Medical image processing device, endoscope system, medical image processing method, and program

1. A medical image processing device comprising:

a processor, configured to:

acquire images at a plurality of in-vivo positions of a subject, from medical equipment that sequentially captures and displays in real time the images;

acquire positional information indicating the in-vivo position of the acquired image;

detect a region of interest from an input image and correspond to the plurality of in-vivo positions, respectively;

select a region-of-interest detection unit corresponding to the position indicated by the positional information from among a plurality of region-of-interest detection units; and

cause the selected region-of-interest detection unit to detect a region of interest from the acquired image,

wherein the processor sequentially acquires a plurality of in-vivo images obtained by chronological imaging,

wherein the processor sequentially acquires a plurality of images captured at a fixed frame rate,

wherein the plurality of region-of-interest detection units are a plurality of learned models.

2. The medical image processing device according to claim 1 ,

wherein the plurality of learned models are models learned using different data sets, respectively.

3. The medical image processing device according to claim 2 ,

wherein the plurality of learned models are models learned using data sets consisting of images captured at different in-vivo positions, respectively.

4. The medical image processing device according to claim 1 ,

wherein the processor comprises a position recognition unit that recognizes the in-vivo position from the acquired image.

5. The medical image processing device according to claim 1 ,

wherein the processor comprises an input unit that receives the positional information through a user's input.

6. The medical image processing device according to claim 1 ,

wherein the processor causes the medical equipment to display the position indicated by the acquired positional information.

7. The medical image processing device according to claim 1 ,

wherein the medical equipment comprises

an ultrasound probe that transmits ultrasonic waves toward an inside of the subject from an outside of the subject, and receives the ultrasonic waves reflected from the inside of the subject,

an ultrasound image generation unit that generates an ultrasound image using the ultrasonic waves received by the ultrasound probe, and

a display unit that displays the ultrasound image.

8. The medical image processing device according to claim 1 ,

wherein the medical equipment comprises

an endoscope to be inserted into a body cavity of the subject,

a camera that images the body cavity, and

a display unit that displays the captured image.

9. An endoscope system comprising:

medical equipment that has an endoscope to be inserted into a body cavity of a subject, a camera that images the body cavity, and a display unit that displays the captured image; and

the medical image processing device according to claim 8 .

10. A medical image processing method comprising:

an image acquisition step of acquiring images at a plurality of in-vivo positions of a subject, from medical equipment that sequentially captures and displays in real time the images;

a positional information acquisition step of acquiring positional information indicating the in-vivo position of the acquired image;

a plurality of region-of-interest detection steps of detecting a region of interest from an input image, the plurality of region-of-interest detection steps corresponding to the plurality of in-vivo positions, respectively;

a selection step of selecting a region-of-interest detection step corresponding to the position indicated by the positional information from among the plurality of region-of-interest detection steps; and

a control step of causing the selected region-of-interest detection step to detect a region of interest from the acquired image,

wherein the image acquisition step sequentially acquires a plurality of in-vivo images obtained by chronological imaging,

wherein the image acquisition step sequentially acquires a plurality of images captured at a fixed frame rate,

wherein the plurality of region-of-interest detection steps are a plurality of learned model.

11. A non-transitory computer readable recording medium causing a computer to realize:

in a case where a command stored in the recording medium is read out by the computer,

an image acquisition step of acquiring images at a plurality of in-vivo positions of a subject, from medical equipment that sequentially captures and displays in real time the images;

a positional information acquisition step of acquiring positional information indicating the in-vivo position of the acquired image;

a plurality of region-of-interest detection steps of detecting a region of interest from an input image, the plurality of region-of-interest detection steps corresponding to the plurality of in-vivo positions, respectively;

a selection step of selecting a region-of-interest detection step corresponding to the position indicated by the positional information from among the plurality of region-of-interest detection steps; and

a control step of causing the selected region-of-interest detection step to detect a region of interest from the acquired image,

wherein the image acquisition step sequentially acquires a plurality of in-vivo images obtained by chronological imaging,

wherein the image acquisition step sequentially acquires a plurality of images captured at a fixed frame rate,

wherein the plurality of region-of-interest detection steps are a plurality of learned model.