Method and system for defining a boundary of a region of interest by applying threshold values to outputs of a probabilistic automatic segmentation model based on user-selected segmentation sensitivity levels

1 . A method comprising:

receiving, by at least one processor, an ultrasound image having a plurality of pixels;

automatically processing, by the at least one processor executing a segmentation model, the ultrasound image to output a probability of each of the plurality of pixels being in a region of interest;

applying, by the at least one processor, a first threshold value to the probability of each of the plurality of pixels output based on the automatic processing of the ultrasound image, wherein the applying the first threshold value determines a boundary of the region of interest, wherein the first threshold value corresponds with a first segmentation sensitivity level of a plurality of segmentation sensitivity levels;

causing, by the at least one processor, a display system to present the ultrasound image with the boundary overlaid on the ultrasound image;

receiving, by the at least one processor, a user selection of a second segmentation sensitivity level from the plurality of segmentation sensitivity levels, wherein the second segmentation sensitivity level corresponds with a second threshold value different from the first threshold value; and

dynamically updating, by the at least one processor, the boundary overlaid on the ultrasound image at the display system by applying the second threshold value to the probability of each of the plurality of pixels output based on the automatic processing of the ultrasound image.

2 . The method of claim 1 , wherein each of the plurality of segmentation sensitivity levels is a preset having a different pre-defined threshold value.

3 . The method of claim 1 , comprising presenting a user interface tool having a list of the plurality of segmentation sensitivity levels at the display system with the ultrasound image, each of the plurality of segmentation sensitivity levels presented in the list being user- selectable.

4 . The method of claim 1 , comprising presenting a user interface tool having a plurality of selectable positions at the display system with the ultrasound image, each of the plurality of selectable positions corresponding with one of the plurality of segmentation sensitivity levels.

5 . The method of claim 1 , wherein the first threshold value corresponding with the first segmentation sensitivity level is a default, and further comprising updating the default to the second threshold value corresponding with the second segmentation sensitivity level based on the user selection.

6 . The method of claim 5 , comprising storing the default in association with one or both of a specific user profile or a specific ultrasound system.

7 . The method of claim 1 , wherein the region of interest is one of an anatomical structure, an artificial structure, or measurement endpoints.

8 . The method of claim 1 , wherein the segmentation model is an artificial intelligence segmentation model.

9 . A system comprising:

at least one processor configured to:

receive an ultrasound image having a plurality of pixels;

automatically process the ultrasound image by executing a segmentation model to output a probability of each of the plurality of pixels being in a region of interest;

apply a first threshold value to the probability of each of the plurality of pixels output based on the automatic processing of the ultrasound image, wherein application of the first threshold value determines a boundary of the region of interest, wherein the first threshold value corresponds with a first segmentation sensitivity level of a plurality of segmentation sensitivity levels;

cause a display system to present the ultrasound image with the boundary overlaid on the ultrasound image;

receive a user selection of a second segmentation sensitivity level from the plurality of segmentation sensitivity levels, wherein the second segmentation sensitivity level corresponds with a second threshold value different from the first threshold value; and

dynamically update the boundary overlaid on the ultrasound image at the display system by applying the second threshold value to the probability of each of the plurality of pixels output based on the automatic processing of the ultrasound image; and

the display system configured to present the ultrasound image with the boundary overlaid on the ultrasound image.

10 . The system of claim 9 , wherein each of the plurality of segmentation sensitivity levels is a preset having a different pre-defined threshold value.

11 . The system of claim 9 , wherein the at least one processor is configured to cause the display system to present a user interface tool having a list of the plurality of segmentation sensitivity levels with the ultrasound image, each of the plurality of segmentation sensitivity levels presented in the list being user-selectable.

12 . The system of claim 9 , wherein the at least one processor is configured to cause the display system to present a user interface tool having a plurality of selectable positions with the ultrasound image, each of the plurality of selectable positions corresponding with one of the plurality of segmentation sensitivity levels.

13 . The system of claim 9 , wherein:

the first threshold value corresponding with the first segmentation sensitivity level is a default; and

the at least one processor is configured to:

update the default to the second threshold value corresponding with the second segmentation sensitivity level based on the user selection, and

store the default in association with one or both of a specific user profile or a specific ultrasound system.

14 . The system of claim 9 , wherein the region of interest is one of an anatomical structure, an artificial structure, or measurement endpoints.

15 . The system of claim 9 , wherein the segmentation model is an artificial intelligence segmentation model.

16 . A non-transitory computer readable medium having stored thereon, a computer program having at least one code section, the at least one code section being executable by a machine for causing an ultrasound system to perform steps comprising:

receiving an ultrasound image having a plurality of pixels;

automatically processing the ultrasound image by executing a segmentation model to output a probability of each of the plurality of pixels being in a region of interest;

applying a first threshold value to the probability of each of the plurality of pixels output based on the automatic processing of the ultrasound image, wherein:

the applying the first threshold value determines a boundary of the region of interest,

the first threshold value corresponds with a first segmentation sensitivity level of a plurality of segmentation sensitivity levels, and

the first threshold value corresponding with the first segmentation sensitivity level is a default;

causing a display system to present the ultrasound image with the boundary overlaid on the ultrasound image;

receiving a user selection of a second segmentation sensitivity level from the plurality of segmentation sensitivity levels, wherein the second segmentation sensitivity level corresponds with a second threshold value different from the first threshold value;

dynamically updating the boundary overlaid on the ultrasound image at the display system by applying the second threshold value to the probability of each of the plurality of pixels output based on the automatic processing of the ultrasound image;

updating the default to the second threshold value corresponding with the second segmentation sensitivity level based on the user selection; and

storing the default in association with one or both of a specific user profile or a specific ultrasound system.

17 . The non-transitory computer readable medium of claim 16 , wherein each of the plurality of segmentation sensitivity levels is a preset having a different pre-defined threshold value.

18 . The non-transitory computer readable medium of claim 16 , comprising presenting a user interface tool having a list of the plurality of segmentation sensitivity levels at the display system with the ultrasound image, each of the plurality of segmentation sensitivity levels presented in the list being user-selectable.

19 . The non-transitory computer readable medium of claim 16 , comprising presenting a user interface tool having a plurality of selectable positions at the display system with the ultrasound image, each of the plurality of selectable positions corresponding with one of the plurality of segmentation sensitivity levels.