Medical image processing apparatus to estimate a shape of a region of interest in an image

1 . A medical image processing apparatus, comprising:

processing circuitry configured to

acquire an input image including a region of interest of a subject,

acquire a partial region including the region of interest from the input image,

acquire information on an existence probability of the region of interest based on the acquired partial region, and

calculate an estimated value of a shape of the region of interest based on the acquired partial region and the information on the existence probability by using a learned model that uses, as input, the acquired partial region and the information on the existence probability, and outputs the estimated value of the shape of the region of interest.

2 . The medical image processing apparatus according to claim 1 , wherein

the region of interest includes a plurality of sites, and

the processing circuitry is further configured to

acquire information on the existence probability related to each of the plurality of sites, and

calculate the estimated value of the shape of the region of interest based on the information on the existence probability related to each of the plurality of sites.

3 . The medical image processing apparatus according to claim 2 , wherein the region of interest is a mitral valve.

4 . The medical image processing apparatus according to claim 3 , wherein the plurality of sites include any of an anterior leaflet, a posterior leaflet, a valve orifice, a valve annulus, and a commissure.

5 . The medical image processing apparatus according to claim 1 , wherein the processing circuitry is further configured to acquire a three-dimensional CT image as the input image.

6 . The medical image processing apparatus according to claim 1 , wherein the processing circuitry is further configured to calculate, as the estimated value of the shape, coordinate values of a plurality of locations corresponding to the region of interest.

7 . The medical image processing apparatus according to claim 6 , wherein the processing circuitry is further configured to calculate mesh information of the region of interest.

8 . The medical image processing apparatus according to claim 1 , wherein the processing circuitry is further configured to calculate, as the estimated value of the shape, information indicating a normal direction of a curved surface in the region of interest.

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

acquire a first partial region and a second partial region different from the first partial region,

acquire the information on the existence probability of the region of interest based on the first partial region, and

calculate the estimated value of the shape of the region of interest based on the second partial region and the existence probability.

10 . The medical image processing apparatus according to claim 9 , wherein the processing circuitry is further configured to acquire the second partial region based on the acquired information on the existence probability of the region of interest.

11 . The medical image processing apparatus according to claim 1 , wherein the processing circuitry is further configured to acquire the information on the existence probability of the region of interest further based on structural information in a vicinity of the region of interest.

12 . The medical image processing apparatus according to claim 1 , wherein the processing circuitry is further configured to acquire the information on the existence probability of the region of interest by a process based on machine learning.

13 . The medical image processing apparatus according to claim 12 , wherein the machine learning is a convolutional neural network.

14 . A medical image processing method, comprising:

acquiring an input image including a region of interest of a subject;

acquiring a partial region including the region of interest from the input image;

acquiring information on an existence probability of the region of interest based on the acquired partial region; and

calculating an estimated value of a shape of the region of interest based on the acquired partial region and the information on the existence probability by using a learned model that uses, as input, the acquired partial region and the information on the existence probability, and outputs the estimated value of the shape of the region of interest.

15 . A non-transitory computer readable medium comprising instructions that cause a computer to execute:

acquiring an input image including a region of interest of a subject;

acquiring a partial region including the region of interest from the input image;

acquiring information on an existence probability of the region of interest based on the acquired partial region; and

calculating an estimated value of a shape of the region of interest based on the acquired partial region and the information on the existence probability by using a learned model that uses, as input, the acquired partial region and the information on the existence probability, and outputs the estimated value of the shape of the region of interest.

16 . A medical image processing apparatus, comprising:

processing circuitry configured to

acquire an input image including a region of interest of a subject,

acquire information on an existence probability of the region of interest based on the input image, and

calculate coordinate values of a plurality of locations of a mesh structure of the region of interest as an estimated value of a shape of the region of interest based on the input image and the information on the existence probability by using a learned model that uses, as input, the input image and the information on the existence probability, and outputs coordinate values at a plurality of locations of a mesh structure of the region of interest.

17 . A medical image processing method, comprising:

acquiring an input image including a region of interest of a subject;

acquiring information on an existence probability of the region of interest based on the input image; and

calculating coordinate values of a plurality of locations of a mesh structure of the region of interest as an estimated value of a shape of the region of interest based on the input image and the information on the existence probability by using a learned model that uses, as input, the input image and the information on the existence probability, and outputs coordinate values at a plurality of locations of a mesh structure of the region of interest.

18 . A non-transitory computer readable medium comprising instructions that cause a computer to execute:

acquiring an input image including a region of interest of a subject;

acquiring information on an existence probability of the region of interest based on the input image; and

calculating coordinate values of a plurality of locations of a mesh structure of the region of interest as an estimated value of a shape of the region of interest based on the input image and the information on the existence probability by using a learned model that uses, as input, the input image and the information on the existence probability, and outputs coordinate values at a plurality of locations of a mesh structure of the region of interest.