Systems and methods for automatic segmentation in medical imaging with multiple anatomical structure segmentation models
Systems and methods for anatomical structure segmentation in medical images using multiple anatomical structures, instructions and segmentation models.
1. A method for automatic image segmentation, comprising:
displaying, on a selection screen, a list of anatomical structures associated with one or more images of a patient;
selecting from the list one or more anatomical structures desired to be contoured;
accessing a library of different image segmentation models;
determining the image segmentation models in the library that support segmentation of the one or more selected anatomical models;
displaying a list of the image segmentation models determined to support segmentation of the one or more selected anatomical structures;
displaying a list of anatomical structures that each of the determined segmentation models supports;
selecting all or a subset of the image segmentation models from the list of segmentation models to be used to contour each of the selected one or more anatomical structures;
applying the selected segmentation models on the selected anatomical structures; and
combining results of different segmentation model outcomes.
2. The method of claim 1 , further comprising:
accessing a database including the one or more medical images associated with the patient;
accessing a database including the library of different image segmentation models; and
determining which ones of the segmentation models in the database support segmentation of at least one of one or more anatomical structures desired to be contoured;
displaying on a selection screen the list of the image segmentation models determined to support segmentation of the one or more of the desired anatomical structures, and the list of the anatomical structures that each of the determined segmentation models supports;
for each of the anatomical structure desired to be contoured, selecting the one or more segmentation models from the displayed list of the anatomical structures;
applying the one or more selected segmentation models to generate corresponding contours for the desired anatomical structures; and
displaying the generated contours on the one or more images for user review.
3. The method of claim 2 , further comprising allowing the user to accept or reject any of the generated contours.
4. The method of claim 3 , further comprising combining the contours obtained for the same anatomical structure using different segmentation models.
5. The method of claim 4 , further comprising combining the contours obtained for different anatomical structures using different segmentation models.
6. The method of claim 5 , wherein the combining is done on a weighted basis.
7. The method of claim 5 , wherein the combining is done using Boolean operations.
8. The method of claim 5 , wherein the combining is done manually by the user.
9. The method of claim 3 , further comprising selecting a different set of segmentation models from the displayed list to generate the contours.
10. The method of claim 2 , further comprising processing the generated contours prior to user review.
11. The method of claim 10 , wherein the further processing includes automatically modifying margins of the one or more anatomical structures.
12. The method of claim 11 , wherein the further processing includes automatically rejecting selected contours.
13. The method of claim 12 , wherein the selected contours include contours of predefined planes of the one or more desired anatomical structures.
14. The method of claim 10 , further comprising allowing the user to further process the generated contours after review, wherein the further processing includes modifying the generated contours.
15. The method of claim 14 , wherein the modifying includes cutting one or more of the generated contours at predefined locations.
16. The method of claim 15 , further comprising locating the predefined locations.
17. The method of claim 2 , wherein the database including the segmentation models also includes model data associated with each of the segmentation models, wherein the model data includes one or more of a list of anatomical structures each segmentation model supports, type of protocol each anatomical structure on the list uses for contouring, and input data that each segmentation model requires to perform segmentation, wherein the input data includes one or more of the type of images the segmentation model requires to perform segmentation, and guidance markers.
18. The method of claim 2 , further comprising:
sending a control signal to a radiotherapy device to deliver radiation treatment to the patient,
wherein the delivery includes irradiating the patient according to a treatment plan based at least on the one or more medical images of the patient and the generated contours.
19. The method of claim 1 , wherein the library of image segmentation models includes segmentation models based on deep learning, and one or more of segmentation models based on atlas, segmentation models based on shape or appearance statistical models, and machine learning based segmentation models.
20. The method of claim 1 , wherein the anatomical structures include one or more of organs, tissues, bones, and blood vessels.