Systems and methods for automated diagnosis and grading of tissue images

• US 7,761,240 B2
• Filed: 08/09/2005
• Issued: 07/20/2010
• Est. Priority Date: 08/11/2004
• Status: Expired due to Fees

• Current Assignee: David Sans
• Original Assignee: Aureon Biosciences, Inc.
• Inventors: Tabesh, Ali, Saidi, Olivier, Teverovskiy, Mikhail
• Primary Examiner(s): Clow; Lori A
• Application Number:
  US11/200,758
• Publication Number:
  US 20060064248A1
• Time in Patent Office: 1,806 Days
• Field of Search: None
• US Class Current: 702/19
• 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: G06V 20/695 Preprocessing, e.g. image segmentation

Abstract

Systems and methods are provided for automated diagnosis and grading of tissue images based on morphometric data extracted from the images by a computer. The morphometric data may include image-level morphometric data such as fractal dimension data, fractal code data, wavelet data, and/or color channel histogram data. The morphometric data may also include object-level morphometric data such as color, structural, and/or textural properties of segmented image objects (e.g., stroma, nuclei, red blood cells, etc.).

Claims

1. An apparatus for evaluating a tissue image for a medical condition, the apparatus comprising:

a model predictive of the medical condition, wherein the model is based on one or more fractal dimension features from one or more binary images, each binary image corresponding to a particular color channel of the tissue image, said model defining a relationship between the expression of said one or more fractal dimension features and the medical condition, wherein the model is configured to;

receive data identifying an expression of said one or more fractal dimension features within the tissue image; and

evaluate said data according to the relationship defined by the model to produce a value indicative of at least one of the presence, absence, or aggressiveness of the medical condition.

2. The apparatus of claim 1, further comprising an image processing tool configured to extract the one or more fractal dimension features from the tissue image, wherein said image processing tool is configured to:

threshold the tissue image at one or more fixed, equally-spaced thresholds in one or more of its red, green, and blue (RGB) color channels to obtain one or more binary images; and

compute a fractal dimension value γ

_kⁱ for each of the one or more binary images.

3. The apparatus of claim 1, wherein the model predicts whether the tissue in the tissue image is cancerous or non-cancerous.

4. The apparatus of claim 1, wherein the model predicts a cancer grade for the tissue in the tissue image.

5. The apparatus of claim 4, wherein the cancer grade comprises a Gleason grade.

6. The apparatus of claim 1, wherein the model is further based on one or more additional features, said model further defining a relationship between the expression of said one or more additional features and the medical condition, wherein said model is further configured to receive additional data identifying an expression of said one or more additional features and to evaluate said additional data according to the relationships defined by the model to produce said value indicative of at least one of the presence, absence, or aggressiveness of the medical condition.

7. An apparatus for evaluating a tissue image for a medical condition, the apparatus comprising:

a model predictive of the medical condition, wherein the model is based on one or more fractal code features from the group of fractal code features consisting of a mean square error (MSE) between a domain block and a range block, a shift parameter of an affine transform, a scaling parameter of an affine transform, a shuffling transform, and a Euclidean distance between a domain block and a range block in an image plane, said model defining a relationship between the expression of said one or more fractal code features and the medical condition, wherein the model is configured to;

receive data identifying an expression of said one or more fractal code features within the tissue image; and

evaluate said data according to the relationship defined by the model to produce a value indicative of at least one of the presence, absence, or aggressiveness of the medical condition.

8. The apparatus of claim 7, further comprising an image processing tool configured to extract the one or more fractal code features from the tissue image, wherein the image processing tool is configured to:

partition the tissue image into non-overlapping blocks of two different sizes;

range blocks B_j and domain blocks Ā

_i; and

search for a domain block Ā

_i and a transformation T_i such that the transformed domain block T_iĀ

_i is an approximation of B_j,

wherein one or more characteristics of the transformation comprise the one or more fractal code features.

9. The apparatus of claim 8, wherein said image processing tool is configured to search for a domain block Ā

_i and a transformation T_i such that a mean square error (MSE) between the domain block and range block is minimized.

10. The apparatus of claim 7, wherein the model is further based on one or more additional features, said model further defining a relationship between the expression of said one or more additional features and the medical condition, wherein said model is further configured to receive additional data identifying an expression of said one or more additional features and to evaluate said additional data according to the relationships defined by the model to produce said value indicative of at least one of the presence, absence, or aggressiveness of the medical condition.

11. An apparatus for evaluating a tissue image for a medical condition, the apparatus comprising:

a model predictive of the medical condition, wherein the model is based on one or more wavelet features comprising one or more measurements of variance of wavelet coefficients for one or more wavelet representation subbands, said model defining a relationship between the expression of said one or more measurements of variance of wavelet coefficients and the medical condition, wherein the model is configured to;

receive data identifying an expression of said one or more measurements of variance of wavelet coefficients within the tissue image; and

evaluate said data according to the relationship defined by the model to produce a value indicative of at least one of the presence, absence, or aggressiveness of the medical condition.

12. The apparatus of claim 11, further comprising an image processing tool configured to extract the one or more wavelet features from the tissue image, wherein the image processing tool is configured to:

generate a wavelet representation of the tissue image; and

compute the variance of the wavelet subband coefficients for at least one of the wavelet representation subbands.

13. The apparatus of claim 11, wherein the one or more wavelet features comprises one or more scalar wavelet transform features.

14. The apparatus of claim 11, wherein the model is further based on one or more additional features, said model further defining a relationship between the expression of said one or more additional features and the medical condition, wherein said model is further configured to receive additional data identifying an expression of said one or more additional features and to evaluate said additional data according to the relationships defined by the model to produce said value indicative of at least one of the presence, absence, or aggressiveness of the medical condition.

15. An apparatus for evaluating a tissue image for a medical condition, the apparatus comprising:

a model predictive of the medical condition, wherein the model is based on one or more color channel histogram features comprising one or more pixel counts, each pixel count corresponding to an intensity of a particular color channel of the tissue image, said model defining a relationship between the expression of said one or more color histogram features and the medical condition, wherein the model is configured to;

receive data identifying an expression of said one or more color histogram features within the tissue image; and

evaluate said data according to the relationship defined by the model to produce a value indicative of at least one of the presence, absence, or aggressiveness of the medical condition.

16. The apparatus of claim 15, further comprising an image processing tool configured to extract the one or more channel histogram features from the tissue image, wherein said image processing tool is configured to:

separate the tissue image into one or more of its red, green, and blue (RGB) color channels to produce one or more color channel images;

separate one or more of the color channel images into bins representing levels of color intensity; and

compute a pixel count for one or more of the bins.

17. The apparatus of claim 15, wherein the model is further based on one or more additional features, said model further defining a relationship between the expression of said one or more additional features and the medical condition, wherein said model is further configured to receive additional data identifying an expression of said one or more additional features and to evaluate said additional data according to the relationships defined by the model to produce said value indicative of at least one of the presence, absence, or aggressiveness of the medical condition.

18. A computer readable storage medium comprising computer executable instructions recorded thereon for performing the method comprising:

receiving data identifying an expression of one or more fractal dimension features from one or more binary images of a tissue image, each binary image corresponding to a particular color channel of the tissue image; and

evaluating said data with a model predictive of a medical condition to produce a value indicative of at least one of the presence, absence, or aggressiveness of the medical condition, wherein the model is based on said one or more fractal dimension features, said model defining a relationship between the expression of said one or more fractal dimension features and the medical condition wherein said evaluating said data comprises evaluating said data according to the relationship defined by the model.

19. The computer readable storage medium of claim 18, wherein the model is further based on one or more additional features, said model further defining a relationship between the expression of said one or more additional features and the medical condition, wherein said computer executable instructions further comprise computer executable instructions for receiving additional data identifying an expression of said one or more additional features and evaluating said additional data according to the relationships defined by the model to produce said value indicative of at least one of the presence, absence, or aggressiveness of the medical condition.

20. A computer readable storage medium comprising computer executable instructions recorded thereon for performing the method comprising:

receiving data identifying for a tissue image an expression of one or more fractal code features from the group of fractal code features consisting of a mean square error (MSE) between a domain block and a range block, a shift parameter of an affine transform, a scaling parameter of an affine transform, a shuffling transform, and a Euclidean distance between a domain block and a range block in an image plane; and

evaluating said data with a model predictive of a medical condition to produce a value indicative of at least one of the presence, absence, or aggressiveness of the medical condition, wherein the model is based on said one or more fractal code features, said model defining a relationship between the expression of said one or more fractal code features and the medical condition wherein said evaluating said data comprises evaluating said data according to the relationship defined by the model.

21. The computer readable storage medium of claim 20, wherein the model is further based on one or more additional features, said model further defining a relationship between the expression of said one or more additional features and the medical condition, wherein said computer executable instructions further comprise computer executable instructions for receiving additional data identifying an expression of said one or more additional features and evaluating said additional data according to the relationships defined by the model to produce said value indicative of at least one of the presence, absence, or aggressiveness of the medical condition.

22. A computer readable storage medium comprising computer executable instructions recorded thereon for performing the method comprising:

receiving data identifying for a tissue image an expression of one or more wavelet features comprising one or more measurements of variance of wavelet coefficients for one or more wavelet representation subbands; and

evaluating said data with a model predictive of a medical condition to produce a value indicative of at least one of the presence, absence, or aggressiveness of the medical condition, wherein the model is based on said one or more wavelet features, said model defining a relationship between the expression of said one or more wavelet features and the medical condition wherein said evaluating said data comprises evaluating said data according to the relationship defined by the model.

23. The computer readable storage medium of claim 22, wherein the model is further based on one or more additional features, said model further defining a relationship between the expression of said one or more additional features and the medical condition, wherein said computer executable instructions further comprise computer executable instructions for receiving additional data identifying an expression of said one or more additional features and evaluating said additional data according to the relationships defined by the model to produce said value indicative of at least one of the presence, absence, or aggressiveness of the medical condition.

24. A computer readable storage medium comprising computer executable instructions recorded thereon for performing the method comprising:

receiving data identifying for a tissue image an expression of one or more color channel histogram features comprising one or more pixel counts, each pixel count corresponding to an intensity of a particular color channel of the tissue image; and

evaluating said data with a model predictive of a medical condition to produce a value indicative of at least one of the presence, absence, or aggressiveness of the medical condition, wherein the model is based on said one or more color channel histogram features, said model defining a relationship between the expression of said one or more color channel histogram features and the medical condition wherein said evaluating said data comprises evaluating said data according to the relationship defined by the model.

25. The computer readable storage medium of claim 24, wherein the model is further based on one or more additional features, said model further defining a relationship between the expression of said one or more additional features and the medical condition, wherein said computer executable instructions further comprise computer executable instructions for receiving additional data identifying an expression of said one or more additional features and evaluating said additional data according to the relationships defined by the model to produce said value indicative of at least one of the presence, absence, or aggressiveness of the medical condition.