Left ventricle segmentation in contrast-enhanced cine MRI datasets
A method for delineating a ventricle from MRI data relating to the heart of a patient, the method comprising: a) providing a contrast-enhanced cine MRI dataset; b) providing one or more additional MRI datasets; c) segmenting one or more features on the additional MRI dataset or datasets; d) mapping the segmented features to the contrast-enhanced cine MRI dataset; and e) using the segmented features as mapped in step d) to assist segmentation of the ventricle on the contrast-enhanced cine MRI dataset. A corresponding device and computer program are also disclosed.
1. A method for delineating a ventricle from Mill data relating to the heart of a patient, the method comprising:
a) providing a contrast-enhanced cine MRI dataset of the heart of the patient with the heart perfused by a contrast agent;
b) providing at least one additional MRI dataset of the heart of the patient, wherein the at least one additional MM dataset is different from the contrast-enhanced cine Mill dataset;
c) segmenting at least one feature of the at least one additional Mill dataset;
d) mapping the at least one feature segmented in c) from the at least one additional dataset to the contrast-enhanced cine MRI dataset, wherein location information for the at least one feature is mapped from the at least one additional dataset to the contrast-enhanced cine Mill dataset; and
e) segmenting the ventricle in the contrast-enhanced cine Mill dataset using the at least one feature mapped in d), wherein the segmenting uses the location information for the at least one feature as mapped in d).
2. A method according to claim 1 , wherein:
the at least one additional Mill dataset comprises image data relating to a short axis view of the heart.
3. A method according to claim 1 , wherein:
in e), the at least one feature mapped in d) is used as a seed for ventricle segmentation on the contrast-enhanced cine MM dataset.
4. A method according to claim 1 , wherein:
in e), the at least one feature mapped in d) is used to refine or correct ventricle segmentation on the contrast-enhanced cine MM dataset.
5. A method according to claim 1 , wherein:
the at least one additional MRI dataset comprises a delayed-enhancement MRI dataset.
6. A method according to claim 1 , wherein:
the at least one additional MRI dataset comprises a first-pass perfusion MRI dataset.
7. A method according to claim 1 , wherein:
the at least one additional MRI dataset of b) comprises a first-pass perfusion MM dataset and a delayed-enhanced MM dataset;
the segmenting of c) involves segmenting a blood-pool cavity on the first-pass perfusion MM dataset, mapping blood pool cavity information to the delayed-enhancement MRI dataset, and segmenting an infarct on the delayed-enhanced MRI dataset using the blood cavity information as mapped;
the mapping of d) involves mapping the blood pool cavity information to the contrast-enhanced cine MM dataset and mapping infarct information to the contrast-enhanced cine MRI dataset; and
the segmenting of e) involves segmenting the ventricle on the contrast-enhanced cine MRI dataset using the blood pool cavity information and the infarct information as mapped.
8. A method according to claim 1 , wherein:
the at least one additional Mill dataset comprises a tissue mapping MM dataset.
9. A method according to claim 1 , wherein the segmenting of c) involves segmenting a plurality of features in a plurality of additional MM datasets of the heart of the patient with the heart perfused by a contrast agent, wherein the plurality of additional Mill data sets are different from the contrast-enhanced cine MM dataset, wherein the plurality of features are segmented by:
segmenting a plurality of features on a first additional MM dataset of the plurality of additional Mill datasets;
mapping the plurality of features to a second additional Mill dataset of the plurality of additional MM datasets; and
segmenting a plurality of features on the second additional MM dataset of the plurality of additional Mill datasets.
10. A method according to claim 9 , wherein:
the plurality of features segmented from the plurality of additional Mill datasets is mapped on the contrast-enhanced cine Mill dataset to obtain mapped segmented features; and
the mapped segmented features are used in e) to segment the ventricle in the contrast-enhanced cine Mill dataset.
11. A method according to claim 1 , wherein:
the ventricle segmentation of e) is performed on the cardiac phase of the contrast-enhanced cine MM dataset matching the cardiac phase of the at least one additional MM dataset.
12. A method according to claim 1 , wherein:
the ventricle segmentation of e) is performed on all cardiac phases of the contrast-enhanced cine MM.
13. A computer product directly loadable into the memory of a digital computer and comprising software code portions for performing the method according to claim 1 when the product is run on a computer.
14. An imaging device for acquiring contrast-enhanced two dimensional or three-dimensional sequences of image frames, the imaging device comprising one or more acquisition modules for obtaining a plurality of image frames of the heart of a patient, such plurality of images being arranged to define a cine Mill dataset of the heart of the patient with the heart perfused by the contrast agent and at least one additional MM dataset of the heart of the patient, wherein the imaging device further comprises a processor programmed to:
a) segment at least one feature of the at least one additional MM dataset;
b) map the at least one feature segmented in a) from the at least one additional dataset to the contrast-enhanced cine MM dataset, wherein location information for the at least one feature is mapped from the at least one additional dataset to the contrast-enhanced cine Mill dataset; and
c) segment the ventricle in the contrast-enhanced cine MM dataset using the at least one feature mapped in b), wherein the segmenting of the ventricle uses the location information for the at least one feature as mapped in b).
15. An imaging device according to claim 14 , wherein the processor is further programmed to:
in c), use the at least one feature mapped in b) to refine or correct ventricle segmentation on the contrast-enhanced cine MM dataset.
16. An imaging device according to claim 14 , wherein:
the images forming the at least one additional Mill dataset are acquired during the same imaging session that acquires the images forming the contrast-enhanced cine MM dataset.
17. An imaging device according to claim 14 , wherein:
the at least one additional MRI dataset is selected from the group consisting of: a delayed-enhanced MRI dataset, a first-pass perfusion MM dataset, a tissue mapping MM dataset, a viability Mill dataset, and an edema Mill dataset.
18. An imaging device according to claim 14 , wherein:
the mapping of b) involves matching slices of the at least one additional MM dataset to slices of the contrast-enhanced cine MM dataset based on slice location and cardiac phase at which the slices are acquired.
19. A method according to claim 1 , wherein:
the mapping of d) involves matching slices of the at least one additional MM dataset to slices of the contrast-enhanced cine MM dataset based on slice location and cardiac phase at which the slices are acquired.
20. A method according to claim 1 , wherein:
the at least one feature of c) includes location information that characterizes a blood pool cavity of the ventricle;
the mapping of d) maps the location information that characterizes the blood pool cavity of the ventricle from the at least one additional dataset to the contrast-enhanced cine Mill dataset; and
the segmenting of e) uses the location information that characterizes the blood pool cavity of the ventricle as mapped in d).
21. A method according to claim 20 , wherein:
the at least one feature of c) further includes location information that characterizes an infarct of the ventricle;
the mapping of d) maps the location information that characterizes the infarct of the ventricle from the at least one additional dataset to the contrast-enhanced cine MM dataset; and
the segmenting of e) uses the location information that characterizes the infarct of the ventricle as mapped in d).
22. A method according to claim 1 , wherein:
the images forming the at least one additional Mill dataset are acquired during the same imaging session that acquires the images forming the contrast-enhanced cine MM dataset.
23. A method according to claim 1 , wherein:
the at least one additional MRI dataset of the heart of the patient is acquired with the heart perfused by a contrast agent.