Color coding an image for identifying anatomy using quantitative transmission ultrasound tomography
The speed of sound, attenuation, and reflection data obtained through quantitative Transmission ultrasound (QTUS) differs by body tissue type. Skin, fat, gland, duct and connective tissues can be classified based on the sound, attenuation, and reflection data. The system can assign coloration to breast images to provide a color-coded breast tissue volume based on the output of the classifier.
1. A method of tissue type identification comprising:
evaluating image data comprising at least a speed of sound image and a reflection image, wherein each pixel of at least the speed of sound image and the reflection image are co-registered, to assign a color to each pixel registration;
distinguishing connective tissue and fat from each other, and from ducts and glands, using both speed of sound data and reflection data, wherein the connective tissue and the fat have speed of sound data greater than the ducts and the glands, and wherein the connective tissue have reflection data greater than the fat;
storing for each pixel registration a color parameter indicating the assigned color for at least one of the connective tissue, the fat, the ducts, and the glands;
in response to receiving a request to display a particular tissue type, identifying each pixel registration storing a corresponding color parameter for the particular tissue type; and
displaying the assigned color for each pixel registration identified as storing the corresponding color parameter for the particular tissue type in a view screen.
2. The method of claim 1 , further comprising: displaying an isolated tissue or tissues as based on a selected tissue type or color.
3. The method of claim 1 , wherein storing the color parameter comprises replacing at least the speed of sound data and reflection data of the pixel registration with the color parameter.
4. The method of claim 1 , wherein storing the color parameter comprises adding the color parameter to at least the speed of sound data and the reflection data of the pixel registration.
5. The method of claim 1 further comprising:
distinguishing the ducts from the glands using speed of sound data.
6. The method of claim 1 , wherein the image data further comprises an attenuation image co-registered with the speed of sound image and the reflection image.
7. The method of claim 6 , wherein storing the color parameter comprises replacing at least the speed of sound data, the reflection data, and the attenuation data of the pixel registration with the color parameter.
8. The method of claim 6 , wherein storing the color parameter comprises adding the color parameter to at least the speed of sound data, the reflection data, and the attenuation data of the pixel registration.
9. The method of claim 1 , wherein distinguishing the tissue types utilizes a decision tree.
10. The method of claim 1 , wherein distinguishing the tissue types utilizes a typing matrix.
11. The method of claim 1 , further comprising:
determining a first range of speed of sound values for the speed of sound image defining a fat sound condition;
determining a second range of speed of sound values for the speed of sound image defining a connective tissue sound condition, wherein the second range of speed of sound values includes higher values than the first range of speed of sound values;
determining a third range of speed of sound values for the speed of sound image defining a glands sound condition, wherein the third range of speed of sound values includes higher values than the second range of speed of sound values;
determining a fourth range of speed of sound values for the speed of sound image defining a ducts condition, wherein the fourth range of speed of sound values includes higher values than the third range of speed of sound values;
determining a first range of reflection values for a reflection image defining a fat reflection condition;
determining a second range of reflection values for the reflection image defining a ducts reflection condition, wherein the second range of reflection values includes higher values than the first range of reflection values;
determining a third range of reflection values for the reflection image defining a glands reflection condition, wherein the third range of reflection values includes higher values than the first range of reflection values;
determining a fourth range of reflection values for the reflection image defining a connective tissue condition, wherein the fourth range of reflection values includes higher values than the third range of reflection values;
combining at least the fat sound condition and the fat reflection condition to create a fat criteria;
combining at least the glands sound condition and the glands reflection condition to create a glands criteria;
combining at least the ducts sound condition and the ducts reflection condition to create a ducts criteria; and
combining at least the connective tissue sound condition and the connective tissue reflection condition to create a connective tissue criteria.
12. The method of claim 11 , further comprising:
determining a first range of attenuation values for an attenuation image defining a glands attenuation condition;
determining a second range of attenuation values for the attenuation image defining a ducts attenuation condition, wherein the second range of attenuation values includes higher values than the first range of attenuation values;
determining a third range of attenuation values for the attenuation image defining a fat attenuation condition, wherein the third range of attenuation values includes higher values than the second range of attenuation values; and
determining a fourth range of attenuation values for the attenuation image defining a connective tissue condition, wherein the fourth range of attenuation values includes higher values than the third range of attenuation values.
13. The method of claim 12 , further comprising:
combining the fat sound condition, the fat reflection condition, and the fat attenuation condition to create the fat criteria;
combining the glands sound condition, the glands reflection condition, and the glands attenuation condition to create the glands criteria;
combining the ducts sound condition, the ducts reflection condition, and the ducts attenuation condition to create the ducts criteria; and
combining the connective tissue sound condition, the connective tissue reflection condition, and the connective tissue attenuation condition to create the connective tissue criteria.
14. The method of claim 1 , further comprising:
in response to receiving a request remove a particular tissue type, identifying each pixel registration storing a corresponding color parameter for the particular tissue type; and
removing the assigned color for each pixel registration identified as storing the corresponding color parameter for the particular tissue type from the view screen.
15. The method of claim 1 , further comprising:
in response to receiving a request remove a particular tissue type, identifying each pixel registration storing a corresponding color parameter for the particular tissue type; and
removing the assigned color for each pixel registration identified as storing the corresponding color parameter for the particular tissue type from the view screen.
16. The method of claim 1 , further comprising:
utilizing at least one of a linear support vector machine, radial basis function support vector machine with a Gaussian kernel, nonparametric Mann-Whitney U-test, Holm correction, discriminant analysis, decision tree, and k-nearest neighbors in assigning the color.
17. The method of claim 1 , the method further comprising:
employing shape-recognition based geometric information to distinguish glands from ducts and classifying the pixels as ducts or maintaining the pixel as glands based on the outcome of employing shape-recognition based geometric information.