Method and system for processing an image of a biological specimen

• US 9,070,006 B2
• Filed: 12/14/2007
• Issued: 06/30/2015
• Est. Priority Date: 12/19/2006
• Status: Active Grant

• Current Assignee: Cytyc Corporation (Hologic Incorporated)
• Original Assignee: Hologic Incorporated
• Inventors: Wong, Kam Lin, Zahniser, David, Mui, Kathy, Linder, Jim, Sheets, Ellen
• Primary Examiner(s): Logie, Michael
• Assistant Examiner(s): MCCORMACK, JASON L
• Application Number:
  US11/957,084
• Publication Number:
  US 20080144915A1
• Time in Patent Office: 2,755 Days
• Field of Search: 382/128
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/10056 Microscopic image
• CPC Class Codes: G06T 2207/30024 Cell structures in vitro; Tissue sections in vitro
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/42 using transform domain methods
• CPC Class Codes: G06V 20/698 Matching; Classification

Abstract

Components, e.g., background, cytoplasm, nucleus and nucleolus, of a biological specimen are identified using multi-wavelength analysis. Specimen components, such as nucleoli, are selected, and a determination is made whether cells having nucleoli are cancer cells or regular repair cells based on one or more physical characteristics of the identified component. The physical characteristics can be one or more of a shape, size, texture and gray value.

Claims

1. A method of processing an image of a biological specimen, the specimen having a plurality of cells, the method comprising:

identifying nucleoli of the cells of the biological specimen in the image based on spectral characteristics of the nucleoli;

and

using a processor to evaluate one or more physical characteristics of the nucleoli in order to determine whether cells having the nucleoli have a likelihood of being cancerous or non-cancerous.

2. The method of claim 1, wherein the nucleoli of the cells of the specimen are identified using one or more light wavelengths in a range from about 400 nm to about 720 nm.

3. The method of claim 1, wherein identifying the nucleoli of the cells of the specimen comprises classifying a pixel of the image as a nucleolus of a cell of the biological specimen based on spectral characteristics of the pixel.

4. The method of claim 1, further comprising determining whether cells corresponding to the nucleoli are cancer cells or normal repair cells.

5. The method of claim 1, wherein the determination of whether cells having the nucleoli have a likelihood of being cancerous or non-cancerous is based on a shape of the nucleoli.

6. The method of claim 1, wherein the determination of whether cells having the nucleoli have a likelihood of being cancerous or non-cancerous is based on a texture of the nucleoli.

7. The method of claim 1, wherein the determination of whether cells having the nucleoli have a likelihood of being cancerous or non-cancerous is based on a size of the nucleoli.

8. The method of claim 1, wherein the determination of whether cells having the nucleoli have a likelihood of being cancerous or non-cancerous is based on a transmittance or gray value of the nucleoli or of an image pixel of the nucleoli.

9. A system for processing an image of a biological specimen having a plurality of cells, comprising:

an imager having a processor configured to identify nucleoli of the cells of the biological specimen in the image based on spectral characteristics of the nucleoli, and evaluate one or more physical characteristics of the nucleoli in order to determine whether cells in the image having the nucleoli are cancerous or non-cancerous.

10. The system of claim 9, wherein the nucleoli of the cells of the specimen are identified using one or more light wavelengths in a range from about 400 nm to about 720 nm, the system comprising a source or sources of the one or more light wavelengths.

11. The system of claim 9, wherein the processor identifies the nucleoli of the cells of the specimen based on classifying a pixel of the image as a nucleolus of a cell of the biological specimen based on spectral characteristics of the pixel.

12. The system of claim 9, the processor further configured to determine whether cells corresponding to the nucleoli are cancer cells or normal repair cells based on one or more of a shape, a texture, a size and a transmittance of the nucleoli.