Systems and methods for semi-automatic tumor segmentation

1. A system for volumetric segmentation in a plurality of planar medical images, comprising:

a display;

a user interface;

an electronic processor coupled to the display and the user interface; and

memory coupled to the electronic processor and storing instructions that, when executed

by the electronic processor, cause the system to:

receive the plurality of planar medical images;

generate a boundary of a candidate structure in the plurality of medical images using a segmentation model;

displaying, on the display, a first planar medical image from the plurality of planar medical images;

receive, by the user interface, a user input indicating a region in the first planar medical image;

generate a first planar contour of the candidate structure;

compare the region to the boundary; and

responsive to the region being at least partially within the boundary:

re-display, on the display, the first planar medical image showing the first planar contour of the structure; and

generate a finding record for the candidate structure including the boundary.

2. The system of claim 1 , wherein the instructions, when executed by the electronic processor, cause the system to:

generate the finding record responsive to receiving a user input accepting the first planar contour.

3. The system of claim 1 , wherein the user input comprises an axis drawn on the first planar medical image.

4. The system of claim 3 , wherein the instructions, when executed by the electronic processor, cause the system to:

compare the region to the boundary by determining if endpoints of the axis are within a predetermined distance from the boundary.

5. The system of claim 3 , wherein the instructions, when executed by the electronic processor, cause the system to:

compare the region to the boundary by determining if endpoints of the axis are within a predetermined distance from the first planar contour.

6. The system of claim 1 , wherein the instructions, when executed by the electronic processor, cause the system to:

generate a second planar contour of the candidate structure; and

display a second medical image of the plurality of medical images showing the second planar contour, wherein the first and second medical images are associated with different plane orientations.

7. The system of claim 6 , wherein the first planar medical image is associated with one of an axial plane, a coronal plane, or a sagittal plane, and the second medical image is associated with a different one of the axial plane, the coronal plane, or the sagittal plane.

8. The system of claim 6 , wherein the first planar medical image is associated with one of an axial plane, a coronal plane, or a sagittal plane, and the second medical image is associated with the same one of the axial plane, the coronal plane, or the sagittal plane.

9. The system of claim 1 , wherein the segmentation model comprises a deep learning segmentation model.

10. The system of claim 1 , wherein the segmentation model comprises an algorithmic segmentation model.

11. A method for volumetric segmentation in a plurality of planar medical images, comprising:

receiving, at an electronic processor, the plurality of planar medical images;

generating a boundary of a candidate structure in the plurality of medical images using a segmentation model;

displaying, on a display, a first planar medical image from the plurality of planar medical images;

receiving, by a user interface, a user input indicating a region in the first planar medical image;

generating a first planar contour of the candidate structure;

comparing the region to the boundary; and

responsive to the region being at least partially within the boundary:

re-displaying, on the display, the first planar medical image showing the first planar contour of the structure; and

generating a finding record for the candidate structure including the boundary.

12. The method of claim 11 , comprising:

generating the finding record responsive to receiving a user input accepting the first planar contour.

13. The method of claim 11 , wherein receiving, by the user interface, the user input comprises receiving an axis drawn on the first planar medical image.

14. The method of claim 13 , wherein comparing the region to the boundary comprises determining if endpoints of the axis are within a predetermined distance from the boundary.

15. The method of claim 13 , wherein comparing the region to the boundary comprises determining if endpoints of the axis are within a predetermined distance from the first planar contour.

16. The method of claim 11 , comprising:

generating a second planar contour of the candidate structure; and

displaying a second medical image of the plurality of medical images showing the second planar contour, wherein the first and second medical images are associated with different plane orientations.

17. The method of claim 16 , wherein the first planar medical image is associated with one of an axial plane, a coronal plane, or a sagittal plane, and the second medical image is associated with a different one of the axial plane, the coronal plane, or the sagittal plane.

18. The method of claim 16 , wherein the first planar medical image is associated with one of an axial plane, a coronal plane, or a sagittal plane, and the second medical image is associated with the same one of the axial plane, the coronal plane, or the sagittal plane.

19. The method of claim 11 , wherein the segmentation model comprises a deep learning segmentation model.

20. The method of claim 11 , wherein the segmentation model comprises an algorithmic segmentation model.