Topological evolution of tumor imagery
Topological evolution of a lesion within a time series of medical imagery is provided. In various embodiments, a time series of medical images is read. Each of the images depicts a subject anatomy and a lesion. The lesion has a size and a contour within each of the medical images. At least one anatomical label is read for the subject anatomy within each of the plurality of images. Based upon the contour of the lesion within each of the medical images and based on the at least one anatomical label, a further contour of the lesion is predicted outside of the time series.
1. A method comprising:
reading a time series of medical images, each of the images depicting a subject anatomy and a lesion, the lesion having a size and a contour within each of the medical images;
reading at least one anatomical label for the subject anatomy within each of the plurality of images;
determining a plurality of disjoint tissue regions around the lesion, wherein the plurality of disjoint tissue regions comprises different tissue types, each different tissue type having different tissue characteristics;
based upon the contour of the lesion within each of the medical images and based on the at least one anatomical label, predicting a further contour of the lesion by:
applying a generalized hidden Markov model to the time series of medical images for each tissue type of the different tissue types.
2. The method of claim 1 , wherein predicting the further contour comprises:
determining a plurality of characteristics of the lesion over the time series.
3. The method of claim 2 , wherein the plurality of characteristics comprise:
size, shape, growth rate, shrinkage rate, intensity, texture, or neighboring tissue characteristics.
4. The method of claim 1 , further comprising:
presenting to a user the further contour overlain on a further medical image corresponding to a time later than the time series.
5. The method of claim 1 , further comprising:
aligning the medical images based on the at least one anatomical label.
6. The method of claim 5 , wherein aligning the medical images comprises performing registration between the medical images.
7. The method of claim 1 , further comprising:
predicting at least one additional lesion.
8. The method of claim 1 , wherein applying the hidden Markov model comprises determining whether the lesion is contained in each of the plurality of disjoint tissue regions.
9. The method of claim 1 , wherein applying the hidden Markov model comprises determining at least one tissue characteristic for each of the plurality of disjoint tissue regions.
10. The method of claim 1 , wherein applying the hidden Markov model comprises applying the hidden Markov model to each of a plurality of tissue types of the time series of medical images.
11. A system comprising:
a data store comprising a time series of medical images, each of the images depicting a subject anatomy and a lesion, the lesion having a size and a contour within each of the medical images;
a computing node comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor of the computing node to cause the processor to perform a method comprising:
reading the time series of medical images;
reading at least one anatomical label for the subject anatomy within each of the plurality of images;
determining a plurality of disjoint tissue regions around the lesion, wherein the plurality of disjoint tissue regions comprises different tissue types, each different tissue type having different tissue characteristics;
based upon the contour of the lesion within each of the medical images and based on the at least one anatomical label, predicting a further contour of the lesion by:
applying a generalized hidden Markov model to the time series of medical images for each tissue type of the different tissue types.
12. The system of claim 11 , wherein predicting the further contour comprises:
determining a plurality of characteristics of the lesion over the time series.
13. The system of claim 12 , wherein the plurality of characteristics comprise:
size, shape, growth rate, shrinkage rate, intensity, texture, or neighboring tissue characteristics.
14. The system of claim 11 , further comprising:
a display, and wherein the method further comprises:
presenting on the display the further contour overlain on a further medical image corresponding to a time later than the time series.
15. The system of claim 11 , further comprising:
aligning the medical images based on the at least one anatomical label.
16. The system of claim 15 , wherein aligning the medical images comprises performing registration between the medical images.
17. The system of claim 11 , wherein applying the hidden Markov model comprises determining whether the lesion is contained in each of the plurality of disjoint tissue regions.
18. The system of claim 11 , wherein applying the hidden Markov model comprises determining at least one tissue characteristic for each of the plurality of disjoint tissue regions.
19. The system of claim 11 , wherein applying the hidden Markov model comprises applying the hidden Markov model to each of a plurality of tissue types of the time series of medical images.
20. A computer program product for predicting topological evolution of a lesion, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method comprising:
reading a time series of medical images, each of the images depicting a subject anatomy and a lesion, the lesion having a size and a contour within each of the medical images;
reading at least one anatomical label for the subject anatomy within each of the plurality of images;
determining a plurality of disjoint tissue regions around the lesion, wherein the plurality of disjoint tissue regions comprises different tissue types, each different tissue type having different tissue characteristics;
based upon the contour of the lesion within each of the medical images and based on the at least one anatomical label, predicting a further contour of the lesion by:
applying a generalized hidden Markov model to the time series of medical images for each tissue type of the different tissue types.