the likelihood that individual objects in the specimen have attributes of cell abnormality justifying further evaluation, said classifying including

i) ranking objects in the specimen in an order according to the likelihood that an object has attributes of cell abnormality, and

ii) identifying locations of one or more of the objects to provide viewing and further classification by a human.

17. The method of claim 16, wherein the step of classifying further includes selecting one or more of the identified locations for review or further classification by a human.

18. The method of claim 16, further comprising the step of presenting images of objects to a human corresponding to one or more of the identified locations.

19. The method of claim 16, further comprising the step of generating a specimen map corresponding to one or more of the identified locations.

20. The method of claim 16, wherein ranking the objects comprises assigning individual objects in the specimen a value according to a likelihood that an object has attributes of cell abnormality.

21. The method of claim 16, wherein said ranking step is performed by a neurocomputer.

22. The method of claim 16, wherein the specimen is a cervical smear.

23. The method of claim 16, wherein said ranking step comprises assigning a value to individual objects in the specimen, said value being on a scale between a first output value associated with a first condition and a second output value associated with a second condition.

24. The method according to claim 23, wherein the first condition is benign and the second condition is non-benign.

25. The method according to claim 16, wherein said classifying step further comprises selecting individual objects to be ranked by analyzing at least one morphological feature of each object.

26. The method according to claim 25, wherein the analyzed morphological feature is selected from the group consisting of cell nucleus size and cell nucleus optical density.