Classification of a health state of tissue of interest based on longitudinal features

Abstract

A method includes determining, with a computer implemented classifier, a health state of tissue interest of a subject in at least one pair of images of the subject based on a predetermined set of longitudinal features of the tissue of interest of the subject in the at least one pair of images of the subject. The at least one pair of images of the subject includes a first image of the tissue of interest acquired at a first moment in time and a second image of the tissue of interest acquired at a second moment in time. The first and second moments in time are different moments in time. The method further includes visually displaying indicia indicative of the determined health state.

1 . A method, comprising:
 determining, with a computer implemented classifier, a health state of tissue interest of a subject in at least one pair of images of the subject based on a predetermined set of longitudinal features of the tissue of interest of the subject in the at least one pair of images of the subject,   wherein the at least one pair of images of the subject includes a first image of the tissue of interest acquired at a first moment in time and a second image of the tissue of interest acquired at a second different moment in time; and   visually displaying indicia indicative of the determined health state.   
     
     
         2 . The method of  claim 1 , further comprising:
 registering the at least one pair of images of the subject;   identifying at least one volume of interest in the registered at least one pair of images, wherein the tissue of interest is located in the at least one volume of interest;   extracting a predetermined set of longitudinal features from the identified at least one volume of interest; and   determining the health state of tissue interest by classifying the extracted set of longitudinal features.   
     
     
         3 . The method of  claim 2 , further comprising:
 obtaining non-image data about the subject; and   determining the health state of tissue interest by classifying the extracted set of longitudinal features and the non-image data.   
     
     
         4 . The method of  claim 3 , wherein the non-image data includes one or more of demographics, medical history, family history, risk factors, molecular test results, or genetic test results. 
     
     
         5 . The method of  claim 1 , further comprising:
 determining a longitudinal feature of the predetermined set of longitudinal features of the tissue of interest from the at least one pair of images.   
     
     
         6 . The method of  claim 5 , wherein the predetermined set of longitudinal features of the tissue of interest includes one or more of a change in a shape of the tissue of interest, a change in a margin of the tissue of interest, a change in a texture of the tissue of interest, a change in a vascularity of the tissue of interest, a change in an architecture of the tissue of interest, a co-existence of the tissue of interest, or a co-morbidity of the tissue of interest. 
     
     
         7 . The method of  claim 1 , further comprising:
 determining a numerical score, in a predetermined range of numerical scores, based on the determined health state in which a first score of the range indicates an absence of an disease and a second score of the range indicates a presence of the disease, wherein visually displaying the indicia includes visually displaying the numerical score.   
     
     
         8 . The method of  claim 7 , wherein the numerical score indicates a severity of the disease. 
     
     
         9 . The method of  claim 7 , further comprising:
 creating a map that maps different sub-ranges of the predetermined range of numerical scores to different levels of severity and different colors; and   visually displaying graphical indicia with a color indicative of the score based on the map.   
     
     
         10 . The method of  claim 9 , wherein the health state distribution is stratified by at least on of geographic regional, age, gender, risk-type, or co-morbidity. 
     
     
         11 . The method of  claim 1 , further comprising:
 receiving a training data set of pairs of images of the tissue, wherein each pair includes images acquired at different moment in time, and wherein a first sub-set of the training data includes a first known health state of the tissue and at least a second sub-set of the training data includes at least a second known health state of the tissue;   registering the images in the training data set;   identifying the tissue of interest in the registered the training data set images;   trending a predetermined set of longitudinal features of the tissue of interest in the registered the training data set images;   selecting a sub-set of the set of features representing a set of relevant features; and   creating and training the classifier based on the training data set and the selected sub-set of the set of features, wherein the classifier includes one or more sub-classifiers.   
     
     
         12 . The method of  claim 11 , further comprising:
 obtaining non-image data about the subject; and   determining the health state of tissue interest by classifying the extracted set of longitudinal features and the non-image data.   
     
     
         13 . The method of  claim 11 , further comprising:
 trending the predetermined set of longitudinal features by determining a difference of a feature between a pair of images.   
     
     
         14 . A computing system, comprising:
 a memory that stores instructions of an image data processing module; and   a processor that executes the instructions, which causes the processor to:
 classify a health state of tissue interest of a subject in at least one pair of images of the subject based on a predetermined set of longitudinal features of the tissue of interest of the subject in the at least one pair of images of the subject, wherein the at least one pair of images of the subject includes a first image of the tissue of interest acquired at a first moment in time and a second image of the tissue of interest acquired at a second different moment in time; and 
 visually display indicia indicative of the determined health state. 
   
     
     
         15 . The system of  claim 14 , the processor that executes the instructions, which further causes the processor to:
 register the at least one pair of images of the subject;   identify at least one volume of interest in the registered at least one pair of images, wherein the tissue of interest is located in the at least one volume of interest;   extract a predetermined set of longitudinal features from the identified at least one volume of interest; and   determine the health state of tissue interest by classifying the extracted set of longitudinal features.   
     
     
         16 . The system of  claim 15 , the processor that executes the instructions, which further causes the processor to:
 obtain non-image data about the subject; and   determine the health state of tissue interest by classifying the extracted set of longitudinal features and the non-image data.   
     
     
         17 . The system of  claim 14 , the processor that executes the instructions, which further causes the processor to:
 determine a numerical score proportional to the health state; and   visually display the numerical score through one or more of alphanumeric indicia or color.   
     
     
         18 . The system of  claim 14 , the processor that executes the instructions, which further causes the processor to:
 receive a training data set of pairs of images of the tissue, wherein each pair includes images acquired at different moment in time, and wherein a first sub-set of the training data includes a first known health state of the tissue and at least a second sub-set of the training data includes at least a second known health state of the tissue;   register the images in the training data set;   identify the tissue of interest in the registered the training data set images;   trend a predetermined set of longitudinal features of the tissue of interest in the registered the training data set images;   select a sub-set of the set of features representing a set of relevant features; and   create the classifier based on the training data set and the selected sub-set of the set of features, wherein the classifier includes one or more sub-classifiers.   
     
     
         19 . The system of  claim 18 , the processor that executes the instructions, which further causes the processor to:
 obtain non-image data about the subject; and   determine the health state of tissue interest by classifying the extracted set of longitudinal features and the non-image data.   
     
     
         20 . The system of  claim 14 , further comprising:
 trending the predetermined set of longitudinal features by determining a difference of a feature between a pair of images.   
     
     
         21 . A computer readable storage medium encoded with computer readable instructions, which, when executed by a processor, causes the processor to:
 train a classifier to determine a health state of tissue of interest based on a change in a feature of the tissue of interest in a pair of images previously acquired at different moments in time for a plurality of pair of images for one or more subjects; and   employ the classifier to classify a health state of the tissue of interest for a patient under evaluation based on a change in the feature of the tissue of interest in pair of images for the patient under evaluation acquired at different moments in time for the patient.   
     
     
         22 . The computer readable storage medium of  claim 21 , wherein the computer readable instructions, when executed by the processor, further causes the processor to:
 receive a training data set of pairs of images of the tissue, wherein each pair includes images acquired at different moment in time, and wherein a first sub-set of the training data includes a first known health state of the tissue and at least a second sub-set of the training data includes at least a second known health state of the tissue;   register the images in the training data set;   identify the tissue of interest in the registered the training data set images;   trend a predetermined set of longitudinal features of the tissue of interest in the registered the training data set images;   select a sub-set of the set of features representing a set of relevant features; and   create the classifier based on the training data set and the selected sub-set of the set of features, wherein the classifier includes one or more sub-classifiers.   
     
     
         23 . The computer readable storage medium of  claim 22 , wherein the computer readable instructions, when executed by the processor, further causes the processor to:
 obtain non-image data about the subject; and   determine the health state of tissue interest by classifying the extracted set of longitudinal features and the non-image data.   
     
     
         24 . The computer readable storage medium of  claim 22 , wherein the computer readable instructions, when executed by the processor, further causes the processor to:
 trend the predetermined set of longitudinal features by determining a difference of a feature between a pair of images.   
     
     
         25 . The computer readable storage medium of  claim 21 , wherein the computer readable instructions, when executed by the processor, further causes the processor to:
 register the at least one pair of images of the subject;   identify at least one volume of interest in the registered at least one pair of images, wherein the tissue of interest is located in the at least one volume of interest;   extract a predetermined set of longitudinal features from the identified at least one volume of interest;   determine the health state of tissue interest by classifying the extracted set of longitudinal features; and   visually display indicia indicative of the determined health state.   
     
     
         26 . The computer readable storage medium of  claim 25 , wherein the computer readable instructions, when executed by the processor, further causes the processor to:
 obtain non-image data about the subject; and   determine the health state of tissue interest by classifying the extracted set of longitudinal features and the non-image data.