Apparatus and trained-model generation device

1. An apparatus comprising processing circuitry configured to:

acquire first image data that is image data obtained during ultrasound scan executed on an object and that is image data before coordinate conversion corresponding to a format of the ultrasound scan,

use a trained model generated through learning using the first image data obtained during the previously executed ultrasound scan and an area including the object in the first image data to estimate the area in the acquired first image data,

extract the estimated area from the acquired first image data to obtain second image data,

execute the coordinate conversion on the second image data to obtain third image data, and

cause a display to display the third image data.

2. The apparatus according to claim 1 , wherein

the trained model relates and outputs, with regard to each pixel of the acquired first image data, a degree of certainty of being the area, and

the processing circuitry is further configured to:

estimate a plurality of areas that are set by using different thresholds in accordance with the degree of certainty,

before the extracting is executed, receive an operator's operation to select one of the areas, and

extract the area selected in accordance with the operator's operation from the first image data to obtain the second image data.

3. The apparatus according to claim 1 , wherein

the trained model relates and outputs, with regard to each pixel of the acquired first image data, a degree of certainty of being the area, and

the processing circuitry is further configured to:

estimate a plurality of areas that are set by using different thresholds in accordance with the degree of certainty,

extract each of the areas from the first image data to acquire a plurality of sets of second image data, and

receive an operator's operation to select, as the second image data, one of the sets of second image data.

4. An apparatus comprising processing circuitry configured to:

acquire first image data that is image data obtained during ultrasound scan executed on an object and that is image data before coordinate conversion corresponding to a format of the ultrasound scan, and

use a trained model generated through learning using the first image data obtained during the previously executed ultrasound scan and an area including the object in the first image data to estimate the area in the acquired first image data,

execute the coordinate conversion on the acquired first image data to obtain second image data,

execute the coordinate conversion on the estimated area to obtain an area having undergone the coordinate conversion,

extract the area having undergone the coordinate conversion from the second image data to obtain third image data, and

causes a display to display the third image data.

5. The apparatus according to claim 4 , wherein the processing circuitry is further configured to:

before the extracting is executed, receive an operator's operation to change an outline of the area having undergone the coordinate conversion, and

extract the area changed in accordance with the operator's operation from the second image data to obtain the third image data.

6. The apparatus according to claim 1 , wherein the first image data is three-dimensional image data.

7. The apparatus according to claim 1 , wherein the processing circuitry is further configured to:

execute rendering on the third image data to generate a rendering image,

generate a cross-sectional image on a predetermined cross-section from the third image data, and

cause a display to display the rendering image and the cross-sectional image.

8. The apparatus according to claim 7 , wherein the processing circuitry is further configured to:

generate a curved line representing an outline of the estimated area on the predetermined cross-section, and

cause the display to display the curved line and the cross-sectional image in a superimposed manner.

9. The apparatus according to claim 1 , wherein the learning and the estimating are executed after the first image data is decimated in a depth direction.

10. An apparatus comprising a processing circuitry configured to:

acquire first image data that is image data obtained during ultrasound scan executed on an object and that is image data different from image data displayed on a display, and

use a trained model generated through learning using the first image data obtained during the previously executed ultrasound scan and an area including the object in the first image data to estimate the area in the acquired first image data,

extract the estimated area from the acquired first image data to obtain second image data,

execute coordinate conversion on the second image data to obtain third image data, and

cause a display to display the third image data.

11. A trained-model generation device comprising a processing circuitry configured to

acquire, as a data set, first image data that is image data obtained during previously executed ultrasound scan on an object and that is image data before coordinate conversion corresponding to a format of the ultrasound scan and an area including the object in the first image data, and

learn a pattern of the area from the acquired data set to generate a trained model that has a function to, in response to an input of the first image data obtained by executing the ultrasound scan, estimate the area in the first image data,

extract the estimated area from the acquired first image data to obtain second image data,

execute the coordinate conversion on the second image data to obtain third image data, and

cause a display to display the third image data.

12. The apparatus according to claim 4 , wherein the first image data is three-dimensional image data.

13. The apparatus according to claim 4 , wherein the learning and the estimating are executed after the first image data is decimated in a depth direction.