Tissue Disorder Imaging Analysis

Abstract

Methods and systems are described for quantitatively determining a score for a patient having a disorder, e.g. a stroke, in a tissue, such as the brain, by analyzing both the extent and location of damage to the tissue caused by the disorder.

What is claimed is: 
     
         1 . A non-patient specific hazard atlas of a disorder in a patient tissue comprising an image of the patient's tissue, wherein the image comprises a plurality of voxels, each voxel representing a hazard value of an extent of deficit caused by damage from the disorder to that voxel of the patient tissue at that location such that the hazard value contains location specific information about the impact that the voxel's death would have on behavior. 
     
     
         2 . The hazard atlas of  claim 1 , wherein the hazard value of each voxel is based on any one or more of anatomical, vascular, and functional regions of tissue scored according to a specific numerical rating scale. 
     
     
         3 . The hazard atlas of  claim 1 , wherein the hazard value of each voxel is based on patient images and recorded patient behavior and outcomes. 
     
     
         4 . The hazard atlas of  claim 1 , wherein the tissue is brain, and the hazard value for each voxel is determined by analyzing a set of images from a group of patients that correlates damage in a specific region of the brain with a degree of loss of function and wherein the hazard value is commensurate with the degree of loss of function. 
     
     
         5 . The hazard atlas of  claim 1 , further comprising a scale that correlates the values of the voxels to a code. 
     
     
         6 . The hazard atlas of  claim 5 , wherein the code is color. 
     
     
         7 . The hazard atlas of  claim 5 , wherein the code is a series of numbers. 
     
     
         8 . The hazard atlas of  claim 5 , wherein the code is a gray scale. 
     
     
         9 . The hazard atlas of  claim 1 , wherein the hazard atlas comprises digital data and is stored on a computer-readable medium. 
     
     
         10 . The hazard atlas of  claim 1 , wherein the image of the tissue is a three-dimensional image. 
     
     
         11 . The hazard atlas of  claim 1 , wherein the tissue is brain. 
     
     
         12 . The hazard atlas of  claim 11 , wherein the disorder is stroke. 
     
     
         13 . The hazard atlas of  claim 1 , further comprising a device arranged to obtain or store the image;
 a memory or computer-readable medium storing the non-patient specific hazard atlas;   an output device; and   a processor linked to the imaging device, memory, and output device, wherein the processor is programmed to
 (i) obtain the image of the patient's tissue; 
 (ii) identify voxels of the patient linage that are damaged by the disorder as damaged patient image voxels; 
 (iii) obtain from the memory or computer-readable medium the hazard atlas of the disorder in the tissue; 
 (iv) compute a hazard score for the patient, wherein the score is the integration of all damaged patient image voxels weighted by the hazard value corresponding to that voxel location; and 
 (v) transmit the hazard score to the output device. 
   
     
     
         14 . The hazard atlas of  claim 13 , wherein the device to obtain the image of the patient's tissue is a magnetic resonance imaging device. 
     
     
         15 . The hazard atlas of  claim 13 , wherein the hazard atlas is an atlas of the brain affected by stroke. 
     
     
         16 . The hazard atlas of  claim 13 , wherein the hazard atlas comprises a scale that correlates the values of the voxels to a code. 
     
     
         17 . The hazard atlas of  claim 13 , wherein the hazard score is computed using the formula:
   Hazard Score=å j=1   N2  å i=1   N1  National Institute of Health Stroke Scale ij /volume i ×inf arct_voxel j  
   where N1 is the number of outlined regions and N2 is the total number of infarct voxels.   
     
     
         18 . The hazard atlas of  claim 13 , wherein damaged patient image voxels are identified using an image segmentation method. 
     
     
         19 . The hazard atlas of  claim 13 , wherein the image of a tissue of the patient comprises a series of images to represent a three-dimensional image. 
     
     
         20 . A method for determining a patient's hazard score for a disorder in a patient tissue, the method comprising:
 obtaining an image of a tissue of the patient, wherein the image comprises a plurality of patient image voxels;   identifying voxels of the patient image that are damaged by the disorder as damaged patient image voxels;   obtaining a non-patient specific hazard atlas of the disorder in a tissue; wherein the hazard atlas comprises a plurality of voxels, each voxel representing a hazard value of an extent of deficit caused by damage from the disorder to that voxel of the tissue at that location such that the hazard value contains location specific information about the impact that the voxel's death would have on behavior; and   computing a hazard score for the patient, wherein the hazard score is the integration of all damaged patient image voxels weighted by the hazard value corresponding to that voxel location; wherein the hazard score determines the patient's prognosis.   
     
     
         21 . The method of  claim 20 , further comprising matching the patient image to a standardized set of anatomical images of that tissue to provide a standardized image. 
     
     
         22 . The method of  claim 20 , further comprising determining a course of treatment based on the hazard score. 
     
     
         23 . The method of  claim 20 , wherein the tissue is brain tissue. 
     
     
         24 . The method of  claim 23 , wherein the disorder is stroke. 
     
     
         25 . A method for determining the efficacy of a treatment method for a patient having a tissue disorder, the method comprising:
 administering a treatment to the patient;   measuring a hazard score using the method of  claim 20  at two or more points in time after administering the treatment; and   determining the efficacy of the treatment based on the hazard scores.   
     
     
         26 . A method of generating an expert hazard atlas, the method comprising:
 identifying one or more of anatomical, vascular, and functional regions on a standard set of anatomical images of a tissue;   assigning weighting factors to the regions according to clinically-relevant assessment methodologies; and   generating a non-patient specific expert hazard atlas of the tissue comprising a plurality of voxels based on the weighting factors for each voxel in the atlas.   
     
     
         27 . A method of generating a data-driven non-patient specific hazard atlas, the method comprising:
 acquiring patient image data;   matching patient image data to a standardized set of anatomical images;   identifying regions of the patient's tissue with damage based on the image data;   assigning weighting factors to the regions based on actual patient outcome data; and   generating a data-driven non-patient specific hazard atlas of the tissue comprising a plurality of voxels based on the weighting factors for each voxel in the atlas.   
     
     
         28 . A computer program for determining a hazard score of a patient having a disorder in a patient tissue, the program residing on a computer-readable medium and comprising instructions for causing a processor to:
 (a) obtain an image of a tissue of the patient, wherein the image comprises a plurality of patient image voxels;   (b) identify voxels of the patient image that are damaged by the disorder as damaged patient image voxels;   (c) obtain from a memory or computer-readable medium a non-patient specific hazard atlas of the disorder in a tissue, wherein the hazard atlas comprises a plurality of voxels, each voxel representing a hazard value of an extent of deficit caused by damage from the disorder to that voxel of the tissue at that location such that the hazard value contains location specific information about the impact that the voxel's death would have on behavior;   (d) compute a hazard score for the patient, wherein the hazard score is the integration of all damaged patient image voxels weighted by the hazard value corresponding to that voxel location; wherein the hazard score determines the patient's prognosis; and   (e) transmit the hazard score to an output device.