Method for determining the efficacy of an anti-cancer treatment using image analysis

• US 7,231,074 B2
• Filed: 05/15/2003
• Issued: 06/12/2007
• Est. Priority Date: 05/17/2002
• Status: Expired due to Fees

• Current Assignee: Pfizer Inc.
• Original Assignee: Pfizer Inc.
• Inventors: Raunig, David L.
• Primary Examiner(s): Johns; Andrew W.
• Application Number:
  US10/437,975
• Publication Number:
  US 20040013292A1
• Time in Patent Office: 1,489 Days
• Field of Search: 382/128, 382/130, 382/131, 382/132, 382/173, 128/922
• US Class Current: 382/128
• CPC Class Codes: G06T 2207/10056 Microscopic image
• CPC Class Codes: G06T 2207/10132 Ultrasound image
• CPC Class Codes: G06T 2207/20104 Interactive definition of region of interest [ROI]
• CPC Class Codes: G06T 2207/20116 Active contour; Active surface; Snakes
• CPC Class Codes: G06T 2207/30096 Tumor; Lesion
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/143 involving probabilistic approaches, e.g. Markov random field [MRF] modelling
• CPC Class Codes: G06T 7/45 using co-occurrence matrix computation
• CPC Class Codes: G06T 7/66 of image moments or centre of gravity
• CPC Class Codes: G06V 20/695 Preprocessing, e.g. image segmentation

Abstract

An apparatus, system, method, and computer readable medium containing computer-executable code for implementing image analysis uses multivariate statistical analysis of sample images, and allows segmentation of the image into different groups or classes, depending on a correlation to one or more sample textures, or sample surface features. In one embodiment, the invention performs multivariate statistical analysis of ultrasound images, wherein a tumor may be characterized by segmenting viable tissue from necrotic tissue, allowing for more detailed in vivo analysis of tumor growth beyond simple dimensional measurements or univariate statistical analysis. Application of the apparatus and method may also be used for characterizing other types of samples having textured features including, for example, tumor angiogenesis biomarkers from Power Doppler.

Claims

1. A method of determining the efficacy of an anti-cancer treatment protocol, the method comprising:

initially segmenting a first image of a tumor represented by a plurality of pixels into necrotic regions and viable tissue regions based upon a vector representation of the plurality of pixels;

directing one or more anti-cancer treatments to the tumor;

subsequently segmenting a second image of the tumor into post-treatment necrotic regions; and

comparing a relative size of the necrotic regions to a size of the post-treatment necrotic regions;

wherein the vector representation of the plurality of pixels at least includes a contrast value and an entropy value for each pixel of the plurality of pixels.

2. The method of claim 1, further comprising adjusting a treatment protocol based upon the comparison result.

3. A method of treating cancer, comprising:

ultrasonically imaging a tumor;

segmenting an image of the tumor based upon one or more image parameters correlated to a texture of tissue within the tumor;

determining necrotic segments of the segmented tumor image based upon the correlated texture of the tissue within the tumor;

providing a cancer treatment regimen;

redetermining necrotic segments of the segmented tumor image after said treatment is provided; and

comparing relative sizes of pre-and post-treatment necrotic segments of the segmented tumor image;

wherein said one or more image parameters includes at least an image contrast value and an image entropy value.

4. The method of claim 3, further comprising adjusting the cancer treatment regiment based upon results from said comparing step.