Interactive and automated tissue image analysis with global training database and variable-abstraction processing in cytological specimen classification and laser capture microdissection applications

• US 20040093166A1
• Filed: 09/15/2003
• Published: 05/13/2004
• Est. Priority Date: 09/13/2002
• Status: Active Grant

• Current Assignee: Life Technologies Corporation (Thermo Fisher Scientific Incorporated)
• Original Assignee: Life Technologies Corporation (Thermo Fisher Scientific Incorporated)
• Inventors: Kil, David H.
• Granted Patent: US 8,346,483 B2
• Time in Patent Office: Days
• US Class Current: 702/19
• CPC Class Codes: G01N 1/06 providing a thin slice, e.g. microtome
• 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 15/1433 using image recognition
• CPC Class Codes: G01N 2001/284 using local activation of adhesive, i.e. Laser Capture Microdissection
• CPC Class Codes: G02B 21/365 Control or image processing arrangements for digital or video microscopes G02B21/361, G02B21/362 take precedence
• CPC Class Codes: G06F 18/211 Selection of the most significant subset of features
• CPC Class Codes: G06F 18/40 Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor
• CPC Class Codes: G06T 2207/10056 Microscopic image
• CPC Class Codes: G06T 2207/20132 Image cropping
• CPC Class Codes: G06T 2207/30024 Cell structures in vitro; Tissue sections in vitro
• CPC Class Codes: G06T 7/00 Image analysis
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/143 involving probabilistic approaches, e.g. Markov random field [MRF] modelling
• CPC Class Codes: G06V 10/771 Feature selection, e.g. selecting representative features from a multi-dimensional feature space
• CPC Class Codes: G06V 10/945 User interactive design; Environments; Toolboxes
• CPC Class Codes: G06V 20/69 Microscopic objects, e.g. biological cells or cellular parts
• CPC Class Codes: G06V 20/695 Preprocessing, e.g. image segmentation
• CPC Class Codes: G06V 20/698 Matching; Classification

Abstract

A system and method for performing tissue image analysis and region of interest identification for further processing applications such as laser capture microdissection is provided. The invention provides three-stage processing with flexible state transition that allows image recognition to be performed at an appropriate level of abstraction. The three stages include processing at one or more than one of the pixel, subimage and object levels of processing. Also, the invention provides both an interactive mode and a high-throughput batch mode which employs training files generated automatically.

Claims

1. A computer method for image analysis, comprising the steps of:

receiving an image;

transforming the image into a feature space;

selecting at least one ROI at a pixel level of processing;

extracting features from the ROI at a pixel level of processing;

selecting at least one non-ROI at a pixel level of processing;

extracting features from the non-ROI at a pixel level of processing;

ranking the extracted features based on feature performance for successful detection of a selected ROI at a pixel level of processing;

recording the ranked extracted features;

selecting a classification algorithm;

running the classification algorithm to classify the image into regions of interest at a pixel level of processing; and

recording the ROIs based on pixel processing.

2. The computer method of claim 1 wherein the step of selecting at least one ROI includes selecting one or more pixels from the image;

and wherein the step of selecting at least one non-ROI includes selecting one or more pixels from the image.

3. The computer method of claim 1 further including the step of transmitting the recorded ROIs at a pixel level of processing for laser capture microdissection.

4. The computer method of claim 1 further including the step of selecting a second level of processing.

5. The computer method of claim 4 wherein the second level of processing is subimage processing.

6. The computer method of claim 5 further including the steps of:

selecting at least one polygonal ROI at a subimage level of processing;

extracting features from the polygonal ROI at a subimage level of processing;

selecting at least one polygonal non-ROI at a subimage level of processing;

extracting features from the non-ROI at a subimage level of processing;

ranking the extracted features based on feature performance for successful detection of a selected ROI;

recording the ranked features based on subimage processing;

selecting a classification algorithm;

running the classification algorithm to classify the image into regions of interest based on subimage processing; and

recording the regions of interest based on subimage processing.

7. The computer method of claim 6 further including the step of transmitting the recorded regions of interest based on subimage processing for laser capture microdissection.

8. The computer method of claim 4 wherein the second level of processing is object processing.

9. The computer method of claim 8 further including the steps of:

selecting at least one polygonal ROI at an object level of processing;

10. The computer method of claim 9 further including the steps of:

recording the at least one polygonal ROI at an object level of processing;

transmitting the at least one polygonal region of interest based on object processing for laser capture microdissection.

11. The computer method of claim 9 further including the steps of:

extracting features from the ROI at an object level of processing;

selecting at least one polygonal non-ROI at an object level of processing;

extracting features from the non-ROI at an object level of processing;

ranking the extracted features based on feature performance for successful detection of a selected ROI;

recording the ranked features based on object processing;

selecting a classification algorithm;

running the classification algorithm to classify the image into regions of interest based on object processing;

recording the regions of interest based on object processing.

13. The computer method of claim 4 further including the step of selecting a third level of processing.

14. The computer method of claim 13 wherein the third level of processing is object level processing.

15. The computer method of claim 14 further including the steps of:

selecting at least one polygonal ROI at an object level of processing.

16. The computer method of claim 15 further including the steps of:

recording the at least one polygonal ROI at an object level of processing;

transmitting the at least one polygonal region of interest based on object processing for laser capture microdissection.

17. The computer method of claim 15 further including the steps of:

extracting features from the ROI at an object level of processing;

selecting at least one polygonal non-ROI at an object level of processing;

extracting features from the non-ROI at the object level of processing;

ranking the extracted features based on feature performance for successful detection of a selected ROI;

recording the ranked extracted features based on object processing;

selecting a classification algorithm;

running the classification algorithm to classify the image into regions of interest based on object processing; and

recording the regions of interest based on object processing.

18. The computer method of claim 17 further including the step of transmitting the regions of interest based on object processing for laser capture microdissection.

12. The computer method of claim 111 further including the step of transmitting the regions of interest based on object processing for laser capture microdissection.

19. A computer method for image analysis, comprising the steps of:

receiving a first image;

transforming the first image into a feature space;

selecting a level of abstraction;

selecting a database containing parameters based on the selected level of abstraction;

classifying the first image into regions of interest employing the parameters from the database based on the selected level of abstraction;

updating the parameters of the database for the level of abstraction with data from the first image;

receiving a second image;

transforming the second image into a feature space;

classifying the second image into regions of interest employing the updated parameters from the database based on the selected level of abstraction;

updating the parameters of the database with data from the second image.

20. The computer method for image analysis of claim 19 wherein the step of selecting a level of abstraction includes selecting pixel processing.

21. The computer method for image analysis of claim 20 further including the step of transmitting the regions of interest obtained from pixel processing for laser capture microdissection.

22. The computer method for image analysis of claim 19 wherein the step of selecting a level of abstraction includes selecting subimage processing.

23. The computer method for image analysis of claim 22 wherein the step of classifying the first image includes classifying the first image into regions of interest employing parameters from the database for pixel processing and classifying the first image into regions of interest employing parameters from the database for subimage processing;

and wherein the step of classifying the second image includes classifying the second image into regions of interest employing parameters from the database for pixel processing and classifying the second image into regions of interest employing parameters from the database for subimage processing.

24. The computer method for image analysis of claim 23 further including the step of transmitting the regions of interest obtained from subimage processing for laser capture microdissection.

25. The computer method for image analysis of claim 19 wherein the step of selecting a level of abstraction includes selecting object processing.

26. The computer method for image analysis of claim 25 wherein the step of classifying the first image includes classifying the first image into regions of interest employing parameters from the database for pixel processing and classifying the first image into regions of interest employing parameters from the database for subimage processing and classifying the first image into regions of interest employing parameters from the database for object processing;

and wherein the step of classifying the second image includes classifying the second image into regions of interest employing parameters from the database for pixel processing and classifying the second image into regions of interest employing parameters from the database for subimage processing and classifying the second image into regions of interest employing parameters from the database for object processing.

27. The computer method for image analysis of claim 26 further including the step of transmitting the regions of interest obtained from object processing for laser capture microdissection.

28. The computer method for image analysis of claim 25 wherein the step of classifying the first image includes classifying the first image into regions of interest employing parameters from the database for pixel processing and classifying the first image into regions of interest employing parameters from the database for object processing;

and wherein the step of classifying the second image includes classifying the second image into regions of interest employing parameters from the database for pixel processing and classifying the second image into regions of interest employing parameters from the database for object processing.

29. The computer method for image analysis of claim 28 further including the step of transmitting the regions of interest obtained from object processing for laser capture microdissection